Business Systems Analysis with Ontologies

Business Systems Analysis with Ontologies Peter Green University of Queensland, Australia Michael Rosemann Queensland U...

Author: Peter Green Michael Rosemann

223 downloads 2615 Views 5MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Business Systems Analysis with Ontologies Peter Green University of Queensland, Australia Michael Rosemann Queensland University of Technology, Australia

IDEA GROUP PUBLISHING Hershey • London • Melbourne • Singapore

Acquisitions Editor: Development Editor: Senior Managing Editor: Managing Editor: Copy Editor: Typesetter: Cover Design: Printed at:

Renée Davies Kristin Roth Amanda Appicello Jennifer Neidig Dawne Brooks Kristin Roth Lisa Tosheff Yurchak Printing Inc.

Published in the United States of America by Idea Group Publishing (an imprint of Idea Group Inc.) 701 E. Chocolate Avenue Hershey PA 17033 Tel: 717-533-8845 Fax: 717-533-8661 E-mail: [email protected] Web site: http://www.idea-group.com and in the United Kingdom by Idea Group Publishing (an imprint of Idea Group Inc.) 3 Henrietta Street Covent Garden London WC2E 8LU Tel: 44 20 7240 0856 Fax: 44 20 7379 3313 Web site: http://www.eurospan.co.uk Copyright © 2005 by Idea Group Inc. All rights reserved. No part of this book may be reproduced, stored or distributed in any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher. Product or company names used in this book are for identification purposes only. Inclusion of the names of the products or companies does not indicate a claim of ownership by IGI of the trademark or registered trademark. Library of Congress Cataloging-in-Publication Data Business systems analysis with ontologies / Peter Green and Michael Rosemann, editors. p. cm. Summary: "This book shows systems analysts and business analysts how ontological thinking can help them clarify requirements analysis tasks in business systems"--Provided by publisher. Includes bibliographical references and index. ISBN 1-59140-339-1 (h/c) -- ISBN 1-59140-340-5 (s/c) -- ISBN 1-59140-341-3 (ebook) 1. Industrial management--Data processing. 2. Information resources management. 3. Ontology. I. Green, Peter, 1958- II. Rosemann, Michael, 1967HD30.2.B879 2005 658'.001--dc22 2004029772 British Cataloguing in Publication Data A Cataloguing in Publication record for this book is available from the British Library. All work contributed to this book is new, previously-unpublished material. The views expressed in this book are those of the authors, but not necessarily of the publisher.

Business Systems Analysis with Ontologies Table of Contents Preface ............................................................................................................. vi Peter Green, University of Queensland, Australia Michael Rosemann, Queensland University of Technology, Australia Introduction: Setting the Scene ................................................................. xii Yair Wand, The University of British Columbia, Canada Ron Weber, Monash University, Australia Chapter I Ontological Analysis of Business Systems Analysis Techniques: Experiences and Proposals for an Enhanced Methodology .................... 1 Peter Green, University of Queensland, Australia Michael Rosemann, Queensland University of Technology, Australia Chapter II Evaluating Conceptual Modelling Practices: Composites, Things, Properties ...................................................................................................... 28 Graeme Shanks, Monash University, Australia Jasmina Nuredini, Monash University, Australia Ron Weber, Monash University, Australia Chapter III Ontological Analysis of Reference Models .............................................. 56 Peter Fettke, Johannes Gutenberg University Mainz, Germany Peter Loos, Johannes Gutenberg University Mainz, Germany

Chapter IV Thinking Ontologically: Conceptual vs. Design Models in UML ........ 82 Jörg Evermann, Victoria University of Wellington, New Zealand Chapter V Template-Based Definition of Information Systems and Enterprise Modelling Constructs ............................................................................... 105 Andreas Opdahl, University of Bergen, Norway Brian Henderson-Sellers, University of Technology, Sydney, Australia Chapter VI A Reflective Meta-Model of Object-Process Methodology: The System Modeling Building Blocks ................................................. 130 Iris Reinhartz-Berger, University of Haifa, Israel Dov Dori, Technion, Israel Institute of Technology, Israel Chapter VII Ontology-Driven Method Engineering for Information Systems Development .............................................................................................. 174 Roland Holten, University of Frankfurt, Germany Alexander Dreiling, Queensland University of Technology, Australia Jörg Becker, European Research Center for Information Systems, Germany Chapter VIII Using a Common-Sense Realistic Ontology: Making Data Models Better Map the World .............................................................................. 218 Ed Kazmierczak, The University of Melbourne, Australia Simon Milton, The University of Melbourne, Australia Chapter IX Applying the ONTOMETRIC Method to Measure the Suitability of Ontologies .............................................................................................. 249 Asunción Gómez-Pérez, Politécnica University of Madrid, Spain Adolfo Lozano-Tello, Extremadura University, Spain

Chapter X A Twofold Approach for Evaluating Inter-Organizational Workflow Modeling Formalisms ............................................................................... 270 Benoit A. Aubert, HEC Montreal and CIRANO, Canada Aymeric Dussart, Robichaud Conseil and CIRANO, Canada Michel Patry, HEC Montreal and CIRANO, Canada Chapter XI Methodological Issues in the Evaluation of System Analysis and Design Techniques .................................................................................... 305 Andrew Gemino, Simon Fraser University, Canada Chapter XII Ontological Foundations of Information Systems Analysis and Design: Extending the Scope of the Discussion ................................... 322 Boris Wyssusek, Queensland University of Technology, Australia Helmut Klaus, Queensland University of Technology, Australia Chapter XIII Some Applications of a Unified Foundational Ontology in Business Modeling ..................................................................................................... 345 Giancarlo Guizzardi, University of Twente, The Netherlands Gerd Wagner, Brandenburg University of Technology, Cottbus, Germany About the Authors ..................................................................................... 368 Index ............................................................................................................ 377

vi

Preface

Ontologies are not a philosophical topic only anymore. For more than 10 years now, researchers in different streams related to information technology have been interested in applying sound ontological foundations to their work. An increasing number of special issues of journals, conference sessions and workshops have been dedicated to the application of ontologies in information systems (IS) and computer science. The best paper at the International Conference of Information Systems (ICIS) 2002 applied an ontology to UML and established academic events such as CAiSE and the ER-Conference include a significant number of papers related to ontologies now. This immense popularity of ontologies hopefully will further contribute to the theoretical foundations of the disciplines of information systems and computer science. However, the popularity also means that we have to be even more careful with our references to ontologies. Already, the type of research work that is conducted under the umbrella term “ontologies” varies significantly. Academics working on the semantic Web, knowledge management, E-business or natural language processing develop, compare, and apply ontologies. However, the understanding of the characteristics of ontology in terms of its scope, details or purpose varies significantly. In 2004, we guest-edited a special issue of the Journal of Database Management titled, Ontological Analysis, Evaluation and Engineering of Business Systems Analysis Methods. It covered the applications of ontologies in the context of methods, techniques and grammars for the purposes of business and systems engineering. Business systems analysis (BSA) grammars were deemed to include data modelling, process modelling and object-oriented modelling techniques. Ontologies are seen as a promising theoretical platform that might be able to provide a valuable reference for the evaluation of the tremendous number of grammars that have been already developed. In that special issue, we were very interested in new results of ontological analyses of different BSA

vii

grammars. Other areas of interest were further theoretical guidance for the process of ontological evaluations of BSA grammars, documentation of ontologies with relevance to the BSA community, or the selection of appropriate ontologies in the first place. Our call for chapters for that special issue received in excess of 15 full chapters from authors in nine different countries. Because of the obvious interest and sound, diverse work in the area, we decided to extend the concept of the special issue and approach Idea Group Publishing about producing an edited research book pulling together more fully the excellent work that is being done by colleagues worldwide in the areas of ontological comparison, evaluation and analysis. We have titled this book Business Systems Analysis With Ontologies. This title reflects the profound influence that the science of ontological analysis and evaluation is having on the development of the grammars, techniques and tools being used by academics and practitioners alike in business systems analysis. We are excited that two of the thought leaders in the development and application of an IS-related ontology provided insights into their current perspective on this topic in the introduction of the book. Yair Wand and Ron Weber outline in Setting the Scene how and why they see theories of ontology being important to the information systems field generally, and particularly, to the area of modelling. Moreover, Wand and Weber are enthused by the work in the area when they maintain, in the introduction of this book, “Conceptual modelling is not a defunct, arcane activity. Rather, in our view it remains a vibrant, central element of information systems development and implementation work.” In Chapter 1, Green and Rosemann reflect on their experiences with the application of the BWW models. Their chapter discusses typical problems in the use of any ontology in the context of business systems analysis. Furthermore, it expands particularly on the problems involved in the process of ontological analysis. The authors propose an enhanced procedural model for the ontological analysis based on the use of meta-models, the involvement of more than one coder and metrics. An overview about previous ontological evaluations of BSA grammars also demonstrates the scope of the related research. Chapter 2 by Shanks, Nuredini, and Weber provides an excellent summary of three years worth of experimental work into how alternative conceptual modelling representations affect end-user understanding of these representations. The researchers find evidence to support better end-user understanding when part-whole relations, things, and properties of things are represented in an ontologically-sound manner. Furthermore, they use a process-tracing technique to explain why the ontologically-sound representation of things and properties is more easily understood. Fettke and Loos, in Chapter 3, begin demonstrating how widely ontological analysis can be applied in the general area of business systems analysis. These authors turn their minds to the analysis of reference models. Reference models

viii

are commonly provided, for example, in enterprise resource package software to provide a starting point by which businesses can understand the business processes that are presumed in the software. Accordingly, their chapter focuses on evaluation of reference models based on a sound theory, namely the ontology proposed by the BWW model. They apply their approach to some parts of Scheer’s reference model for production planning and control. The results demonstrate that the modelling grammar used to represent the reference models has ontological deficiencies. These deficiencies lead to several problems in the reference model, for example the meaning of some modelling constructs is vague and some aspects of a reference model are redundant. In Chapter 4, Evermann explores the idea that languages such as UML currently used for conceptual modelling possess no real-world business or organizational meaning. His chapter discusses how such meaning can be assigned to languages like UML. It provides an example that demonstrates the differences between a software design model and a conceptual model in UML. He demonstrates how ontology can assist the modeller to not confuse software aspects with aspects of the real world being modelled. Opdahl and Henderson-Sellers have used an ontology, the BWW representational model, as a basis for developing a template for defining enterprise and IS modelling constructs in a way that facilitates language integration. In their Chapter 5, they have clarified the template further by formalising the meta-model through semi-formal constraints expressed in the object constraint language (OCL) and by populating the meta-model with definitions of example constructs from the UML version 1.4. The purpose was to make the template easier to understand, to validate the template, to pave the way for stronger tool support for the template, and to further our work on providing a complete, template-based definition of the UML. The authors of Chapter 6 focus on the ontologically complete object process model (OPM) for conceptual modelling. A comprehensive reflective meta-model of OPM is presented, using a bimodal representation of object-process diagrams and object-process language paragraphs. The meta-model of the UML industry standard depicts only the language part, leaving the (software or any other) system development processes informally defined as a “unified process”. In sharp contrast to this, OPM, being an object-process approach, enables reflective meta-modelling of the complete methodology, including its language (with both its conceptual-semantic and notational-syntactic aspects) and the OPM-based system development process. This ability to create a reflective meta-model of OPM is indicative of OPM’s expressive power, which goes hand in hand with OPM’s ontological completeness according to the Bunge-WandWeber (BWW) evaluation framework. Holton, Dreiling, and Becker, in Chapter 7, have used several philosophical and linguistic foundations, such as Kamlah and Lorenzen’s language critique approach, Morris’ findings on semiotics, de Saussure’s findings on signs, and

ix

Bunge’s research in ontology to produce an ontology-driven method for information system development. The authors show that ontologies are created and maintained by language communities using linguistic actions and how new concepts can be created to handle new situations. Furthermore, they demonstrate their ontology-driven method to information systems development by introducing an ontology for the domain of management information systems. Chapter 8 begins work on the vexed question of which ontology to use as a basis for the analytical and evaluative work on business systems analysis grammars. Only a few ontologies that tend to be more general in nature are popular in the analysis of business systems analysis grammars. One of these ontologies comes originally from the work by Chisholm and it forms the reference for the study by Kazmierczak and Milton. Their chapter and the work reported in it are driven by an interest in the fundamental nature of data modelling languages. In this research, the ontology helps us to understand, compare, evaluate, and strengthen data modelling languages. This work on which ontology to use is continued in Chapter 9 by Gómez-Pérez and Lozano-Tello. Many researchers tend to select a familiar ontology rather than carefully evaluating different ontologies. ONTOMETRIC is an adaptation of the AHP method to help knowledge engineers to choose the appropriate ontology for a new project; in order to do this, the engineer must compare the importance of the objectives, and study carefully the characteristics of ontologies. The framework provides a useful schema to carry out complex multicriteria decision-making. However, the evaluators need to specify in detail the aims of their analysis. Aubert, Dussart, and Paltry, in Chapter 10, demonstrate another area of application of ontology within business systems analysis: the semantic specification of inter-organizational workflow. Moreover, their chapter aims at determining if the ontological validity of available formalisms is sufficient to represent workflows crossing organizational boundaries. A review of several formalisms reveals that the UML fulfils essential representation criteria related to B2B workflows. Moreover, it possesses several extension possibilities that make it a powerful — and popular — language for business modelling. Andrew Gemino, in Chapter 11, provides a refreshing and contrasting point-ofview on the question of the effectiveness of the ontology selected and used as the analytical basis. He reverts to tried and tested economic theory espoused by Friedman to advocate that the first test of any ontology or meta-model is logical completeness and consistency. This should be a relatively objective exercise. Once an ontology or meta-model has passed this logical test, it can then be used to identify differences among modelling techniques. The impact that these differences have on participants can then be hypothesized using cognitive theory and eventually tested empirically. The ontology (meta-model) that is “better” is the ontology that provides us with differences that lead to “useful” empirical results.

x

The researchers in Chapter 12, Wyssusek and Klaus, take a very philosophical reflection on the process of using an ontology as a basis for analysis, evaluation and development of information systems. The authors try to establish that when dealing with fundamental issues of theory and practice it is advisable to create an awareness of the potential and limitations of our knowing and doing. This entails considering marginalised positions in a critical discussion of approaches toward information systems analysis and design. Finally, in Chapter 13, Guizzardi and Wagner attempt to draw on all the previous research in the area of ontological foundations to produce a unified foundational ontology UFO 0.2. They have stratified UFO into three ontological layers in order to distinguish its core, UFO-A, from the perdurant extension layer UFO-B and from the agent extension layer UFO-C. The researchers claim that, although there is not much consensus yet in the literature regarding the ontology of agents, such an ontology is needed for building the foundation of conceptual business process modelling. We hope that you will enjoy this research book as much as we have enjoyed the work involved in preparing it. May this book and the work reported in it be of guidance and stimulation for your own research.

Peter Green UQ Business School, University of Queensland, Australia Michael Rosemann Queensland University of Technology, Australia

xi

Acknowledgments

The fact that this book is able to provide a comprehensive, detailed, and current overview about the utilization of ontologies in the context of Business Systems Analysis is due to the excellent contributions we received by academics who are globally perceived as the thought-leaders in this area. We are very grateful to those authors who were willing to revise, update, and extend their papers as they were published originally in our Special Issue (Vol. 15, Nr. 2, 2004) for the Journal of Database Management. Moreover, we are thankful to those authors who followed our invitation and submitted a book chapter. Each chapter in this book has been evaluated by at least two experienced reviewers and carefully revised based on these comments. We are indebted to our international and national colleagues who selflessly provided comprehensive and insightful reviews through which the contributing authors could improve their chapters. We acknowledge the related workloads of all concerned and we believe that this rigorous process contributed significantly to the overall quality of this publication. A particular note of thanks must go to Ron Weber and Yair Wand. Without their original ideas, unflagging support, and exemplary academic professionalism, we would not have been inspired to start and complete this project. Furthermore, we like to express our appreciation for the excellent support we received from IDEA Publishers. It has been a well-managed process that kept the entire endeavor on track at all times. Finally, we like to take the opportunity to dedicate this book to our families, to Barbara, Brendan and Daniel, to Louise, Noah and Sophie, who carry too often the burden of two easily over-committed academics. Peter Green & Michael Rosemann

xii

Introduction:

Setting the Scene

Ontology is the branch of philosophy that deals with theories about the structure and behaviour of the worlds that humans perceive. Ontologists seek to articulate the fundamental types of phenomena that exist in the world and the relationships that can arise among these different types of phenomena. Ontologies can be proposed at various levels of abstraction. At the most-general level, an ontology articulates the fundamental constructs we need to be able to describe any phenomenon in the world. At any intermediate level, an ontology articulates the constructs needed to describe particular types of phenomena that occur in some domain — for example, architecture, law, nursing, and carpentry. At lower levels, ontologies articulate the constructs needed to describe specific worlds — for example, the world faced by a particular business as it attempts to survive in a particular context. Why are theories of ontology relevant to the information systems field? The answer is that the essence of an information system is that it is intended to be a faithful representation of a world that a human or group of humans perceives. Theories of ontology provide us with an artifice for describing a perceived world. Our descriptions will only be as good as our ontologies. Accordingly, our information systems will only be as good as our ontologies. In the mid-1980s, we happened on the field of ontology by chance. We were seeking to identify the core — the essence — of an information system and to determine whether we had any theories of this core (whatever the core might be). After substantial discernment, we had concluded the core pertained to “representation” of some world. Thus, we began to seek theories that would account for the nature of good and poor representations. In part, work that had been done by semantic modelling researchers seemed relevant. We found this work inherently unsatisfying, however, because it was not grounded in rigorous theory, nor did it seem complete. One of us (Weber) was visiting the University of British Columbia on sabbatical leave at the time. I (Weber) had been allocated an office next to professor Richard Mattesich, who is both an eminent

xiii

accounting researcher and philosopher of science. During a conversation with Mattesich where I explained the fundamental problem that Wand and I were addressing, he simply went to his bookshelf, selected the two volumes of Mario Bunge’s Treatise on Basic Philosophy that deal with ontology, handed them to me, and suggested I read them. A new world began to unfold for us. We first tried to apply Bunge’s ontological model to the formal analysis of control and audit procedures in information systems. As our work progressed, we realised ontological theories could be used in several ways. First, ontology provides a set of “benchmark” concepts to evaluate models used in systems development — notably, conceptual models of some application domain. Second, ontology provides a set of concepts to model systems and reason about their characteristics (this was our first use of Bunge’s ontology). Third, a specialised ontology can be used to define the meaning of information that will be available in an information system. In this latter role, ontologies often have been used in the artificial intelligence and (recently) semantic Web contexts. Since the early 1990s, we are delighted to see that a growing number of researchers have started to use ontological theories as a basis for their work on conceptual modelling. Much has been done. In our view, however, much still remains to be done. For instance, witness the problems currently being faced by researchers who are trying to find ways to model the world that will allow information systems interoperability to be achieved. Indeed, we are convinced that we have only commenced to scrape the surface of an immense, difficult research area. In terms of theory, for example, it is clear that even well-developed ontologies like Bunge’s need considerable extension and refinement to address the needs of information systems scholars and practitioners. For instance, Bunge’s ontology provides only a small number of constructs to describe processes — albeit fundamental constructs. In contrast, information systems scholars have devised much more-extensive ontologies to describe process phenomena. Unfortunately, these latter ontologies are not always rooted in a sound foundation of more fundamental constructs like things and properties. In short, we see substantial opportunities for philosophers interested in ontology and information systems scholars to work together to develop high-quality, comprehensive ontological theories. In this regard, information technology and its applications have taken the development of ontological theories from an abstruse, esoteric pursuit to an activity with important, high-value practical applications. Theories of ontology also have a curious status. Conventionally, theories provide a means to explain or predict some phenomena. For the most part, however, theories of ontology provide a means of describing rather than explaining or predicting some phenomena. In this light, they function more like a taxonomy than a conventional theory because they provide a set of constructs for classifying and relating phenomena in the world rather than predicting or explaining them. Nonetheless, they still have predictive and explanatory overtones. They

xiv

imply that describing phenomena in the world via the constructs they provide somehow has value. Presumably, if phenomena are classified correctly according to the theory, humans will be better able to understand and predict the phenomena and thus work more effectively and efficiently with the phenomena. This assumption underlies the work we have undertaken to map between ontological constructs and the constructs provided in different conceptual modelling grammars. Our motivation was the recognition that most modelling methods have emerged (and continue to emerge!) without much theoretical grounding. We believe this situation has been a major reason for the proliferation of modelling methods (a phenomenon given the pejorative nickname YAMA — yet another modelling method). To the extent a one/one mapping exists between ontological constructs and grammatical constructs, the implication is that conceptual models will somehow be better. For instance, users of conceptual models will be better able to understand them and work more effectively and efficiently with them. In the interests of parsimony, for the most part we have eschewed employing sophisticated psychological and social theories to provide an account of why a one/one mapping between ontological constructs and grammatical constructs is desirable. Like many economic theories, we simply employ broad assumptions about human behaviour in the hope that detailed, complex accounts of why ontological theories are useful can be avoided. Thus, it is an empirical question whether the explanations and predictions we make based on the (usually implicit) assumption that a given ontology reflects the way humans perceive reality are valid. In terms of practice, we have barely begun to explore the implications of ontological theories for how we undertake conceptual modelling work. In this regard, the chapters of this research book provide excellent examples of the sorts of work that might be done. Ultimately, our concern is to build better conceptual models and devise better tools to assist our conceptual modelling work. In our view, to date ontological theories have shown the most potential for informing practice and the design of conceptual modelling tools. For too long, we have proceeded without the benefit of theory. We have designed and built conceptual modelling tools and undertaken conceptual modelling work using too much of a “pure” engineering strategy — construct the artifact and, if we have time, test the artifact. In the absence of good theory, however, we have been unable to predict the likely strengths and weaknesses of our conceptual modelling tools and practices. As a result, we have a mishmash of views of what constitutes good conceptual modelling tools and practice. We also have a large number of different conceptual modelling grammars that have been devised, and the relationships among these grammars are unclear. For instance, if UML is a comprehensive modelling grammar, why is the W3C® developing the Web ontology language called OWL? Is UML deficient in some way? If so, how is it deficient? Long ago, many scholars in the conceptual modelling area deplored this

xv

state of affairs and underscored the need for good theory to inform our work. We believe that finally we are starting to see theory-driven conceptual modelling work, primarily via the articulation of ontological theories. Recently, we have encountered colleagues who argue that work on conceptual modelling (and thus ontologies) is no longer important. With the development of and widespread deployment of enterprise systems and their embedded “bestpractice” business models, why, they ask, would we bother to build conceptual models of some domain? These arguments are reminiscent of those made about the principles of good programming when fourth-generation languages first appeared. Structured programming precepts, for instance, allegedly were no longer important when fourth-generation languages were used to develop programs. Of course, the disasters that ensued with fourth-generation languages when good programming principles were ignored were an acid reminder that good theory transcends technologies. So it is, we believe, with good conceptual modelling principles, especially in the complex environments of enterprise systems. In such environments, conceptual models enable us to represent both the business and the software in a common way and to compare them. The extent to which misfits arise between the business models employed by an organization and the business models engaged within an enterprise system seems to be a good predictor of the likely success that an organization will enjoy when it implements an enterprise system. Conceptual modelling is not a defunct, arcane activity. Rather, in our view it remains a vibrant, central element of information systems development and implementation work.

Yair Wand The University of British Columbia, Canada Ron Weber Monash University, Australia

Ontological Analysis of Business Systems Analysis Techniques

1

Chapter I

Ontological Analysis of Business Systems Analysis Techniques: Experiences and Proposals for an Enhanced Methodology Peter Green, University of Queensland, Australia Michael Rosemann, Queensland University of Technology, Australia

Abstract For many years in the area of business systems analysis and design, practitioners and researchers alike have been searching for some comprehensive basis on which to evaluate, compare, and engineer techniques that are promoted for use in the modelling of systems’ requirements. To date, while many frameworks, factors, and facets have been forthcoming, most of them appear not to be based on a sound theory. In light of this dilemma, over the last 10 years, attention has been devoted by researchers to the use of ontology to provide some theoretical basis for the advancement of the business systems modelling discipline. While the selected ontologies Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

2 Green & Rosemann

are reasonably mature, it is the actual process of an ontological analysis that still lacks rigour. The current procedure leaves room for individual interpretations and is one reason for criticism of the entire ontological analysis. This chapter proposes an enhanced procedural model for the ontological analysis based on the use of meta-models, the involvement of more than one coder and metrics. This model is explained with examples from various ontological analyses.

Introduction As techniques for conceptual modelling, enterprise modelling, and business process modelling have proliferated over the years (e.g., Olle et al., 1991), researchers and practitioners alike have attempted to determine objective bases on which to compare, evaluate, and determine when to use these different techniques (e.g., Karam & Casselman, 1993; Gorla, Pu, & Rom, 1995). Throughout the 80s, 90s, and into the new millennium, however, it has become increasingly apparent to many researchers that without a theoretical foundation on which to base the specification for these various modelling techniques, incomplete evaluative frameworks of factors, features, and facets would continue to proliferate. Furthermore, without a theoretical foundation, one framework of factors, features, or facets is as justifiable as another for use (e.g., Bansler & Bodker, 1993). Ontologies and ontological engineering have received much attention in the business systems analysis and design literature over the last decade. Ontology is a well-established theoretical domain within philosophy dealing with identifying and understanding elements of the real world and their meaning. Given that IS professionals create computer systems that depict a portion of the real world, IS professionals might look to ontology to provide the conceptual underpinning that has been missing for so long from the IS modelling discipline. Wand and Weber (1989, 1990a, 1993, 1995) have adapted an ontology proposed by Bunge (1977) in order to provide a foundation for understanding the process in developing an information system. A popular application area of this ontology has been conceptual modelling. Today however, interest in, and the applicability of, ontologies extend to areas far beyond modelling. As Gruninger and Lee (2002) point out, “a Web search engine will return over 64,000 pages given ‘ontology’ as a keyword … the first few pages are phrases such as ‘enabling virtual business’, ‘gene ontology consortium’ and ‘enterprise ontology’” (p. 13). The usefulness of ontology as a theoretical foundation for knowledge representation and natural language processing is a fervently debated topic at the present time in the artificial intelligence research community (Guarino & Welty, 2002). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


3

Accordingly, this chapter has two main objectives (Rosemann, Green, & Indulska, 2004). First, we aim to identify comprehensively the shortcomings in the current practice of ontological analysis. The identification of such shortcomings will provide a basis upon which the practice of ontological analysis can be improved. Second, we want to develop several propositions and methodology extensions that enhance the ontological analysis process by making it more objective and structured. There are several contributions that this chapter aims to make. They are based on previous experiences with ontological analyses as well as observations derived from published analyses. First, the work presents a detailed analysis of the actual process of performing an ontological evaluation. The presented work identifies eight shortcomings of the current ontological analysis process — that is, lack of understandability, lack of comparability, lack of completeness, lack of guidance, lack of objectivity, lack of adequate result representation, lack of result classification and lack of relevance. Each of the identified shortcomings is classified then as belonging to one of three phases of analysis — that is, input, process and output. Second, the chapter presents recommendations on how each of the shortcomings in the three phases can be overcome. The recommendations, among other things, include an extended methodology for the improvement of the objectivity of the analysis, as well as a weighting model that aims to improve the classification of the results of any ontological analysis. This chapter unfolds in the following manner. The next section provides an overview about the basic concepts of applying ontologies for the purposes of evaluating modelling techniques and the related work. The third section identifies eight current shortcomings of ontological analyses of modelling techniques that are classified with respect to the three phases of analysis — that is, input, process and output. The fourth section provides recommendations concerning how to overcome the identified shortcomings in each of the three phases. The final section provides a brief summary of this work and outlines future research in this area.

Ontological Analysis of Modeling Techniques The ontological analysis of modelling techniques is a popular application of ontologies in information systems. The aim of these analyses is to evaluate the “goodness” of representations that can be produced by a particular modelling technique from the viewpoint of a selected ontology. The ontology forms in this process the “benchmark” against which the constructs of the modelling techniques are evaluated. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

4 Green & Rosemann

Weber (1997) distinguishes the following two major situations that may occur when a modelling technique is analysed in such a way. After a particular modelling technique has been analysed, predictions on the modelling strengths and weaknesses of the technique can be made according to whether some or any of these situations arise out of the analysis. 1.

Ontological completeness exists if there is at least one modelling grammatical construct for each ontological construct.

2.

Ontological clarity is determined by the extent to which the modelling technique does not exhibit one or more of the following deficiencies:

•

Construct overload exists in a modelling technique if one grammatical construct represents more than one ontological construct.

•

Construct redundancy exists if more than one grammatical construct represents the same ontological construct.

•

Construct excess exists in a modelling technique when a grammatical construct is present that does not map into any ontological construct.

The popularity of using ontologies as a basis for the analysis of Business Systems Analysis techniques has been growing steadily. The Bunge-Wand-Weber (BWW) ontological models (Weber, 1997), for example, have been applied extensively in the context of the analysis of various modelling techniques. Wand and Weber (1989, 1990b, 1993, 1995) and Weber (1997) have applied the BWW representation model to the “classical” descriptions of entity-relationship (ER) modelling and logical data flow diagramming (LDFD). Weber and Zhang (1996) also examined the Nijssen Information Analysis Method (NIAM) using the ontology. Green (1997) extended the work of Weber and Zhang (1996) and Wand and Weber (1993, 1995) by analysing various modelling techniques as they have been extended and implemented in upper CASE tools. Furthermore, Parsons, and Wand (1997) proposed a formal model of objects and they use the ontological models to identify representation-oriented characteristics of objects. Along similar lines, Opdahl and Henderson-Sellers (2001) have used the BWW representation model to examine the individual modelling constructs within the OPEN Modeling Language (OML) version 1.1 based on “conventional” objectoriented constructs. Green and Rosemann (2000) have extended the analytical work into the area of integrated process modelling based on the techniques presented in Scheer (2000a). The BWW models also have been applied in the context of Enterprise Resource Planning (ERP) Systems. Sia and Soh (2002) utilise the BWW models to propose

Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


5

a theoretically grounded framework for assessing the severity of ERP misalignment in organisations. The authors demonstrate the application of the proposed framework by applying it to a hospital case study, in which significant ERP misalignment is identified as a result. Shanks, Tansley, and Weber (2003) utilise the application of the BWW model in order to investigate the representation of part-whole relationships in conceptual modelling grammars. The authors use the BWW model to support their argument for representation of part-whole relationships as entities as opposed to relationships or associations. Their argument is further supported by an empirical study that concludes that using entities to represent part-whole relationships leads to an improvement in the level of the user’s understanding of the domain. Davies, Green, and Rosemann (2002) demonstrate the potential usefulness of meta-models for comparing and evaluating ontologies.1 The authors focus on the analysis of the meta-models of the BWW representation model and Chisolm’s Ontology, concentrating on ontological equivalence, depth of structure, and comprehensiveness of scope of the models. The findings of the work revealed that the two models were not completely ontologically equivalent, with the BWW model being more comprehensive in scope and Chisolm’s Ontology having a deeper structure than that of the BWW model. Davies, Green, Milton, and Rosemann (2004) extend the work to include a detailed discussion of the benefits of the use of meta-models for evaluating ontologies. Fettke and Loos (2003) discuss the process of BWW ontological evaluation of reference models and identify a number of possible application areas. The authors suggest that the proposed method may be used for evaluation of reference models, comparison of two or more reference models, representation of reference models in model repositories, and describing the key characteristics of reference models in order to facilitate selection of appropriate models in specific situations Most recently, Green, Rosemann, Indulska, and Manning (2004) have extended the use of this evaluative base into the area of enterprise systems interoperability using business process modelling languages like ebXML, BPML, BPEL4WS and WSCI. Table 1 provides an overview of the related work performed to date involving the Bunge-Wand-Weber models. Indeed, much of this work has involved evaluations based on Weber’s (1997) two situations. A mismatch between ontological and modelling constructs however does not necessarily indicate weaknesses of the target modelling technique. Rather, as Rosemann and Green (2002) point out, it could indicate misspecification in the ontology used for the evaluation.


6 Green & Rosemann

Table 1. Ontological analysis of modelling techniques using the BWW models Ontological Ontological Empirical

Business Systems Analysis Grammar Study

Traditional Structured

Data-

O-O Process Completeness Clarity

Tests

Other Purpose

Centred Wand & Weber (1989)

• (LDFD)

Wand & Weber (1993, 1995)

• (ER)

•

• (ER)

•

Weber (1997)

•

Sinha & Vessey (1995)

• (Relational)

Weber & Zhang (1996)

•

Green (1997)

•

•

•

•

•

Parsons & Wand (1997)

•

Opdahl & Henderson-Sellers (1999)

• (OML)

Wand, Storey & Weber (1999)

• •

•

•

•

•

•

•

•

•

• •

•

•

Rosemann & Green (2000)

•

•

•

•

Green & Rosemann (2002)

•

•

Sia & Soh (2002)

•

Green & Rosemann (2000) Bodart et al (2001)

• (ARIS) •

•

Opdahl & Henderson-Sellers (2002)

• (UML)

Rosemann & Green (2002)

•

Davies, Green & Rosemann (2002) Shanks et al (2003)

•

• (ARIS & UML Class)

• • (Enterprise Interoperability) • (ERP Systems)

•

•

•

•

•

•

• (UML Class)

• (ActivityBased Costing)

•

• (Other Ontology) •

Davies et al. (2004)

•

•

• (Other Ontology)

Fettke & Loos (2003)

•

•

• (Reference Process Models)

Green, Rosemann, Indulska & Manning (2004)

•

•

• (Interoperability Standards)



7

•

It could be that the ontology is over-engineered. The ontology may include constructs that are not relevant. The ontological analyses of various modelling techniques to date have consistently identified certain ontological constructs that do not have representations in the grammars examined, for example, conceivable state space, conceivable event space and lawful event space. The ontological analyses to date in themselves form an empirical study around this possibility of over-engineering. One conclusion then could be the identification of the need for a reduction in the number of constructs thought to be sufficient and necessary in the ontology.

•

Even if the ontology is not over-engineered, most modelling techniques usually focus on modelling particular aspects of the real-world, for example, statics, dynamics, processes, data, actors, actions, goals and the like. Apparently, the objectives of the modelling grammar need to be taken into account during the ontological analysis. Such work suggests a need for individualization of the ontology by means of not only designing subsets but also specializations of the ontology — a focused ontology.

•

Finally, there may be a need for extending the ontology. Weber (1997), for example, has already extended the understanding of the ontological construct, property, by explaining the various types of property, for example, property in general, property in particular. The growing importance of strategic enterprise modelling might lead to the explication of the BWW model to incorporate for example business objectives, strategies, goals or knowledge.

While there may be misspecification in ontologies, such a problem cannot be verified without substantial empirical research based on the theory being performed. In any case, ontology is seen as a potential fruitful theoretical basis on which to perform analyses of modelling techniques. However, while ontological analyses are frequently utilised, particularly in the area of analysing conceptual modelling techniques, the actual process of performing the analysis remains problematic. The current process of ontological analysis is open to the individual interpretations of the researchers who undertake the analysis. Consequently, such analyses are criticised as being subjective, ad hoc, and lacking in relevance. There is a need, therefore, for the systematic identification of shortcomings of the current ontological analysis process. The identification of such weaknesses, and their subsequent mitigation, will lead to a more rigorous, objective and replicable analytical process.


8 Green & Rosemann

Shortcomings of Current Ontological Analyses An ontological analysis is in principle the evaluation of a selected modelling grammar from the viewpoint of a defined and well-established ontology. The current focus of ontological analyses is on the bi-directional comparison of ontological constructs with the elements of the modelling technique that is under analysis. Weber (1997) defines ontological clarity and completeness as the two main perspectives of an ontological analysis. Though this type of ontological analysis is widely established, it still has a range of issues. These issues can be categorised into the three main phases of an ontological analysis — that is, preparation of the input data, the process of conducting the analysis and the evaluation and interpretation of the results. The first two identified shortcomings refer to the quality of the input data.

Lack of Understandability Several ontologies that are currently used for analysis of modelling grammars have been specified in formal languages. While such a formalisation is beneficial for a complete and precise specification of the ontology, it is not a very intuitive specification. An ontology that is not clear and intuitive can lead to misinterpretations as the involved stakeholders might have problems with the specifications. Furthermore, it forms a hurdle for the application of the ontology as it requires a deep understanding of the formal language in which it is specified. Moreover, it is not only the meta-model and the notation that is used for the specification of the ontology, but also the selected terminology. In our own applications, for example, we realised that elements of the BWW model such as “conceivable state space” are not self-explanatory to members of the modelling community.

Lack of Comparability The specification of an ontology requires typically a formal syntax that allows the precise specification of the elements and their relationships of the ontology. Consequently, textual descriptions of the ontology in “plain English” often extend the formal specification. However, even if an ontology is specified in an intuitive and understandable language, the actual comparison with the selected modelling grammar remains a problem. Unless the ontology and the grammar are specified in the same language or a precise mapping of the two languages exists, it will be up to the Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


9

coder to “mentally convert” the two specifications into each other, which adds a subjective element to the analysis. Different languages can also lead easily to different levels of detail and further complicate the analysis. In any case, they make a more automated comparison practically impossible. This situation is typical in many previous analyses. The further three shortcomings identified below are related to the process of the ontological analysis and refer to what should be analysed, how it should be analysed, and who should conduct the analysis.

Lack of Completeness The first decision that has to be made in the process of an ontological analysis is the scope and depth of the analysis. Even if most ontologies have been discussed for many decades, they still undergo modifications and extensions. It is up to the researcher to clearly specify the selected version of the ontology and the scope and level of detail of the analysis. In our work in the area of Web services standards, for example, it was often not clear what constructs form the core of the selected Web services standard. Two researchers, who conducted independent analyses of the same Web services standard, selected consequently a different number of constructs. Moreover, many ontological analyses solely focus on the constructs of the ontology and the constructs of the grammar, but do not sufficiently consider the relationships between these constructs. The difficulty in clearly specifying the boundaries of the analysis, as well as the limited consideration of relationships between the ontological constructs, lead to a potential lack of completeness.

Lack of Guidance After the scope and the level of detail of the analysis have been specified, it is typically up to the coder to decide on the procedure of the analysis — that is, in what sequence the ontological constructs and relationships will be analysed? Currently, there are hardly any recommendations on where to start the analysis. This lack of procedural clarity underlies most analyses and it has two consequences. First, a novice analyst lacks guidance in the process of conducting the ontological evaluation. Thus, the application of ontological analyses is potentially limited to experts in both the selected ontology and the modelling technique. Second, the procedure of the analysis can potentially have an impact on the results of the analysis. Consequently, it is possible that two analyses that follow a different process may lead to different outcomes.


10 Green & Rosemann

Lack of Objectivity An ontological analysis of a modelling technique requires not only detailed knowledge of the selected ontology and technique, but also a good understanding of the languages in which the ontology and the grammar are specified. This requirement explains why most analyses are carried out by single researchers as opposed to research teams. Consequently, these analyses are based on the individual interpretations of the involved researcher, which adds significant subjectivity to the results. This problem is further compounded by the fact that, unlike other qualitative research projects, ontological analyses typically do not include attempts to further increase the validity of the results. The five shortcomings identified above have a common flavour in that they heavily depend on the researcher conducting the ontological evaluation. Three further shortcomings have been identified — that is, lack of result representation, lack of result classification and lack of relevance. These shortcomings are detailed below and refer to the outcomes of the analysis.

Lack of Adequate Result Representation The results of a complete ontological analysis — that is, representation mapping and interpretation mapping, are typically summarised in two tables. These tables list all ontological constructs (first table) and all grammatical constructs (second table) and the corresponding constructs. Such tables can become quite lengthy and are typically not sorted in any particular order. They do not provide any insights into the relative importance of identified deficiencies. Furthermore, the findings are not clustered typically allowing related deficiencies to appear more apparent. In doing such clustering, the relative importance of the related deficiencies is made clearer as well.

Lack of Result Classification It is common practice to derive ontological deficiencies based on a comparison of the constructs in the ontology and the modelling technique. Ontological weaknesses are identified when corresponding constructs are missing in the obtained mapping between the ontology and the technique, or 1-many (or many1 or even many-many) relationships exist. Such identified deficiencies are the typical starting point for the derivation of propositions and then hypotheses. In general, the ontological analysis does not make any statements regarding the relative importance of these findings in comparison with each other. Though this seems to be the established practice, it lacks more detailed insights into the Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


11

significance of the results. It is expected, however, that the missing support for a core construct of an ontology can be rated higher than a missing corresponding construct for a minor ontological construct or a relationship. This lack of a more detailed statement regarding the significance of a potential shortcoming makes it difficult to judge quickly the outcomes of the results of two different sets of analyses, for example, an ontological analysis of ARIS in comparison with an ontological analysis of UML.

Lack of Relevance Finally, the results of an ontological analysis should be perceived as relevant by the related stakeholders. However, if an ontological analysis leads, for example, to the outcome that entity relationship models do not support the description of behaviour, then such an outcome needs a clarification. It seems that an ontological analysis has to consider the purpose of the grammar as well as the background of the modeller who is applying this grammar. The application of a high-level and generic ontology does not consider this individual context and there is a danger that the outcomes can be perceived as trivial or non-relevant.

A Reference Methodology for Conducting Ontological Analyses The shortcomings identified above motivated the development of an enhanced methodology for ontological analyses. The main purpose of this methodology is to increase the rigour, the overall objectivity and the level of detail of the analysis. The proposed methodology for ontological analyses is structured in three phases — that is, input, process and output.

Input The formal specification of ontologies, together with the differences in the languages used to specify the ontologies and the grammars under analysis, have been classified as issues pertaining to the lack of understandability and comparability. In order to overcome this shortcoming, we have worked on converting existing specifications for our selected ontology to a more commonly used language — that is, to a more intuitively understandable meta-model. There are several Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

12 Green & Rosemann

motivations for converting current specifications of ontologies into meta-modelbased specifications. First, the development of a meta-model that describes and clarifies the current understanding of the ontological constructs facilitates the use of ontologies in other related areas such as information systems education. Second, a formal meta-model that clearly describes the elements and relationships within an ontology can help to identify inconsistencies and anomalies in an ontology itself. Third, it can be used for the ontological analysis of modelling techniques (grammars) that are specified in the same metalanguage. In this case, the analysis turns into a pattern matching exercise. Fourth, a meta-model can be used to improve existing techniques and derive new modelling techniques (i.e., ontology-based method engineering). Fifth, it can also be applied for the comparison of different ontologies, if they are specified in the same metalanguage (Davies et al., 2002). Finally, based on the outcomes of the evaluation and comparison of ontologies, a meta-model can be used to develop and specify a new ontology. Figure 1 outlines these application areas for a meta-model of ontological constructs. In order to overcome the lack of understandability and comparability, the first step is to convert the ontology, as well as the selected modelling grammar, to meta-models using the same language (e.g., ER models or UML class diagram). This conversion facilitates a pattern-matching approach towards the ontological analyses of completeness and clarity of a grammar. We converted the BungeWand-Weber ontology into an ER-based meta-model. This meta-model includes 50 entity types and 92 relationship types. It has clusters such as system, property or class/kind. Such a meta-model explains, in a language familiar to the information systems community, the core constructs of the ontology. It also highlights the underlying focus of the ontology. In the case of the BWW model, for example, the visual inspection of the meta-model indicates that the ontology is centred around the existence of a thing, which is the central entity type in the meta-model. Figure 2 provides, as an example, an impression of the size and complexity of the meta-model for the BWW ontology. We used a modern version of the entity relationship (ER) language as the metamodelling language. The version of the ER approach used in our work is based on the original ER specification from Chen (1976) with extensions made by Scheer (2000a). This version is called the extended ER model. This selection was made for the following reasons: 1.

Since Chen (1976) introduced the original ER approach, it has undergone intensive discussions and further developments. It is realistic therefore to expect that solutions for special methodological problems that could occur during the process of designing the meta-model are already available in most cases.



13

Figure 1. Application areas of a meta-model for ontological constructs

1a) Facilitates communication about the ontology 2) Clarifies inconsistencies and anomalies

1b) Simplifies teaching the ontology 5) Streamlines the comparison of ontologies

Q

Ontology B

Meta Model for ontological constructs

3) Streamlines the ontological analysis of grammars

Ontology C Grammar A

Grammar B

4) Enables ontology-based method engineering

6) Enables ontology

engineering

New Grammar

New Ontology

Figure 2. The BWW meta-model R eal W o r ld

B W W m eta m o de l v e r4 - 2 3 /9 /20 0 2

1,n

A ut ho r : Is la y D a v ie s m ade up of

P R O P ER TY

T H I N G / C L A S S / K IN D

SYS TE M

S y s te m S t r u c tu r e

0,n

2,n

as s o c i a te d i n to

0,n

has

C o m p o s i te Th i n g

0,n

d,t

1,n has

1,n d,t

S im p l e Th i n g

1,n

P r op e r ty i n general

0,n

P r op e r ty i n p a r t ic u la r

0,1

0,n obs erv ed as is subset of

ca u s e s

c on t ain s

in h e re n t ly p o s se s s e s

1,n

i nit ia t e d by

1,n

is n o t i n c lu d e d in 0,n

has

S y s te m E n vi ro n m e n t

1,n

has

1,n 1,n

I n tr in si c P r op e r ty

n,p 1,n a ffe c t e d by

2,n 2,n

0,n

s hares

1,n

0,n

d,t

1,n

B in d in g M u tu a l P r op e r ty

2,n

H e r e d it a r y P r op e r ty

1,n

1,n

N o n - b in d in g M u tu a l P r op e r ty

1,n

U n s ta b l e S ta t e

1,n

0,n

1,n

E x te r n a l E ve n t

M u tu a l P r op e r ty

not a ffe c t e d by

i n te r a c ts w it h

is subset of

oc c urs on

1,n

E mer gent P r op e r ty

bel ongs to

i n h e ri te d by

d,t 1,n 0,n is subset of

d oe s n o t o c cu r o n

bel ongs to

x2,n

remai ns

S ta b l e S ta t e

d,t x1 + x 2 = 1 ; x1 , 2 i s e le m e n t o f { 0 , 1 }

1,n

x1,n

produc es

2,n 0,n x6,n

1,n

I n te r n a l E ve n t

0,n generat es

S y s te m C o m p o s i ti o n

x5,n

0,n

0,n

oc c urs on

has

2,n

m akes up

assum es

C la s s

1,n

1,n 0,n

2,n

x5 + x 6 = 1 ; x5 , 6 i s e le m e n t o f { 0 , 1 }

is i n c lu d e d in

1,n

1,n

K in d

1,n

0,1 0,n 0,n has

2,n

m akes up

i nit ia t e d by

1,n

2,n c h ar a c te r iz e d by

c on t ain s

di s p la y s

assum es

is subset of c h ar a c te r iz e d by

x3 + x 4 = 1 ; x3 , 4 i s e le m e n t o f { 0 , 1 }

1,1 S y s te m

0,n 0,n 2,n

1,1

0,1 1,1 1,n

1,1

x4,n x3,n

0,n

i s d e fin e d as

0,n

1,1

p a r t ia l o r d e r d e f in e d in

1,n

C o u p li n g

I n te r n a l C o u p li n g

L e ve l S t r u c tu r e

d,t

1,n

di s p la y s

0,n

0,n

0,n Th i n g

0,n

0,n 0,n

0,n

0,n

0,n

1,n

1,n

1,1

0,n

0,n

0,n

0,n

f o rm e d by

1,n 1,n

P r op e r ty 0,n

1,n

1,n

1,1 0,1

0,n

0,n

2,n 1,n 0,1

0,n

0,n

0,n

decom posed i n to

E x te r n a l C o u p li n g

1,n S ub s y s t e m

1,1

m o d e ll e d as

E V E N T / T R A N S F O R M A T IO N

1,n

E ve n t

oc c urs on

1,1

1,n

ST A TE

p o s se s s e s

1,n 1,n

1,1

0,n

0,n

r e p r e s e n te d by

A tt ri b u te

0,n

2,2

can occur on 1,1

C ro s s P ro d u c t

1,1

c o n t a in e d in

f orms c o n t a in e d in

'h a s ' r e la t io n s h i p = 'se t o f in d i vid u a l s t a te f u n c ti o n s '

0,n

0,n 1,n L a w fu l E ve n t S pace

K EY :

is subset of

1,1

C o n c e iva b le E ve n t S pace

0,1

1,n

com posed of

0,n

0,n

has

V a lu e

f orms

M o d el o b je c ts :

c o n s is ts of

c o n s is ts of

'a s s u m e s ' r e la t io n s h ip = 't o t al s t a t e f un c t io n ' 0,n

1,n

1,n

0,n

1,n

0,n 0,n

assum es

1,n

oc c urs at

1,1

Ti me I n s ta n t

0,1 1,1

0,1

T r a n s fo r m a t io n 1,1

c o n t a in e d in

can assum e 0,n

rec orded in

is

= B W W co re co n s tr uct s

0,n

0,n 1,n

1,n 0,n

prec edes

0,n

a f fe c t s

K now n S ta t e

S ta t e

= o th e r B W W c o ns tru c t s

has

1,n n,p H is t o ry

P re d i c ta b l e S ta t e

0,n

o rd e r e d by

s uc c e e d s 0,n 0,n

= no n -B W W - co ns t ru c t e nt ity ty p es W e l l- d e fi n e d E ve n t

1,n

0,n

0,n

is c o n t a in e d in

d,t

= n o n - BW W -co n st r uc t r e la t io ns h ip t y p es P o o rl y -d e fin e d E ve n t

1,n

is not 1,n

0,n

c o n t a in e d in

is a

= int r o du c e d c o n ce p ts o r ter m s n o t e x plic it ly s t a te d in W e be r ( 1 9 9 7) o r G r e en & R o se m a n n ( 2 0 0 0 ) , w h ic h h a v e be e n a d d ed in o r d er to fu lly de sc ri b e th e B W W c o n s tr u c ts w ith in th e c o n s tr a in ts o f eE R m o de lling

0,n

L a w fu l S ta t e S pace

1,1 0,n

is subset of

0,n

C o n c e iva b le S ta t e S pace

1,n

1,n

r e q u ir e s

C ro s s P ro d u c t

G e ne r a liza tio n /Sp e c ia liza t io n: d = d is jo in t (X O R ) n = n o n -d is jo in t (in clus iv e O R ) t = to t a l (a ll s ub ty p es w h ic h e x is t a r e dip ic t ed ) p = p a rtia l (fu rt he r su b t y p es n o t de p ict e d in th e m o d el e x is t)

0,n T r a n s fo r m a t io n L aw

0,n

0,n

1,n 1,n a l lo w e d by

1,n

0,n enabl es

0,n

1,1

V a lu e C hange

0,n

1,1 0,n enabl es

1,n

0,n

a l lo w e d by

1,1

0,n

S ta t e L aw

d,t

N a tu r a l L aw

0,n

s p e c i if e d by H uman L aw

1,n

C o rr e c t iv e A ct io n 1,n

in s ti g a te d by 1,n 1,1

L a w fu l T r a n s fo r m a t io n

0,n s p e c i if e d in

s p e c i if e s

0,n

0,n

0,n S t ab i lit y C o n d it io n

s p e c i fi e s

s p e c i fi e d in

1,n

1,1

is subset of


14 Green & Rosemann

2.

Even though many potential meta-languages are available, the ER approach is widely accepted as a de facto standard for (data) modelling.

3.

Several meta-models based on the ER approach are already available (e.g., ARIS [Scheer, 2000b]) and object-oriented schemas).

The obtained meta-model can be used now for a variety of ontological analyses. Moreover, it allows a critical review of the BWW model by a wider community. The approach, however, is not without its limitations. Commonly used modelling techniques, such as ER or UML, are often widely accepted, but they have not been designed for the purposes of meta-modelling. Thus, they occasionally lack the required expressiveness. While an ER-based meta-model helps to overcome issues related to the understandability of an ontology, a corresponding meta-model of the analysed grammar is required to deal with the lack of comparability issue. Many popular modelling techniques (e.g., ARIS or UML, and also interoperability standards such as ebXML) are already specified in meta-models using ER-notations or UML class diagrams. If the meta-models for the ontology and the modelling technique are specified in the same language, the ontological analyses turns into a comparison of two conceptual models. As part of the analyses, corresponding entity types and relationship types in both models need to be identified. It also becomes immediately obvious whether the focus of the analysed grammar differs from the ontology. In the case of ARIS or many Web services standards, for example, the meta-models are centred around functions or activities instead of being centred around things. As an example of constructs from a particular ontology, Table 2 provides some core ontological constructs defined in plain English and adapted to the IS discipline by Wand and Weber (1995). An extract of the meta-model for a set of selected BWW constructs is described in Figure 3. All object types in this model described as nouns correspond with constructs in the BWW representation model. The basic elements in the BWW representation model are things and their properties. Every thing possesses at least one property and every property belongs to at least one thing. Consequently, a mutual existential dependency exists. Things often consist of other things or they are part of other things. These composite things can be depicted by a recursive relationship type. While thing, composite thing, and property exist in the real world, for modelling purposes, it is necessary to define ways of concentrating the focus in order to reduce complexity. Things together with their properties can be classified in classes by identifying a characteristic property that all the involved things have in common. Each class has at least one relationship to a thing-property couple. Classes (e.g., human beings) may possess subtypes (e.g., man and woman) Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


15

Table 2. Core ontological constructs in the BWW representation model Ontological Construct

Explanation

THING*

A thing is the elementary unit in the BWW ontological model. The real world is made up of things. Two or more things (composite or simple) can be associated into a composite thing.

PROPERTY*: IN GENERAL IN PARTICULAR HEREDITARY EMERGENT INTRINSIC NON-BINDING MUTUAL BINDING MUTUAL ATTRIBUTES

Things possess properties. A property is modelled via a function that maps the thing into some value. For example, the attribute “weight” represents a property that all humans possess. In this regard, weight is an attribute standing for a property in general. If we focus on the weight of a specific individual, however, we would be concerned with a property in particular. A property of a composite thing that belongs to a component thing is called an hereditary property. Otherwise it is called an emergent property. Some properties are inherent properties of individual things. Such properties are called intrinsic. Other properties are properties of pairs or many things. Such properties are called mutual. Non-binding mutual properties are those properties shared by two or more things that do not “make a difference” to the things involved; for example, order relations or equivalence relations. By contrast, binding mutual properties are those properties shared by two or more things that do “make a difference” to the things involved. Attributes are the names that we use to represent properties of things.

STATE*

The vector of values for all property functions of a thing is the state of the thing.

TRANSFORMATION*

A transformation is a mapping from one state to another state.

STABLE STATE*

A stable state is a state in which a thing, subsystem, or system will remain unless forced to change by virtue of the action of a thing in the environment (an external event).

Figure 3. Thing, property, class, kind, attribute 1 ,n

T h in g

C la ss p osse sse s 0 ,n

0 ,n

C h ara cte ristic P rop e rty

1 ,n 1 ,1

P rop e rty 0 ,n

0 ,n C o m p osite T h in g

K in d

1 ,n

is a

0 ,1

C la ss

is m o d elle d as

1 ,n A ttrib u te

called kinds. Through attributes the context-relevant properties can be modelled and they become more easily understood. In contrast, an attribute requires the existence of at least one property, as it cannot exist on its own. The development and applicability of the full meta-model is reported in Rosemann and Green (2002). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

16 Green & Rosemann

Figure 4. Comparison of the BWW meta-model and the ARIS meta-model BWW Model 0,n

precedes

Transformation 0,n

0,n 0,n

State

succeeds

ARIS 0,1

Function

precedes

0,1

0,1 0,1

Event

succeeds

Figure 4 depicts an example that shows how meta-models can facilitate the ontological analysis of a modelling grammar. The excerpt of the BWW metamodel depicts the dynamic part that constitutes a process in which states and transformations are strictly alternate. Both constructs together form, in the terminology of the BWW models, an event. The bottom portion of Figure 4 includes the corresponding part of the meta-model of the Architecture of Integrated Information Systems (ARIS). In the modelling technique, eventdriven process chains (EPC), of ARIS, each process consists of an alternate sequence of events and functions. Thus, functions (events) of the EPC modelling technique can be mapped to the transformations (states) of the BWW models. Corresponding mappings are possible for the relationship types. Such a model comparison allows an objective ontological analysis and easily facilitates the identification of weaknesses such as ontological overlap, excess or redundancy (Green & Rosemann, 2000). Furthermore, this approach helps to identify synonyms (e.g., function and transformation) as well as homonyms (e.g., event).

Process Issues related to the process of conducting an ontological analysis have been described as lack of completeness, lack of guidance and lack of objectivity. Based on the assumption that corresponding meta-models for the ontology and the analysed grammar are available, it is possible to clearly specify the scope of an analysis using those meta-models. Such a selection of clusters, entity types and relationship types would define all elements that are to be perceived of relevance for a complete analysis. An analysis of an ER-based notation, for example, could be focused on the BWW clusters thing, system, and property



17

and could exclude the more behavioural-oriented clusters event and state. Such boundaries of an analysis could be easily visualised in the meta-model and would provide a clear description of the comprehensiveness of the analysis — thus, mitigating the completeness criticism. The existence of two corresponding meta-models and a clear definition of the scope of the analysis are necessary, but not sufficient, criteria for a well-guided process. Further guidelines are required regarding the starting point of such a process and the actual sequence of activities. Based on our experiences, we recommend starting with the representation mapping — that is, selecting the meta-model of the ontology and subsequently identifying the corresponding elements in the modelling grammar. The first construct to be analysed should be the most central entity type — that is, in the case of the BWW models the entity type thing. Our previous work provides a strong argument that this analysis should be followed by a cluster-by-cluster approach. Starting with the core constructs in a cluster, this approach allows a more structured and focused analysis of the completeness of a modelling grammar. The analysis of the entity types is followed by the relationship types and the cardinalities. Constructs in the meta-model that only have been introduced for the correctness of the metamodel, but that do not reflect ontological constructs are excluded from the analysis. The representation mapping is followed by an analysis of the clarity — that is, the interpretation mapping. In this case the meta-model of the grammar under analysis is the starting point. The general procedure is similar. A main advantage of a cluster-based analysis is that the structure of the two metamodels provides valuable input for the ontological analysis. In addition to the cluster-based analysis, a further guideline in the process relates to generalisation-specialisation relationships in the meta-model of the grammar. We propose to classify ontologically the super-type first and then to inherit this ontological classification to all sub-types. These guidelines streamline the process of the analyses and increase the consistency. The lack of objectivity issue, on the other hand, stems frequently from the analysis being performed by a single researcher. The situation results in an analysis that is almost certainly biased by the researcher’s background as well as their interpretation of the specification of the grammar. In order to improve the validity of the analysis, a research methodology can be adopted that undertakes individual analyses of a particular grammar by at least two members of a research team, followed by consensus as to the final analysis by the entire team of researchers. The methodology consists of three steps: 1.

Using the specification of the grammar in question, at least two researchers separately read the specification and interpret, select and map the ontological constructs to candidate grammatical constructs to create individual first drafts of the analysis.


18 Green & Rosemann

2.

The researchers involved in step 1 of the methodology meet to discuss and defend their interpretations of the modelling technique analysis. A concurrence score is determined then from their initial analyses. This meeting leads to an agreed second draft version of the analysis that incorporates elements of each of the researchers’ first draft analyses. The overlap in the selection of the constructs and in the actual ontological analysis can be quantified by concurrence/agreement scores that are used in content analysis and other more qualitative research.

3.

The second draft version of the analysis of the modelling technique is used as a basis for defence and discussion in a meeting involving the entire research team. The outcome of this meeting forms the final analysis of the grammar in question.

Such a methodology was employed in a project that sought to apply the BWW representation model analysis to a number of the leading potential Web service standards — that is, ebXML, BPML, BPEL4WS, and WSCI. The project team was composed of four researchers and the standards were analysed in the order: ebXML → BPML → BPEL4WS → WSCI. Two researchers were involved in steps 1 and 2 of the methodology — that is, the individual analysis of a standard followed by a meeting of the two researchers in order to obtain an agreed mapping. This phase was followed by a meeting of the entire team in order to discuss the mapping and arrive at the final analysis. The process was performed for each of the four standards. Table 3 shows the recorded agreement statistics at the second step of the applied methodology, while Table 4 shows the recorded agreement statistics at the third step of the methodology. Meta data of the ontological analysis such as the mapping ratio provides valuable information in addition to the actual outcomes of the analysis. In the case of the analysis of the Web services standards, for example, these figures give insight into how difficult or easy these standards are to understand. The adoption of such a methodology is seen to have improved significantly the objectiveness of the analyses.

Table 3. Summary of step 2 mapping agreement between both researchers Web Service Language ebXML BPML BPEL4WS WSCI

Representation Mapping agreed upon by both researchers 43 36 30 39

Total number of specification constructs identified 51 46 47 49

Mapping ratio

84% 78% 63% 79%



19

Table 4. Summary of step 3 mapping agreement Web Service Language ebXML BPML BPEL4WS WSCI

Representation Mapping agreed upon by the team 49 41 42 46

Total number of specification constructs identified 51 46 47 49

Mapping ratio

96% 89% 89% 94%

Output The three main shortcomings related to the outcome of an ontological analysis have been characterised as the lack of adequate result representation, lack of result classification and the lack of relevance. The meta-models that have been used as input for the ontological analyses are also an appropriate medium to visualise the outcomes of the entire analysis process. In our work on the analysis of ARIS, we derived a meta-model of the BWW model that highlighted all constructs of the ontology that do not have a corresponding construct in the grammar under analysis — that is, we visualised incompleteness in the model using simple colour coding. In a similar way, we derived three ARIS meta-models that highlighted excess, overload and redundancy in ARIS. Such models form a very intuitive way of representing the identified ontological shortcomings. The underlying clustering of the models also helps to quickly comprehend the main areas of shortcomings. At the present time, the process of an ontological analysis results in the identification of ontological incompleteness and ontological clarity through the identification of missing, overloaded or redundant grammatical constructs. While the end result identifies such problems, it fails to account for their relative importance. For example, thing is one of the fundamental constructs of the BWW model. The lack of mapping for the construct should, therefore, be considered more important than the lack of mapping for the well-defined event construct for example. There is a need for the development of a scoring model that enables the calculation of the ‘goodness’ of a grammar with respect to the ontology. In such a scoring model, each of the ontological constructs has a value assigned to it that reflects the relative importance of the construct in the ontology. Core constructs would therefore have high weightings whereas less important constructs would attract lower values of weightings. Following an ontological analysis of a particular grammar, the weighting of all missing constructs would be calculated to arrive at one value that generally reflects the outcome of the analysis.


20 Green & Rosemann

An example for such a classification could have for example the following structure. All core constructs of an ontology (and the modelling grammar) would get the value 1. All other constructs represented as an entity type in the metamodel of the ontology would receive the value 0.7, and all remaining constructs get the value 0.3. Such a weighting would then be applied to the outcomes of the ontological analysis. The scores would be aggregated across the ontology and modelling grammar. They also would be calculated separately for completeness, excess, overload and redundancy. Furthermore, they could be aggregated per cluster that allows a more differentiated view on the particular strengths of a modelling grammar. Though the consolidated score of such an evaluation should not be overrated, it provides better insights into the characteristics of the ontological deficiencies and provides a first rating of the significance and importance of the identified shortcomings. It can also be used for the design of the subsequent empirical studies. Apart from the lack of result classification that is addressed by the scoring model, another problem with the outcome of the analyses has been the perceived lack of relevance. The merit of a foundational ontology — that is, its generic nature and its completeness, can also be seen as a shortcoming — the ontology might cover more than what one single modelling technique can support and its level of abstraction is too high in order to form a specific benchmark. Thus, three activities seem to be required in order to convert foundational ontologies into focused ontologies.

•

First, since most modelling grammars concentrate on modelling a sub-set of the phenomena that occurs in the real world, it would follow that not all constructs of an ontology are necessary in order to analyse such a grammar. If the full ontology is used in the analysis, the result may identify potential problems that would not, in reality, occur, because the modelling grammar is not used to model any phenomena described by the missing constructs. Consequently, a focused ontology can be derived by deleting constructs from the selected ontology. Indeed, the outcomes of the ontological analyses of different modelling grammars to date appear to support the need for a focused ontology that consists of different subsets of the ontological constructs for different domains. The analyses of process modelling grammars consistently show that the constructs conceivable state space, conceivable event space and lawful event space, for example, have no representation constructs in the grammars. Such missing constructs, if identified to be unnecessary for the particular domain, can be ignored leading to a simpler analysis that does not consider phenomena that are deemed to be outside of the scope of the domain.



21

•

Second, there may also be a need for specialisation of some of the ontological constructs in order to enhance analysis of a grammar pertaining to a particular domain. For example, our analyses of Web services standards such as ebXML, BPEL4WS or BPML included the mapping of various activity types to the ontological construct transformation. Such findings could motivate the derivation of relevant sub-types of transformation when it comes to the context of business process management.

•

Third, the derivation of a focused ontology will require adapting the terminology of the analysed domain for two reasons. On the one side, the terms of the ontology might not be intuitive (e.g., conceivable state space within the BWW ontology). On the other side, the analysed domain might have its own established terminology. An example is the area of workflow modelling techniques, in which the Workflow Management Coalition had a significant impact with its glossary.

The argument for a focused ontology might be quite convincing and even seen as trivial. However, the development of focused ontologies faces a major challenge. The decisions about deleting constructs, adding sub-types and renaming constructs have to be based on a substantial number of ontological analyses before they can be justified. Thus, such focused ontologies are not readily available. In general, current ontological analyses focus on the selection of an adequate ontology and the evaluation of modelling grammars against that ontology. Ontological weaknesses are often interpreted as a weakness of the ontology or a weakness of the analysed grammar. It might be however a weakness of the comparison as the ontology and the analysed grammar do not fit. This situation can be explained by the highly interdisciplinary history of most ontologies and it has motivated our extension of the process of ontological analysis by adding a dimension that expresses the relevance of the results. The main advantages of this kind of analysis are that the identified weaknesses are relevant weaknesses and that the focused ontology is based on a well-discussed ontology with philosophical foundations. This use of the focused ontology in an analysis integrates the type of user and his/her relevant purpose. The purpose describes the objectives of the modelling tasks and is used to focus the modelling process at an early stage. For example, many workflow management systems include their own approach to describing the workflows. They are designed for exactly one purpose — the design and support of the execution of workflows. Nevertheless, a traditional ontological analysis would identify certain weaknesses. Possibly however, the developer and the ensuing users of this particular workflow modelling language


22 Green & Rosemann

Figure 5. An extension of ontological analysis through the use of focused ontologies Chosen Ontology

Focused Ontology

Elimination and Specialization

Modeling Grammar

Focused Ontological Analysis

do not care about such weaknesses, and never intended to provide a language that covers all constructs of the ontology. Besides the purpose, the type of user impacts the requirements of a situation. The user can be classified principally by their role within a modelling project, their role within the modeled domain, their knowledge of the domain, their experience with modelling, and/or their position in the organization. So far, we have only focused on the relevant purpose aspect. To this end, we have examined activity-based costing (ABC) (Rosemann & Green, 2000) and interoperability standards (Green & Rosemann, 2002; Green et al., 2004). We have used ABC (in its classical specification) first to develop a focused ontology because it is now well known and well specified in the business costing literature. One of our near-future directions for research is to test this focused ontology with ABC users to determine if the focused ontology better explains the constructs really required in the target technique.

Lessons Learned There has been a marked increase in the popularity of the application of ontologies for the purposes of modelling grammar analysis. For example, a literature review identified more than 25 papers that applied the Bunge-WandWeber ontology for the analysis of modelling grammars such as ER (e.g., Wand & Weber, 1989, 1993, 1995), OMT, UML (e.g., Opdahl & Henderson-Sellers, 2002; Shanks et al., 2003), Petri-Nets, ARIS (e.g., Green & Rosemann, 2000, 2002; Rosemann & Green, 2002) or Web services standards such as ebXML, BPEL4WS, BPML or WSCI (e.g., van der Aalst, Dumas, ter Hofstede, & Wohed, 2002; Wohed, van der Aalst, Dumas, & ter Hofstede, 2003; Green et al., Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


23

2004). Over the last five years, our understanding of ontologies and the contribution that they can make to requirements modelling and conceptual modelling has increased greatly. We have learned a number of important lessons. 1.

The understandability and the applicability of the selected ontology must be clear for IS professionals otherwise they will find it difficult to see the net benefits in the use of the analytical work. Accordingly, we have focused our efforts on developing a more intuitive meta-model for our preferred ontology and using this meta-model as the basis for explaining and applying the constructs of the ontology.

2.

Hypothesized weaknesses in a particular target modelling technique may not be in fact weaknesses of the technique but rather a misspecification in the adaptation of the preferred ontology to the IS modelling discipline. The adapted ontology may be over-engineered, under-engineered, and/or misspecified. In our work over the last five years in using our preferred ontology to analyse a range of techniques, we have noted on several occasions a core of ontological constructs whose representations in the target techniques have been absent. It would appear that our preferred ontology might be over-engineered in some respects. That is, the benefits of having representations in the target techniques for these particular ontological constructs do not appear to outweigh the costs of providing those representations irrespective of the type of user or business purpose of the modelling.

3.

We have perceived the need for a focusing of the ontology dependent on the type of user and the relevant business purpose. Accordingly, as an initial attempt in this direction, we have selected activity-based costing as a relatively well-defined business purpose and we are developing a focused ontology for this technique.

In general, selected ontologies and their interpretations, from an information systems viewpoint, are reasonably advanced. However, the actual process of conducting an ontological analysis is still rather premature. At this stage, the process is focused on the identification of the cardinality of the relationships between corresponding elements in the ontology and the modelling grammar under analysis. In total, eight shortcomings of the current process of ontological analysis have been identified and categorised into issues related to the input, process and output of the analysis. This chapter proposed to enhance further the current methodology of ontological analyses. The objectives of such a methodology are:


24 Green & Rosemann

•

To provide guidance for researchers who are interested in conducting ontological analyses.

•

To add rigour to the entire process and reduce the dependence on the subjective interpretations of the involved researcher.

•

To increase overall the credibility of the ontological analysis.

Examples from our ontological analyses of ARIS and various Web services standards have been used to exemplify this methodology. As a consequence, we hope that the presented more rigorous process will increase the overall acceptance of using ontologies for the analysis, comparison and engineering of various grammars. Our future work is continuing and developing in four principal directions. First, we are converting our meta-model to a UML-based definition. In this way, where there are UML-based meta-models for other grammars, we can make our analyses more objective. Second, we are using our meta-model work to provide a basis on which to compare ontologies. In this way, we can provide some theoretical guidance for the selection of an ontology for an evaluative/analytical task. Third, we continue to investigate different business purposes for the production of relevant focused ontologies for the evaluation/engineering of modelling methods that are popularly used in that area. For example, we are currently working on a focused ontology for business process management that will be derived from the BWW ontology. Finally, we continue to empirically test the predictions of our ontologically based evaluations. In this way, we can contribute to the development of the BWW theoretical foundation for business and information systems modelling techniques.

References Bansler, J. P., & Bodker, K. (1993). A reappraisal of structured analysis: Design in an organizational context. ACM Transactions on Information Systems, 11(2), 165-193. Bunge, M. (1977). Treatise on basic philosophy: Volume 3: Ontology I: The furniture of the world. Boston: Reidel. Burton-Jones, A., & Meso, P. (2002). How good are these UML diagrams? An empirical test of the Wand and Weber good decomposition model. In L. Applegate, R. Galliers & J. I. DeGross (Eds.), Proceedings of the 23rd International Conference on Information Systems (ICIS 2002), Barcelona, 15-18 December, (pp. 101-114). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


25

Chen, P. P-S. (1976). Entity-relationship model: Towards a unified view of data. ACM Transactions on Database Systems 1(1), 9-36. Davies, I.G., Green, P., Milton, S., & Rosemann, M. (2004). Analysing and comparing ontologies with meta-models. In J. Krogstie, T. Halpin, & K. Siau (Eds.), Information modeling methods and methodologies (pp. 116). Hershey, PA: Idea Group Publishing. Davies, I., Green, P., & Rosemann, M. (2002). Facilitating an ontological foundation of information systems with meta models. In A. Wenn, M. McGrath, & F. Burstein (Eds.), Proceedings of the 13th Australasian Conference on Information Systems (ACIS 2002), Melbourne, 4-6 December, (pp. 937-948). Fettke, P., & Loos, P. (2003). Ontological evaluation of reference models using the Bunge-Wand-Weber model. In Proceedings of the 9th Americas Conference on Information Systems, Tampa, (pp. 29-44). Gorla, N., Pu, H. C., & Rom, W. O. (1995). Evaluation of process tools in systems analysis. Information and Software Technology, 37(2), 119126. Green, P., & Rosemann, M. (2004). Applying ontologies to business and systems modelling techniques and perspectives: Lessons learned. Journal of Database Management, 15(2), 105-117. Green, P., Rosemann, M., Indulska, M., & Manning, C. (2004). Candidate interoperability standards: An ontological overlap analysis. Submitted to Data & Knowledge Engineering, April, 2004. Green, P. F. (1997). Use of information systems analysis and design (ISAD) grammars in combination in upper CASE tools — An ontological evaluation. In Proceedings of the 2nd CaiSE/IFIP8.1 International Workshop on the Evaluation of Modeling Methods in Systems Analysis and Design, Barcelona, (pp. 1-12). Green, P. F., & Rosemann, M. (2000). Integrated process modelling: An ontological evaluation. Information Systems, 25(2), 73-87. Green, P. F., & Rosemann, M. (2002). Perceived ontological weaknesses of process modeling techniques: Further evidence. In Proceedings of the 10th European Conference on Information Systems, Poland, (pp. 312321). Green, P. F., & Rosemann, M. (2002). Usefulness of the BWW ontological models as a “core” theory of information systems. In Proceedings of Workshop on Information Systems Foundations: Building the Theoretical Base, Australian National University: Canberra, (pp. 147-164). Gruninger, M., & Lee, J. (2002). Ontology: Applications and design. Communications of the ACM, 45(2), 39-41. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

26 Green & Rosemann

Guarino, N., & Welty, C. (2002). Evaluating ontological decisions with OntoClean. Communications of the ACM, 45(2), 61-65. Karam, G. M., & Casselman, R. S. (1993). A cataloging framework for software development methods. IEEE Computer, 34-46. Olle, T. W., Hagelstein, J., Macdonald, I. G., Rolland, G., Sol, H. G., Van Assche, F. J. M., & Verrijn-Stuart, A. A. (1991). Information systems methodologies: A framework for understanding. Wokingham, England: AddisonWesley. Opdahl, A. L., & Henderson-Sellers, B. (2001). Grounding the OML metamodel in ontology. Journal of Systems and Software, 57(2), 119-143. Opdahl, A. L., & Henderson-Sellers, B. (2002). Ontological evaluation of the UML using the Bunge-Wand-Weber model. Software and Systems Modeling Journal, 1(1), 43-67. Parsons, J., & Wand, W. (1997). Using objects in systems analysis. Communications of the ACM, 40(12), 104-110. Rosemann, M., & Green, P. F. (2000). Integrating multi-perspective views into ontological analysis. In W. Orlikowski, S. Ang, P. Weill, H. Krcmar, & J. deGross (Eds.), Proceedings of the 21st International Conference on Information Systems, Brisbane, 10-13 December, (pp. 618-627). Rosemann, M., & Green, P. F. (2002). Developing a meta model for the BungeWand-Weber ontological constructs. Information Systems, 27(2), 75-91. Rosemann, M., Green, P. F., & Indulska, M. (2004). Towards an enhanced methodology for ontological analyses. In J. Grabis, A. Perrson, & J. Stirna (Eds.), Proceedings of the CAiSE ’04 Forum, Riga, June, (pp. 112-121). Scheer, A. W. (2000a). ARIS – Business process modeling. Springer: Berlin. Scheer, A. W. (2000b). ARIS – Business process frameworks (3rd ed.). Berlin: Springer. Shanks, G., Tansley, E., & Weber, R. (2003). Using ontology to validate conceptual models. Communications of the ACM, 46(10), 85-89. Sia, S. K., & Soh, C. (2002). Severity assessment of ERP-organization misalignment: Honing in on ontological structure and context specificity. In L. Applegate et al. (Eds.), Proceedings of 23rd International Conference on Information Systems (ICIS2002), Barcelona, December. (pp. 723729). van der Aalst, W. M. P., Dumas, M., ter Hofstede, A. H. M., & Wohed, P. (2002). Pattern based analysis of BPML (and WSCI) (Technical report No. FIT-TR-2002-050). Brisbane, Australia: Queensland University of Technology.



27

Wand, Y., & Weber, R. (1989). An ontological evaluation of systems analysis and design methods. In E. D. Falkenberg & P. Lindgreen (Eds.), Information system concepts: An in-depth analysis (pp. 79-107). Amsterdam, Netherlands: North-Holland. Wand, Y., & Weber, R. (1990a). Mario Bunge’s ontology as a formal foundation for information systems concepts. In P. Weingartner & G. J. W. Dorn (Eds.), Studies on Mario Bunge’s Treatise (pp. 123-149). Atlanta: Rodopi. Wand, Y., & Weber, R. (1990b). An ontological model of an information system. IEEE Transactions on Software Engineering, 16(11), 1281-1291. Wand, Y., & Weber, R. (1993). On the ontological expressiveness of information systems analysis and design grammars. Journal of Information Systems, 3(4), 217-237. Wand, Y., & Weber, R. (1995). On the deep structure of information systems. Information Systems Journal, 5, 203-223. Wand, Y., & Weber, R. (2002). Information systems and conceptual modelling: A research agenda. Information Systems Research, 13(4), 363-376. Weber, R. (1997). Ontological Foundations of Information Systems (Monograph No. 4). Melbourne, Australia: Melbourne, Vic., Coopers & Lybrand and the Accounting Association of Australia and New Zealand. Weber, R., & Zhang, Y. (1996). An analytical evaluation of NIAM’s grammar for conceptual schema diagrams. Information Systems Journal, 6(2), 147-170. Wohed, P., van der Aalst, W.M.P., Dumas, M., & ter Hofstede, A. (2003). Analysis of Web service composition languages: The case of BPEL4WS. Proceedings of 22 nd International Conference on Conceptual Modelling (ER) (pp. 200-215), Chicago, October.

Endnote 1

In [5] a meta-model can be distinguished from a grammar, for the purposes of this work, as a model of how the constructs of a grammar are related.


28 Shanks, Nuredini & Weber

Chapter II

Evaluating Conceptual Modelling Practices: Composites, Things, Properties Graeme Shanks, Monash University, Australia Jasmina Nuredini, Monash University, Australia Ron Weber, Monash University, Australia

Abstract This chapter examines how ontological theory can be used to predict how alternative conceptual modelling representations affect end-user understanding of these representations. Specifically, it examines how ontological theory can be used to show how part-whole relations (composites) and things and properties can be best represented to enhance understanding of these real-world phenomena. We report the outcomes of two experiments that provide evidence to support the ontologically sound representation of part-whole relations and things and properties. We also discuss the outcomes of a cognitive process tracing study that explains why the ontologically sound representation of things and properties is more easily Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

Evaluating Conceptual Modelling Practices

29

understood. In essence, our empirical research provides evidence to support the use of ontology as a theoretical basis to guide conceptual modelling practices.

Introduction The representation of real-world phenomena as conceptual models has been a concern of information systems practitioners and researchers for some time. For example, Wand, Storey, and Weber (1999) have sought to build a rigorous ontological theory to provide a model of the structure and dynamics of some facets of the real world in general. Their goal has been to provide a theoretical basis for evaluating conceptual modelling practices. Their theory is an adaptation and extension of an ontological theory proposed by Bunge (1977). Bunge’s theory was selected because of its rigour and comprehensiveness. It provides thorough articulation of constructs such as things (entities), properties of things, states of things, and compositions of things — phenomena that are of major interest to conceptual modelling practitioners. In this chapter, we focus on two features of the real world that conceptual modellers encounter — namely, the existence of things that are part of another thing and the distinction between things and properties. The notions that one thing may be part of another thing (e.g., a wheel is part of a bicycle) and the distinction between things and properties (e.g., a person is a thing with properties such as height and weight) are fundamental to the way people perceive and understand the world. In the context of conceptual modelling, these notions are problematic because alternative representations have been proposed and substantive theoretical issues remain unresolved. To illustrate, Rumbaugh, Jacobson, and Booch (1999, p. 146) state: “The aggregation (part-whole) relationship is transitive and antisymmetric across all aggregation links, even across those from different aggregation associations”, yet Winston, Chaffin, and Herrman (1987, pp. 431432) argue that not all part-whole relations are transitive. Furthermore, composite things are sometimes represented explicitly as entities (e.g., Kilov & Ross, 1994, pp. 96-97) and sometimes implicitly as relationships between the components of the composite (e.g., Chen, 1976, p. 31). In terms of distinguishing between things and properties, proponents of the object-role approach to conceptual modelling claim the distinction is unimportant (Halpin, 1995). They model things and properties of things using the object symbol in a conceptual schema. In the entity-relationship model (Chen, 1976), however, things are represented as entity types, and properties are represented as attribute types. In our view, conceptual models should be used to discover and document stakeholder perceptions of a domain to provide a basis for informed discernment Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


about how phenomena should be represented in an information system (Hischheim, Klein & Lyytinen, 1995) rather than being driven by database design considerations (Simsion & Witt, 2001, p. 101). For this reason, we argue that the representation of part-whole relations and things and properties in conceptual models should be based on a sound underlying theory of how the world is structured. To the best of our knowledge, however, no rigorous empirical evaluation of alternative representations of part-whole relations and things and properties has been undertaken. In the absence of such research, we undertook to empirically evaluate alternative representations. Our research had several motivations. First, the cost of fixing errors increases the later they are discovered in the system development process (e.g., Boehm, 1981). Because, conceptual modelling work is undertaken early in the system development process, improvements in conceptual modelling practice potentially will lead to high payoffs (Moody & Shanks, 1998). Second, we sought to test prior theoretical work undertaken to predict how well different types of representations facilitate or inhibit human understanding of real-world phenomena. If accurate predictions about the types of conceptual modelling practices that are likely to be effective can be made, the high cost of learning the strengths and weaknesses of different practices through experience can be avoided. Third, we seek to improve user understanding of conceptual models. When conceptual models are prepared initially (e.g., by systems analysts), the users of an information system are asked to validate them to determine how accurately and completely the models represent their perceptual worlds. Finally, we sought to contribute to improved conceptual modelling practice. Numerous varying and sometimes ambiguous guidelines for representation of part-whole relations and things and properties exist in the literature. These guidelines tend to confuse rather than assist practitioners (Simsion & Witt, 2001). We aim to help practitioners by developing improved conceptual modelling rules for part-whole relations and things and properties. In this chapter we report the results of three empirical studies we undertook to examine user understanding of conceptual models. We used ontological theory to predict how part-whole relations and things and properties are best represented to enable user understanding of these phenomena. We also discuss the research and practical implications of our findings as well as future work that might be undertaken.

Theoretical Background Little theory exists that can be used to predict or explain why any particular conceptual modelling notation or representation is better understood by end Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


31

users. Furthermore, there is no empirical evidence to explain which representation of part-whole relations and things and properties is better. In this light, we relied on Wand et al.’s (1999) arguments about which representation is better. They use Bunge’s (1977) ontological theory as the basis for their analysis. In brief, their arguments run as follows: 1.

“The world is made of things that possess properties” (p. 497). Things and properties are the two atomic constructs needed to describe the world.

2.

Every thing in the world possesses one or more properties (there are no bare things).

3.

Properties themselves cannot have properties. Moreover, properties cannot exist by themselves. They must attach to some thing.

4.

Two types of properties that exist in the world are intrinsic properties, which depend on one thing only, and mutual properties, which depend on two or more things.

5.

Two things interact (are coupled) when a history of one thing (manifested as a sequence of the thing’s states) would be different if the other thing did not exist.

6.

The existence of a mutual property between two things can indicate that they interact with each other. Mutual properties that manifest interactions between two things are called binding mutual properties.

7.

“Two things may associate to form another thing.” A thing is a composite if and only if it is formed from the combination of at least two other things. Otherwise, it is a simple thing. (p. 504).

8.

Every composite thing possesses emergent properties — properties that are not possessed by the components of the composite. (p. 504).

In the context of Bunge’s (1977) ontological theory, a composite can not be represented as a relationship because a) relationships themselves represent mutual properties and b) every composite must possess at least one emergent property. Figures 1 and 2 illustrate an example of how part-whole relations may be represented in a UML diagram. Figure 1 depicts an association between student, subject and term. An enrollment associates a student with a subject in a term; therefore, each separate enrollment must contain a student, a subject and a term. In Figure 2, the example is expressed as a part-of association between student, enrollment, subject and a term. Here, enrollment is represented as an object, where it is strongly aggregated with student and term and weakly aggregated with subject. If the ontological principles are contravened and composites are represented as relationships, the resulting conceptual schema diagram is limited. Users will Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Figure 1. UML association class Student Student 1..1

Enrollment Enrolment

1..1

Subject Subject

1..*

Subject Subject

1..1

Term Term

Figure 2. UML entity class Student 1..* 0..*

Enrollment Enrolment

0..*

0..* 1..*

Term Term

employ tacit knowledge to determine whether the relationship represents a composite thing or a mutual property of two or more things. For example, in Figure 1, enrollment could be interpreted as a mutual property or relationship (association) between student, term, and subject classes. If intrinsic and/or mutual properties were attached to the relationship, it would be unclear whether the properties were intended as properties of the relationship or properties of the composite. Also, in the context of Bunge’s (1977) ontological theory, a property cannot be represented as an entity type because construct (semantic) overload will arise. This outcome occurs when the same grammatical construct (entity type symbol) is used to represent two ontological constructs (things and their properties). Figures 3 and 4 demonstrate how entities and properties may be represented in ER models. Figure 3 shows things represented as entity types (student, address) and properties as entity types (student address). Figure 4 shows an alternative representation that we propose where things are represented as entities and properties are represented as attributes. If the ontological principles are contra-



33

vened and properties are represented as entities, the conceptual schema model becomes limited. Users will resort to tacit knowledge to determine whether the entity type represents a thing or a property. For example, in Figure 3, student address may be interpreted as a thing when instead it is a mutual property of student and address. In addition to Bunge’s theory, this research relies on theoretical work on cognition. Extensive research reveals that humans cognitively cluster phenomena that they perceive to be related (e.g., Bousfield, 1953). Clustering appears to provide a means for humans to deal with the complexity they often encounter in their perceptual worlds (Miller, 1956). By focussing on clusters, they reduce cognitive load and enhance their abilities to understand the world. Properties of things naturally cluster with the things to which they belong. Perceiving the world in terms of things and their properties, therefore, helps humans to mitigate the cognitive problems they experience when they perceive phenomena to be complex. To maximise our contribution to conceptual modelling practices, we based our empirical studies on the widely used UML and ER modelling notations. Part-

Figure 3. Practice student address ER model

Student

has

has

Student Number Student Name Student DOB Student Address

Address Address Number Street Number Street Name Suburb Country Post Code

Student Number Address Number From Date To Date

Figure 4. Ontologically sound student address ER model

Student Student Number Student Name Student DOB {Address Number From Date To Date}*

has

Address Address Number Street Number Street Name Suburb Country Post Code {Student Number From Date To Date}*



whole relations feature in object-oriented conceptual modelling approaches (e.g., Rumbaugh et al., 1999); therefore, we used UML to test the representation of part-whole relations (Appendices A and B). Part-whole relations also feature in the ER modelling notation (Simsion & Witt, 2001). Entity and attribute types are fundamental constructs in ER modelling approaches; therefore, we used ER modelling to test the representation of things and properties (Appendices C, D, E, and F). Things and properties also appear in the UML modelling notation. We contend that the choice of representation in conceptual modelling is important in terms of users’ ability to elicit the meaning of the phenomena described via the representation. Hence, the following propositions motivate the two experiments we undertook:

•

Proposition 1: Conceptual models that use entity class constructs to represent composites will enable users to better understand the semantics associated with the composite than conceptual models that use a relationship class construct.

•

Proposition 2: Conceptual models that use an attribute construct to represent properties will enable users to better understand the semantics associated with the model than conceptual models that use an entity class construct.

In order to better understand the outcomes of our second experiment, we undertook an exploratory cognitive process tracing study motivated by the following proposition:

•

Proposition 3: Cognitive behavior patterns of users explain the differences in their ability to understand conceptual models with different representations of things and properties.

Representing Part-Whole Relations The first experiment investigated user understanding of the representation of part-whole relations and was reported in detail in Shanks, Tansley, Nuredini, Tobin, and Weber (2002). A summary of the design and outcomes follows.



35

Research Method and Design Design and Measures The experiment was conducted in a laboratory setting to allow for control of external factors that might confound the results. One between-groups factor was used. This factor, “type of representation”, had two levels: the ontologically sound level which had both composites and components in part-whole relations represented as entity classes, and the ontologically unsound level which had composites represented as relationship classes. Both levels were represented in a UML class diagram. The dependent variable was the performance of on problem-solving questions. This variable was selected to evaluate how well participants in the experiment understood the project-planning domain represented in the UML class diagram. It is deemed to provide a better indicator of “deep” understanding (Mayer, 1989; Bloom, 1956), and it has been used before in the information systems field (Geminio, 1999; Bodart, Sim, Patel, & Weber, 2001). Performance on problem-solving questions was measured according to solution accuracy and time taken.

Materials Four sets of materials were used in the experiment — namely, a summary of the UML symbols with definitions, two UML class diagrams, eleven problemsolving questions, and a personal profile questionnaire. The symbol summary was designed to inform participants of the meaning of each symbol used in the experimental diagrams. One model, the ontologically sound model (Appendix A), had both components and composites represented explicitly as entity classes and were linked via associations. The other model, the ontologically unsound model (Appendix B), showed components as classes whereas composites were implied via the links between component classes. A set of problem-solving questions was also developed to encourage participants to use the UML class diagrams and to avoid use of tacit knowledge. Finally, participant demographic data was collected using a personal profile questionnaire.

Participants Participants were selected based on their ability and experience to act as surrogate end users. Thirty took part in the study, all of whom were working in non-technical roles and had little or no modelling experience.



Procedures Participants were run individually through the experiment to enable detailed observation of their problem-solving behavior. They were assigned randomly to one of the two treatments. They also signed a consent form and provided demographic and experiential information about themselves. Next, they were given symbol summary of UML notation that they retained and could refer to during the experiment. They were then given either the ontologically sound or ontologically unsound UML class diagram together with the problem-solving questions. Participants were asked to speak aloud as they worked through each question so that their utterances could be audio recorded and documented. The time taken to answer each question was recorded, and we made notes on their reactions and queries in relation to each problem-solving question.

Outcomes To evaluate the results of the experiment, we analysed both the scores for problem-solving questions and the transcriptions of the audio recordings.

Quantitative Analysis The scoring of the problem-solving task was calculated as follows:

•

Answer where one mark was given if the answer (“possible” or “not possible”) was correct; zero was given if the answer was incorrect.

•

Explanation where a judgment was made based on the participant’s explanations, researchers’ notes, and the audio recording. Clear explanations supporting an answer were awarded one mark, and moderately clear explanations were awarded a half mark. If explanations were unclear, zero marks were awarded.

•

Interpretation where a judgment was made using participant explanations, notes, and audio recordings. Clear interpretations of the domain received one mark, while moderately clear interpretations received a half mark. Any unclear interpretations were awarded zero marks.

Table 1 presents the statistics for the total accuracy, total time and t-tests. These figures suggest that participants who received the ontologically sound model scored much higher in terms of accuracy than those who received the ontologically



37

Table 1. Problem solving summary statistics Group / Measure Sound Model Unsound Model

Accuracy (percentage) t=7.450, df= 28, pforAll(s2 | s.stateName”” and s.stateName=s2.stateName implies s=s2) The next constraint is again similar and we do not write it out in OCL.


Template-Based Definition of Information Systems

•

115

Two “Events” with non-empty “eventNames” cannot have the same “eventName”.

Constraints on “Classes” and “RepresentedClasses” The three constraints in this group deal with the uniqueness of “Class roleNames” within “ConstructDefinitions”, with the uniqueness of the set4 of “characteristic Properties” of a “Class” and with “Class” specialisation/generalisation.

•

If a “ConstructDefinition” contains more than one “RepresentedClass”, each of them must have a “roleName” that is unique to the “ConstructDefinition .

context ConstructDefinition inv: representedClass->forAll(rc | representedClass->forAll(rc2 | rcrc2 implies rc.roleName”” and rc2.roleName”” and rc.roleNamerc2.roleName))

•

Two different “Classes” cannot be associated with the same sets of characteristic “Properties”.

context c : Class inv: Class->forAll(c2 | c c2 implies c.representedProperty c2.representedProperty)

•

If the set of “characteristic Properties” of one “Class” is a subset of that of another “Class”, the first “Class” must generalise the second.

context c : Class inv: Class->forAll(c2 | c.characteristic->includesAll(c2.characteristic) implies c.generalisation->includes(c2))


116

Opdahl & Henderson-Sellers

Hence, at the conceptual level described in the meta-model, generalisations are used as an aid to validate partially the sets of “characteristic Properties” that are associated with each “Class”. At the implementation level, when managing the taxonomy of “Classes”, “Properties”, “Events”, and “States”, generalisations can instead be used to limit the number of “Properties” that have to be explicitly associated with each “Class”, that is, by not explicitly associating a “characteristic Property” with a “Class” whose “generalisation” already possesses that “Property”. Otherwise, adding new “Classes” to the taxonomy would quickly become cumbersome because each new “Class” would have to be explicitly associated with an unfeasibly large number of “characteristic Properties”.

Constraints on “Properties” and “RepresentedProperties” The first four constraints in this group ensure that the “RepresentedClasses” and “RepresentedProperties” contained in a “ConstructDefinition” match one another, that is, that all the necessary “Classes” and “Properties” are contained in the “ConstructDefinition” and that the “Properties” belong to the “Classes” and vice versa.

•

If a “ConstructDefinition” contains a “RepresentedClass” that has a “RepresentedProperty”, the “ConstructDefinition” must also contain the “RepresentedProperty”.

context ConstructDefinition inv: representedClass->forAll(rc | representedProperty->includesAll(rc.representedProperty))

•

Conversely, if a “ConstructDefinition” contains a “RepresentedProperty” that has a “RepresentedClass”, the “ConstructDefinition” must also contain the “RepresentedClass”.

context ConstructDefinition inv: representedProperty->forAll(rp | representedClass->includesAll(rp.representedClass))



•

117

If a “RepresentedClass” has a “RepresentedProperty”, the corresponding “Class” must have the corresponding “Property” as “characteristic”.

context rc : RepresentedClass inv: representedProperty->forAll(rp | rc.class.characteristic->includes(rp.property))

•

Conversely, if a “RepresentedProperty” has a “RepresentedClass”, the corresponding “Property” must be “characteristic” of the corresponding “Class”.

context rp : RepresentedProperty inv: representedClass->forAll(rc | rp.property.class->includes(rc.class)) The next constraint deals with the uniqueness of “roleNames” of “RepresentedProperties”.

•

If a “RepresentedClass” has more than one “RepresentedProperty”, each of them must have a “roleName” that is unique relative to the “RepresentedClass”.

context RepresentedClass inv: representedProperty->forAll(rp | representedProperty->forAll(rp2 | rprp2 implies rp.roleName”” and rp2.roleName”” and rp.roleNamerp2.roleName)) The next constraint defines precedence between “Properties”.

•

If a “Class” has a “characteristic Property” that is “preceded” by another “Property”, then the “Class” must also have the second “Property” as “characteristic”.


118


context Class inv: characteristic->forAll(cp | characteristic.includesAll(cp.preceded)) Like “Class” generalisations, at the conceptual level described in the metamodel, precedence associations are used as an aid to validate partially the sets of “characteristic Properties” that are associated with each “Class”. At the implementation level, when managing the taxonomy of “Classes” and “Properties”, precedence associations can instead be used to limit the number of “Properties” that have to be explicitly associated with each “Class”, that is, by not associating a “characteristic Property” with a “Class” when the “Class” is already associated with another “characteristic Property” that is preceded the first one.5 The final two constraints in this group together ensure that precedence associations between “Properties” are acyclic. This first constraint reflects that the precede/preceding association is transitive the second that it is irreflexive.

•

If a “Property” is “preceded” by a second “Property” and the second “Property” is “preceded” by a third, then the first “Property” must also be “preceded” by the third “Property”.

context p : Property inv: Property->forAll(p2 | Property->forAll(p3 | p.preceded->includes(p2) and p2.preceded->includes(p3) implies p.preceded->includes(p3)))

•

A “Property” cannot be “preceded” by itself.

context Property inv: preceded->excludes(self) Together, the two constraints ensure that precedence associations are acyclic because, if there were a cycle of associations, each “Property” in the cycle would be “preceded” by itself by the first constraint, thereby violating the second constraint.



119

Constraints on “RepresentedSegments”, “States”, and “Events” The final and most complex group of constraints deal with “RepresentedSegments” and the (“Represented-”)“States” and “Events” they contain. The first constraint reflects that every BWW state must be a state in a particular thing or class.

•

If a “State” has a set of “Properties”, there must be a “Class” whose set of “characteristic Properties” is a (possibly improper) superset of the first set.

context s : State inv: Class->exists(c | c.characteristic->includesAll(s.Property)) The following two constraints ensure that “ConstructDefinitions” that represent “States” also contain matching “Classes” and “Properties”.

•

If 1) a “ConstructDefinition” contains a “RepresentedSegment” and 2) the “RepresentedSegment” contains a “RepresentedState” and 3) the corresponding “State” has a set of “Properties”, then there must be a “Class” whose set of “characteristic Properties” is a (possibly improper) superset of the first set and the “ConstructDefinition” must contain the corresponding “RepresentedClass”.

context ConstructDefinition inv: representedSegment.representedState->forAll(rs | Class->exists(c | c.characteristic->includesAll(rs.state.property) and representedClass->exists(rc | rc.class=c)))

•

If 1) a “ConstructDefinition” contains a “RepresentedSegment” and 2) the “RepresentedSegment” contains a “RepresentedState” and 3) the corresponding “State” has a “Property”, then the “ConstructDefinition” must also contain a corresponding “RepresentedProperty”.


120


context ConstructDefinition inv: representedSegment.representedState->forAll(rs | rs.state.property->forAll(rsp | representedProperty->exists(rp | rp.property=rsp))) The next two constraints ensure that all “States” and “Events” in the common taxonomy are unique.

•

Two distinct “States” cannot have the same “invariant”.

context s : State inv: State ->forAll(s2 | s.invariant=s2.invariant implies s=s2)

•

Two distinct “Events” cannot have identical “from-” and “toStates”.

context e : Event inv: Event ->forAll(e2 | e.fromState=e2.fromState and e.toState=e2.toState implies e=e2) The next four constraints restrict which “RepresentedStates” and “-Events” that a “RepresentedSegment” can contain.

•

If a “RepresentedSegment” has “segmentType”=lifetime it cannot contain a “RepresentedState” or a “RepresentedEvent”.

context RepresentedSegment inv: segmentType=lifetime implies representedState->isEmpty() and representedEvent->isEmpty() The three other constraints are so similar to this one that we do not write them out in OCL:

•

If a “RepresentedSegment” has “segmentType”=state it must contain one “RepresentedState” and it cannot contain a “RepresentedEvent”.



121

•

If a “RepresentedSegment” has “segmentType”=event it must contain one “RepresentedEvent” and two “RepresentedStates”.

•

If a “RepresentedSegment” has “segmentType”=process it must contain at least three “RepresentedStates” and at least two “RepresentedEvents”.

The next constraint ensures that the “States” and “Events” in a “RepresentedSegment” match.

•

If a “RepresentedSegment” contains a “RepresentedEvent”, then the “RepresentedSegment” must also contain a “RepresentedState” so that the corresponding “State” has the corresponding “Event” as “exitEvent”.

context RepresentedSegment inv: representedEvent->forAll(re | representedState->exists(rs | rs.state.exitEvent->includes(re.event)))

•

If a “RepresentedSegment” contains a “RepresentedEvent”, then the “RepresentedSegment” must also contain a “RepresentedState” so that the corresponding “State” has the corresponding “Event” as “entryEvent”.

We do not write out this analogous constraint in OCL. The converse of the two previous constraints is not a constraint. Although an “Event” cannot be specified without its “from-” and “toStates”, a “State” can be specified without its “entry-” and “exitEvents”, as indicated by the cardinality constraints. The reason is that whereas a “State” is fully defined by its association with one or more “Properties” and by its invariant over those “Properties”, an “Event” is defined by its “from-” and “toStates”.6 The final constraint in this group cannot be expressed using OCL, because it is a constraint about the OclExpression that defines a “State invariant”. OCL expressions currently cannot constrain other OCL expressions.

•

The invariant of a “State” can only refer to “Properties” that are associated with the “State” and it must refer non-trivially to all such “Properties”. In other words, the invariant of a “State” 1) can only constrain and 2) must constrain all the “Properties” that determine the “State”.


122


By non-trivially we mean that, for each “Property” of a “State”, the invariant must disallow at least one potential value for the “Property” for at least one combination of values of the other “Properties” of the “State”.

Populating the Template Opdahl and Henderson-Sellers (2004) use modelling constructs from the unified modeling language (UML) (OMG, 2001) as examples and present in table form initial template-based definitions of 58 UML construct. The examples and the table are based on Opdahl, Henderson-Sellers, and Barbier (1999) and Opdahl and Henderson-Sellers, (2001, 2002), which have analysed and evaluated the UML and a related language in terms of the BWW model. They have also been inspired by Evermann and Wand’s (2001a, 2001b) related work. In addition, Opdahl and Henderson-Sellers (2001) have compared the underlying ontological assumptions of the BWW model with those of object-oriented modelling in general. This section presents more detailed definitions of selected UML construct using the template. The constructs have been selected either because they are central to the UML or because they illustrate important ideas behind the template. The purpose is to make the template easier to understand by providing concrete examples, to validate the meta-model by instantiating it, and to further our work on providing a complete template-based definition of the UML by investigating selected UML construct in more detail. Because the UML has weak semantics in relation to concrete problem domains today, some of the definitions are interpretations and proposals that must be evaluated in further work. Figure 2 shows a UML object diagram of the definition of the UML’s object construct using the template. A UML object is at the instance level and may represent any instance of the class of “AssociativeThings”, a very general class defined by the characteristic property of being able to associate (Bunge, 1977).7 A few other UML construct have very similar definitions. For example, UML active objects differ from objects only in that they represent instances of the class of “ActiveThings”. In turn, UML swimlanes differ from UML active objects only in that they represent processes instead of lifetimes, that is, they do not represent active objects from creation all the way to destruction. UML types differ from objects only in that they belong to the type level. Figure 3 shows a UML object diagram of the definition of the UML’s property construct using the template. Even this definition is very similar to the definition of UML objects, but in addition a UML property may represent “anyRegularProperty”, that is, any intrinsic property that is not a law or a



123

Figure 2. A UML object diagram of the template-based definition of UML objects

Figure 3. A UML object diagram of the template-based definition of UML properties

whole-part relation. Whereas all the “Properties” we have encountered so far have been real properties, that is, properties that belong to real things, “anyRegularProperty” is abstract in the sense that it is represented in the common taxonomy as a “propertySet” of “setMember Properties”. Moreover, “anyRegularProperty” is not itself “characteristic” of any “Class”, it has no precedence associations and it does not determine a “State”. (Only real “Properties” can play these three roles in a definition.) The definition of UML properties reuses the “Class” “AssociativeThings” and its “characteristic


124


Figure 4. A UML object diagram of the template-based definition of UML multiplicities

Property” “abilityToAssociate” from Figure 2. UML attributes differ from properties only in that they belong to the type level. Figure 4 shows a UML object diagram of the definition of the UML’s multiplicity construct using the template. UML multiplicities differ from UML attributes only in that they may represent different “Properties”. A “multiplicityStateLaw” contains three “Properties” : a “minimumCardinality”, a “maximumCardinality”, and “anyRegularProperty”, indicating that the number of regular properties must be between the minimum and maximum cardinalities. The “multiplicityConstraint” “Property” also has a law attribute (left out of Figure 4 for space reasons), which is an OclExpression describing the constraint in detail: minimumCardinality.size()=1 and maximumCardinality.size()=1 and minimumCardinality.value()>=0 and maximumCardinality.value()>=minimumCardinality.value() and



125

anyRegularProperty.size()>=minimumCardinality.value() and anyRegularProperty.size() A process (class) A set of objects / states linked to a process by a condition / event / effect / consumption / instrument link. The process is linked to another set of objects / states by an effect / result link Object / state → event link → process Process → effect / result link → object / state A persistent state, or any other state, which is not unstable (see below)

State A in the sequence <state A → condition / event / consumption link → process → result link → state B> is an unstable state An object class, which is related to another class by a specialization link Composition and decomposition are given by the sequence . The composite thing is linked at the vertex of the aggregation symbol and its components at the bottom


170 Reinhartz-Berger & Dori

Appendix B: OPM Concepts and Symbols Table 3. Entities: things and states Entity Type

Entity Symbol

Systemic, informatical object Object

Environmental, informatical object Systemic, physical object Environmental, physical object Systemic, informatical process

Process

Environmental, informatical process Systemic, physical process Environmental, physical process

State

Regular state Initial state Final state Default state

Table 4. Structural relations, their OPD symbols, and OPL sentences Structural Relation Name

OPD Symbol

OPL Sentence

Aggregation-Participation

A consists of B.

Exhibition-Characterization

A exhibits B.

Generalization-Specialization

B is an A.

Classification-Instantiation

B is an instance of A.

Tagged Structural Link

A relates to B. A and B are equivalent.

XOR relation

E.g., A relates to either B or C.

OR relation

E.g., A relates to B or C.


A Reflective Meta-Model of Object-Process Methodology 171

Table 5. Procedural links, their OPD symbols, and OPL sentences Type

Link Name

Semantics

Enabling Links

Instrument

The process requires the entity, but does not change it during execution.

OPD Symbol

OPL Sentence

P requires A.

Transforming Links Conditional Links

Consumption

The process consumes the entity.

P consumes A.

Result

The process generates (creates) the entity.

P yields A.

Effect

The process changes (affects) the thing.

P affects A.

Instrument

The process occurs if the entity exists (in some state). The process requires the entity.

Consumption

The process occurs if the entity exists (in some state). The process consumes the entity.

Effect

The process occurs if the thing exists. The process changes (affects) the thing.

P occurs if A exists. P requires A.

c

P occurs if A exists. P consumes A.

c

c

P occurs if A exists. P affects A.

Logical Relations

XOR relation

E.g., P affects either A or B.

OR relation

E.g., P affects A or B.


172 Reinhartz-Berger & Dori

Table 6. Event links: their semantics and symbols

Event Type Agent

Semantics

OPD Symbol

The process is triggered by the intelligent object.

State Change

The process is triggered when the object enters or exits the state. The object may be changed.

General Event

The process is triggered when the object or process is changed or cause external stimuli. The object may be consumed or changed.

A handles P.

Enter: Exit: Switch: Any:

e

,

Invocation

The process is triggered when the source process starts or ends.

Start: End: Border: Any:

Minimal or Maximal State Timeout

The process is triggered when the object violates its minimal or maximal time constraints for staying at the state.

Min: Max: Extreme:

Minimal or Maximal Process Timeout

The process is triggered when the process violates its minimal or maximal execution time constraints.

Reaction Timeout

XOR relation OR relation

OPL Sentence

The process is triggered when the event link violates its minimal or maximal reaction time constraints.

Any:

Min: Max: Extreme: Any:

Min: Max: Extreme: Any:

e e

e e

e

, , , ,

A triggers P when it enters/exists/either enters or exists st. St A triggers P.

e e e

e

,

e

A triggers P.

P invokes P1 when it starts/ends/ either starts or ends. P invokes P1. A triggers P when st lasts less than Time/ more than Time/less than Time or more than Time. Timeout of st A triggers P. P1 triggers P when it lasts less than Time/ more than Time/ either less than Time or more than Time. Timeout of P1 triggers P. This link triggers P when its reaction time lasts less than Time/ more than Time/ either less than Time or more than Time. This link timeout triggers P. E.g., A triggers either P

or Q when it changes.

E.g., A triggers P or Q

when it changes.

Note: The OPL sentences in this table are for the event aspect of the link. For state change and general event links, an additional OPL sentence, which represents its procedural aspect, should be added.


A Reflective Meta-Model of Object-Process Methodology 173

Appendix C: Abstraction Order of Procedural Links Table 7. Abstraction order of procedural links

c

c

c

c

c

c

c

c

c

c

e

e

e

e

e

e

e

e

e

e

c

e

c

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e

e e

e

e

e

e

e

e

e e e e

e

e

e

e

e

e

e

e e

e e

e

e

e e

e

e

e

e e

e

e


174 Holten, Dreiling & Becker

Chapter VII

Ontology-Driven Method Engineering for Information Systems Development1 Roland Holten, University of Frankfurt, Germany Alexander Dreiling, Queensland University of Technology, Australia Jörg Becker, European Research Center for Information Systems, Germany

Abstract Information systems development has to deal with evolving technologies and changing environments. Therefore, the engineering of methods as the problem of creating suitable instruments for new situations is critical to information systems development. The failure of IS development projects shows that method engineering is an open field. The question is if and how research on ontology can contribute to overcome the current situation. We show, based on linguistic and philosophical findings, how ontology can be used as linchpin in method engineering. We found that the language critique approach of Kamlah and Lorenzen (1984) provides the means to


Ontology-Driven Method Engineering 175

create ontologies by linguistic actions and that ontologies are always related to language communities sharing the knowledge of using a common language in communication processes. We present an ontology-driven engineering method for information systems development. Our method helps to create required language constructs to handle new situations. The ontology-driven engineering method is demonstrated using an elaborate example case.

Introduction An ongoing discussion on the business value of IT (Hitt & Brynjolfsson, 1996; Im, Dow, & Grover, 2001; Mukhopadhyay, Kekre, & Kalathur, 1995; Subramani & Walden, 2001; Tam, 1998), the role of IT in creating competitive advantage (Johnston & Vitale, 1988), and the perception that IT has changed from a simple administrative support tool to the vital backbone of an organization (Henderson & Venkatraman, 1999; Li & Chen, 2001; Venkatraman, 1994) clearly indicate that the role and impact of IT in contemporary organizations has changed significantly. In order to cope with the increased pressure on IT (Mukhopadhyay et al., 1995) as a result of these developments, the implementation of business solutions needs, more than ever to be effective, that is to meet business requirements exactly. Moreover, it needs to be increasingly efficient, requiring shorter development cycles, increased quality, and lower development costs. However, even if information systems research and practice have reached an advanced stage, there are still serious concerns about the effectiveness and efficiency of IT projects. Keil states that a significant number of IT projects will ultimately escalate and fail, if they have not reached their objectives within predefined time restrictions and allocated resources (Keil, 1995). Empirical results from Keil, Mann, and Rai suggest that 30 to 40 percent of all IT projects are subject to project escalation (Keil et al., 2000). Even if, in this survey, not all projects that exhibited some degree of escalation failed, we can assume that the frequency of IT project failure rates will be not significantly below the statistical average escalation rate, because escalation is not the only reason for project failure. Obtaining an exact frequency of IT project escalation rates is very difficult. The Standish Group’s Extreme Chaos research report revealed that only 28 percent of several thousand software development projects were completed on budget, on time, and with all features and functions originally specified (Standish Group International, 2001). Twenty-three percent were never implemented or canceled before the development was completed, 49 percent were completed and operational, but over budget, behind schedule, and with fewer functions and Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


features than originally specified. Furthermore, the failure rate rises with the complexity of the project (Standish Group International, 2001). Reports from different IS domains suggest, that one of the reasons for the failure of IT projects is inadequate communication between business and IT people. Within the field of data warehousing, Wixom and Watson state that the involvement of management, as targeted users of the system, or management support both contribute to system quality and system success (Wixom & Watson, 2001). If non-IT personnel is to be involved in an IS development project, especially the specification phase is important, because it includes the creation of a catalogue of functions that a future system needs to be able to perform. From an IT perspective, a broad variety of methods, architectures, and applications aim at supporting the IS development process (Hirschheim, Klein, & Lyytinen, 1995). However, the high failure rate especially of complex IT projects, indicates that some so-called best practices for IT development are inadequate. There is a continually increasing need for methodical approaches that are sufficiently theoretically well-founded to handle complex IT projects (Jiang, Klein, & Discenza, 2001). By using approved methods, systems engineers gain several advantages, such as learning effects, through repeated tasks within IT projects, or time reduction corresponding with cost reduction resulting from learning effects (Standish Group International, 2001). On the other hand, technology is continuously evolving, demanding IT project management permanently to deal with new situations. To be in the position of being able to handle new situations requires languages to communicate about these situations. For example, in data warehouse environments a common language of IT and non-IT personnel is required to avoid misunderstandings while specifying the requirements of the system (Holten, 2003a). The methodical problem of creating new languages and methods is crucial for IT project success. In this chapter, we propose a method for engineering information systems featuring the development of an ontology as a core concept. Especially, we show its role during the process of creating a specification language for information systems. Our method integrates well-founded concepts such as meta-models, ontology and language abstraction to increase the probability of successful information systems projects. As an introduction to the topic, we first analyze research positions and provide an overview of related work on information modeling, meta-model-based methods and ontology. The next section gives an introduction to the fundamentals of method engineering to define our ontology-driven method for information systems development. Using our developed method, we present an elaborate example and create an ontology for the specification of information systems for the purpose of analysis, as well as a model of this ontology. In the next section, we introduce a modeling case that describes how managerial objectives can be



used to derive information models that can be easily transformed into data structures. Finally, a summary and future prospects will conclude the chapter.

Scientific Position and Related Work In this section, a survey of related work is presented. Contributions on the general topic of information systems and method engineering deal with information modeling, ontology and meta-modeling. First, our scientific position is clarified.

Ontological and Epistemological Position Our method as well as our research is based on certain assumptions concerning two fundamental questions: 1) Does a world exist (ontological question)? 2) Is objective recognition possible (epistemological question)? The answers to these questions mark the researcher’s position and limit the interpretability of scientific work. Our position has been described in detail elsewhere (Niehaves, Ribbert, Dreiling, & Holten, 2004; Ribbert, Niehaves, Dreiling, & Holten, 2004). As basis for our argument we first repeat our main assumptions and sketch our position. 1.

Existence of a real world. The first epistemological assumption of our research approach is the existence of a real world beyond the realms of pure imagination of a subject. We presume a world that exists independently of cognition, thought and speech processes. Thus, we assume the position of (ontological) realism. Hereby, we refer to ontology as the analysis or the theory of “what is” and “how it is” (von Foerster, 1996; Bunge, 1977).

2.

Subjective cognition. The second epistemological assumption of our research is concerned with the question whether things in the real world can at least in principle be recognized as objective. This is neglected by constructivists who assume that cognition is subjective (or “private”). The relationship of cognition to the object of cognition is thus determined clearly by the subject.

Taking into account the first epistemological assumption “existence of an objective world” and the second epistemological assumption “subjective cognition” we consider our research approach as belonging to the position of interpretivism (cp. Figure 1) which is to separate from the positivist and radical Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


constructivist positions (Weber, 2004; Dubé & Paré, 2003). Due to the high level of subjectivity in the process of cognition, the process of cognition prevails as the (re)construction of reality through (predominantly linguistic) action. The quality of such a (re)construction is determined by the extent to which it can be aligned with the individual’s own immediate cognition. The positivists position is popular, for example, in IS case research (Dubé & Paré, 2003).

Information Modeling The terms software engineering and systems engineering emphasize engineering-related methods featuring a strong theoretical foundation for developing information systems. Information models can be used as a basis for systems engineering (Kottemann & Konsynski, 1984; Karimi, 1988). In order to develop high quality IT solutions, business requirements need to be identified and modeled from a business perspective. After having defined the business requirements, an information system that can subsequently be implemented must be specified. The Object Management Group (OMG) addresses the problem of information system engineering by proposing the so-called Model Driven Architecture (MDA) (Soley & the OMG Staff Strategy Group, 2000). Various modeling techniques are used to develop vendor- and middleware-neutral information models. In a second step, these information models are used to design middleware concepts. After selecting a language, the implementation of information systems based on the middleware design, can be initiated. The Architecture of Integrated Information Systems (ARIS) presented by Scheer (2000), is another approach for specifying information systems. The four

Figure 1. Ontological and epistemological positions, cp. (Niehaves et al., 2004; Ribbert et al., 2004) Epistemological position with respect to the relationship of cognition and the object of cognition

Ontological position

Objective cognition is impossible.

Objective cognition is possible.

A real world is existent.

Interpretivism

Positivism

No real world exists.

Radical Constructivism



different perspectives of data, functions, organization, and control, each consisting of three layers of conceptual model, technical model, and implementation, can be used to model different aspects of a software system from a business perspective as well as from an IT perspective. All these models correspond with one another. Language constructs from one perspective, can be integrated into models with a different perspective, which ensures that the information models are highly integrated. Furthermore, some perspectives support a certain degree of automation. The data model can be transformed automatically into a relational model from which a physical data warehouse schema can be derived. A shared domain knowledge between business and IT executives, positively influences an improved alignment of business and IT objectives and thus enhances the quality of IT solutions (Reich & Benbasat, 2000). If the business and IT staff can work collaboratively on IT specifications using the same information modeling method, it is a reasonable assumption that the requirements engineering of information systems can be simplified. Information models defined from a business perspective, can be used to specify information systems from a more technical perspective (Holten, 2003b, 2003a; Becker, Dreiling, Holten, & Ribbert, 2003).

Meta-Model-Based Methods Meta-modeling is a popular approach for analyzing information system methods. Based on models related to real-world objects, meta-models are used to specify modeling languages (Nissen, Jeusfeld, Jarke, Zemanek, & Huber, 1996; Holten, 2000; Strahringer, 1996). Recently, Rosemann and Green presented (2002) a meta-model of the Bunge-Wand-Weber ontology. Meta-models have also been developed within the field of decision support systems (van Hee, Somers, & Voorhoeve, 1991). In the approach by van Hee et al., both the meta-model and language developed, are specified by formal expressions. Modeling techniques using user appropriate concepts and representations, simplify the modeling process and thus help to align further business and IT objectives (Reich & Benbasat, 2000). Based on a thorough analysis of concepts relevant to and used by management, the MetaMIS approach consistently integrates a meta-model and a formal graphic representation, to support the specification of managerial views in information warehouse projects (Becker et al., 2003; Holten, 2003b, 2003a; Holten et al., 2002).



Ontology in the Information Systems Field Information modeling as well as meta-modeling are instruments in the IS domain. The analysis of a modeling language and a metalanguage leads to the concept of an ontology. Campbell and Shapiro (1995) explain that an ontology “consists of a representational vocabulary with precise definitions of the meanings of the terms of this vocabulary plus a set of formal axioms that constrain the interpretation and well-formed use of these terms.” Rosemann and Green (1999) state that “ontology is a well-established domain within philosophy dealing with models of reality” (p. 41). In accordance with our scientific position (cp. Figure 1) an ontology structures the object of the ontological question (“What is reality?”) and defines elements of it and relationships between these elements (Uschold, King, Moralee, & Zorgios, 1998). This procedure can be observed in several ontology construction approaches (Bunge, 1977; Green & Rosemann, 2000a; Uschold et al., 1998; Wand & Weber, 1990b). According to Bunge (1977, p. 21), an ontology cannot be tested empirically, because it has a constitutional character. Thus, it is an interpretivist or constructivist approach referring our framework of epistemological positions (cp. Figure 1). From a cognitive perspective, the construction of an ontology could be supported by the repertory grid technique introduced by Kelly (1955), which can be used to visualize cognitive knowledge. Tan and Hunter (2002) stress the relevance of the repertory grid technique for information systems research, describing several approaches where the technique has been applied. They explain further that cognitive knowledge consists of constructs of realworld elements that are connected by certain links. Thus, the repertory grid technique could be used to construct domain elements and their explanation, which could then be labeled unequivocally to define an ontology. The analysis of ontology is popular in the field of IS modeling and development. Based on the work of Bunge (1977), Wand and Weber (1989, 1990a, 1990b, 1993, 1995) introduced the ontological approach to the field of information systems development. The authors defined “a set of core concepts that can be used to describe the structure and behavior of an information system” (Wand & Weber, 1990b, p.1282). The aim of this systematic definition approach, was to structure the field of information systems and to “better understand the static and dynamic properties of information systems” (p. 1282). Recently, Wand and Weber’s work has been applied to the analysis of information system method engineering. Green and Rosemann present an analysis of the ARIS approach using the ontological framework of Wand and Weber (Green & Rosemann, 2000a, 2000b; Rosemann & Green 1999, 2002). Since formal and domain specific languages are the core of IS modeling and development in evolving environments, their creation and maintenance is a methodological



problem. In the next section we show how linguistic research can help to deal with situations where new concepts are required. Especially the language perspective of Kamlah’s and Lorenzen’s language critique approach can be used to create an ontology. According to this approach language to communicate about an IS domain, needs to be reconstructed critically using linguistic actions. The task of creating an ontology is then to be understood as creating a common understanding of a system of symbols. This approach and required background are discussed in the next section.

Fundamentals of Method Engineering Whereas a model describes a real-world object, a meta-model is usually referred to as a model of a language that describes real-world objects (Holten, 2000; Nissen et al., 1996; Strahringer, 1996). Thus, model and meta-model are related to the same real world object. This kind of meta-model is called a language based meta-model (Strahringer, 1996). Related to the real-world object which is to be modeled, the meta-model is defined in the metalanguage (Guarino & Welty, 2002; Holten, 2000). Holten (2000) depicted the interdependencies of the meta level, type level, and instance level on three layers. He stated that a model M1 of a real-world object is described in a language L1 which itself is described in a model M2 (meta-model of the real world object). These relationships are assigned at the language abstraction levels shown in Figure 2. In Figure 2, model M1 relies on language L1 in the sense that model M1 cannot be understood without knowing language L1. Referring our interpretivist position (cp. Figure 1) language L1 is the basis for recognizing real world’s aspects and arranging them in model M1. From a methodical point of view, a core question is how a language as L1 can be created and maintained to handle new situations.

Languages as Matter of Linguistics In order to understand how languages as sets of symbols are created and a common understanding of symbols can be established, the work on language critique, a branch of constructive philosophy known as the “Erlangen School”, of Wilhelm Kamlah and Paul Lorenzen, provides useful insight (Kamlah & Lorenzen, 1984; Lorenzen, 1987). This approach is related to other research on the nature languages. Languages as matter of subject are analyzed in linguistics. One of the early researchers involved in forming the modern view on languages was Ferdinand de Saussure. He conceptualized a linguistic sign as a union of a concept, or alternatively the signified (signifié) and a sound image, or Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Figure 2. Arrangement of models and meta-models on language abstraction levels (Holten, 2000, p. 142)

M2

represented in

L2

meta level language based meta model of

M1

model of

represented in

L1

type level model of

part of the real or perceived world

instance level

alternatively the signifier (significant) (de Saussure, 1974, p. 66). De Saussure (1974, p. 67) states that a linguistic sign has two basic characteristics represented by two principles. Principle I describes that the combination of concept and sound image is arbitrary. Thus, a language consisting of linguistic signs is based on conventions. Principle II focuses on the linearity of auditory sound images (auditory signifier). The auditory signifier is fugacious. Each element is represented successively. If auditory signifiers are written down, the line by which they are to be interpreted substitutes the time dimension of auditory sound signifiers (p. 70). In this case the signifiers are still represented linearly. After de Saussure, one of the most influential research projects in linguistics was conducted by Charles Morris. According to Morris (1974, p. 24), a language consists of a set of interrelated signs. In contrast to de Saussure’s sign, Morris’ sign only addresses what de Saussure calls the signifier. In order to clearly distinguish de Saussure’s sign from Morris sign we will denote Morris’ sign as symbol. The basis of this understanding of symbols goes back to the ancient Greeks. As a science semiotic states facts about symbols and is divided into three subordinate branches, syntactics, semantics, and pragmatics (Morris, 1971, p.



23). As a more commonly-used term, syntactics, which we will further refer to as syntax, deals with the relations of symbols to one another (p. 28). Language communities need such syntactical conventions in order to create a common understanding of interrelated symbols. Semantics deals with the relation of symbols to concepts or objects which they may or do donate (p. 35). Such conventions are necessary within a language in order to address one real-world object with the same symbol. Finally, pragmatics deals with the relation of symbols to their interpreters (p. 43). It addresses the understanding of symbols to users. Both introduced conceptualizations are based on conventions. However, de Saussure did not explicitly address the problem as to how the relationship between a signifier and a signified were bound, thus creating a sign. Morris’ semantics and syntax are based on conventions as well, because a community needs a shared understanding of symbols and their relationships as a prerequisite for meaningful communication. Pragmatics, however, focus on the relation of symbols and their interpreters. It deals with an individual who understands a symbol, which has a meaning to a group of people the same way these people understand it. In order to analyze how a common understanding of symbols is created, Language Critique, known as the “Erlangen School”, of Wilhelm Kamlah and Paul Lorenzen, is helpful. Kamlah and Lorenzen (1984, p. 33)show that language is used to disclose the world and is thus a proper instrument in relation to our interpretivist position (cp. Figure 1). The language critique approach integrates two linguistic abstractions: first, the abstraction from discourse to language as a system of signs, and second the abstraction from sign to concept (cp. Figure 3). The first abstraction separates language and discourse in the sense of schema versus linguistic action. By this, Kamlah and Lorenzen provide a means of separating signs (founding a language and thus a schema) from their linguistic usage (discourse) (p.44). Discourse means the repeatedly actualized usage of signs in changing combination and variation. Thus, discourse is an actualized activity, whereas language comprises potential activities, defined as activityschema (p. 45). The second abstraction in the language critique approach deals with the relation of sign and meaning. De Saussure, Morris, and Kamlah and Lorenzen divide between a concept and its representation, which is called sound pattern by Kamlah and Lorenzen (1984, p. 72). Given a term, concept is the meaning of this term. A concept “is at first no more than a term; however we abstract from the arbitrary sound-pattern of a term when we call it a ‘concept’” (p. 72). On the other hand, if statements are made about signs that are invariant with respect to the changing meaning of these signs, these statements deal with the soundpattern. So, to reach the sound-pattern of a sign, disregarding the meaning as an abstraction is required (p. 73). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Figure 3. Agreements and abstractions in the language critique approach (Holten, 2003a) 1. agreement discourse

1. abstraction from accidental properties of activities

actualized activity, spoken

term, word as schema

2. agreement 2. abstraction from arbitrary soundpattern

sound pattern, sign

meaning, intension

concept, state-ofaffairs

The separation of concept and representation is common to all linguistic approaches mentioned. Whereas de Saussure does not address how combinations of concept and representation are established, Morris addresses this problem explicitly with the pragmatics dimension of symbols. In both conceptualizations, the combination is arbitrary. Using Morris’ terminology symbols have syntax, semantics and pragmatics. But how does this happen and how are these dimensions related? Where do the conventions making syntax, semantics and pragmatics of these symbols come from? Seen through the eyes of Kamlah and Lorenzen, these questions can be answered using the construct language community: A new term is introduced by explicit agreement with respect to its usage and meaning (Kamlah & Lorenzen, 1984, p. 57). This agreement leads to a relation of concept and term and is shared by a language community as the knowledge of using this term (p. 45). In the words of Kamlah and Lorenzen, Since discourse as actualized activity pursues the particular end of mutual understanding, we may say of language … that as a system of signs it promotes mutual understanding. For this very reason it is, in a unique way, a ‘know-how’ held in common, the possession of a “language community”. (p. 47)

An Example on Language Critique The following example is intended to demonstrate the importance of Kamlah’s and Lorenzen’s concepts, language critical re-construction and language community, for information systems modeling problems. Morris explicitly divides semantics and pragmatics as the general meaning of one symbol and the meaning of a symbol to a user. If, for instance, hieroglyphs of ancient cultures are found, the symbols would have a linguistic meaning nobody understands any Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


longer. However, scientists may intensively research the hieroglyphs and relearn the language by relearning what the symbols mean (or what they think they mean). Using Morris’ terminology hieroglyphs as symbols then would have syntax, semantics, and pragmatics, again. But how does this happen? Where do the conventions making syntax, semantics, and pragmatics of these hieroglyphs come from? Seen through the eyes of Kamlah and Lorenzen, these questions can be answered using the concepts language critical re-construction and language community: Semantics and pragmatics are directly linked to each other in Kamlah’s and Lorenzen’s approach, which means that the ancient language symbols have no meaning once the last ancient language user has died. The scientists researching the hieroglyphs can develop an understanding of these symbols, which creates a new language community understanding these symbols in same way. It is possible that the ancient language users belonged to the same language community. However, this may not be the case if they understood them differently. The approach implies that there is no clear existence of languages without anyone currently using this language. If members of a group of people communicate and each has an aligned semantic and pragmatic dimension of a symbol in mind, then this group of people forms a language community.

Language Critique Summary and Implications De Saussure, Morris, and Kamlah and Lorenzen divide between a concept and its representation. Whereas de Saussure does not address how combinations of both are established, Morris addresses this problem explicitly with the pragmatics dimension of symbols. In both conceptualizations, the combination is arbitrary. With Kamlah and Lorenzen’s language critique approach, the combination of a signified and a signifier from de Saussure or semantics and syntax from Morris can be created deliberately. In Kamlah’s and Lorenzen’s world there are no languages without users. Language communities have to be created by introducing symbols and explaining them. The implications for our work are that the semantic and pragmatic dimensions of symbols need to be introduced together. In order to create a language for the specification of information systems, language constructs need to be introduced and explained. By studying this work, the reader will become a member of the language community sharing the language that will subsequently be introduced. If a language community has been created based on a critical language (re)construction of a domain, the members of this language community share the pragmatic dimension of a symbol. All members have the same concept in mind if they are confronted with a symbol of the language and vice versa. In turn, nonmembers of the language community do not understand the language symbols or Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


they understand them differently. In order to become a member of a language community, an individual must align the pragmatic dimension of language signs with that of the language community. Once inside a language community, the border between pragmatics and semantics of a symbol disappear for an individual. The symbol becomes a sign for de Saussure, where all individuals relate a concept to the sign and vice versa. In summary, linguistic actions used to establish language communities are the instruments to create ontologies and — with respect to our interpretivist position (cp. Figure 1) — ontologies are related to language communities.

Operationalization: Types of Linguistic Actions for Information Systems Engineering Wedekind (1981) defines a set of construction operations (linguistic actions) to create systems of concepts relevant for the development of database systems derived from the work of Kamlah and Lorenzen. These linguistic actions are called subsumption, subordination, and composition. They are defined as follows (Holten, 1999; Wedekind, 1981):

•

Subsumption. A concept is created by statements. By subsumption, object types are created in the sense of an instance-of relation. An object type defines a set of objects. Concepts created by subsumption are modeled with the entity type symbol.

•

Subordination. A set of concepts is subordinated to a higher concept by statements. By subordination, is-a relations are defined between object types. Is-a relations are modeled with a triangle.

•

Composition. Two (or more) concepts are related by statements. By composition relationships, types are created. Concepts created by compositions are modeled with relationship type symbols and cardinalities in minmax notation. Cardinalities define the complexity of relationship types. For any concept used to define the meaning of the composition, the complexity of minimum and maximum values of the elements is given as zero, one, or many. If composed concepts are required to compose further concepts, this is modeled by surrounding the respective relationship type symbol by an entity type symbol.

Using these linguistic actions as an example, we are able to define a system of concepts relevant for the investigated domain. Ontology is clearly a synonym for system of concepts specified based on these construction operations. The implications on the method engineering approach proposed here are as follows: Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


In order to approach the field of information systems development methodically, we first need to define an ontology. As already stated, an ontology is a constructivist or interpretivist approach to creating terms and explaining the concepts relating to an information systems domain. The ontology will be defined by critically reconstructing the language of the considered domain with Wedekind’s linguistic actions. This will create signs in de Saussure’s understanding. The necessary pragmatic alignment of the symbols created, will be achieved by common language explanations of the introduced symbols. Syntactic conventions for the introduced symbols will be defined partially by creating an entityrelationship model. This model, combined with explanations, can be seen as ontological foundation. It enables the user to avoid misunderstandings with a structured domain description. Based on the ontological foundation, we are able to develop a consistent representation language for the investigated domain. These three steps to approaching information systems development methodically will be demonstrated in a case in the next section. Resulting from the linguistic analysis of languages, Figure 2 needs to be extended. The modeling language L1 consists of two main parts which are symbols that represent something for users of this language on the one hand and rules how these symbols are interrelated on the other hand. From the work of Kamlah and Lorenzen, it is evident that semantics and pragmatics are closely related to each other. This part of the language can be modeled in a terminological model. Syntactic alignment of the use of language symbols in this approach is as necessary as semantic and pragmatic alignment, in order to create language communities. However, this part of the language can be formalized in a syntactical model that is different from the terminological model. Both are metamodels related to the relevant aspect of the real or perceived world. They are modeled in two modeling languages L2 and L3, for which the same differentiation (semantics/pragmatics and syntax) can be applied as for L1. L2 may be different from L3. Nevertheless, this differentiation is not necessary for our further work and will therefore not be examined. Figure 4 depicts the arrangement of models and meta-models at language abstraction levels using the linguistic foundations from this paragraph.

Creation of an Ontology Using Wedekind’s three introduced linguistic actions subsumption, subordination, and composition we will now (re-)construct a language to make statements about the domain of management information systems. We introduce language constructs for specifying information spaces and aspects within information



Figure 4. Arrangement of models and meta-models at language abstraction levels considering linguistic foundations

M2 syntactical part of the model

L3

terminological part of the model

L2

meta level language based meta model of

model of

model of

syntax M1

L1

semantics / pragmatics (Ontology)

type level model of

part of the real or perceived world

instance level

spaces. Furthermore, business objectives will be added to the language to assist in providing a specification of MIS with the introduced language.

Definition of Information Spaces The first base concept of the modeling approach is Dimension. Dimensions are necessary to span information spaces. From a managerial point of view, they are orthogonal. There are mandatory dimensions, because every managerial view must have a temporal reference, or, for example, a reference to an optimistic or pessimistic planning scenario of the enterprise. Additionally, dimensions used to define a managerial view have to be explicitly compatible. Dimensions are represented by (red) rectangles. In order to integrate hierarchical views on identical leaf objects consistently, the concept Dimension Grouping is introduced. Dimensions defining views on identical leaf objects are subsumed in the same dimension grouping. Dimensions belonging to the same dimension grouping have the same set of leaves. For example, in the retailing business different aspects of stores are relevant. Stores Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


are classified according to competitive environments, sites (e.g. downtown, suburb or village) or appearance (age and degree of modernization). All attributes relate to the same set of business objects (set of stores). Since it is meaningful to combine these aspects with one another, different dimensions are introduced. These dimensions are integrated in one dimension grouping (“Store”), because their leaves are identical. Separating dimensions and dimension groupings avoids parallel branches within dimensions. Dimensions with parallel branches (networks) found in the literature (Blaschka, Sapia, & Dinter, 1998; Bulos, 1996; Golfarelli, Maio, & Rizzi, 1998; Golfarelli & Rizzi, 1998, 1999; Sapia, Blaschka, & Hofling, 2000; Lechtenbörger, 2001) correspond in part to the concept of dimension grouping introduced here. Nevertheless, dimensions within dimension groupings are connected by leaves only, leading to strict hierarchies (see Example 1). Dimensions consist of dimension objects. Referring to Riebel’s enterprise theory, management’s arrangements and examinations deal with dimension objects (Riebel, 1979; Holten, 1999). Riebel’s (1992) theory defines management decisions as vital elements. Each activity is produced and maintained by decisions. Therefore, decisions are the only sources of cost, outcome and liquidity. Following Riebel, the language concept Dimension Object is introduced. Dimension objects are organized in hierarchies (concept DO-Hierarchy) as part of a dimension’s definition. The concept of DO-Hierarchy enables the construction of, for example, product hierarchies or regional hierarchies. Each dimension object is associated with an unequivocal hierarchy level (concept Hierarchy Level). Dimension objects at the lowest hierarchical level are called Leaves, all other dimension objects are called Non-Leaves. Referring to abstraction and aggregation, dimension objects at the same hierarchical level must build homogeneous sets (Böhnlein, 2001). Dimension object hierarchies are represented by hierarchical structures. Squares represent hierarchical levels of non-leaves. Leaf objects have no square as Example 1. Conceptual Language Aspect Dimension

Dimension Grouping

D-DG-As (Dimension Dimension Grouping Association)

Linguistic action and statement Meta Model Component Object Language Symbols

Subsumption: Used to create and organize the space of which a managerial view is composed. Subsumption: A specific object type for which different dimensions can be used to characterize the aspects relevant for management. Composition: Relationship between concepts Dimension and DimensionGrouping. A certain dimension belongs to one unequivocal dimension grouping (cardinalities (1,1)). A certain dimension grouping comprises at least one dimension, but may comprise many dimensions (cardinalities (1,n)).

Dimension Object

Dimension Grouping

Dimension Grouping

"Dimension Grouping"

(1,n)

"Dimension" D-DG-As

Dimension

(1,1)



prefix. Every dimension object has an identifier. Hierarchy level identifiers are related to levels by dotted lines. Indentations are used to increase the clarity of the diagram. Lower level objects are placed to the right of higher-level objects. Squares labeled with “+” indicate subordinate dimension objects not shown to improve clarity. In the case of squares labeled with “-”, all dimension objects of the succeeding hierarchy level are visible (see Example 2). To prevent information overflow, managerial users need segments from dimension hierarchies. Task specific views can be combined, based on dimension hierarchy segments. For this purpose the concepts Dimension Scope and Dimension Scope Combination are introduced. Dimension scopes are sub trees of dimensions. Dimension scopes are represented by (white) rectangles with (red) triangles inside. The combination of dimension scopes, defines a space of multi-dimensional objects relevant for managerial users. This space is also called navigation space (see Example 3). Referring to Riebel’s enterprise theory the concept Reference Object denotes vector types within a navigation space spanned by dimension scopes. Reference objects are “measures, processes and states of affairs which can be subject to arrangements or examinations on their own” (Riebel, 1979, p. 869) (see Example 4).

Aspects Within Information Spaces The modeling constructs introduced so far, allow for constructing navigation spaces and creating reference objects within these information spaces. In the next steps, we introduce constructs that enable us to fill this navigation space with information. Generally, every piece of information within an information space is seen as an aspect (concept Aspect). It comprises axiomatically defined aspects (concept Basic Aspect) or calculated aspects (concept Calculated Aspect). Another specialization of Aspect comprises qualitative (concept Qualitative Aspect) and quantitative aspects (concept Quantitative Aspect). For qualitative aspects a definition of value elements is mandatory (concept Value Element). Defined value elements are assigned to qualitative aspects. For instance, the qualitative aspect “project status” can have either one of the values “started”, “pending”, or “finished”. Aspects can be grouped within aspect systems (concept Aspect System), to allow for analyses that require the use of more than one aspect. In a balanced scorecard (Kaplan & Norton, 1992, 1993, 1996) scenario, different performance measures are presented as comprehensively grouped together so as to evaluate business developments holistically. Aspects within aspect systems can be arranged hierarchically (concept Aspect-Aspect System-Association Hierarchy). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Example 2. Conceptual Language Aspect Dimension Object

Leaf, Non-Leaf

D-DO-As (Dimension Dimension Object Association)

DO-Hierarchy (Dimension Object Hierarchy)

Hierarchy Level

D-HL-As (Dimension Hierarchy Level Association)

D-HL-Sequence (Dimension Hierarchy Level Association Sequence) DO-DHL-As (Dimension Object Dimension Hierarchy Level Association Association)


Subsumption: Entities relevant for managerial investigations or analysis; part of the definition of dimensions in the sense that they have strong relationships to each other, from a managerial point of view. Subsumption and Subordination: The concept Dimension Object is unequivocally and totally (symbols u, t) specialized in the concepts Leaf and NonLeaf. Leaves are at the lowest level of the dimension hierarchies. Non-Leaves are at all other levels. The set of leaves is the same for all dimensions which belong to the same dimension grouping. Composition: Relationship between concepts Dimension and Dimension Object. A dimension requires a (possibly empty) set of dimension objects for its definition (cardinalities (0,n)) and any dimension object requires a relationship to at least one dimension (cardinalities (1,n)). Leaves are related to all dimensions of a dimension grouping. All other dimension objects (non-leaves) are related to exactly one dimension. Composition: Recursive relationship from concept Dimension Object to itself. For dimension objects a hierarchical order is required. Any dimension object may have zero or one superior imension object (cardinalities (0,1)) and zero or many subordinated ones (cardinalities (0,n)). Subsumption: Dimensions consist of hierarchical levels. Dimension objects are necessarily assigned to these levels within one dimension. Composition: Relation between concepts Dimension and Hierarchy-Level. Any Dimension is composed of one or many hierarchical levels (cardinalities (1,n)); a hierarchical level as an abstract object can be related to one or many dimensions (cardinalities (1,n)). Composition: There is an unequivocal order of hierarchy levels associated to a dimension. Each hierarchical level of a dimension has zero or one predecessor and zero or one successor. (cardinalities (0,1) on either side). Composition: Relationship between the concepts Dimension-Object and D-HLAs. Each dimension object must unequivocally be associated to one hierarchy level of the dimension to which it belongs (cardinalities (1,1)) and each hierarchy level of a dimension must contain at least one or many dimension objects (cardinalities (1,n)).

"Dimension Object"

Dimension Object

"Non-Leaf" Dimension Object

"Non-Leaf" Leaf

"Leaf"

D,T

Non-Leaf

Dimension

"Dimension"

(0,n)

"Non-Leaf" D-DO-As

Dimension Object

"Leaf"

(1,n)

"Non-Leaf" DO-Hierarchy (0,1) Dimension Object

"Non-Leaf" "Leaf"

(0,n)

"Hierarchy Level" Hierarchy Level

Dimension

"Hierarchy Level"

(1,n)

"Hierarchy Level" "Hierarchy Level"

D-HL-As

Hierarchy Level

"Dimension"

(1,n)

"Non-Leaf" "Non-Leaf" D-HLSequence (0,1) D-HL-As

"Leaf" "Leaf"

(0,1)

"Non-Leaf" "Non-Leaf" "Non-Leaf"

Dimension Object

(1,1) DODHLAs

D-HL-As

(1,n)

A reference object from a defined navigation space can be assigned to an aspect. This leads to a fact (concept Fact). Facts make statements about the referenced objects. They may state the turnover achieved by a business unit, the performance of an organizational member, or the status of a current business process. Each of these facts necessarily comprises a reference object (business unit, Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Example 3. Conceptual Language Aspect Dimension Scope

DO-DS-As (Dimension Object Dimension Scope Association)

Dimension-ScopeCombination DS-DSC-As (Dimension Scope Dimension ScopeCombination Association)


Subsumption: Used to define scopes of dimensions relevant for a managerial view. Composition: Relationship between concepts Dimension-Object and Dimension-Scope. Any dimension object may or may not be member of a dimension scope (cardinalities (0,n)). Any dimension scope is composed of one or more dimension objects (cardinalities (1,n)). Subsumption: Used to identify combinations of dimension scopes while defining managerial views. Composition: Relationship between concepts Dimension-Scope and Dimension-Scope-Combination. Any dimension scope combination may contain one or many dimension scopes (cardinalities (1,n)) whereas any dimension scope can be a member of zero or many dimension scope combinations (cardinalities (0,n)).

Dimension Scope

Dimension Object

"Dimension Scope"

(0,n)

"Non-Leaf" "Non-Leaf"

DO-DS-As

Leaf Dimension Scope

(1,n)

Dimension Scope Combination

Dimension Scope

"Dimension Scope Combination"

(0,n)

"Dimension Scope" "Dimension Scope"

DS-DSC-As

Dimension Scope (1,n) Combination

Example 4. Conceptual Language Aspect Reference Object

Combined Reference Object C-RO-Coordinates (Combined Reference Object Coordinates)

Reference Object, Combined ReferenceObject, DimensionObject

RO-Structure (Reference Object Structure)

Linguistic action and statement Meta Model Component Subsumption: Reference objects are defined by RIEBEL as all “measures, processes and states of affairs which can be subject to arrangements or examinations on their own” (Riebel (1979), p. 869). Subsumption and Subordination: A combined reference object is a reference object interpreted as a vector. Composition: Relationship between concepts Combined-Reference-Object and Dimension-Object. Dimension objects are used as coordinates to specify combined reference objects. Any dimension object can be used as a coordinate for one or many combined reference objects (cardinalities (1,n)) and any combined reference object has one or many coordinates (cardinalities (1,n)). Subordination: A reference object is a vector and then specialized as a combined reference object. Additionally, a reference object can have the role of a dimension object. In this case, it is used to define dimensions and as coordinates for combined reference objects. Nevertheless, any dimension object is a reference object. The specialization of reference objects is thus not unequivocal (symbol n), but total (symbol t). Composition: Recursive relationship from concept the Reference Object to itself. Logically, this relationship defines the space of all reference objects of which managerial views can be composed. Any reference object may have zero or many higher reference objects (cardinalities (0,n)) and zero or many subordinate ones (cardinalities (0,n)).

Reference Object

Combined Reference Object

Reference Object

Combined (1,n) Reference Object C-ROCoordinates

Dimension Object

(1,n)

Reference Object

n,t

Combined Reference Object

(1,n)

C-ROCoordinates

Dimension Object

(1,n)

RO-Structure (0,n) Reference Object

(0,n)



organizational member, business process) and an aspect (turnover, performance, status). Facts and aspects are generalized to the construct Operand. As soon as they are seen as operands, calculations based on these operands may be performed. These calculations are defined by means of calculation expressions, which can generally be distinguished by their purpose. If they are defined for a set of reference objects, the calculation is an aggregation. Aggregations play an important role in so-called drilling operations. For quantitative aspects such as a turnover, aggregations can be made over a set of products groups, business units or shops. The turnover of product groups can be drilled down to the turnover of products within the same period of time, requiring exactly the same calculation expression, applied to a different set of reference objects. Another example relates to warehouse stocks that may require the aggregation to average stocks. Drilled down to units of the warehouse again, the calculation expression stays untouched. Only the reference objects change. If calculation expressions are defined for single reference objects, these calculations have no further implications. Profits are calculated by subtracting costs from turnovers. The implication of this calculation is that only one reference object is used (the profits of a company unit within a set time period are the turnover reduced by the costs of exactly this company unit in the set time period). The concept Calculation Expression is thus specialized into the concepts Calculation Expression Element and Calculation Expression Set. Finally an Information Object combines a dimension scope combination with an aspect system. This implies the creation of facts, if elements of the navigation space are selected (reference objects) and valued with an aspect of the assigned aspect system. If calculations of facts are necessary (for instance for plannedactual-analyses or variance analyses of profits within two time periods), fact calculations can be included in information objects (see Example 5).

Business Objectives to Derive Information Spaces and Create Planning Scenarios At this point, we are able to define navigation spaces filled with measures. These measures can be processed by various means, facilitating the definition of information requirements of organizational members. The remaining problem is that information requirements cannot usually be given by organizational members, because they are implicitly embedded in their objectives or strategies. A small extension to the above defined meta-model constructs can help to close this gap and transform business objectives into the modeling constructs introduced so far (Becker et al., 2003).



Example 5. Conceptual Language Aspect Aspect

Basis Aspect; Calculated Aspect

Qualitative Aspect

Value Element QA-VE-As (Qualitative Aspect Value Element Association)

Quantitative Aspects (Ratio)

Aspect; Qualitative Aspect; Quantitative Aspect (Ratio)

Fact

Calculation Expression; Calculation Expression Set; Calculation Expression Element

Operand CE-On-As (Calculation Expression Operand Association)


Subsumption: Aspects are attributes concerning the dynamics of reference objects. Subsumption and Subordination: The concept Aspect is unequivocally and totally (symbols u and t) specialized through the concepts Basis Aspect and Calculated Aspect. Basis aspects are defined by means of statements. The definition of calculated aspects additionally requires logical or algebraic expressions. Each aspect used to define a calculated aspect must be defined in advance. Subsumption: Qualitative aspects are attributes concerning the dynamics of reference objects which are not measured exactly. Each qualitative aspect comprises a set of discrete values characterizing states of processes. Subsumption: The concept Value Element comprises elements which set ranges for qualitative aspects. Composition: The set of value elements defining the range for a qualitative aspect is assigned explicitly to this aspect. Value elements can be part of zero or many sets (cardinalities (0,n)) and every qualitative aspect must comprise at least two different value elements (cardinalities (2,n)). Subsumption: Quantitative aspects (Ratios) are attributes concerning the dynamics of reference objects which are measured exactly. They quantify managerially relevant aspects such as the value of an enterprise, business performance and the financial situation. Subordination: The concept Aspect is unequivocally and totally (symbols u and t) specialized though the concepts Qualitative Aspect and Quantitative Aspect. The concept Quantitative Aspect is a synonym for the concept Ratio. Composition: Relationship between concepts Reference Object and Aspect. Any reference object can be combined with zero or many aspects and vice versa (cardinalities (0,n) on either side). Subsumption and Subordination: The concept Calculation Expression is unequivocally and totally (symbols u and t) specialized though the concepts Calculation Expression Set and Calculation Expression Element. The latter comprises every calculation expression specifying the value for one element of concept Reference Object. The concept Calculation Expression Set is required to logically or algebraically calculate values of a set of elements. Subsumption: Operands are part of logical and algebraic calculation expressions. Composition: Each calculation expression is related to at least one operand (cardinalities (1,n)) and each operand becomes part of zero or many calculation expressions (cardinalities (0,n)).

"Aspect" Aspect

Using Rows in Tables Aspect

u,t

Basic Aspect

Calculated Aspect

"Qualitative Aspect" Qualitative Aspect

"Value" Value Element

Value Element

Using Rows in Tables

(0,n)

QA-VE-As

Qualitative Aspect

(2,n)

"Ratio"

Ratio (Quantitative Aspect)

Using Tables Aspect

u,t

Qualitative Aspect

Ratio (Quantitative Aspect)

Reference Object

(0,n)

Fact

Aspect

(0,n)

Calculation Expression

u,t

Calculation Expression Element

Formal Expression; Column in Table

Calculation Expression Set

Operand

Calculation Expression

(1,n)

CE-On-As

Operand

(0,n)



Example 5. (Continued) Conceptual Language Aspect Aspect System

A-AS-As (Aspect-Aspect System Association)

Linguistic action and statement Meta Model Component Object Language Symbols Subsumption: An aspect system is a set of aspects which enables the analysis of business situations. Composition: Relationship between concepts Aspect and Aspect System. An aspect system is composed of one or many aspects (cardinalities (1,n)) and an aspect may be a member of zero or many aspect systems (cardinalities (0,n)).

A-AS-As-Hierarchy Composition: Recursive relationship (Aspect-Aspect System- from concept A-AS-As to itself. Aspects Association Hierarchy) within an aspect system are organized hierarchically. Each aspect may have zero or many higher aspects (cardinalities (0,1)) and zero or many subordinate ones (cardinalities (0,n)). Operand; Aspect; Fact; Subsumption and Subordination: The Fact Calculation Operand concept is unequivocally and totally (symbols u and t) specialized through the concepts Aspect and Fact. Quantitative aspects (ratios), qualitative aspects and facts can be involved in logical and algebraic calculation expressions. If facts are used as operands the calculation expression is called a fact calculation. The concept Fact Calculation is defined implicitly and does not become part of the meta model. Information Object Composition: Relationship between concepts Aspect System and Dimension Scope Combination. Set of facts relevant to a management user. One aspect system can be combined with none or many dimension scope combinations and vice versa (cardinalities (0,n) on either side).

1

Aspect System

67 1

Aspect

67

(0,n)

34

29

34

29

9

"Aspect System"

5 57

5 57

9

"Aspect" "Aspect"

A-AS-As

"Aspect" Aspect System

(1,n)

"Aspect"

"Superordinate Aspect"

A-AS-AsHierarchy

"Subordinate Aspect"

(0,1) A-AS-As

"Subordinate Aspect"

(0,n)

Operand

u,t

+

Fact

%

+ %

-

-

"Fact Calculation Expression"

Aspect

"Dimension Scope Combination" 1 67

34

29

5 57

9

"Aspect System"

"Fact Calculation" := Expression

Aspect System

(0,n)

Information Object

"Information Object“ "Dimension Scope Combination"

Dimension Scope (0,n) Combination 1

34

29

67

5 57

+ %

-

9

"Aspect System" "Fact Calculation"

The first necessary construct is Objective. According to Porter (1979), the most abstract and general business objectives are defined in a business strategy. A business strategy deals with defending and strengthening a competitive business position. A major difficulty of business strategies (concept Strategy, General Condition, and Guideline) is their non-operational character. Objectives need to be defined operationally in order to be manageable (Latham & Kinne, 1974). Operational objectives (concept Operational Objective) are defined by a certain measure (concept Objective Measure), level (concept Objective Level), reference (concept Reference Object), and time frame (part of concept Reference Object) (Adam, 1996). Objective levels combine one reference object with an objective measure. For instance, the “profit margin” (objective measure) of a business unit in 2005 (reference object) needs to be at least 10 percent (objective level). The objective level is assigned to an operational objective. None of these concepts are part of the object language. They are necessary when constructing models with the above introduced language, by providing indications as to how to construct dimensions (from objective refer-



ences) and aspects (from objective levels). Furthermore, these constructs allow for creating planning scenarios (Becker et al., 2003) (see Example 6).

Example Modeling Case Business Objectives After having introduced an ontology that consists of a (re-)constructed language to make statements about information spaces and aspects within information spaces we will now introduce a comprehensive modeling case. The starting point is the definition of business objectives which will be transformed into information Example 6. Conceptual Language Aspect Objective

Objective; Objective Structure

Objective; Strategy, General Condition and Guideline; Operational Objective

Objective Measure; Reference Object; Objective Level

Objective Measure; Quantitative Measure; Qualitative Measure

Operational Objective; Objective Level; OOOL-AS (Operational Objective, Objective Level Association)

Linguistic action and statement Meta Model Component Subsumption: Objectives are planned achievements of future time periods. They may have the more general character of a business strategy or the detailed character of a sales plan for a forthcoming month. Composition: Objectives can be arranged hierarchically, placing more general objectives above detailed objectives. Detailed objectives can help to achieve general objectives, if the objective system is free of conflict. Subsumption and Subordination: Objective is specialized into the concepts Strategy, General Condition and Guideline (general, non-operational objectives) and Operational Objective (manageable objectives). Manageable objectives give more hints about the construction of navigation spaces and measures within these navigation spaces. Composition: As each operational objective must comprise a reference object and an objective measure, the concept Objective Measure is assigned to the concept Reference Object. This composition constitutes an Objective Level and can be seen as equivalent to a fact. Subsumption and Subordination: Similar to aspects, an objective measure can be defined qualitatively or quantitatively. Both types of objective measures will later be transformed into a qualitative or quantitative aspect as a counterpart for specifying information requirements. Composition: a defined objective level ultimately will be assigned to the operational objective from which it has been derived. Only operational objectives feature objective levels. Business strategies may contain references, time frames, or measures, but they do not necessarily have to, thus they do not necessarily have an objective level.

Objective

Objective Structure (0,n) Objective

(0,n)

Objective

u,t

Strategy, General Condition, and Guideline

Operational Objective

Objective Measure

(0,n)

Objective Level

Reference Object

(0,n)

Objective Measure

u,t

Quantitative Measure

Qualitative Measure


(0,n)

Objective Level

(0,n)

OO-OL-AS



spaces and planning scenarios. We use a method which has been introduced by Becker et al. (2003). It first decomposes operational objectives into their defining parts. These defining parts will be transformed into information spaces in the second step. The derived information spaces will be filled with planning scenarios in a third step. Our sample company is part of a supply chain that decided to decrease delivery times, in order to increase customer satisfaction. Products are shipped directly from company warehouses on the basis of customer orders. The company stocks a small number of products and attempts mainly to produce just in time. Most supply chains can potentially achieve higher customer satisfaction by reducing the delivery times of ordered products. The additional effort required to decrease the delivery times, can be justified with the savings from decreased stocks along the supply chain. The savings can be passed on to the customer, invested in improving customer service or in strengthening the supply chain. To decrease delivery times of ordered products, the efficiency of operative business processes along the entire supply chain needs to be increased. A major managerial responsibility is to define business objectives and undertake the necessary steps to deploy improved business processes. Furthermore, control mechanisms need to be implemented to monitor the degree to which business objectives have been reached. The main objective of our example company focuses on profiting from the positive effects of information sharing along supply chains. It is consistent with the general goal of long-term profitability:

•

Objective Delivery Time Reduction of Business Unit Automotive Supplies: decrease the average delivery time of all products of business unit Automotive Supplies to a maximum of 24 hours within the next year.

The time frame for the main objective is next year. Furthermore, it refers to all products of business unit Automotive Supplies. The time frame combined with the reference constitutes the reference object. Average delivery time is a quantitative measure. If a temporal value (not a time frame) is assigned to the reference object, this value becomes a business fact. Since delivery time is composed of production time and shipment time, the objective is broken down into two sub-objectives. The first sub-objective has been set as follows:

•

Objective Increase Production Efficiency: increase production efficiency at assembly line V8 engine in factory alpha from level 8 to level 9 within the next year.



As in the case of the main objective, the time frame is next year. The reference of the objective is assembly line V8 engine in factory alpha. Efficiency is a qualitative measure, which can be expressed by the values (categories) 0 to 10. The efficiency categories can be calculated by algorithms, which consider various influencing variables or are derived by an auditing process, where trained personnel set the efficiency based on their observations. The second subobjective to decrease delivery times refers to the shipping efficiency: •

Objective Increase Shipping Efficiency: increase shipping efficiency of products shipped out of factory alpha by any logistic partner from level 8 to level 9 within the next year.

The time frame again is next year. It refers to factory alpha, any logistic partner, and any product and is measured by the qualitative measure efficiency. Both sub-objectives are measured qualitatively. In order to derive the efficiency measures for both sub-objectives deterministically, each is split up again into three sub-objectives. Production efficiency is described by the following objectives:

•

Objective Rejection Rate Reduction: decrease the average rejection rate of product group Original Equipment — Engines products at assembly line V8 engine in factory alpha from 0.4 to 0.2 percent within the next year, without increasing the rejection rate of other product group’s products assembled at this line,

•

Objective Machine Defect Rate: decrease average machine defect rate of machines at assembly line V8 engine in factory alpha from 0.7 class A defects per week to 0.3 within the next year,

•

Objective Lead Time Reduction: achieve an average lead time reduction during production of any single product of product group Original Equipment — Engines at assembly line V8 engine in factory alpha from 256 minutes to 240 minutes within the next year.

On the other hand, shipment efficiency is broken down into these three objectives:

•

Objective Decrease Just-In-Time Deviation of Logistic Partners: decrease the average just-in-time deviation of any logistic partner for any product shipped from factory alpha with an appropriate transportation to five minutes within the next year (Just-In-Time deviation is the time



difference between planned and actual collection of a customer order by a logistics partner),

•

Objective Decrease Packing Time: decrease the average packing time of factory alpha warehouse workers for any customer order to one hour within the next year,

•

Objective Reduce Warehousing Costs: reduce the total costs of factory alpha warehouse to • 500,000 within the next year.

These six objectives can each be decomposed into their defining components. Table 1 gives an overview of the entire objective system by decomposing each objective to objective reference, time frame, objective measure, and objective level. The defining components of decomposed operational objectives are structured according to the model constructs introduced in the last section. In order to monitor the degree to which an objective has been accomplished, operational objectives need to be transformed into planning scenarios. After the end of the planning period has been reached, deviation analyses help to compare these planning scenarios to the actual business development. The next section shows how objectives can be transformed into planning scenarios.

Table 1. Operational objective components Main Objective

Sub-Objective Level 1

Sub-Objective Level 2

Delivery Time Reduction of Business Unit Automotive Supplies Increase Production Efficiency Rejection Rate Reduction Machine Defect Rate Lead Time Reduction Increase Shipping Efficiency

Objective Reference

Time Frame

business unit automotive supplies, next year any product assembly line V8 engine, factory alpha

next year

products of product group Original Equipment - Engines, assembly next year line V8 engine, factory alpha machines, assembly line V8 engine, factory alpha

next year

Objective Measure

Objective Level

average delivery time

24 hours

efficiency

9

average rejection rate

0.2 percent

average machine 0.3 class A defect rate defects per week

single products of product group Original equipment, assembly line next year average lead time V8 engine, factory alpha efficiency

9

Decrease Just-Infactory alpha, any logistics partner, Time Deviation of next year product Logistics Partners

average just-intime deviation

five minutes

Decrease Packing factory alpha warehouse workers, Time customer, order

next year

average packing time

one hour

next year

total costs

500,000 •

Reduce Warehousing Costs

factory alpha, any logistics partner next year

240 minutes

factory alpha warehouse



Deriving Information Models from Business Objectives Constructing Dimensions Having defined operational objectives and structured them hierarchically, we are now able to create a conceptual model of the information system supporting managerial analysis. We first need to define dimensions that consist of hierarchically structured dimension objects. As a first step, the initial set of objective references taken from the definitions of operational objectives can be decomposed. The objects of the Reference column in Table 1 represent such decomposed objective references, which will be redefined as dimension objects and structured hierarchically. They thus form the basic structure of what will be a dimension. This process is complex creative work. Even so, without a methodological approach such as the one presented here, no assistance with this process would be available. Questioning managers on basis of the specified operational objectives is imperative for deriving further insights into the structures of the information systems supporting managerial analysis. Our example objective Rejection Rate Reduction states that rejection rates of other product group’s products must not increase. This inevitably leads to the question as to which other product groups should be considered for managerial analysis. The planning scenario that needs to be set up, will include the objective level of the product group Original Equipment — Engines, which needs to be decreased according to the objective. Furthermore, it includes the objective levels of all other product groups that must not exceed the respective levels from the previous year. The identification of dimensions can be assisted by answering the question of whether the elements of operational objective references are structured in an n:m relationship or in a 1:m relationship. The first case implies the modeling of two dimensions (because dimensions are hierarchical constructs of dimensions objects) whereas in the latter case, only one dimension is modeled. This decision needs to be made carefully. It needs to be identified whether this 1:m relationship occurs only temporarily, just as objective references of operational objectives, or generally. If it occurs generally, it is imperative to know, if the relationship might be changed by an ongoing business strategy. As mentioned above, identifying dimensions is a complex process that directly influences data warehouse structures. It can be seen as a strategic decision during the MIS specification process. In our example objectives from Table 1, there are 11 types of fundamentally different entities, business units, assembly lines, warehouses, factories, Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


product groups, workers, products, logistic partners, time entities, orders, and customers. Now, does an assembly line always belong to one factory or can it be spread over more than one factory? Is it possible that a factory runs more than one assembly line? Do workers work in one factory (at one assembly line) or are they allocated to more factories (assembly lines)? Is a product always assigned to exactly one product group? Questions like these have been made possible by the definition of operational objectives with the proposed method. They need to be answered by responsible personnel from business domains to specify the management-supporting information system. Implying 1:m relationships between business units, product groups, and between product groups and products, these three different entity types can be aggregated within one dimension Product. If, furthermore, all other entity types are bound by n:m relationships, each will be structured in one dedicated dimension. Only warehouses and assembly lines have been aggregated within one dimension, because allowing analysis between these entity types would serve no purpose. To distinguish planning scenarios from actual business developments, we need the dimension Version. Version is a dimension consisting of the dimension objects Actual, and several plans such as Plan, Plan optimistic, Plan pessimistic, or Forecast. Due to the fact that we transform objectives into planning scenarios to compare them to future business development, we need to add a dimension object of Version to each business fact. If it is a planned fact, a reference to a plan-version is necessary. In case of actual business facts, the Version dimension object Actual is referenced. Deviation analysis later compares business facts that differ only in the reference component of the dimension Version. Figure 5 contains all dimensions necessary to build the MIS environment, which allows for the managerial activity monitor delivery time. After the identification of dimensions, their basic structure of dimension objects that have been derived from operational objectives needs to be completed. Other dimension objects that will further be necessary to answer the managers’ questions need to be added. Basically, this means that all relevant products of all product groups (product group Original Equipment — Engines and all others obtained from the answer to the question derived from the operational objective Rejection Rate Reduction) are added to the product dimension. In this case, the product dimension would be extended by the products, product groups, and business units shown in Figure 5. This procedure needs to be repeated for every identified dimension.



Figure 5. Set of dimensions for managerial activity monitor delivery time Models of Dimensions Product

Production and Storing Facilities

Automotive Supplies Original Equipment OE Engines

Assembly Line V8 Engine Machine V8 - JH7765K Machine V8 - HJG5RF4

OE Chassis Components OE Electronic Components

Assembly Line Chassis Components Factory Alpha Warehouse

Replacement Electronic Components Engine Parts

Personnel Assembly Line V8 Engine Foreman

Electronic Parts

Machine V8 - JH7765K Foreman Workplace 1

Industrial Supplies

Workplace 2

Services Machine V8 - HJG5RF4 Foreman Time by Month January 2004

Assembly Line Chassis Components Foreman

February 2004

Factory Alpha Warehouse Foreman

Logistics Partner Partners for Engines Partners for Chassis Components

Customers by CRM Class Class A Customers Class B Customers

Factory Factory Alpha Factory Beta

Order Order 0000001 Order 0000002

Version Plan Actual

Legend <non-opened non-leaf dimension object identifier>

Constructing Navigation Spaces for Managerial Activities The definition of business objectives first needs to be followed by the managerial activity of undertaking the necessary steps to implement improved business processes. Secondly, management must monitor the degree to which the defined business objectives have been achieved. To address the problems of information overflow and information misuse, we need to define dimension scopes for specific managerial monitoring activities. To monitor the objective Increase Production Efficiency introduced above, we need to define six dimension scopes. As time can be limited to all time dimension Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


objects of the sub-hierarchy 2004, the first dimension scope Time by Month → Year 2004 consists of all days and months in 2004 and the year 2004 itself. All other time entities are blanked out. Five more dimension scopes are built similarly. The dimension scope Factory → Factory Alpha reduces all factories of the dimension Factory to Factory Alpha, Product → Product Group Automotive Supplies — Original Equipment — OE Engines focuses on engines, and Production and Storing Facilities → Assembly Line V8 Engine reduces the total set of warehouses and assembly lines to assembly line V8 Engine. Version is reduced to Plan in one dimension scope (Version → Plan) and Actual in another (Version → Actual), which allows for comparing the business facts based on these two valuations. The dimension scope combination Production Efficiency joins all six dimension scopes. It creates a navigation space for the required information of the managerial monitoring activity corresponding to the objective Increase Production Efficiency. This navigation space consists of all combined reference objects that are necessary to monitor the objective Increase Production Efficiency itself, and all of its sub-objectives once the respective qualitative and quantitative measures have been assigned to them. The dimension scope combination features two hierarchy levels. To create a combined reference object, one dimension object of each dimension scope of the first hierarchy level needs to be selected. Version is split up into two dimension scopes, which means that one of its dimension scopes needs to be picked for the valuation of business facts. This is necessary, because no information would be aggregated from the versions, Actual and Plan (Holten & Dreiling, 2002; Holten et al., 2002). Figure 6 contains the dimension scopes and the dimension scope combination for the managerial activity monitor production efficiency. The second sub-objective Increase Shipping Efficiency of the main objective Delivery Time Reduction of Business Unit Automotive Supplies requires the construction of a different set of dimension scopes and a different dimension scope combination. Four existing dimension scopes can be used for the managerial activity monitor shipment efficiency, which are Time by Month → Year 2004, Factory → Factory Alpha, Version → Plan, and Version → Actual. Additionally, five new dimension scopes are necessary for customers, logistic partners, orders, personnel, and production and storing facilities. Each reduces the total set of its corresponding dimension’s dimension objects to the relevant one for the managerial activity. As for the managerial activity monitor production efficiency, a dimension scope combination joins all of these dimension scopes (Shipping Efficiency). In order to create combined reference objects, again one dimension object from the first hierarchy level of the dimension scope combination, needs to be selected as well as one element of either one the version dimension scopes. Figure 7 contains the dimension scopes and the



Figure 6. Set of dimensions scopes and dimension scope combination for managerial activity monitor production efficiency Models of Dimension Scopes for Objective Increase Production Efficiency Time by Month Æ Year 2004

Product Æ Product Group Automotive Supplies - Original Equipment - OE Engines

January 2004

Automotive Supplies

2004-01-01

Original Equipment

2004-01-02

OE Engines Production and Storing Facilities Æ Assembly Line V8 Engine

February 2004

Assembly Line V8 Engine

December 2004

Machine V8 - JH7765K Factory Æ Factory Alpha

Machine V8 - HJG5RF4

Factory Alpha Version Æ Plan

Version Æ Actual

Plan

Actual

Model of Dimension Scope Combination Production Efficiency Production Efficiency Time by Month Æ Year 2004 Product Æ Product Group Automotive Supplies - Original Equipment - Engines Production and Storing Facilities Æ Assembly Line V8 Engine Factory Æ Factory Alpha Version Version Æ Plan Version Æ Actual Legend

dimension scope combination for the managerial activity monitor shipment efficiency. The introduced dimension scopes and dimension scope combinations from Figure 6 and Figure 7, correspond to two managerial activities of the managers responsible for production and logistics. Both managerial activities serve the purpose of reducing the delivery time of business unit Automotive Supplies as introduced with the main objective above. The activities of the higher management may just require the information, whether the delivery times have been reduced or not. The determining factors for this reduction are clear to the production and logistics managers, but in order to minimize information overflow, they are not part of upper management’s view on business processes. Also, the hierarchical depth of the dimensions Product and Time by Month have been reduced. In contrast to the horizontal reduction of dimensions, this reduction is made vertically. It is no longer possible to drill down from months and product groups to more detailed dimension objects such as products or days. Figure 8 contains the dimension scopes and the dimension scope combination for the



Figure 7. Set of dimensions scopes and dimension scope combination for managerial activity monitor shipment efficiency Models of Dimension Scopes for Objective Increase Shipping Efficiency Time by Month Æ Year 2004 January 2004

Factory Æ Factory Alpha Factory Alpha

2004-01-01

Customers by CRM Class Æ Any Customer

2004-01-02

Class A Customers Class B Customers

February 2004 December 2004

Order Æ Any Order

Logistics Partner Æ any Logistics Partner Partners for Engines

Order 0000001 Order 0000002

Partners for Chassis Components Personnel Æ Factory Alpha Warehouse Workers Production and Storing Facilities Æ Factory Alpha Warehouse Factory Alpha Warehouse

Factory Alpha Warehouse Foreman Version Æ Actual Actual

Version Æ Plan Plan Model of Dimension Scope Combination Shipping Efficiency Shipping Efficiency Time by Month Æ Year 2004 Logistics Partner Æ any Logistics Partner Factory Æ Factory Alpha

Personnel Æ Factory Alpha Warehouse Workers Customers by CRM Class Æ Any Customer Order Æ Any Order Production and Storing Facilities Æ Factory Alpha Warehouse Version Version Æ Plan Version Æ Actual Legend

managerial activity monitor delivery time of business unit automotive supplies.

Constructing Aspect Systems All defined business objectives have now been decomposed and used to construct dimensions. Furthermore, navigation spaces have been created to monitor if the business objectives have been accomplished. In the next step, aspect systems will be defined which will be assigned to navigation spaces, allowing for the construction of business facts. The decomposition of facts, led to measures that have been used to quantify or qualify the references. These Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Figure 8. Set of dimensions scopes and dimension scope combination for managerial activity monitor delivery time of business unit automotive supplies Models of Dimension Scopes for Objective Delivery Time Reduction of Business Unit Automotive Supplies Product Æ Business Unit Automotive Supplies Automotive Supplies Original Equipment OE Engines

Time by Month Æ Year 2004 January 2004 February 2004 December 2004

OE Chassis Components OE Electronic Components

Version Æ Actual Actual

Replacement

Version Æ Plan

Electronic Components

Plan

Engine Parts Electronic Parts Model of Dimension Scope Combination Delivery Time of Business Unit Automotive Supplies Delivery Time of Business Unit Automotive Supplies Product Æ Business Unit Automotive Supplies Time by Month Æ Year 2004 Version Version Æ Plan Version Æ Actual Legend

measures will be transformed either into quantitative or qualitative aspects, depending on the nature of their values. To monitor the objective Increase Production Efficiency several aspects are necessary. First, production efficiency is a qualitative aspect. Levels from 0 to 10 can be used to value production efficiency. The objective Increase Production Efficiency has been broken down into three sub-objectives, which have been transformed into quantitative aspects. The measures of the three subobjectives are average rejection rate, average defect rate, and average lead time. All three aspects will be organized into an aspect system Production Efficiency Measurement as sub-aspects of the aspect production efficiency. For analytical purposes, the production efficiency is significant and used as a starting point. In case something is wrong, it is possible to drill-down to the influencing aspects average rejection rate, average defect rate, and average lead time. The construction of the second aspect system for the objective Increase Shipping Efficiency is similar to the construction of the aspect system for the objective Increase Production Efficiency. Shipping efficiency is the most significant aspect. Three sub-aspects are derived from the sub-objectives of the



Figure 9. Aspect systems for the managerial activities monitor production efficiency, monitor shipment efficiency, and monitor delivery time of business unit automotive supplies Models of Aspect Systems 1 34 67

29

Production Efficiency Measurement

5 57

9

Production Efficiency Average Rejection Rate Average Defect Rate Average Lead Time 1 34 67

29

Shipment Efficiency Measurement

5 57

9

Shipment Efficiency Average Just-In-Time Deviation Average Packing Time Costs 1 34 67

29

Delivery Performance Measurement

5 57

9

Average Delivery Time Shipment Efficiency Production Efficiency Legend 1 67

34

29

5 57

9

<super-aspect (hierarchically)>

<sub-aspect (hierarchically)

objective Increase Production Efficiency, which are average just-in-time deviation, average packing time, and costs. The construction of the main objective’s aspect system Delivery Performance Measurement differs from the first two aspect systems. It is constructed from three aspects, which are average delivery time, shipment efficiency, and production efficiency. Shipment efficiency and production efficiency are taken from the first two aspect systems, but in contrast to the production and logistics management, there are no drill-down possibilities for the aspects Shipment Efficiency and Production Efficiency. This again is due to avoid information overflow. All aspect systems for the three managerial activities monitor production efficiency, monitor shipment efficiency, and monitor delivery time of business unit automotive supplies are shown in Figure 9.

Constructing Information Objects As pointed out previously, a managerial activity, which monitors if a business objective has been accomplished, needs to compare planning scenarios with actual business developments. The nature of such analyses is that the reference objects of compared business facts differ only in a value of dimension Version



(Holten, et al., 2002; Holten & Dreiling, 2002). To compare planned facts with actual business facts, fact calculations need to be defined. Any business fact of a dimension scope combination that features the included dimension of the fact calculation can be calculated according to the calculation expression. Figure 10 shows the fact calculation Plan Variance. It calculates a percentage, which represents the deviation by which planned aspects differ from actual aspects. The fact calculation abstracts from aspects. It can be assigned to each dimension scope combination, where part of it equals to definition of version in Figure 10. Finally, we are able to construct information objects for the three main objectives. Each information object in our example consists of a dimension scope combination, an aspect system, and a fact calculation expression. The information object Production Efficiency assigns the aspect system Production Efficiency Measurement to the dimension scope combination Production Efficiency. Furthermore, the deviation analysis of planned and actual aspects is rendered possible by the fact calculation Plan Variance. Both other information objects are structured similarly. Figure 11 contains the information objects Production Efficiency, Shipment Efficiency, and Delivery Performance Measurement. The constructed information objects consist of planned and actual business facts. Besides the planned facts that arise from planning scenarios defined by the introduced business objectives, other facts are included within these information objects. Examples are production efficiency of factory alpha, shipment efficiency of logistic partners, or the average delivery times for customer orders within the year 2003. These other facts are part of a dynamic managerial analysis, which aims at detailing or generalizing the examined aspect of the business.

Conclusions and Outlook This chapter introduced an ontology-driven method for information systems development. Based on the analysis of epistemological positions, the interpretivists position — explained as a combination of ontological realism and subjective cognition — was chosen for the analysis. The ontology-driven method for IS development incorporates an ontology as the core concept and is based on several philosophical and linguistic foundations such as Kamlah and Lorenzen’s language critique approach, Morris’ findings on semiotics, de Saussure’s findings on signs, or Bunge’s research in ontology. We showed that ontologies are created and maintained by language communities using linguistic actions and how new concepts can be created to handle new situations.



Figure 10. Fact calculation expression plan variance Model of Fact Calculation Plan Variance Plan Variance

+ %

-

Version Version Æ Plan Version Æ Actual Plan Variance := (Plan/Actual)*100 Legend + %

-

Figure 11. Information objects for the managerial activities monitor production efficiency, monitor shipment efficiency, and monitor delivery time of business unit automotive supplies Models of Information Objects Production Efficiency Production Efficiency 1 34 29

67

9

Production Efficiency Measurement

-

Plan Variance

5 57

+ %

Shipment Efficiency Shipment Efficiency 1 34 29

67

9

Shipment Efficiency Measurement

-

Plan Variance

5 57

+ %

Delivery Performance Measurement Delivery Time of Business Unit Automotive Supplies 1 34

29 67

9

Delivery Performance Measurement

-

Plan Variance

5 57

+ %

Legend + %

-

Furthermore, we have applied our ontology-driven method to information systems development by introducing an ontology for the domain of management information systems. A comprehensive business case indicates the usability of this specification language. By using Wedekind’s three introduced linguistic actions subsumption, subordination, and composition we have (re-)constructed a language to make statements about the domain of management information Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


systems. We have introduced language constructs for specifying information spaces and aspects within information spaces. Furthermore, business objectives have been added to the language to assist a specification of MIS with the introduced language. There remains, however, significant room for extension and development. The language needs to be applied to different modeling cases for gaining experience with it and improving it. To better profit from the language we will work on an automated, respectively semi-automated generation of information systems based on the specification models as shown by Holten (2003b). Furthermore, we will examine different domains such as Customer Relationship Management and create an ontology for this domain as well (respectively extent the one used here). Again, this will result in a specification language (or a language extension, this time focusing on CRM.

References Adam, D. (1996). Planung und Entscheidung. Wiesbaden, Germany: Gabler. Becker, J., Dreiling, A., Holten, R., & Ribbert, M. (2003). Specifying information systems for business process integration — A management perspective. Information Systems and e-Business Management, 1(3), 231-263. Blaschka, M., Sapia, C., Höfling, G., & Dinter, B. (1998). Finding your way through multidimensional data models. International Workshop on Data Warehouse Design and OAP Technology, Vienna, Austria, (pp. 198203). Böhnlein, M. (2001). Konstruktion semantischer data-warehouse-schemata. Dissertation thesis, Deutscher Universitäts-Verlag, Wiesbaden, Germany. Bulos, D. (1996). OLAP database design: A new dimension. Database Programming & Design, 9(6), 33-37. Bunge, M. A. (1977). Ontology I: The furniture of the world: Treatise on basic philosophy. Dordrecht, Netherlands: Reidel. Campbell, A. E, & Shapiro, S. C. (1995). Ontologic mediation: An overview. IJCAI Workshop on Basic Ontological Issues in Knowledge Sharing, (pp. 16-25). de Saussure, F. (1974). Course in general linguistics. London: Peter Owen. Dubé, L. & Paré, G. (2003). Rigor in information systems positivist case research: Current practices, trends, and recommendations. MIS Quarterly, 27(4), 597-635.



Golfarelli, M., Maio, D., & Rizzi, S. (1998). The dimensional fact model: A conceptual model for data warehouses. International Journal of Cooperative Information Systems, 7(2/3), 215-247. Golfarelli, M. & Rizzi, S. (1998). A methodological framework for data warehouse design. ACM First International Workshop on Data Warehousing and OLAP (DOLAP), ACM, Washington, D. C. Golfarelli, M. & Rizzi, S. (1999). Designing the data warehouse: Key steps and crucial issues. Journal of Computer Science and Information Management, 2(3). Green, P. & Rosemann, M. (2000a). Integrated process modeling: An ontological evaluation. Information Systems, 25(3), 73-87. Green, P. & Rosemann, M. (2000b). Usefulness of the BWW ontological models as a “core” theory of information systems. Information Systems Foundation Workshop: Building the Theoretical Base, (pp. 147-164). Guarino, N. & Welty, C. (2002). Evaluating ontological decisions with OntoClean. Communications of the ACM, 45(2), 61-65. Henderson, J. C. & Venkatraman, N. (1999). Strategic alignment: Leveraging information technology for transforming organizations. IBM Systems Journal, 38(2/3), 472-484. Hirschheim, R., Klein, H. K., & Lyytinen, K. (1995). Information systems development and data modeling. Cambridge, U.K.: Cambridge University Press. Hitt, L. M. & Brynjolfsson, E. (1996). Productivity, business profitability, and consumer surplus: Three different measures of information technology value. MIS Quarterly, 20(2), 121-142. Holten, R. (1999). Entwicklung von Führungsinformationssystemen. Ein methodenorientierter Ansatz. Wiesbaden, Germany: Gabler. Holten, R. (2000). Framework and method for information warehouse development processes. In R. Jung & R. Winter (Eds.), Data Warehousing 2000 — Methoden, Anwendungen, Strategien Heidelberg, Germany: Springer. (pp. 135-163). Holten, R. (2003a). Integration von informationssystemen. Theorie und anwendung im supply chain management. Habilitation thesis, University of Münster, Department of Information Systems, Münster, Germany. Holten, R. (2003b). Specification of management views in information warehouse projects. Information Systems, 28(7), 709-751. Holten, R. & Dreiling, A. (2002). Specification of fact calculations within the MetaMIS approach. Muenster, Germany: University of Muenster.



Holten, R., Dreiling, A., & Schmid, B. (2002). Management report engineering — A Swiss Re business case, In Data Warehousing 2002 Conference, Friedrichshafen, Germany, pp. 421-437. Im, K. S., Dow, K. E., & Grover, V. (2001). Research report: A reexamination of IT investments and the market value of the firm — An event study methodology. Information Systems Research, 12(1), 103-120. Jiang, J. J., Klein, G., & Discenza, R. (2001). Information system success as impacted by risks and development strategies. IEEE Transactions on Engineering Management, 48(1), 46-55. Johnston, H. R. & Vitale, M. R. (1988). Creating competitive advantages with interorganizational information systems. MIS Quarterly, 12(2), 153-165. Kamlah, W. & Lorenzen, P. (1984). Logical propaedeutic. Pre-school of reasonable discourse. Lanham, MD: University Press of America. Kaplan, R. S. & Norton, D. P. (1992). The balanced score card. Measures that drive business performance. Harvard Business Review, 70, 71-79. Kaplan, R. S. & Norton, D. P. (1993). Putting the balanced scorecard to work. Harvard Business Review, 71, 134-142. Kaplan, R. S. & Norton, D. P. (1996). Translating strategy into action. The balanced scorecard. Boston, MA: Harvard Business School Press. Karimi, J. (1988). Strategic planning for information systems: Requirements and information engineering methods. Journal of Management Information Systems, 4(4), 5-24. Keil, M. (1995). Pulling the plug: Software project management and the problem of project escalation. MIS Quarterly, 19(4), 421-447. Keil, M., Mann, J., & Rai, A. (2000). Why software projects escalate: An empirical analysis and test of four theoretical models. MIS Quarterly, 24(4), 631-664. Kelly, G. A. (1955). The psychology of personal constructs. New York: W. W. Northern & Company. Kottemann, J. E. & Konsynski, B. R. (1984). Information systems planning and development: Strategic postures and methodologies. Journal of Management Information Systems, 1(2), 45-63. Latham, G. P. & Kinne, S. B. (1974). Improving job performance through training in goal setting. Journal of Applied Psychology, 59, 187-191. Lechtenbörger, J. (2001). Data warehouse schema design. Berlin, Germany: Akademische Verlagsgesellschaft. Li, E. Y. & Chen, H.-G. (2001). Output-driven information system planning: A case study. Information & Management, 2001(38), 185-199.



Lorenzen, P. (1987). Constructive philosophy. Amherst, MA: The University of Massachusetts Press. Morris, C. (1971). Writings on the general theory of signs. The Hague, Netherlands: Mouton. Mukhopadhyay, T., Kekre, S., & Kalathur, S. (1995). Business value of information technology: A study of electronic data interchange. MIS Quarterly, 19(2), 137-156. Niehaves, B., Ribbert, M., Dreiling, A., & Holten, R. (2004). Conceptual modeling — An epistemological foundation. 2004 Americas Conference on Information Systems, New York, NY (pp. 4232-4242). Nissen, H. W., Jeusfeld, M., Jarke, M., Zemanek, G. V., & Huber, H. (1996). Managing multiple requirements perspectives with metamodels. IEEE Software, 13(3), 37-48. Porter, M. E. (1979). How competitive forces shape strategy. Harvard Business Review, 57, 137-145. Reich, B. H. & Benbasat, I. (2000). Factors that influence the social dimension of alignment between business and information technology objectives. MIS Quarterly, 24(1), 81-113. Ribbert, M., Niehaves, B., Dreiling, A., & Holten, R. (2004). An epistemological foundation of conceptual modeling, 12th European Conference on Information Systems, Turku, Finland. Riebel, P. (1979). Gestaltungsprobleme einer zweckneutralen Grundrechnung. Zeitschrift für betriebswirtschaftliche Forschung, 31, 863-893. Riebel, P. (1992). Einzelerlös-, Einzelkosten- und Deckungsbeitragsrechung als Kern einer ganzheitlichen Führungsrechung. In W. Männel (Ed.), Handbuch Kostenrechnung, Wiesbaden: Gabler (pp. 247-299). Rosemann, M. & Green, P. (1999). Enhancing the process of ontological analysis — The “who cares” dimension. Information Systems Foundations Workshop — Ontology, Semiotics and Practice, (pp. 39-54). Rosemann, M. & Green, P. (2002). Developing a meta model for the BungeWand-Weber ontological constructs. Information Systems, 27(2), 75-91. Sapia, C., Blaschka, M., & Höfling, G. (2000). GraMMi: Using a standard repository management system to build a generic graphical modeling tool. Hawaii International Conference on System Sciences (HICSS-33), Vol. 33, Maui, Hawaii. Scheer, A. W. (2000). ARIS — Business process modeling. Berlin, Heidelberg, Germany: Springer. Soley, R. & The OMG Staff Strategy Group (2000). Model driven architecture. Needham, MA: Object Management Group. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Standish Group International. (2001). Extreme CHAOS (Research report). Available for ordering at http://www.standishgroup.com Strahringer, S. (1996). Metamodellierung als instrument des methodenvergleichs. Eine evaluierung am beispiel objektorientierter analysemethoden. Herzogenrath, Germany: Shaker. Subramani, M. & Walden, E. (2001). The impact of e-commerce announcements on the market value of firms. Information Systems Research, 12(2), 135-154. Tam, K. Y. (1998). The impact of information technology investments on firm performance and evaluation: Evidence from newly industrialized economies. Information Systems Research, 9(1), 85-98. Tan, F., B. & Hunter, M. G. (2002). The repertory grid technique: A method for the study of cognition in information systems. MIS Quarterly, 26(1), 3957. Uschold, M., King, M., Moralee, S., & Zorgios, Y. (1998). The enterprise ontology. The Knowledge Engineering Review, 13(1), 13-90. van Hee, K. M., Somers, L. J., & Voorhoeve, M. (1991). A modeling environment for decision support systems. Decision Support Systems, 7(3), 241251. Venkatraman, N. (1994). IT-enabled business transformation: From automation to business scope redefinition. Sloan Management Review, 35(2), 73-87. von Foerster, H. (1996). Wissen und Gewissen. Versuch einer Brücke. Frankfurt a. M., Germany: Suhrkamp. Wand, Y. & Weber, R. (1989). An ontological evaluation of systems analysis and design methods. In E. Falkenberg & P. Lindgreen (Eds.), Information systems concepts: An in-depth analysis (pp. 79-107). Amsterdam, Netherlands: North-Holland. Wand, Y. & Weber, R. (1990a). Mario Bunge’s ontology as a formal foundation for information systems concepts. In P Weingartner & G. Dorn (Eds.), Studies on Mario Bunge’s Treatise (pp. 123-149). Atlanta: Rodopi. Wand, Y. & Weber, R. (1990b). An ontological model of an information systems. IEEE Transactions on Software Engineering, 16(11), 1282-1292. Wand, Y. & Weber, R. (1993). On the ontological expressiveness of information systems analysis and design grammars. Journal of Information Systems, 3(4), 217-237. Wand, Y. & Weber, R. (1995). On the deep structure of information systems. Information Systems Journal, 5(3), 203-223. Weber, R. (2004). The rhetoric of positivism versus interpretivism: A personal view. MIS Quarterly, 28(1), 3-xiii. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Wedekind, H. (1981). Datenbanksysteme I. Eine konstruktive Einführung in die Datenverarbeitung in Wirtschaft und Verwaltung. Mannheim, Germany: B.I.-Wissensverlag. Wixom, B. H. & Watson, H. J. (2001). An empirical investigation of the factors affecting data warehousing success. MIS Quarterly, 25(1), 17-41.

Endnote 1

This work has been funded by the German Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung), record no. 01HW0196.



Appendices Appendix A. Meta-model segments for defining information spaces and aspect within information spaces Meta Model (extented Entity Relationship Model)

Reference Object Structure (0,m) (0,m) Combined Reference Object

CRO-Coordinates

(0,m)

Dimension Object Hierarchy Operator

(0,m)

(0,m) (0,m)

CE-Ot-As

(0,m)

Dimension Object

(0,1) (1,1)

(1,m)

(1,m) Calculation Expression

Calculation Expression Element

u,t

(1,m)

DO-DS-AS

(1,m) Operand

(0,m)

Calculation Expression Set

CE-On-As

Dimension Scope

u,t

Fact (0,m) (0,m)

Aspect

DS-DSC-AS u,t

(0,m)

Quantitative Aspect (Ratio) u,t

(1,m)

Basic Aspect Dimension Scope Combination

Qualitative Aspect

(0,m)

Calculated Aspect (2,m) A-AS-As

(0,m) Aspect System

Dimension Grouping

Value Element

(0,m)

A-AS-As

(0,m)

Information Object

(1,m)

D-DG-AS

(1,1) Dimension

(1,m)

DO-D-AS

(1,m)

D-HL-AS

Hierarchy Level

(1,m)

DO-D-HL-AS

(1,1)

Legend

Entity Type

Relationship Type

(min,max)

Reinterpreted Relationship Type

Specialization (Types: - u unequivocally, e equivocally - t total, p partial)

Connector ( - min minimum cardinality, - max maximum cardinality)



Appendix B. Meta-model segment for business objectives Meta Model (extented Entity Relationship Model)

Objective Structure (0,m)

(0,m)

Objective

u,t

Strategy, General Condition, and Guideline


Quantitative Measure (Aspect Ratio)

u,t

(0,m)

Objective Measure

(0,m)

OO-OL-AS

Objective Level

Qualitative Measure (Aspect Category)

Reference Object

(0,m)

(0,m)

Legend

Entity Type

Relationship Type

(min,max)

Reinterpreted Relationship Type

Connector ( - min minimum cardinality, - max maximum cardinality)

Specialization (Types: - u unequivocally, e equivocally - t total, p partial)


218 Kazmierczak & Milton

Chapter VIII

Using a Common-Sense Realistic Ontology: Making Data Models Better Map the World

Ed Kazmierczak, The University of Melbourne, Australia Simon Milton, The University of Melbourne, Australia

Abstract This chapter examines the following question: “How well do data models map the world?” Data modelling languages are used in today’s information systems engineering environments to model reality. Many have a degree of hype surrounding their quality and applicability with narrow and specific justification often given in support of one over another. We want to more deeply understand the fundamental nature of data modelling languages. We thus propose a theory, based on ontology, that should allow us to understand, compare, evaluate, and strengthen data modelling languages. We then introduce Chisholm’s ontology and apply methods to analyse some data modelling languages using it. We find a good degree of overlap between all of the data modelling languages analysed and the core concepts of Chisholm’s ontology, and conclude that the data modelling


Using a Common-Sense Realistic Ontology

219

languages investigated reflect an ontology of commonsense-realism. Critical common-sense realism more generally due to its perspectival nature and its implicit recognition of institutional and social reality has the potential to dramatically improve our ability to better map the world.

Introduction Data models have been used in information engineering environments for many decades for the precise purpose of building representations of reality. Data models are used in organizations to represent reality at three levels. First they are used to establish the highest level of description of an organisation’s reality to guide strategic information systems development and management and for high level data management. The model is used to drive information systems management and development and for the implementation or management of databases. Second, they are used to construct a description of the reality surrounding a proposed information system. The description is used in systems analysis and design. This is often called “conceptual modelling” although the name does not describe well the purpose of this activity. This facilitates the accurate and timely implementation of a system by helping establish relevant shared understandings of reality and in implementing some specific aspects of the system in technology such as databases. An increased degree of detail is required compared with corporate data modelling. Significant attributes of things found in the reality are required together with relationships between entities. Finally, they are used to model parts of an organisation’s reality leading to implementation in an operational database into which facts about that reality are stored. Such databases may serve several information systems within an organization. This description is the most detailed but assumes only enough detail for all applications relying on it to function. To date, there have been many different data modelling languages proposed with the most popular being the entity-relationship model (Chen, 1976) but also including the functional data model (Kerschberg & Pacheco, 1976; Shipman, 1981), the semantic data model (Hammer & McLeod, 1981), NIAM (Nijssen & Halpin, 1989), and object modelling technique (Blaha & Premerlani, 1998) that later became the basis for the unified modelling language (UML). Each new modelling language has often been accompanied with claims of its superiority and at times hype when compared with the others. There has been little beyond opinion to substantiate such claims and yet all notations purport to do similar things. We have two research questions:



1.

How well do data models represent reality?

2.

What are the similarities and differences between data modelling languages?

We need a theory to help us answer these questions. The mature philosophical study of ontology has been used as a source of theory to investigate tools and techniques used in the analysis and design of information systems. A key development in the use of ontology for the study of information systems has been the work of Wand and Weber (Weber, 1997), based on Bunge’s (1977, 1979) ontology. Part of the focus of this research has been to investigate the representational power of data modelling languages (Green, 1996; Rohde, 1995; Wand, 1996; Wand & Weber, 1989, 1990, 1993; Weber, 1997). Our work is motivated by the search for semantic methods to answer the research questions just mentioned. The work of Wand and Weber, while ground breaking, is based on structural comparison of elements of grammar and concludes only presence or absence of a “construct”. The conclusions drawn are very much based upon whether or not the data modelling language supports the ontological construct. Our work seeks to develop semantic methods that not only detect the presence or absence of a construct but also allow us to judge the level of agreement or disagreement between a data modelling language and an ontology. The contribution of this chapter is threefold. First, we develop qualitative methods: 1) “the method of conceptual comparison”, for conceptually evaluating individual data modelling languages through ontologies and 2) “the method for conceptual comparison”, for comparing a range of data modelling languages with an ontology based on a number of individual evaluations. These methods help answer the two research questions and are detailed in the method section. Our methods are, to some extent, independent of the individual ontology chosen as the basis of comparison. As a by-product we are starting to investigate the dominant ontology within data modelling languages. Second, we apply the methods using Chisholm’s (1996) ontology to a representative range of data modelling languages. We follow this introduction with a deeper discussion of exactly what ontology is before relating ontologies and data modelling languages. We then examine the realism assumed in Chisholm’s ontology and relate it to that contained within Bunge’s ontology (the ontology upon which BWW is based.) Following this we describe Chisholm’s ontology, which is the ontology used in this study, before describing the methods applied. The methods can be applied to ontologies other than Chisholm’s. Finally we present the results and conclude. Our conclusions also comment on the more fundamental influence common-sense realism is likely to have in data modelling specifically, and information systems more generally. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


221

Ontology Defined: A Stratified Domain Ontology is the study of “what there is”. An ontology written by a philosopher can be described as “a particular system of categories accounting for a certain vision of the world. As such, this system does not depend upon a particular language: Aristotle’s ontology is always the same, independently of the language used to describe it” (Guarino, 1998). An ontology describes what is fundamental in “what there is” or “what is real”. It defines the terms used to construct a description of reality in its most general sense and how the terms are related. It must be able to describe a reality without specifying particulars of any category. It must be able to be used to describe reality at any point in time (either well into the future, or into the past). A highlevel philosophical ontology must be able to describe reality in this way. “[Ontology is] the study of being in so far as this is shared in common by all entities, both material and immaterial. It deals with the most general properties of beings in all their different varieties” (Kim & Sosa, 1995, p. 373). Metaphysics can also be understood in a more definite sense, suggested by Aristotle’s notion (in his Metaphysics, the title of which was given by an early editor of his works, not by Aristotle himself) of “first philosophy,” namely, the study of being qua being, i.e. of the most general and necessary characteristics anything must have in order to count as being an entity (ens). Sometimes “ontology” is used in this sense, but this is by no means common practice, “ontology” being often used as a synonym for metaphysics. (Audi, 1995, p. 490) For our purposes, we find the following definition most helpful, and we adopt it in this work: Definition 1. Reference Ontology Ontology, understood as a branch of metaphysics, is the science of being in general, embracing such issues as the nature of existence and the categorical structure of reality. … Different systems of ontology propose alternative categorical schemes. A categorical scheme typically exhibits a hierarchical structure, with “being” or “entity” as the topmost category, embracing everything that exists. (Honderich, 1995, p. 634)



Essentially, this is in alignment with our interests, because ontology of this nature represents a framework using which the building blocks of reality are described, in a way that is divorced from any specific situation or state of affairs. This agrees with Guarino’s definition we gave earlier. It encompasses everything that exists but is general. In contrast, other parts of informatics, notably but not restricted to artificial intelligence (AI) (Vet & Mars, 1998; Vickery, 1997) and more generally, computer scientists use ontology in a highly pragmatic way. On the other hand, in its most prevalent use in AI, an ontology refers to an engineering artifact, constituted by a specific vocabulary used to describe a certain reality, plus a set of explicit assumptions regarding the intended meaning of the vocabulary words. (Guarino, 1998) This is what we adopt as the meaning for a “domain-specific ontology”. For example, microeconomics, or a specific plant taxonomy, each has its own categories of terms and intended meaning for terms used in these fields. To help clarify the distinction between the two meanings Guarino finds in computing, he continues by saying that the two readings of “ontology”… are indeed related [to] each other, but in order to solve the terminological impasse we need to choose one of them, inventing another name for the other: we shall adopt the AI reading, using the word conceptualization to refer to the philosophical reading. Specifically, two ontologies can be different in the vocabulary used (using English or Italian words, for instance) while sharing the same conceptualization. (sic) (Guarino, 1998) We, in line with philosophy, maintain the term “reference” ontology for the philosophical meaning and use “domain-specific” ontology for this latter, much more recently adopted term. However, the two should be related. A sensibly constructed domain-specific ontology should map back to a reference or philosophical ontology. Also each may be used to specify specific members of categories that account for a certain reality. Let’s consider examples. A “reference” ontology has extremely general terms (individual, attributes, class, set, etc.) and describes reality “in its most general sense”. A “domain specific” ontology, while not necessarily committing to specific instances of the ontology (you, me, my cat), may commit to categories Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


223

such as “pet”, “bridge”, “animal”, “election campaign”, and also remain broad in its coverage of reality. It is clearly becoming more specific, but without committing to specific individuals; but it is not, in its fullest definition, general enough to “stand the test of time” (there was a time when bridges did not exist). Some “domain specific” ontologies are expressed in terms of a “reference ontology” but some have grown “bottom up” without a deep consideration of the upper most categories. A domain specific ontology is also likely to be restricted in the breadth of reality it describes, such as ontologies restricted to “engineering”, or “medicine”. Finally, we may be able to commit ourselves to actual members of each category (you, me, and my cat). This specificity of reality is not of interest to us. We can represent these ideas diagrammatically and see the definition of ontology stratified as we have just described (see Example 1). In the data-modelling world, a reference ontology can be understood as a data modelling language such as a crude object model consisting of “objects” that are described using “properties” and where relationships relate objects with one another. A domain-specific ontology is paralleled in data modelling by a data model, for example consistent with the crude object model, showing a university with entities such as “department”, or “faculty”, and “examination” with abstracted relationships between the entities. An example of a specific reality would commit to student “Tony Blair” and subject “AKA100 International Politics 1”. A domain specific ontology may extend to a certain reality such as this. In data modelling, the language used commits to a reference ontology and we want to examine this commitment in more detail. However, it can be said that in AI not all domain specific ontologies map back to coherent reference models. Some highly pragmatic examples of domain specific ontologies (such as CYC and Semantic Web) do not attempt to define high-level categories that are philosophically consistent.

Example 1. A stratification of “ontology” philosophical (Reference), computer science (Domain-specific), and specific reality.

Reference ontology

Domain-specific ontology specific reality



In researching data modelling languages it is reference ontology that is of interest. Examining the ontological commitments at this level is likely to have a profound effect on our deeper understanding of the role and purpose of data modelling tools. In the following discussions it is in this sense that “ontology” will be intended.

Relating Ontology and Data Modelling Languages Any ontology uses technical “expressions” to define what is real and it is these same “expressions” that are used to describe a specific “state of affairs”. Each expression has meaning and may be defined formally or informally. It is important therefore to distinguish between the labels or words given to each technical expression and the meaning behind that expression. For example individual is a label given to a technical expression in an ontology and has a specific meaning in that ontology. For the purpose of distinguishing expressions from their intended interpretation we will use Term to mean the label used in the ontology and Concept to mean the interpretation that the ontology gives to that term. A concept is defined as follows. Definition 2. Concept “[A] concept is a way of thinking about something — a particular object, or property, or relation, or some other entity” (Dancy & Sosa, 1992, p. 74). Now, to describe an ontology, we require the following elements: 1.

A categorisation of what constitutes reality, and what fundamental categories of things exist. As an example Chisholm presents his categorisation in a taxonomy (shown later in Figure 1).

2.

A set of terms, each associated with a concept that fully defines the term, with which to construct descriptions of “what there is”.

3.

A set of fundamental terms labelling the fundamental categories (in 1) and a means of relating all terms back to those labelling the fundamental categories.

For example, Chisholm’s ontology uses the terms individual, attribute, event, relation and set/class but only individual, attribute, and event are fundamental and label categories in Chisholm’s taxonomy. Set/class and relation derive their Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


225

definitions from individual and attribute and are thus related back to the fundamental categories. Data modelling languages also use terms, to signify modelling elements. For example, the Entity Relationship model (Chen, 1976) uses the terms entity, relationship, and attribute to signify specific modelling concepts used for modelling information systems. The terms, and their underlying concepts, used to describe a given data modelling language constitute an ontology of the type described above. We claim that the terms and concepts used to define a data modelling language constitute an ontological meta-model for the modelling language. We can now compare the ontological meta-model for a data modelling language with a philosophical ontology (acting as an external, or reference, ontology) thus yielding a qualitative semantic analysis of the similarity and difference between the world view that can be captured by the data model and that embodied in the ontology. Concepts are the basis for comparing an ontology with an ontological metamodel for a data modelling language. A concept is expressed using a possession condition. Definition 3. Possession Condition A possession condition is “[a] statement which individuates a concept by saying what is required for a thinker to possess it” (Dancy & Sosa, 1992, p. 75). Some concepts are compound. A compound concept has a core without which the entire concept is meaningless. In this sense the core of a concept is necessary for the concept’s meaning to be preserved (Honderich, 1995). Definition 4. Core Concept A core concept is one such that its absence would render the definition of an entire concept meaningless (Milton, 2000; Milton, Kazmierczak, & Keen, 2001). For example, the concept of a knife would be meaningless without a blade and some handle that fits within the palm of one’s hand. If absent, other parts of the concept do not have the effect of rendering the concept meaningless. By way of example, consider the term entity as used in the Entity-Relationship (ER) model. The concept for this term in ER is “something which involves information. It is usually identifiable. Each entity has certain characteristics, known as attributes. A grouping of related entities becomes an entity set” (Thalheim, 2000). Other similar definitions can be found in many texts, including the classical chapter by Chen (1976). For data modelling languages, the concept associated with a term can be synthesised from seminal sources. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Table 1. The concept “entity” from the classic ER modelling language Concept Entity

Core

Identity

Description ER allows for significant entities, or objects (either physical or conceptual) to be modelled. These must be grouped into entity classes. Each entity cannot depend upon other entities to be classed as an entity. Each member of an entity class must have an identity (a key).

In the case of entity from the classic ER modelling language, we can show the compound concept in Table 1. We can observe if the models built using our data modelling languages exhibit the world view embodied by an ontology by observing how well concepts found in the data modelling language’s ontological meta-model matches concepts from the ontology. This in turn is achieved by comparing the concepts in the ontological meta-model with those of the ontology. An examination of the results would reveal the overlap between the ontology and the data modelling language. The method for conducting such conceptual evaluations and comparisons is described in the method section. Before this, we discuss the philosophical heritage and content of Chisholm’s ontology, and its categories. We do this respectively in the following two sections.

Commonsense Realism: Chisholm’s Philosophy In this section we discuss the objectives for a philosopher when writing an ontology. Chisholm’s ontology is then discussed and found to be one of Commonsense Realism. We conclude by placing Bunge’s ontology (used as a basis for BWW used in information systems) in context with Chisholm’s ontology. Ontology defines the “sum total of reality” (Honderich, 1995). It examines the nature of existence (what exits) and the categories into which these things fit. Existence is what exists outside an individual’s mind and asks “What is real?” Things that exist may be concrete (physical) or abstract. Each ontologist must answer these questions when defining their categories. They do so by stating their philosophical approach and by relating their ontology to the stream of philosophical arguments.



227

Chisholm (1996) asserts that his approach is one of critical commonsensism: “Our approach to philosophy is what Charles Sanders Peirce has called ‘critical commonsensism’. This approach is based on faith in one’s own rationality. Reason, as Peirce put it, not only corrects its premises, ‘it also corrects its own conclusions’” (p. 4). “Commonsensism is the view that we know, most, if not all, of those things which ordinary people think they know and that any satisfactory epistemological theory must be adequate to the fact that we do know such things” (Dancy & Sosa, 1992). Critical commonsensism differs from commonsensism in that it demands a more rigorous standard of support for knowledge to be acquired, requiring the term “critical”. Chisholm’s (1996) ontology is also categorised as being one of “extreme realism” in that in addition to individuals, abstract things such as attributes are also real. Realism in any area of thought is the doctrine that certain entities allegedly associated with that area are indeed real. Common sense realism — sometimes called ‘realism’, without qualification — says that ordinary things like chairs and trees and people are real. Scientific realism says that theoretical points like electrons and fields of force and quarks are equally real. And psychological realism says mental states like pains and beliefs are real. (sic) (Dancy & Sosa, 1992) However, Chisholm’s critical commonsensism combined with his form of extreme realism means that his ontology adheres to a philosophy that may be a little different from commonsense realism. Barry Smith (1995), also a prominent realist and who refers to Chisholm in his discussion, defines the school more clearly than Dancy and Sosa (1992), and in his article he outlines his support for the thesis that commonsense realism in its various guises seems to be useful in cognitive science. The thesis that there is only one world towards which natural cognition relates is a central plank of what philosophers in the course of history have identified as the doctrine of common-sense realism. It is a doctrine according to which: (a) we enjoy in our everyday cognitive activities a direct and wideranging relational contact with a certain stable region of reality called the commonsense world



(b) our everyday cognitive activities rest upon a certain core of interconnected beliefs called “common sense” which is in large part true to the common-sense world as it actually is, not least in virtue of the fact that such beliefs and our associated cognitive capacities have arisen through interaction with this world; (c) this common-sense world exists autonomously, which is to say independently of our cognitive relations to it. Indeed from the perspective of common-sense realism the common-sense world exists entirely independently of human beings. Partial evidence for this thesis is provided by the fact that palæontology and related disciplines describe this world as it was before human beings existed. Of course this world would lack theoretical interest in a universe populated exclusively by creatures with cognitive capacities radically different from those of human beings. But what these disciplines describe is, nonetheless, such as to exist independently. (Smith, 1995, p. 644) A careful reading of the extract reveals that it leaves room to subsume the scientific reality or outlook while still allowing for a commonly held view or socially agreed reality and it reassures us that we do not require human cognition for this world to exist. Additionally, this school of philosophy allows for a difference between the reality and the appearance of reality. Often this is called the error that is involved in making sense of reality. Thus common-sense is not, in spite of its reputation, naïve; it draws a systematic distinction between reality and appearance, or in other words between the way the world is and the way the world seems or appears via one or other of the sensory modalities and from the perspective of one or other perceiving subject in one or other context. The thesis that there is only one world towards which natural cognition relates must thus be understood as being compatible with the thesis that there are many different ways in which the world can appear to human subjects in different sorts of circumstances. (Smith, 1995, p. 645) It is important, however, not to ignore the success in describing reality through science, and so we need to describe the relationship between the world as seen in common sense and that described by physics, the most closely related science. Later in this chapter, Smith relates commonsense realism with physics.



229

The common-sense realist must confront the question of the relation between the common-sense world and the world that is described in the textbooks of standard physics. Here again a number of different philosophical alternatives have been mapped out in the course of philosophical time, including that view that it is the common-sense world that is truly autonomous while the world of physics is to be awarded the status of a cultural artifact. Here in contrast, we assume a thesis to the effect that the commonsense world overlaps substantially with physical reality in the more standard sense. (Smith, 1995, p. 647) This is important because physics, and indeed science generally, cannot be discounted in the quest for determining what there is. Science has been very successful in determining what there is in the physical world and is in part responsible for constructing models of physical reality. As paradigm shifts in science come and go, so will the models we use to describe physical reality. Paradigm shifts clarify our understanding of the physical world often importantly at the margins. We are experiencing the ramifications of just such a paradigm shift that began a little under a century ago. Issues such as string theory, quarks, and quantum mechanics generally have arisen from this paradigm shift, just as notions of mass, force, momentum, gravity, and planetary motion accompanied the Newtonian paradigm shift of several centuries ago. One therefore needs to be careful not to overstate the importance of science when examining ontology. Nevertheless, commonsense realism has a place for such development and clearly refuses to discard scientific realities in order to allow for social structures and understanding. To date, the only ontology that has been adapted and used in information systems is a scientific ontology by Mario Bunge (1977, 1979). It is scientific in that it uses the results of the natural sciences and of systems theory in its design. Consequently, it requires reworking when paradigm shifts occur in science or when our understanding of how the natural world works changes significantly. It also takes a mechanistic or deterministic view towards the world and its physical and social structures. Being a scientific ontology, Bunge’s ontology is set at a level much more finely grained than an ontology based on common-sense realism. The two are clearly related as we explained earlier in this section. Both are likely to have a role in information science research with commonsense realism likely to be influential when considering socio-technical or social design issues and scientific realism is likely to be influential when considering purely technical or computerised parts of information systems.



Nevertheless the two ontologies have different but complementary realisms and it is therefore valuable to consider Chisholm as an alternative ontology as invited by earlier researchers.

Chisholm’s Ontology Roderick Chisholm (1957, 1976, 1979, 1982, 1989a, 1989b, 1996) has written extensively in the areas of ontology, metaphysics, and epistemology and these works provide a backdrop to his 1996 monograph. We are obviously restricted in covering Chisholm’s ontology in this chapter. Interested readers are encouraged to refer to the monograph by Chisholm (1996), or his earlier chapter (1992), for a comprehensive treatment of the ontology. Chisholm’s categories are organised into the taxonomy shown in Figure 1. Chisholm adheres to the theory that establishes the dichotomy dividing the world into entities that are “contingent” and do not have to exist, and those that are “necessary” entities and must exist in order for the theory to be consistent (Honderich, 1995). This latter group is also often referred to as abstract entities that nevertheless exist. This is part of his realism and is reflected in the first branch in Figure 1 where the entire universe consists of entities divided in those that are contingent and those that are necessary. The fundamental categories that are relevant to all types of modelling in information systems are shown in bold typeface. Additional terms are real and are defined in the ontology (relation, set/class) and are related back to the terms that label the fundamental categories. In studying data modelling (a sub-set of information systems modelling) states and events are not relevant. These are relevant for studies of the process models of the Object Modelling Technique (UML) and other modelling languages that model states and changes in states. Necessary substance and its state (God and His state, Chisholm, 1996) are also clearly not in the realm of this study. Similarly, we do not need to consider issues of boundaries between spatial individuals in data modelling (Where do I end and the chair upon which I sit begin?) and are therefore outside the scope of this work. Consistent with our earlier discussion, the nodes in the taxonomy are labelled by terms forming a subset of those found in the ontology, for example, the terms individual and attribute. In fact individuals and attributes are central to Chisholm’s ontology. Further, other terms that Chisholm’s ontology requires to make sense of “what there is” are defined with reference to these fundamental terms. For example, “individual” and “attribute” have descriptions that show not only their own nature, the terms class and relation and related (and defined) in Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


231

Figure 1. The categories for Chisholm’s (1996) ontology Entities Contingent

State

Event

Individuals

Necessary

States

Boundaries Substances

Non-states

Attributes Substance

terms of attributes. In this section we introduce the terms from Chisholm’s ontology that we use in ontological studies of data modelling languages. The paragraphs describing each term convey the concept associated with the term. We conclude with a tabular summary of the concepts.

Individual An individual is a discernible and transient object. It need not be material (or physical) in nature. Examples of individuals are an accountant named Freda, the annual financial statements for Ericsson, and Orly International Airport. Individuals are identified using attributes that only they exemplify, and may have constituents thereby giving them structure. This is called mereology (Honderich, 1995). Constituents may be other individuals (called parts) or may be boundaries (the other constituents). For example, consider Orly Airport. It has several renta-car franchises, bars, restaurants, and departure gates. Each of these is a part of Orly Airport and each is also an individual. In this example, most of these parts can be further sub-divided. On the other hand spatial substances have boundaries. A boundary is a surface, line, or point. For example, Orly Airport may have as its constituent surfaces that help to identify it as a spatial object. That surface is a boundary and is in turn made up of a number of surfaces, lines, and points. Boundaries of spatial substances are not of interest in data modelling.

Attribute An individual may exemplify attributes. Each attribute may be exemplified by many individuals. Orly Airport is very busy; Nokia’s balance sheet is good; Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Freda, our accountant, is of age 43. Some attributes may never be exemplified and others cannot be exemplified. For example, Orly Airport may never be green. We can be sure that Orly Airport cannot be a liquid. If two attributes are considered to be equivalent then it is the case that where one attribute is exemplified by an individual then so is the other. This is called conceptual entailment in the ontology. This can be illustrated by considering Orly Airport. The attribute very busy may involve a conceptual entailment with the attribute of having over a certain number of aircraft movements an hour. Chisholm allows for compound attributes, which in turn may consist of other compound attributes or simple attributes. He suggests that an attribute may be the conjunction or disjunction of several attributes. For example, the attribute of “being good” with respect to Nokia’s financial statements may be the conjunction of being in surplus (profit) and being of good credit rating. Chisholm also indicates that there may be alternative mechanisms for providing compound attributes, other than conjunction and disjunction. Philosophically and logically it makes little sense to talk about when an attribute came into being. In Chisholm’s ontology, attributes are enduring, thus avoiding the problem of declaring when an attribute comes into being. For example, when did the attribute “being green” first come into being? Since we cannot know and since raising its genesis brings about certain problems it is better to adopt the position that attributes are non-contingent, they exist perpetually.

Classification Classes and sets may be part of a state-of-affairs. In Chisholm’s ontology, attributes are used to restrict membership of sets and classes. Indeed, Chisholm’s ontology reduces the discussion of classes to the discussion of attributes by adopting Russell’s (1908) reduction of classes to attributes. This has the effect of building classes and sets from individuals through the exemplification of an individual’s attributes and not by constructing elaborate class structures. For example, suppose we are maintaining a taxonomy of plants. Periodically, the taxonomy may change quite drastically without a change in the majority of attributes exhibited by the plants involved. Using Chisholm’s ontology classes can change radically through a change in membership criteria based on attribute exemplification. Classes and sets can be selected based upon attributes that are conjunctions and disjunctions of other attributes, and in this sense complex class relationships can be realised, that are essentially class structures. The central point remains that individuals come together to form classes and are fundamental to the ontology. Classes are reflections of attributes exemplified by individuals due to the fact that they exemplify the attributes that are used to select the class. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


233

Relation Chisholm’s ontology allows for relations between individuals. Chisholm (1996) says, “To know what relations are, we must understand the concept of the direction of relations” (p. 51). Chisholm means that relations may not be reciprocated, or alternatively must be carefully considered from the viewpoints of all individuals concerned. For example, I may be interested in a job with Nokia. Nokia may not be interested in employing me. It is for this reason that relations are unidirectional. Further, in the ontology, relations are be represented by ordered pairs of identifying individuals. This comes from the fact that ordered pairs are related to sets of a specific form and therefore can be reduced to a discussion of attributes in the manner noted above. Therefore relations, despite being needed to describe a state-of-affairs, do not constitute a separate fundamental category but instead are related to attribute. For an ordered pair to represent unidirectional relations, attributes need to be found that uniquely describe and thereby identify each individual. For example, suppose that Freda (our accountant) is recruited to audit Nokia’s books then an attribute being an ordered pair of identifying attributes for Freda and Nokia would

Table 2. Relevant concepts from Chisholm’s ontology Concept Description Individual Chisholm allows for discernable and transient objects. These are Core called individuals. Individuals come into being (are created) and pass away (destroyed). In this sense they are transient. Identity Each individual possesses an attribute (or several attributes) that identifies it. Structure Individuals may have constituents. These are either other individuals (known as parts) or boundaries (the other constituents.) Individuals that make up parts of others are still thought of as being individuals. Attribute Attributes are exhibited by individuals. They are central to Chisholm’s Core ontology, after individuals. Further, attributes are enduring, in the sense that they do not come into being and do not pass away. Further, attributes must be loosely coupled with individuals. Conceptual Attributes can be equivalent in the sense that if something exhibits one Entailment attribute then it exhibits the other. Complexity Attributes may be simple or complex. Complex attributes are combinations of either simple or other complex attributes. The mechanism suggested by Chisholm is one involving conjunction and disjunction of attributes. He feels there may be other ways of providing for this complexity. Classification Classes and sets are provided using attributes, in the ontology. Core Specifically, it is through the attributes that membership of classes is determined. Relation Individuals may be related. Specifically, relations are attributes (an Core ordered pair). The ontology requires that attributes that identify the participating individuals are required. The relations are unidirectional (not bi-directional).



have to be exhibited by Freda that in turn represents the relation. A corresponding attribute representing the reverse relation would need to be exhibited by Nokia, if the relation were to be reciprocal. In the simplest case an individual may be related to another (binary). More complex relations between three individuals (ternary) or more (n-ary) are allowed. Mathematically it is proven that these all can be reduced to a series of binary relations (Quine, 1960).

Method The role of our methods is to compare and to contrast the ontological meta-model embodied in the data modelling languages with a reference ontology (in this case Chisholm’s ontology). In the discussion below we assume that we have not yet selected the ontology to be used and refer to a fixed but arbitrary ontology as our reference ontology. The reader can however, consider Chisholm’s ontology as our intended target. We present two methods for evaluating data modelling languages against a specific ontology: 1) the method of conceptual evaluation which can be applied to each specific data modelling language, and in turn forms the basis of the second method which is, 2) the method of conceptual comparison. The relationship between the two is explained in this section. In this chapter we present the results of an application of the latter method.

The Method of Conceptual Evaluation The aim of the method of conceptual evaluation is to compare the ontology embodied in a data modelling language with the reference ontology selected from the range of ontologies available. In conducting a conceptual evaluation, we are seeking to provide qualitative answers for specific data modelling languages to questions such as:

•

How well does the data modelling language capture reality relative to an ontology?

•

How similar are a range of data modelling languages?

As indicated in Figure 2 the inputs to the method of conceptual evaluation are the reference ontology and the ontology derived from the meta-language of the data modelling language. The output of the method is a list of similarities and



235

Figure 2. The method of conceptual evaluation

Reference ontology Conceptual Evaluation Data model ontology

Qualitative assessment of similarities and differences

differences between the two sets of concepts and a qualitative analysis of those similarities and differences. It is not necessary for either the ontology or the data modelling language to be described using a mathematical formalism. It is possible that both are at best semi formal with natural language descriptions of concepts found in each. The method of conceptual evaluation has four basic steps.

•

Step 1. Determine the set of concepts from the reference ontology to be used in a forward evaluation. This set of concepts we call the reference concepts.

•

Step 2. Determine the set of concepts from the ontology embodied in the data modelling language to be used in a backward evaluation. This set of concepts we call the data modelling concepts.

•

Step 3. Perform a forward and backward evaluation of the two sets of concepts and tabulate the results.

•

Step 4. Perform the analysis step in which the results are analysed.

The first step is to determine the basic set of concepts on which the forward evaluation will be based. The method does not prescribe which set of concepts from the reference ontology should be chosen. For example, one may wish to study the concept of state in UML (Rumbaugh, Blaha, Premerlani, Eddy, & Lorensen, 1991) by reference to Chisholm’s ontology, and consequently the fundamental concepts of states and events from Chisholm’s ontology may be the only relevant concepts that need to be considered for such a limited study. Nevertheless, the chosen concepts must be appropriate for the modelling language under study. In this study, for example, only the static or structural



concepts are required because that is the common nature of the data modelling languages under examination. The second step resembles the first and involves determining the set of concepts from a particular data modelling language. Each data modelling language will have a different group of concepts using which sense is made of reality. For example, the ER model uses different terms from that used by UML. It is likely that there will be a degree of similarity in the concepts associated with terms from those languages. The third step involves the comparison of concepts from each of the reference ontology and the ontology embodied in concepts from a data modelling language. It is performed utilising concepts from the data modelling language as well as the reference ontology according to our philosophy — the reference ontology is not the only ontology that could be chosen and is not the only theory in that other reference ontologies could be used. Nor is the ontology necessarily better than that embodied in any data modelling language. Further, the comparison is at the level of concepts thus moving beyond the specific names or terms used to signify the concepts. Additionally, this step is highly subjective — there is no other way to undertake a conceptual evaluation of this nature. The presentation of the results of the evaluation utilises semiotic theory for two reasons. First, terms and concepts are clearly semiotically related. Second, comparison of concepts is semantic with semiotic theory providing an ideal basis for explaining semantic differences in terms. The relationship between terms in an ontology and their concepts are explained through semiotics: each term, through its associated concept in a reference ontology or the ontology of a data modelling language, spans part of a semantic field (Eco, 1976), or conceptual plane (Cruse, 2000; Culler, 1976). Alternatively, each term from an ontology possesses an essential depth (Liska, 1996) which similarly evokes the conceptual span of a term. In this chapter we adopt the term “semantic field” to label these ideas and use it to express the similarities and differences between concepts in the reference ontology and those embodying the ontology from the data modelling language. Specifically, we use a graded indicator to express the similarities and difference. When comparing a concept c (from the ontology) with a specific data modelling language, there are three broad categories of results. These categories are consistent with the semiotic texts we quoted above. First, the data modelling language may have total overlap with respect to c. Total overlap may be provided by one concept (for example, d) or perhaps by several concepts (for example, two concepts d and e). That is, there may be one concept or several concepts that together provide total overlap, in terms of semantic field, with the concept from the ontology. The second possibility is where the overlap is partial. Finally, it may be that there is no overlap at all between the data modelling language and c from the ontology. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


237

Figure 3 shows the three categories of results pictorially. While the coverage of a specific concept is depicted in this figure as a sharp rectangle, the nature of semantic fields dictates that the boundaries between semantic fields are quite imprecise. This emphasises the fact that the comparison is conceptual and that concepts may be partially covered and that a simple presence absence is not ideal for ontological evaluations of this nature. Each of these categories of results can be indicated using symbols so that an idea of the results of the comparison can be conveyed easily in tabular form. This is called the indicative results. The three symbols for full coverage, partial coverage, and no coverage are (√), ( √p), and (X) respectively. For compound concepts a summary for the complete concepts can be calculated. This is shown in Table 3. The summary (concept level) result is shown as the “additive of results” for the parts of the complex concept made up of parts a and b. The second dimension of final step in the method is the qualitative result of evaluating a data modelling language using an ontology revealing the story behind the indicative results from step 3. The analysis of the qualitative results presents issues beyond the direct comparison of concepts and discusses issues such as the nature of the gaps in coverage that are evident from the results as presented in step 3 and the implications of these on the data modelling language under study.

Figure 3. Degree of overlap in coverage of semantic field

c

d

d

c

c

d

c

d

c

e

e Ãp

Ã

X

Table 3. Results for a compound concept, with parts a and b Concept Result Part A Part B Key: X – no coverage

X X X

√p √ √p

√p √ X

√p √p √p

√ p – partial coverage

√ √ √

√ – full coverage



Our method is related to work by Wand and Weber (1989) where they undertook an informal comparison between tools and an ontology. This informal comparison later progressed to be a more formalised (1993) understanding of the representational clarity with which a model of reality is created from human perceptions using a specific tool. In this, they examine the grammars that information systems analysis and design methods provide to describe aspects of the real world. A grammar in this context “generates a language, which is a set of strings over some alphabet.…In these grammars, sentences provide a graphical representation of some real-world phenomena” (Wand & Weber, 1993). In their latter more formal treatment, Wand and Weber use ideas of ontological clarity and ontological completeness to establish a measure of the ontological expressiveness of an analysis and design tool as represented by its grammar as it compares with the grammar representing the ontology. These measures are defined using construct (or term) mapping to and from the ontology for clarity and completeness respectively. Their measures are based on the presence or absence of terms in a grammar representing the modelling tool when compared with terms in a grammar from the ontology. Both grammars and terms are required to be expressed mathematically. These concepts bear a degree of similarity to those used in our method. However, they do not explore the more fundamental question of the qualitative differences or similarities in world-view between the ontology and the various tools under examination that are uncovered by examining the subtle differences in meaning of the various terms found in the ontology and in the tools. Instead their comparison is based on the mapping (or failure in mapping) of mathematical constructs to and from the ontology. Their approach also requires that the ontology selected be precise and defined mathematically. Not all ontologies are capable of being defined mathematically due to mathematics failing to adequately represent reality. The modelling tool under investigation must similarly be precisely defined. Formalizing the modelling tool in a grammar compromises the meaning attached to terms. Terms may have their meaning restricted by formalising them. Data modelling languages often lack formality. It is for these reasons that we have included semiotic theory to express the results of each ontological evaluation of data model. We further note that as a result, and of necessity, the method is highly qualitative and has a degree of subjectivity.

The Method of Conceptual Comparison The method of conceptual comparison seeks to compare a number of data modelling languages by analysing the results of conducting a series of conceptual evaluations against the selected reference ontology. The method consists of Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


239

repeated applications of the method of conceptual evaluation against a number of data modelling languages. The results indicate the degree to which the reference ontology is reflected in the ontology of a range of data modelling languages and utilises the reference ontology as a benchmark against which the data modelling languages can be assessed. In conducting the series of conceptual evaluations we are testing each language against the selected and independent view of reality as represented by a reference ontology. As a direct consequence, the method of conceptual comparison can be used to determine how wide spread and to what degree a reference ontology is reflected in the ontologies implicit in a range of data modelling languages. The analysis of results sheds light on the ontological overlap or dominance of the reference ontology with a range of data modelling languages. A reference ontology is also a measure of how well something represents reality in the way encapsulated in the ontology. In the following section we present the results of such a conceptual comparison with a range of data modelling languages the range of languages spanning 20 years of scholarship. In accordance with the method, it is constructed from a number of conceptual evaluations using Chisholm’s ontology and the ontology implicit in the respective data modelling language.

Results We have selected five data modelling frameworks from the literature to use in this work. These data modelling languages span the period from the beginning of the semantic data modelling to its extension into the world of object data modelling. First, the Entity-Relationship (ER) model (Chen, 1976). It is one of the most important entity-relationship-attribute style of modelling framework and is still extremely popular in industry. Second, the Functional Data Model (FDM) (Kerschberg & Pacheco, 1976; Shipman, 1981) is the cleanest example of the functional-style of semantic data model. Third, we include the Semantic Data Model (SDM) (Hammer & McLeod, 1981) recognised to be an important model (Hull & King, 1987; Peckham & Maryanski, 1988) and may be considered to be a significant model with respect to object data model development. Fourthly, NIAM (Nijssen & Halpin, 1989). Finally, the object model UML (Blaha & Premerlani, 1998) is included as a significant object model that is used in contemporary information systems development yet, in part, originating from the semantic data modelling stream. Table 4 shows the indicative results for the comparison of Chisholm’s ontology with the data modelling languages.



Table 4. Indicative results of the comparison of selected data modelling languages using Chisholm’s ontology Ontological Concept

ER

FDM

Individual

√p √ √ X √p √p X √p √p √p

√p √ √ √p √p √ X √ √ √

Core Identity Structure Attribute Core Conc. Entailment Complexity Classification Relation Key: X – no coverage

SDM √ √ √ √ √p √p X √ √p √

√p – partial coverage

NIAM √p √ √ X √p √ X √p √p √p

UML √ √ √ √ √p √p X √ √p √p

√ – full coverage

Summarising from an earlier section, Chisholm’s ontology views the world as a collection of individuals and relations between them. Individuals have structure and represent ontologically distinct entities. Attributes in the ontology are used to describe and identify individuals and using identity of individuals, describe relations. Further, attributes are universals in a philosophical sense and endure, and, by inference they are loosely coupled with individuals. Attributes are also used to determine class and set membership. Relations are also seen as being uni-directional. Our conceptual comparison, summarised by Table 4, suggests that the worldview described by the reference ontology is to a large extent a similar world-view to those imparted by the languages and there is a significant level of agreement with the ontology and the modelling languages that we’ve studied, but the data modelling languages lack the full generality of Chisholm’s ontology. The departures are in the structural aspects of individuals and more subtly, attitudes to the nature of attributes and relations and the implications of a lacking of loose coupling between individuals and attributes (particularly implications concerning classification). We examine each in turn. Individual structure represented by part-whole relationships defined at the individual level is supported by some but not all data modelling languages. Those modelling languages not showing support provide a crude mereology at the class level. This approach is useful in cases where structure cannot be generalised to the class level due to a high degree of diversity in the structure of individuals grouped. Apart from this difference the remainder of the concept “individual” is supported. The core of the concept “attribute” is partly supported by most data modelling languages. This is due to the lack of support by most data modelling languages of loose coupling between attributes and individuals by three data modelling Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


241

languages. Loose coupling allows for a high degree of diversity in the types of attributes that similar individuals exemplify. Consequently, individuals where such diversity is evident are no longer constrained to classes of like individuals all exemplifying a fixed range of types of attributes. Some data modelling languages (FDM and NIAM) support the loose coupling evident in Chisholm’s approach; the remainder do not. Nevertheless most have a high degree of coverage of the core and complexity parts of attribute. Attribute equivalence is completely absent from all data modelling languages. Classification and relations are concepts recognised by all data modelling languages. Classification in the ontology is evident through the attributes exemplified by members of classes. In the ontology, classes are related to each other by the intersections and unions of the attributes used to select them and thereby can simulate class hierarchies. This approach to classification structure is entirely different from the most common classification approaches used by most data modelling languages where instead, rich and rigid class hierarchies are prevalent. Essentially, in the ontology there exists a sea of individuals from which classes are built. This provides flexibility in cases where the classes of individuals that are required change drastically without very much changing in the nature of the individuals themselves. The concept of relations between individuals is supported by all data modelling languages. However, the reference ontology also requires relations to be unidirectional, thus allowing for non-reciprocation of relations. We have found that relations are not bi-directional in several of the data modelling languages but the concept is fully supported by FDM and SDM. The consequence of the departures from the ontology by the data modelling languages that we have observed is that it is likely one can model a narrower range of situations using the studied data modelling languages than Chisholm’s ontology. Further, Chisholm’s ontology has the potential to change our view of data modelling by its increased flexibility achieved through bi-directional relations and through its loose-coupling of attributes with respect to individuals. In turn, this has positive implications for the flexibility of models that are subject to radical change. It is the formation of classes through attributes as a direct consequence of loose coupling that is most beneficial for flexibility. We can see from the results that the modelling languages share the world-view of the ontology to a large degree. The areas of departure are of the nature of a difference in emphasis rather than complete absence of support. In the case of complex concepts all modelling languages support the core to a high degree of coverage. On the basis of this, we can say that the world-view held through the ontology is substantially similar to that held by this representative range of data modelling languages.



Conclusions, Future Work, and Reflection In this chapter we have compared Chisholm’s ontology with the ontologies implicit in a representative range of data modelling languages using qualitative methods to see how well data models “map the world”. We have found that there is a significant degree of overlap between Chisholm’s ontology and the data modelling languages selected. Indeed, we have found that the ontology’s core elements are reflected in the range of data modelling languages selected. Recall that Chisholm’s ontology consists of individuals and their structure, the attributes that they exemplify and the relationships that exist between individuals. At the beginning of the chapter we posed two research questions. They were:

• •

How well do data models represent reality? What are the similarities and differences between data modelling languages?

We discuss these in the following sub-sections.

Representing Reality We can conclude based on the results presented in this chapter that the data models generally overlap with the core concepts of Chisholm’s ontology to a large degree and that the world-view encapsulated in Chisholm’s ontology is broadly consistent with the world-view implicit in the data modelling languages at least as far as the terms used for comparison are concerned. Chisholm’s ontology is one of commonsense realism and is categorised by Chisholm as being a realistic ontology. We have found, in our reading (Audi, 1995; Flew, 1989; Honderich, 1995), that the terms used by Chisholm are widely supported, and the style of realism upon which his ontology is based has a high degree of consensus between philosophers. We are confident that this core of consensus, which is considerable, forms a good starting point from which to progress towards a detailed view of reality that is shared between the data modelling languages. We have found the following forms such a consensus and can be summarised as a realistic core: 1.

Individuals that are ontologically independent.

2.

Attributes that are exemplified by individuals.



3.

Part-whole relationships exist between individuals (mereology).

4.

Relations exist between individuals.

5.

Classes of individuals are selected on the attributes exemplified.

243

All of the data modelling languages showed good support for all of these concepts. Consequently, we conclude that the data modelling languages studied, by having a large degree of overlap with a typical commonsense realist ontology, represent reality well. This is because an ontology (such as Chisholm’s or Bunge’s Ontology) is a theory that discusses what there is in reality and in the most general sense.

Similarities and Differences Between Data Modelling Languages The similarities between the data modelling languages is summarised by the realistic core (the aforementioned list). However, we can see three areas where there are differences between the modelling languages and the view of Chisholm’s ontology. First, in the ontology, attributes are considered to be quite separate from the individuals that exemplify them, and they endure. One can describe this as loose coupling between individuals and attributes. Some, but not all, of the data modelling languages contrastingly group individuals into homogeneous classes and by so doing restrict the range of attributes that each individual can exhibit thus exhibiting a tight coupling. This tight coupling is also counter to the realistic core. Second, in data modelling languages the data modelling equivalent of individuals are seldom allowed to be members of several classes simultaneously. The only exception to this is in cases where a class hierarchy around a common family of classes (with UML) is established or where individuals are distributed or dispersed across several classes (with SDM). Both of these are different from the ontology in which individuals can be member of more than one class simultaneously because classes are established based upon attributes that member individuals exhibit. The classes are not repositories for individuals, and instead they filter individuals. Third, in contrast to class hierarchies there is support for part-whole (or aggregation) structures in the ontology, known in philosophy as mereology. This also supports related research that uses a different ontology. Many of the modelling frameworks supported this to at least a crude degree, such as with the ER modelling framework where the part-whole relationships are expressed at the class level. Others had more sophisticated methods of providing this, such as Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


with most object modelling frameworks or with SDM where individual-based mereology is present.

Future Work and Critical Analysis There are some issues that need further investigation. The importance of conceptual entailment and the implications of the enduring and abstract nature of attributes need to be investigated in practical situations because of their potential to influence efficiency and effectiveness of implemented databases. There is also a question about the nature and usefulness of rigid class hierarchies found in many modelling languages. The ontological evaluation of modelling languages requires a deep understanding of the ontology and of each data modelling language. Further, it is on the basis of the terms described in the ontology that the comparison with each candidate modelling language is undertaken. It takes time to understand an ontology as condensed as Chisholm’s and to relate each term and its associated concept to modelling languages. We cannot conceive of an easier way in which to undertake such a work. Further, there is clearly a critically important interpretive dimension to the methods and consequently the results of the comparison may vary to a degree between researchers. This, however, is the very nature of the type of research and is not grounds enough to doubt its applicability. There is no algorithmic comparison of concepts. However, given that we are interested in comparing, contrasting, evaluating, and ultimately improving our data modelling languages, our approach is quite appropriate. Further, the approach has the potential to greatly enhance our understanding of the nature of data modelling languages by permitting analysis using qualitative ontologies such as Chisholm’s.

How to Better Map the World There is a much deeper reflection possible from applying Chisholm’s ontology to data modelling. This reflection cuts to the core of the discipline of information systems and brings together reflection in philosophy into the nature of reality and our interaction with it. Chisholm’s ontology is one of common-sense realism. This realism has as one of its traits the recognition that not only does the natural world, as understood by science, exist outside our minds, so too does the social world we inhabit. Further, we can find a social reality that is to a large extent arbitrary (in the sense that two alternative realities could be conceived of and implemented for social existence) and made manifest at least partly through institutions, their rules, and objects. Examples of social reality and institutional reality abound from currenCopyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


245

cies to accounting practice and from social etiquette to political systems. These are somewhat arbitrary and can come into conflict with one another. However, they are also real for people. The very subject we study as information systems researchers and practitioners, the company, or organization, is not real in a naturalist sense yet is very real for the people for whom we design information systems. We can go further and say that in many cases, there is not physical reality that corresponds to it. It then seems very reasonable, indeed of critical importance, to venture beyond naturalistic ontologies when considering such issues. Chisholm’s ontology, being one of common-sense realism, has a clear potential to help data models better map the world inhabited by customers of information systems in three ways. First, and as we discussed at the beginning of this chapter, it subsumes a naturalist ontology such at that expounded by Bunge and thus is equally powerful compared with Bunge’s ontology. Second, in-line with its common-sense roots, it allows for a social reality (Searle, 1995) and, importantly for us, institutional reality (in a social sense of institution) and demonstrates much more clearly its ability to handle social reality than Bunge’s ontology does. Finally, it brings to the discussion a perspectival realism that goes beyond the frames of reference found in Bunge’s ontology. Perspectival realism is important in that it gives us the ability to view reality from multiple perspectives and also gives us the ability to explain appearances of reality. Given the current interest in ontologies for achieving a range of computerised functions, common-sense realism also provides the best style of reference ontology for use in information systems foundational research. By applying a realism to information systems that is broader than Bunge’s realism, we have the tools by which we can help data models “better map the world”.

References Audi, R. (Ed.). (1995). The Cambridge dictionary of philosophy. Cambridge, MA: Cambridge University Press. Blaha, M. & Premerlani, W. (1998). Object-oriented modeling and design for database applications. Upper Saddle River, NJ: Prentice Hall. Bunge, M. (1977). Treatise on basic philosophy: Vol. 3: Ontology I: The furniture of the world. Boston: Reidel. Bunge, M. (1979). Treatise on basic philosophy: Vol. 4: Ontology II: A world of systems. Boston: Reidel. Chen, P. (1976). The entity-relationship model — Toward a unified view of data. ACM Transactions on Database Systems, 1(1), 9-36. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Chisholm, R. (1957). Perceiving: A philosophical study. Ithaca, NY: Cornell University Press. Chisholm, R. (1976). Person and object: A metaphysical study. La Salle, IL: Open Court. Chisholm, R. (1979). Objects and persons: Revisions and replies. Grazer Philosophische Studien, 7/8, 317-388. Chisholm, R. (1982). The foundations of knowing. Minneapolis, MN: University of Minnesota Press. Chisholm, R. (1989a). On metaphysics. Minneapolis, MN: University of Minnesota Press. Chisholm, R. (1989b). Theory of knowledge (3rd ed.). Englewood Cliffs, NJ: Prentice-Hall. Chisholm, R. (1992). The basic ontological categories. In K. Mulligan (Ed.), Language, truth, and ontology (p. 211). Dordrecht, Germany: Kluwer Academic Publishers. Chisholm, R. (1996). A realistic theory of categories — An essay on ontology (1st ed.). Cambridge, MA: Cambridge University Press. Cruse, D. A. (2000). Meaning in language: An introduction to semantics and pragmatics. Oxford: Oxford University Press. Culler, J. (1976). Saussure. Fontana. Dancy, J. & Sosa, E. (Eds.). (1992). A companion to epistemology. Oxford: Blackwell Publishers. Eco, U. (1976). A theory of semiotics. Bloomington, IN: Midland. Flew, A. (1989). An introduction to Western philosophy: Ideas and arguments from Plato to Popper (fully revised edition of the original 1971 volume ed.). London: Thames and Hudson. Green, P. (1996). An ontological analysis of information systems analysis and design (ISAD) grammars in upper case tools. Unpublished PhD thesis, The University of Queensland. Guarino, N. (1998). Formal ontology and information systems. Paper presented at the Formal Ontology in Information Systems, Trento, Italy. Hammer, M. & McLeod, D. (1981). Database description with SDM: A semantic database model. ACM Transactions on Database Systems, 6(3), 351-386. Honderich, T. (Ed.). (1995). The Oxford companion to philosophy. Oxford: Oxford University Press. Hull, R. & King, R. (1987). Semantic database modelling: Survey, applications, and research issues. ACM Computing Surveys, 19(3), 201-260. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


247

Kerschberg, L. & Pacheco, J. E. S. (1976). A functional database model. Rio de Janeiro, Brazil: Pontificia Univ. Catholica do Rio de Janeiro. Kim, J. & Sosa, E. (Eds.). (1995). A companion to metaphysics. Oxford: Blackwell Publishers. Liska, J. J. (1996). A general introduction to the semeiotic of Charles Sanders Peirce. Bloomington, IN: Indiana University Press. Milton, S. K. (2000). Ontological studies of data modelling languages. Unpublished PhD dissertation, The University of Tasmania. Milton, S. K., Kazmierczak, E., & Keen, C. (2001). Data modelling languages: An ontological study. Paper presented at the 9th European Conference on Information Systems, Bled, Slovenia. Nijssen, G. M. & Halpin, T. A. (1989). Conceptual schema and relational database design: A fact oriented approach. New York: Prentice-Hall. Peckham, J. & Maryanski, F. (1988). Semantic data models. ACM Computing Surveys, 20(3), 153-189. Quine, W. V. O. (1960). Word and object. Cambridge, MA: MIT Press. Rohde, F. (1995). An ontological evaluation of Jackson’s system development model. Australian Journal of Information Systems, 2(2), 77-87. Rumbaugh, J., Blaha, M., Premerlani, W., Eddy, F., & Lorensen, W. (1991). Object-oriented modeling and design. Englewood Cliffs, NJ: PrenticeHall. Russell, B. (1908). Mathematical logic as based on the theory of types. American Journal of Mathematics, 222-263. Searle, J. (1995). The construction of social reality. New York: The Free Press. Shipman, D. W. (1981). The functional data model and the data language DAPLEX. ACM Transactions on Database Systems, 6(1), 140-173. Smith, B. (1995). Formal ontology, commonsense and cognitive science. International Journal of Human-Computer Studies, 43(12), 641-667. Thalheim, B. (2000). Entity-relationship modeling: Foundations of database technology. Berlin, Germany: Springer-Verlag. Vet, P. E. v. d. & Mars, N. J. I. (1998). Bottom-up construction of ontologies. IEEE Transactions on Knowledge and Data Engineering, 10(4), 513526. Vickery, B. C. (1997). Ontologies. Journal of Information Science, 23(4), 277-286. Wand, Y. (1996). Ontology as a foundation for meta-modeling and method engineering. Information and Technology Software, 38, 182-287. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Wand, Y. & Weber, R. (1989). An ontological evaluation of systems analysis and design methods. In E. D. Falkenberg & P. Lindgreen (Eds.), Information systems concepts: An in-depth analysis (pp. 79-107). Amsterdam: Elsevier Science Publishers B.V. Wand, Y. & Weber, R. (1990). An ontological model of an information system. IEEE Transactions on Software Engineering, 16(11), 1282-1292. Wand, Y. & Weber, R. (1993). On the ontological expressiveness of information systems analysis and design grammars. Journal of Information Systems, 1993(3), 217-237. Weber, R. (1997). Ontological foundations of information systems (Vol. Monograph #4). Blackburn, Victoria: Buscombe Vicprint.


Applying the ONTOMETRIC Method 249

Chapter IX

Applying the ONTOMETRIC Method to Measure the Suitability of Ontologies Asunción Gómez-Pérez, Politécnica University of Madrid, Spain Adolfo Lozano-Tello, Extremadura University, Spain

Abstract In the last years, the development of ontology-based applications has increased considerably, mainly related to the Semantic Web. Users currently looking for ontologies in order to incorporate them into their systems, just use their experience and intuition. This makes it difficult for them to justify their choices. Mainly, this is due to the lack of methods that help the user to determine which are the most appropriate ontologies for the new system. To solve this deficiency, the present chapter proposes a method, ONTOMETRIC, which allows the users to measure the suitability of existing ontologies, regarding the requirements of their systems. ONTOMETRIC, based in the analytic hierarchy process, can be used to select the most Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

250 Gómez-Pérez & Lozano-Tello

appropriate ontology among various alternatives. This chapter describes the main techniques and activities to apply the method.

Introduction: The Problem of Selecting Ontologies In 1991, the ARPA Knowledge Sharing Effort (Neches, 1991) revolutionized the way in which intelligent systems were built in artificial intelligence when proposing the construction of knowledge-based systems by means of the “assembling” of reusable components. Reusable components become the base (or skeleton) of the new system, to which are added specialized knowledge and specific reasoning methods, depending on the task that the system attempts to solve. This vision allows the building of bigger and more potent systems. The ontologies, used to represent the “static” knowledge of a domain, and the problem solving methods, used to carry out reasoning, become the key pieces that allow the reuse of knowledge and problem-solving methods (Gómez-Pérez, 1999a). The saving in costs and time that is obtained in the software reuse (Bollinger, 1990; Poulin, 1997) is achieved in more scope in the reuse of this knowledge (ontologies and problem-solving methods), due to the enormous effort in the processes of knowledge acquisition of a domain, conceptual model’s construction, formalization, and implementation of such knowledge. At the moment, the ontologies are implemented in a great variety of languages. At the beginning of the 90s, a group of languages was designed and used for the implementation of ontologies. The most representative languages are Ontolingua (Gruber, 1993), LOOM (McGregor, 1991), OCML (Motta, 1999), FLogic (Kifer, 1995), and so forth. These languages receive the name of “classic languages” (Corcho, 2000), they follow a syntax based on LISP (to exception of FLogic), and they are in a phase of stable development. Recently, XML has been adopted as a standard language to exchange information on the Web. In the field of the ontologies, several languages have been created based on XML to implement ontologies. For example RDF (Lassila, 1999), RDF Schema (Brickley, 1999), XOL (Karp, 1999), SHOE (Luke, 2000), OIL (Horrocks, 2000), DAML+OIL (Horrocks, 2001) and OWL (Dean, 2003). These languages, called “Web-based languages”, are still in development phase and in continuous evolution. Equally, methodologies for building ontologies have been numerous. Already in 1990, Lenat and Guha (1990) published some methodological considerations related with the development of the CYC ontology. Some years later, in 1995, Uschold and King (1995) published the main steps in the development of the Enterprise ontology. In the same year, Grüninger and Fox (1995) showed the Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


methodology used in the development of the TOVE ontology (Virtual Toronto Enterprise). One year later, Uschold (1996) carried out a proposal of unification of both methodologies. In the 12th European Conference for Artificial Intelligence, the methodology used to build the project Esprit KACTUS project’s ontologies (Bernaras, 1996) was presented. In 1997, METHONTOLOGY, appeared (Fernández, 1997), which was extended later (Fernández, 1999a; Fernández, 2000). It proposed the steps that should be continued to build ontologies, some guides to carry out ontologies reengineering (Gómez-Pérez, 1999b), and ontologies evaluation (Gómez-Pérez, 1999c). Also in 1997, the methodology used to build domains ontologies from the SENSUS ontology was presented (Swartout, 1997). All these methodologies do not consider the cooperative development of ontologies. The first methodology that includes development aspects in-group is Co4 (Euzenat, 1995). A comparative study of some of these methodologies appears in Fernández (1999b). Since 1996, there was an important increase in the development of technological platforms related with the ontologies. The first ontology site was the Ontolingua Server (Farquhar, 1996), of the Knowledge Systems Laboratory (KSL) at Stanford University. In 1997, Ontosaurus appeared (Swartout, 1997), developed by the Information Sciences Institute (ISI) in the University of South California. Later, several tools were created based on Java technology: WebOnto (Domingue, 1998) developed in the Knowledge Media Institute (KMI) of the Open University (UK); OILed (Bechhofer, 2001), developed in the IST OntoKnowledge project; OntoEdit (Staab, 2000) developed by the AIFB of the Karlsrhue University; Protégé2000 (Noy, 2001) developed by the Stanford Medical Informatics (SMI) at Stanford University; and WebODE (Arpírez, 2001) developed at the Politécnica University of Madrid. In spite of the great increase that the use of ontologies has acquired, nowadays, the knowledge engineers need to look for ontologies dispersed in quite a few Web servers. When they find several that can be adapted, they should examine their characteristics attentively and decide upon the best ontologies to incorporate them to their system. This election procedure usually depends on the experience and the engineer’s intuition. If the system is being developed with commercial goals, it will be very difficult for them to justify the taken election. Although most of the methodologies for building ontologies (Fernández, 1999b) propose a phase of ontology reuse, there are no works that show the users how to choose ontologies for a new project, and there are no methodologies that quantify the suitability of these ontologies for the system. This election problem would be palliated if there existed a metric that quantified, for each one of the candidates (ontologies), how appropriate they are for a new system. The method that is described in this work (ONTOMETRIC) presents the set of processes that the user should carry out to obtain these measures.



This chapter is organized as follows. The second section gives a general overview of ONTOMETRIC method. The next section presents existing frameworks of characteristics to compare ontologies. After this, the multilevel framework of characteristics to select ontologies is described. The next section exposes the main steps of the analytic hierarchy process (AHP). The section after this describes how we have adapted AHP in the choice of ontologies (ONTOMETRIC method). Finally, the last section indicates the evaluation processes to prove the applicability of ONTOMETRIC.

Overview of the ONTOMETRIC Method The ONTOMETRIC method allows users to measure the suitability of the existent ontologies, regarding requirements of their systems. We have developed the following tasks to complete the method.

•

Identification of a multilevel framework of characteristics. The framework consists of 160 characteristics that describe the ontology domain. They are classified in the following dimensions: the content represented in the ontology, the language in which the ontology is implemented, the methodology followed to develop it, the tools used for building it, and the costs of using the ontology in the system. This framework provides: a) the outline to represent the information of existing ontologies, b) the comparison of ontologies, and c) the choice of the most convenient according to the requirements of the new system.

•

The building of the conceptual model of a domain ontology about ontologies, the reference ontology (RO), based on the multilevel framework of characteristics. The conceptual model of the RO was built following METHONTOLOGY methodology and WebODE platform. The dimension “content” of this ontology is instantiated with information coming from ontologies stored on ontology servers and available on the Web. The other dimensions were instantiated taking information from previous works, as well as analyzing the languages, methodologies, and tools, directly.

•

Design of a method that measures the suitability of a set of candidate ontologies that could be incorporated into a new system. The method uses: a) the multilevel framework of characteristics, b) the conceptual model of the RO and its instances, and c) an adaptation of the analytic hierarchy process. For each candidate ontology, the method gets a quantitative measure of its suitability. It is useful to decide, in a justified way,



what ontologies are the most appropriate for the application that the user is going to develop.

•

The creation of the technological support to assist the method. We have developed the OntoMetric Tool to calculate (using a modification of the analytic hierarchy process) the measurement of suitability for the candidate ontologies. This tool assists users in all the processes of the ONTOMETRIC method.

Existing Studies and Frameworks of Characteristics There are different studies on identifying characteristics for designing, comparing and classifying ontologies. The more elaborated proposals tend to organize the groups of characteristics in a taxonomical fashion. A summary of the proposals, with the number of characteristics and the purpose for which were created, is shown in the Table 1. On the one hand, the five characteristics of Gruber (1995) and the three characteristics pointed out by Uschold and Grüninger (1996) are very general features and were described as fundamental properties that should be considered in the design of ontologies and that should be kept in mind in the reuse process. The comparison framework of Noy and Hafner (1997) gathers 28 characteristics about ontologies, although the definition of some characteristics is not very precise and some features can include others. With this framework, they indicated the differences and similarities among 10 chosen ontologies. The aim of this study was to compare the different alternatives and designs of ontologies to clarify the description of a standard of ontology construction. However, some characteristics can have quite confusing values. Also, this framework is not appropriate for classifying ontology according to the identified characteristics.

Table 1. Summary of works related to characteristics of ontologies Author Gruber Uschold and Grüninger Noy and Hafner Hovy Uschold

Year Number of characteristics 95 5 96 3 97 28 97 36 98 10

Purpose To establish design criteria To establish design criteria To study several ontology designs To compare linguistic ontologies To identify the ontology roles in applications



Hovy’s (1997) framework proposes a taxonomy of 36 characteristics about different aspects of ontologies. This framework was outlined for the purpose of comparing ontologies for natural language processing. The study was carried out like an unfinished technical report in the ISI. For this reason, some of the characteristics are not defined, some characteristics are not clearly defined, some definitions contradict the examples, and several characteristics are only relevant to natural language issues. The 10 dimensions of characteristics indicated by Uschold (1998) were gathered from the point of view of the role that the ontologies play in an ontology-based application. The purpose of this framework was that new developers of ontology-based applications could use this information to build applications with the same requirements. The definitions of the characteristics of this framework are quite imprecise and, with the proposed framework, it is difficult to classify the ontologies appropriately. To conclude, although there are some frameworks that identify characteristics related with the ontologies, they are only adequate to classify them partially. In addition, they are not useful to compare the suitability of several alternative ontologies with regard to the necessities of an application, because these frameworks were not conceived for this purpose. So, none of the aforementioned frameworks are appropriate for deciding which of the different ontologies is the best for using in a system. For these reasons, we have developed a framework that allows for the description and the comparison ontologies.

A Multilevel Framework of Characteristics to Compare Ontologies In order to solve the aforementioned problem, we propose a complete taxonomy of 160 characteristics, multilevel framework of characteristics, which provides the outline for choosing and comparing existing ontologies. The framework is used, on the one hand, as a representation template of the information about existing ontologies. On the other hand, it helps the user select the necessary requirements that should complete the ontologies that will be considered candidates. Finally, it is the skeleton used to build the multilevel tree of characteristics used in the processes of the ONTOMETRIC method for the evaluation of candidate ontologies. The multilevel framework of characteristics possesses, in the superior level of the taxonomy, five basic aspects on the ontologies that are denominated dimensions. The dimensions are defined through a set of factors, as shown in the next tables. The factors are the fundamental elements that should be analyzed Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


to obtain the value of the dimensions. These factors are defined through a group of characteristics that allow the calculating of the value of their suitability. These characteristics can be defined, recurrently, by means of other characteristics, or more specific sub-characteristics, so that they describe them with more detail. It allows the analyzation of the criteria with more depth. Thus, the multilevel framework of characteristics is organized taxonomically, with dimensions described by factors, the factors by characteristics, and these by other more specific sub-characteristics. The term “criterion” refers indistinctly to any element of the multilevel framework, that is: dimensions, factors, characteristics, and sub-characteristics of the taxonomy. The multilevel framework of characteristics can be represented like a tree, so that the criteria placed in son-nodes describe and represent the father-node’s properties. Thus, the users will be able to extend or prune the criterion that they consider opportune, so that the new tree depends on the particularities of the project, the business, and the organization that will reuse the ontology. This tree is called the multilevel tree of characteristics (MTC) and can be represented in a graphic form as appears in the Figure 1. It should be kept in mind that the framework is subject to the conceptual and technological novelties that will appear in the future in the ontology field. In this sense, the MTC constitutes a set of “living” criteria that should be actualized according to the produced changes.

Figure 1. Representation of the multilevel tree of characteristics Ontology suitability Content

Language

Domain knowledge

Methodology

Costs

Tool

Inference mechanism

st

1 level: dimensions nd

2 level: factors

rd

Reasoning potential

Inference engine

...

3 level: characteristics

... level n: leafcharacteristics



We have identified the factors and characteristics of the dimensions (content, language, tool, methodology, and costs) from the referenced papers in the third section and from works about content of the ontologies, implementation languages, development methodologies and environments of ontologies, and costs. We have added others to complete the framework. The result of this task is a comparison framework of ontologies that serves as baseline work for the proposed method. Tables 2, 3, 4, 5, and 6 show the identified characteristics for the dimensions “content”, “language”, “tools”, “methodology”, and “costs”. Detailed definitions and examples of these characteristics can be found in Lozano-Tello (2002). This multilevel framework of characteristics is the base to build an ontology in the ontology domain called reference ontology. The conceptual model of this ontology gathers the mentioned characteristics expressed in concepts. In order to build the reference ontology, we have used the methodology METHONTOLOGY (Fernández, 1997; Gómez-Pérez, 1998) and WebODE platform (Arpírez, 2001). The reference ontology gathers instances of these concepts from 140 existing ontologies, and this knowledge is then used by the OntoMetric Tool to assist users in applying the method.

Table 2. Characteristics related with the dimension “content” DIMENSION: CONTENT FACTOR CONCEPTS

RELATIONS

TAXONOMY

AXIOMS

CHARACTERISTIC Essential_Concepts Essential_Concepts_In_Superior_Levels Concepts_Properly_described_In_NL Formal_Specification_Of_Concepts_Coincides_With_NL Attributes_Describe_Concepts Number_Of_Concepts Essential_Relations Relations_Relate_Appropriate_Concepts Formal_Specification_Of_Relations_Coincides_With_NL Arity_Specified Formal_Properties_Of_Relations Number_Of_Relations Several_Perspectives Appropriate_Not-Subclass-Of Appropriate_Exhaustive-partitions Appropriate_Disjoint-partitions Maximum_Depth Average_Of_Subclasses Axioms_Solve_Queries Axioms_Infer_Knowledge Axioms_Verify_Consistency Axioms_Not_Linked_To_Concepts Number_Of_Axioms



Table 3. Characteristics related with the dimension “language” DIMENSION: LANGUAGE FACTOR DOMAIN KNOWLEDGE

CHARACTERISTIC CONCEPTS/ INSTANCES/ FACTS/ CLAIMS

ATTRIBUTES

FACETS

RELATIONS

TAXONOMIES

AXIOMS

PRODUCTION RULES

INFERENCE MECHANISM REASONING POTENTIAL

INFERENCE ENGINE

SUB-CHARACTERISTIC Allows_Instances_Of_Class Has_Metaclasses Can_Define_Classes_Without_Metaclasses Allows_Facts Allows_Claims Can_Define_Class_Attributes Can_Define_Instance_Attributes Can_Define_Local_Attributes Can_Define_Global_Attributes Can_Define_Polymorph_Attributes Can_Define_Exceptions_In_Attributes Has_Default_Attribute_Values Has_Attribute_Types Can_Define_Cardinality_Of_Attributes Allows_Define_Procedural_Knowledge Allows_New_Facets Allows_Definition_Of_Functions Arbitrary_N-ary_Relations Allows_Define_Ad-hoc_Relations Can_Constrain_The_Type_In_Relations Can_Constrain_The_Value_In_Relations Has_Operational_Definition Can_Declare_Properties_In_Relations Contain_-SubclassOf-_Relation Contain_-NotSubclassOf-_Relation Can_Define_Exhaustive_Decomposition Can_Define_Disjoint_ Decomposition Multiple-Subclass-of_In_Classes Multiple-Instance-of_In_Instances Allows_Axioms_Embedded_In Terms Allows_Independent_Axioms Allows_Axioms_In_First_Order_Logic Allows_Axioms_In_Second_Order_Logic Allows_Disjuntives_In_PRs Allows_Conjuntives_In_PRs Each_Rule_Has_Defined_A_Chaining_Mechanism Each_Rule_Has_Defined_A_Priority Procedures_In_The_Consequent_In_PRs Certainty_Values_In_PR Allows_Multiple_Inheritance Allows_Monotonous_Reasoning Allows_Non-Monotonous_Reasoning Makes_Exceptions_In_Inheritance Axioms_Keep_The_Consistency Execute_Procedures Inference_Mechanism_In_PR IE_Is_Sound_and_Complete IE_Performs_Automatic_Clasifications IE_Deals_ Exceptions IE_Deals_Multiple_Inherance Allows_New_Inference_Engine



Table 4. Characteristics related with the dimension “tool” DIMENSION: TOOL FACTOR CAPABILITIES

VISUALIZATION

EDITION

INTERACTION

METHODOLOGICAL ASPECTS

COOPERATIVE ASPECTS

TRANSLATION

INTEGRATION

CHARACTERISTIC Local_Use Network_Use Internet-based_Use Clarity_Of_User_Interface Response_Time Reliability Browsers_Shows_Whole_Information_Of_Terms Browser_Allows_Selection_Of_Detail_Level Browser_Shows_Taxonomy Browser_Shows_Ad-hoc_Relations Tool_Builds_The_Same_Of_Language Tool_Allows_Edition_In_Any_Time Tool_Shows_Taxonomy_Graphically Tool_Allows_Definition_Of_New_Relations Tool_Allows_Independent_Use Tool_Supplies_Access_Interfaces Documentation_Using_Access_Interfaces Access_Interfaces_Are_OpenSource Documentation_Programming_Access_Interfaces Tool_Supports_Whole_Life_Cicle Tool_Supports_Important_Development_Activities Tool_Supplies_Documentation_About_Built_Products Tool_Checks_Consistency Tool_Creates_Work_Groups Tool_Allows_Simultaneous_Working Tool_Looks_Edited_Ontologies Tool_Looks_Edited_Terms Tool_Notifies_The_Changes_To_Group Tool_Identifies_The_User_Changes Tool_Imports_From_Others_Langs Tool_Imports_From_Markup_Langs Tool_Exports_To_Langs Tool_Exports_To_Markup_Langs Translations_Lose_Minimun_Semantic Translation_Is_Supervised Ease_Of_Integration Difficulty_Of_Refering_New_Terms Tool_Allows_Selection_Of_Terms_To_Integration Tool_Checks_Consistency_In_Integration_Or_Merge Assistance_For_Manual_Merge Semi-automatic_Merge

Table 5. Characteristics related with the dimension “methodology” DIMENSION: METHODOLOGY FACTOR PRECISION

USABILITY MATURITY

CHARACTERISTIC Delimitation_Of_Phases Specification_Of_Activities_By_Phases Specification_Of_Personnel_By_Phases Specification_Of_Techiques_By_Phases Specification_Of_Finished_Products_By_Phases Clarity_Of_Activities_and_ Techniques _Description Quality_Of_Manuals Manuals_With_Complete_Examples Number_Of_Developed_Ontologies Number_Of_Different_Domains Importance_Of_Developed_Ontologies



Table 6. Characteristics related with the dimension “costs” DIMENSION: COSTS FACTOR

Use_Licences_of_the_Ontology Estimated_costs_of_hw_and_sw Costs_of_access_interfaces Use_Licences_of_the_ontology_tools

The Analytic Hierarchy Process There are several methods and tools available to aid decision-making (Triantaphyllou, 2002), some of which are used in software projects (Fenton, 1996). One kind in particular, the multi-criteria decision methods, is useful in comparing several alternatives when, at the same time, several objectives need to be kept in mind. In these methods, the evaluator can directly assign a normalized weight to a criterion that will indicate the importance that that criterion has for the final objective. The analytic hierarchy process constitutes one of the best options to aid multi-criteria decision-making. AHP was devised by Thomas L. Saaty (1977) in the early 70s. It is a powerful and flexible tool for decision-making in complex multi-criteria problems. This method allows people to gather knowledge about a particular problem, to quantify subjective opinions, and to force the comparison of alternatives in relation to established criteria. The method consists of the following steps:

•

Step 1. Define the problem and the main objective to make the decision. For example, “Buy a car”.

•

Step 2. Build a hierarchy tree (as shown in Figure 2) in this way: the root node is the objective of the problem, the intermediate levels are the criteria, and the lowest level contains the alternatives. This hierarchical organization is used to obtain a general overview of the criteria and their relations.

•

Step 3. For each level, build a pairwise comparison matrix with the brothers (sons of the same node). The matrix (like example in table 7) contains the weights of pairwise comparisons between brother nodes. For each comparison matrix, an eigenvector must be calculated, using the equation: |A - λI| = 0, where A is the comparison matrix, I is the identity matrix and λ is the eigenvector. This calculus must be performed for each level of the tree. The entire process can be studied in Saaty (1990). The obtained weights are the importance of each criterion related with brother nodes.



Figure 2. Hierarchical structure of the AHP method, and example of the objective “Buy a car” Objective

Criterion 1

Buy a car

Subcrit-1

Alternative 1

...

consumption, price, security...

Criterion N

...

...

ABS, Airbag, ...

Subcrit-K

Volvo,

Alternative M

Mercedes,

...

Table 7. Example of comparison matrix for the first level “Buy a car” (the user considers “Consumption” six times less important than “Price”, twice as important as “Security”, etc.) Consumption Price Security ...

Consumption 1 6 1/2

Price 1/6 1 1/3

Security 2 3 1

...

•

Step 4. Value each alternative (leaf nodes) with a fixed scale previously. The scales for rating characteristics should be established and described in a precise way.

•

Step 5. Determine the value of each leaf node using a weighted addition formula, with the weights from step 3 and the values from step 4. These results ascend up the tree to calculate the final value of the objective (root). These final values are used to make a decision about the objective.

ONTOMETRIC: A Method for Choosing Ontologies The AHP model can be applied to decide whether or not to reuse ontologies with the ONTOMETRIC method. In order to decide upon the reuse in a new project, Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


ONTOMETRIC can be used to: 1) select the most appropriate ontology among various alternatives or, 2) decide the suitability of a particular ontology for the project. We have developed a software tool (OntoMetric Tool) to assist users in applying the method. Taking into account the general steps of AHP, we have adapted the method for use in the reuse of ontologies:

•

Step 1. Specify the objectives of the project. The engineers should know the exact guidelines of their company and available resources in relation to the new business. They must decide on the importance of the terms of the ontology, the precision of the definitions, the suitability of relations between concepts, the reliability of the methodology used to build the ontology, and so forth.

•

Step 2. Build the decision tree from the MTC (shown in Figure 3), so that the objective, “suitability value of the ontology for a new project”, is placed at the root node; the dimensions (content, language, methodology, tool, and costs) are placed at the first level; the factors of each dimension at the second level; and underneath these factors, the sub-trees of specific characteristics of the particular evaluation project. The general characteristics of all types of ontologies (shown in Tables 2, 3, 4, 5, and 6) should be specialized according to the particular ontology, the specific target project, and the organization that will develop the project.

•

Step 3. For each set of brother nodes, make the pairwise comparison matrixes with the criteria of the decision tree. These comparisons depend on the objectives and aims identified in step 1. The eigenvectors (Saaty, 1990) are calculated from these matrixes. These weights represent the relative importance between brother criteria. For example, in Table 4, you can see the assessments from a user that considers C2 (Quality of manuals) three times more important than C1 (Clarity of activities and techniques), and so forth.

Figure 3. Representation of part of decision tree from MTC Suitability of the ontology

Content

...

Language

...

Methodology Tools

...

Costs

...

Precision

C1: Clarity of Activities and Tech.

Usability

C2: Quality of Manuals

Maturity

C3: Quality of Examples

…



Table 8. Example of the comparison matrix for the “Usability” factor Usability

C1

C2

C3

C1

1

1/3

5

C2

3

1

2

C3

1/5

1/2

1

Figure 4. Example of a linguistic scale with five values 1

very low low medium high very high

0 1.2 3.4 5.6 7.8

0 2.2 4.4 6.6 8.8

1.2 3.4 5.6 7.8 10

2.2 4.4 6.6 8.8 10

medium

high

0 0

1

2

3

4

5

6

7

8

9

10

The eigenvector from the comparison matrix of Table 8 is (C1: 0.3420, C2: 0.5241, C3: 0.1339), and these values are linked to the MTC.

•

Step 4. For each alternative ontology, assess its characteristics. These values will (always multiplying by the weights calculated in step 3) ascend up to the superior nodes of the tree until the node root is calculated. For each one of these characteristics, the engineer should establish a scale of appropriate ratings.

•

Step 4.1. ONTOMETRIC assigns linguistic values (non-numbers) to the alternatives because human beings, in their daily activities, usually make this type of judgement. For example, if users evaluate the “essential relations for the system are defined in the ontology”, they can assess this quality using the following linguistic scale: very_low, low, medium, high, and very_high (Figure 4). This assessment is more intuitive than a numeric scale from 0 to 10. In this process, it is important that the groups of the linguistic values are precisely defined. However, it is not possible to perform calculations with linguistic values. One possible representation of these linguistic values is with fuzzy intervals (Gómez-Pérez, 1997). Their angular points in a scale from 0 to 10, as shown in Figure 4, determine the fuzzy intervals. By assigning linguistic values with fuzzy intervals, we can perform basic



mathematical operations for intervals. This way, the operations of the sum of intervals can be defined as 1) product by a constant and 2) useful to apply the method: (a1, a2 , a3, a4)+(b1, b2, b3, b 4) = (a 1+b1, a2+b2, a3+b3, a4+b4) (1) n * (a1, a2, a3, a4) = ( n*a1, n*a2, n*a3, n*a4)

•

•

(2)

Step 4.2: With these established linguistic scales for each one of the criteria, the user will proceed to study each one of the ontologies that have been considered as alternatives, and to value them using these scales. Table 9 shows an example of assessment for two ontologies (O1 and O2) in every characteristic (C1, C2, and C3) of the “Usability” factor.

Step 5. Lastly, combine the vectors of weights W obtained in step 3 with the values of the alternatives V, using the formula Σ n wi vi.. As appears in Table 9, the weights of each criterion are multiplied by the linguistic valuation (with the corresponding numeric relationship indicated in Figure 4), being obtained by the fuzzy intervals of the column Weighted Value (Wi Vi). The result is calculated by means of the combination of weighted addition (Σn Wi Vi). This value multiplied by their weight (that was calculated in the corresponding comparison matrix), enables us to find the father value “Methodology” (i.e., the weighted addition with the brother nodes: “Precision” and “Maturity” will be calculated). The value obtained for the node “Methodology” (together with their brother nodes) helps us obtain the final valuation of the objective of the suitability of the ontologies O1 and O 2. The final results are shown in Figure 5.

Table 9. Example of calculation values for two ontologies in “Usability” factor Usability

O1

vi C1: (w=0.3420) High(5.6, 6.6, 7.8, 8.8)

O2

wi vi

vi

wi vi

(1.9, 2.2, 2.6, 3.0) Medium(3.4, 4.4, 5.6, 6.6)

(1.1, 1.5, 1.9, 2.2)

C2: (w=0.5241) Medium(3.4, 4.4, 5.6, 6.6) (1.7, 2.3, 2.9, 3.4) Low(1.2, 2.2, 3.4, 4.4)

(0.6, 1.1, 1.5, 2.3)

C3: (w=0.1339) VeryLow(0, 0, 1.2, 2.2)

(0.7, 0.8, 1, 1.1)

(0, 0, 0.1, 0.2)

Usability of O1 Σ n wi vi:(3.6, 4.5, 5.6, 6.6)

High(5.6, 6.6, 7.8, 8.8)

Usability of O2 Σn w i vi:(2.4, 3.4, 4.4, 5.6)



Figure 5. Representation of the final assessment of two ontologies Low

1

Very Low Low Medium High Very High

0 1.2 3.4 5.6 7.8

0 2.2 4.4 6.6 8.8

1.2 3.4 5.6 7.8 10

2.2 4.4 6.6 8.8 10

Medium

High

O1 O2 0

0

1

2

3

4

5

6

7

8

9

10

Figure 6. Comparison of two ontologies (Documents and Document.o) in the “taxonomy” factor using OntoMetric Tool

The obtained suitability values relating it with the linguistic scale chosen by the user (Figure 4) indicates that both ontologies are close to Medium (as shown in Figure 5). With this result, the user can decide to use the best ontology (O1) in the project. Also, with the partial values, it allowed the project manager to obtain a detailed report for the directive of the company about the taken decision. Although in this case only the suitability of two ontologies was valued, for other candidates, steps 4.2 and 5 should be performed repeatedly, since the compariCopyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


son frame (designed in the previous steps) has been established. We must remember that OntoMetric Tool assists users in applying all these steps of the method: it proposes a default tree of characteristics, it uses the “pairwise comparison matrixes” module to do the assessment, it calculates partial and final values, and it represents graphically the obtained results. Figure 6 shows an example of OntoMetric Tool with a partial assessment for two ontologies.

Evaluation of the ONTOMETRIC Method In order to evaluate the completeness and consistency of the multilevel framework of characteristics, 10 expert developers of ontologies (from iSOCO and Knowledge Reuse Group and from AI Lab of UPM) have validated it by means of questionnaires. They identified some conceptual errors and mistakes in the taxonomy of characteristics. The questionnaire and the experts’ opinions appear in appendices I and III of (Lozano-Tello, 2002). On the other hand, we have made several study cases to evaluate the decision capability of the ONTOMETRIC method. Firstly, we have carried out 15 study cases to measure the suitability of each dimension (contents, languages, methodologies, and tools). In addition, we have carried out 15 study cases to measure the suitability of ontologies, taking into account jointly the five dimensions of the framework. For each evaluation process, we have made these tasks: 1) specification of necessary values for the characteristics of the framework; 2) identification of alternative ontologies that we hope to find; 3) searching of alternative ontologies using the OntoMetric Tool; 4) assignment of importance weights using pairwise comparison matrixes; and 5) calculation of suitability measures for each ontology and drawing of the comparison graphics. These evaluation processes have been carried out for several expert ontologists and, with the evaluation results, we can ratify that it is possible to make a justified decision in order to select one ontology from another.

Conclusions and Future Work ONTOMETRIC is an adaptation of the AHP method to help knowledge engineers choose the appropriate ontology for a new project; in order to do this, the engineer must compare the importance of the objectives and carefully study the characteristics of ontologies. Although the specialization of the characterisCopyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


tics and the assessment of the criteria of a particular ontology require considerable effort, the mentioned framework provides a useful schema to carry out complex multi-criteria decision-making. However, the evaluators need to specify in detail the aims of their analyses. Feedback from project managers who have used the method reveals that specifying the characteristics of a certain ontology is complicated, takes time, and its assessment is quite subjective; however, they state that, once the framework has been defined and if it is applied to one particular type of ontology, ONTOMETRIC helps to justify decisions taken, to “clarify ideas”, and to weigh up the advantages and the risks involved in choosing one ontology from other options. The software tool, OntoMetric Tool (http://158.49.116.183:8000/ ontometric), assists the knowledge engineer in applying the method. Shortly, it will be integrated in WebODE platform (Arpírez, 2001). Future work will consist of adapting the method to different ontology scenarios (Uschold, 1999) and establishing formal metrics to assess the suitability of instances in knowledge-based systems for different domains.

Acknowledgments This work has been developed with the support of CICYT under contract TIC2002-04309-C02-01 and the IST Programme of the Commission of the European Communities as project number IST-2000-29243.

References Arpírez, J., Corcho, O., Fernández-López, M., & Gómez-Pérez, A. (2001). WebODE: A workbench for ontological engineering. First International Conference on Knowledge Capture (K-CAP’01), Victoria, B.C. (pp. 613). Bechhofer, S. (2001) OilEd. University of Manchester, OntoWeb Workshop, Universal Ontology Editor Workshop, Vrije Universiteit, Amsterdam. Bernaras, A., Laresgoti, I., & Corera, J. (1996). Building and reusing ontologies for electrical network applications. European Conference of Artificial Intelligence (ECAI’96.), (pp. 298-302). Wiley and Sons. Bollinger, T. & Pfleeger S. (1990). The economics of reuse: Issues and alternatives. Proceedings of the 8th Annual National Conference on Ada Technology, Atlanta, Georgia, (pp. 436-447). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Brickley, D. & Guha, R. (1999). Resource description framework (RDF) schema specification. W3C proposed recommendation. Retrieved from http://www.w3.org/TR/PR-rdf-schema Corcho, O. & Gómez-Pérez, A. (2000). A roadmap to ontology specification languages. Proceedings of 11th European Workshop on Knowledge Acquisition, Modelling and Management (EKAW’00), (pp. 80-96). Jean-Les-Pins. Dean, M., Connolly, D., van Harmelen, F., Hendler, J., Horrocks, I., McGuinness, D.L., Patel-Schneider, P.F., & Stein, L.A. (2003). OWL Web ontology language 1.0 reference. W3C working draft, February 21, 2003. Domingue, J. (1998) Tadzebao and WebOnto: Discussing, browsing, and editing ontologies on the Web. Proceedings of the 11th Knowledge Acquisition Workshop, (KAW’98), (pp. 18-23). Banff, Alberta, Canada: B. Gaines and M. Musen. Euzenat, J. (1995). Building consensual knowledge bases: Context and architecture. Towards very large knowledge bases. 2nd International Conference on Building and Sharing Very Large-Scale Knowledge Bases (KBKS), (pp. 143-155). Enschede: Amsterdam IOS Press. Farquhar, A., Fikes, R., & Rice, J. (1996). The ontolingua server: A tool for collaborative ontology construction. Proceedings of the 10th Knowledge Acquisition for Knowledge-based Systems Workshop, 19, (pp. 1-44). Banff, Alberta, Canada. Fenton, N. & Pfleeger, L. (1996). Software metrics: A rigorous & practical approach (2nd ed.). London: International Thomson Press. Fernández, M. (1999b). Overview of methodologies for building ontologies. Proceedings of the IJCAI’99, Workshop on Ontologies and PSMs, Stockholm, Sweden. (4:1-4:13). Fernández, M. (2000). Método bi-fase para la conceptualización de ontologías basado en meta-modelos. Doctoral thesis, Facultad de Informática. Universidad Politécnica de Madrid. Fernández, M., Gómez-Pérez, A., & Juristo, N. (1997). METHONTOLOGY: From ontological art toward ontological engineering. Spring Symposium Series on Ontological Engineering, (AAAI’97), (pp. 33-40). Stanford. Fernández, M., Gómez-Pérez, A., Pazos, A., & Pazos, J. (1999a). Building a chemical ontology using METHONTOLOGY and the ontology design environment. IEEE Intelligent Systems, 14(1), 37-46. Gómez-Pérez, A. (1998). Knowledge sharing and reuse. In J. Liebowitz (Ed.) The handbook of applied expert systems. New York: CRC Press.



Gómez-Pérez, A. (1999b). Evaluation of taxonomic knowledge in ontologies and knowledge bases. Proceedings of the Knowledge Acquisition Workshop (KAW’99), (6:1:1-6:1:18). University of Calgary/Stanford University. Gómez-Pérez, A. & Benjamins, R. (1999a). Applications of ontologies and problem-solving methods. AI Magazine, 20(1), 119-122. Gómez-Pérez, A. & Rojas-Amaya, M. (1999c). Ontological reengineering for reuse. Proceedings of the European Knowledge Acquisition Workshop (EKAW’99), Dagstuhl Castle, Germany, (pp. 139-156). Gruber, T. (1993). A translation approach to portable ontology specifications. Knowledge Acquisition, 5(2), 199-220. Gruber, T. (1995) Toward principles for the design of ontologies used for knowledge sharing. International Journal of Human-Computer Studies, 43, 907-928. Gruninger, M. & Fox, M. (1995). Methodology for the design and evaluation of ontologies. Proceedings of the IJCAI’95, Workshop on Basic Ontological Issues in Knowledge Sharing, Montreal, (pp. 19-21). Horrocks, I., Fensel, D., Harmelen, F., Decker, S., Erdmann, M., & Klein, M. (2000). OIL in a nutshell. Workshop on Applications of Ontologies and PSMs, European Conference of Artificial Intelligence (ECAI’00), (pp. 20-25). Horrocks, I. & van Harmelen, F. (2001). Reference description of DAML+OIL ontology markup language. Draft report, 2001. Hovy, E. (1997). What would it mean to measure an ontology? Internal paper, ISI of the University of Southern California, Marina del Rey, CA. Karp, R., Chaundhri, V. & Thomere, J. (1999). XOL: An XML-based ontology exchange language. Technical report, Artificial Intelligence Center, SRI International. Kifer, M., Lausen, G., & Wu, J. (1995). Logical foundations of object-oriented and frame-based languages. Journal of the ACM, 42(4), 741-843. Lassila, O. & Swick, R. (1999). Resource description framework (RDF). Model and syntax specification. W3C recommendations. Retrieved from http://www.w3.org/TR/PR-rdf-syntax Lenat, D. & Guha, R.V. (1990). Building Large Knowledge-Based Systems: Representation and Inference in the CYC Project. Reading, MA: Addison-Wesley. Lozano-Tello, A. (2002). Métrica de idoneidad de ontologías. PhD thesis. Departamento de Informática, Universidad de Extremadura. Luke, S. & Heflin, J. (2000). SHOE 1.01. Proposed specification. SHOE project. Retrieved from http://www.cs.umd.edu/projects/plus/SHOE/ spec1.01.htm Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


McGregor, R. (1991). Inside the LOOM classifier. SIGART Bulletin, 2(3), 7076. Motta, E. (1999). Reusable components for knowledge modelling. Amsterdam: IOS Press. Neches, R., Fikes, R., Finin, T., Gruber, T., Patil, R., Senator, T., & Swartout, W. (Fall 1991). Enabling technology for knowledge sharing. AI Magazine, 36-56. Noy, N. F., et al. (2001). Creating semantic Web contents with Protege-2000. IEEE Intelligent Systems, 16(2), 60-71. Noy, N. F. & Hafner, C. (Fall 1997). The state of the art in ontology design. AI Magazine, 53-74. Poulin, J. S. (1997). Measuring software reuse: Principles, practices, and economic models. Boston: Addison-Wesley Longman. Saaty, T. (1977). A scaling method for priorities in hierarchical structures. Journal of Mathematical Psychology, 15, 234-281. Saaty, T. (1990). How to make a decision: The analytic hierarchy process. European Journal of Operational Research, 48, 9-26. Staab, S. & Maedche, A. (2000). Ontology engineering beyond the modeling of concepts and relations. Proceedings of the ECAI’00 Workshop on Applications of Ontologies and PSMs, Berlin, Germany, (pp. 110-117). Swartout, B., Patil, R., Knight, K., & Russ, T. (1997). Towards distributed use of large-scale ontologies. AAAI’97 Spring Symposium Series on Ontological Engineering, (pp. 138-148). Triantaphyllou, E. (2002). Multi-criteria decision making methods: A comparative study. London: Kluwer Academic Publishers. Uschold, M. (1996). Building ontologies: Towards a unified methodology. Proceedings of Expert Systems (ES’96), Cambridge, (pp. 3-50). Uschold, M. (1998). Where are the killer applications? Workshop on Applications of Ontologies and PSMs. (ECAI’98), Brighton, (pp. 107-111). Uschold, M. & Grüninger, M. (1996). Ontologies: Principles, methods and applications. Knowledge Engineering Review, 2, 93-155. Uschold, M. & Jasper, R. (1999). A framework for understanding and classifying ontology applications. Proceedings of the IJCAI’99, Workshop on Ontologies and PSMs, Stockholm, Sweden, (pp. 21-32). Uschold, M. & King, M. (1995). Towards a methodology for building ontologies. Proceedings of the IJCAI’95, Workshop on Basic Ontological Issues in Knowledge Sharing, Montreal, (pp. 20-25).


270 Aubert, Dussart & Patry

Chapter X

A Twofold Approach for Evaluating Inter-Organizational Workflow Modeling Formalisms Benoit A. Aubert, HEC Montreal and CIRANO, Canada Aymeric Dussart, Robichaud Conseil and CIRANO, Canada Michel Patry, HEC Montreal and CIRANO, Canada

Abstract This chapter presents a twofold methodology for the evaluation of interorganizational workflows modeling formalisms. The first approach is ontological and based on the Bunge-Wand-Weber models. The second is based on prototyping and consists in the development of a WFMS for language evaluation. The dual evaluation methodology is then applied to the UML with a practical example from the aerospace industry. Both convergent and divergent results are found from the two validations. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

Evaluating Inter-Organizational Workflow Modeling Formalisms

271

Possible enhancements to the UML formalism are suggested from the convergent results. On the other hand, the divergent results suggest the need for a contextual specification in the BWW models.

Introduction Transactions have been traditionally managed either through organizations or through markets. With advances in electronic commerce and in information systems, this distinction is getting blurred. For example, the last years have seen the development of electronic intermediaries, also known as electronic marketplaces (e-marketplaces), which aim at concentrating transactions made within, or across, industrial sectors through a limited number of virtual intermediaries. These virtual markets enhance transactional efficiency through the aggregation of trading partners (Lucking-Reiley & Spulber, 2001) and through a reduction in asymmetrical information. It is clear that electronic business has penetrated business to business (B2B) processes and consequently spurred a transformation of the traditional organizational boundaries (Zwass, 1998). Since technology has made possible the participation of several partners in shared business processes, these have been crossing organizational boundaries to an extent never experienced before (van der Aalst, 2000). Research on inter-organizational workflow technology is facing an important problem. It has essentially focused on technical issues and has almost ignored language structure (van der Aalst, 2000). This is a classical case of a “technology seducer” problem, very present in the Information Systems (IS) discipline, which has been criticized by Weber (1997). This chapter assesses the adequacy of the unified modeling language (UML) for inter-organizational business processes. There is no question that having adequate language structures for representation is a fundamental requirement for adequate development. The evaluation methodology is based on ontology, using Wand and Weber’s models (1990), and prototyping. Since little empirical validation work has been done on Wand and Weber’s models, ontological analysis will be combined with a prototypical validation that will consist in comparing the process language used in a workflow management system to the process language used for modeling business processes. By combining the two approaches, convergent results are expected to be found to validate the language. The chapter is organized as follows. First, workflows are defined. Then, a literature review is presented to introduce the ontological evaluation framework Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


and to motivate the choice of the UML as a candidate for thorough ontological and prototypical evaluation. Then, the ontological and the prototypical validations are developed. A discussion of the results follows using the convergent and divergent elements from both validations.

Definition of Workflows Originally, workflow appeared from attempts to automate administrative tasks by storing digital copies of bureaucratic documents such as invoices or customer letters (Chaffey, 1998). It has since evolved into a more complex tool for coordinating groups and individuals working in organizations. Recently, workflow technology has been presented as a new way to support inter-organizational business processes (Gartner group, 1999a, b). With leading e-business software vendors such as IBM, BEA systems, Oracle, Vignette.com and Microsoft (with Biztalk Server) offering workflow solutions, workflow technology can no longer be purely considered as hype. There are over 200 products available today (van der Aalst, 2000). A workflow can be defined as “the computerized facilitation or automation of a business process, in whole or in part” (WfMC, 1995, p. 6) and a workflow management system (WFMS) as a system that defines, creates and manages the execution of workflows through the use of software, running on one or more workflow engines, which is able to interpret the process definition, interact with workflow participants and, where required, invoke the use of IT tools and applications. (WfMC, 1995, p. 6) Three types of workflow are generally recognized in workflow practitioneroriented literature (Leymann & Roller, 2000; Chaffey, 1998): ad hoc workflows, which possess a low potential to add value and which generally consist of nonrepetitive tasks; administrative workflows, which are also of the low addedvalue type, but which are composed of highly repetitive tasks; and, finally, production workflows, which are similar to administrative workflows but correspond to critical business processes for the organization with important value-added potential.



273

Inter-Organizational Workflows and Language Convergence Inter-organizational workflows differ from standard workflows in their feature of crossing organizational boundaries. This particularity has consequences on a technical level in order to define how distributed workflow will inter-operate (i.e., in a loosely coupled manner, through capacity sharing) and be coordinated (van der Aalst, 1999). It also has important consequences on an organizational level, as business partners must now consider new aspects related to security, integration with internal processes, strategic alignment, trust, conflict, and lockin between partners participating in the shared workflow. Workflow technology has been presented as a possible tool for integration for a few years now (van den Heuvel & Weigand, 2000; Muth et al., 1999) and already, several software packages such as Microsoft Biztalk Server offer integration solutions that allow workflow-based modeling. More recently, this tendency in business process management has evolved towards the possibility of having a common language for both business and systems analysis in order to attain a more direct conversion of business process models to application integration models (Zetie, 2003). For example, the Business Process Management Initiative proposes the Business Process Modeling Notation along with a formal mapping to an executable format as a possible tool for both business process modeling and execution (BPMI, 2002). Among the objectives of this convergence is the narrowing of the gap between business requirements and IT solutions in a process-centric manner (Smith & Fingar, 2003). The generalization of business process modeling for application integration and the aforementioned emerging language convergence is bringing new challenges for language evaluation. Indeed, if similar languages are now to be used for both business and systems analysis it is essential to define a methodology for their validation and, to date, research on workflow integration has mainly focused on technical issues and not language structure (van den Heuvel & Weigand, 2000). As stated by van der Aalst (2000), “The semantics of the constructs needed to model inter-organizational workflows should be defined before solving the technical issues (which are mainly syntactical)” (p.68). This chapter aims at bringing some elements of explanation to this problem by proposing a methodology for evaluating the ontological validity of available formalisms in order to represent workflows that cross organizational boundaries in contexts such as emarketplaces. The benefits of this research are numerous. First, few or no efforts are reported in the literature on the different workflow modeling formalisms for interorganizational processes. Second, this work will bring more formal basis to the development of e-marketplaces. And finally, finding an adequate common Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


language will allow a common denominator representation for translation from a language to another as defined by Curtis, Kellner, and Over (1992).

Evaluation Framework The idealist that does not distinguish a thing from any of its models cannot account for the multiplicity of schemata of one and the same thing. Consequently, he cannot understand the history of theoretical science, which consists partly in the replacement of some schemata by others. (Bunge, 1977, p.121) In this section, ontology-related definitions are first presented followed by an ontological evaluation framework for IS languages. Afterwards, several process formalisms are briefly reviewed in order to select the most appropriate one for an inter-organizational workflow context. Finally, an evaluation model combining both ontological and prototypical analyses is presented.

Single Grammar Evaluation The number of different business modeling languages and the necessity of interoperability between modeling tools has brought a debate on the need for common languages, the importance of model engineering as a part of software engineering, and on the advantages of ontology-driven modeling (Bézivin, 1998). Ontology, in the context of business modeling, refers to meta-models that define or constrain the model. For more consistency, a precise terminology, as used in the work of Weber (1997) and in ontology-related literature, will be used for the remaining of this chapter. The basic concept of a language consists in its grammar. A grammar can be defined as a set of constructs that include all fundamental objects of the language plus all higher-level constructs that can be generated using those objects. A grammar is composed of grammatical constructs. Constructs represent the building blocks of the grammar. Finally, grammars are used to generate scripts. Scripts represent a meaningful representation of reality. To evaluate a language, we need to determine if the grammar is appropriate for representation of a real-world phenomenon. Wand and Weber (1990) have developed a set of models, usually referred to as the Bunge-Wand-Weber (BWW) models, based on the work on ontology by Bunge (1977, 1979). They have been used to evaluate different grammars such Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


275

as data flow diagrams, entity relationship or object-oriented diagrams (Green & Rosemann, 2000). The BWW models are described in Chapter 1 of this book (see Ontological Analysis of Business Systems Analysis Techniques: Experiences and Proposals for an Enhanced Methodology, by Green and Rosemann). The choice of the BWW models, relying on the ontology developed by Bunge (1977), is justified by two elements. First, since this particular ontology has been often utilized in the information systems field, its use enables comparisons with previous research and the establishment of cumulative knowledge. Second, the Bunge (1977) Ontology has been regarded as very clear, precise and formal (Weber, 2003).

Multiple Grammar Evaluation There are situations in which an information systems analyst would use multiple grammars to represent the real world. For example, if he uses the unified modeling language (UML), he would have to use the different grammars included in the different diagrams of the language. It may be in order to compensate for the weaknesses present in the initially chosen grammar. Green (1996), in a study of 168 users of computer-aided software engineering (CASE) packages, found that users were five times more likely to use multiple grammars than a single one. Weber and Zhang (1996) hypothesized that users rely on multiple grammars in order to minimize ontological overlap. This hypothesis was supported by Green’s (1996) findings. Apart from minimizing the ontological overlap, Green (1996) also identified the goal of achieving maximum ontological completeness. Users should choose their grammars in a combination in order to leave the smallest possible number of ontological constructs uncovered by grammatical constructs. Table 1, reproduced from Green and Rosemann (2000), reviews ontological analyses done on modeling grammars. Green and Rosemann (2000) present the only process related grammar for BWW analysis used in the ARIS toolset, the event-driven process chain (EPC). In this evaluation, all four situations of ontological deficiencies were identified, raising concerns by the authors of possible misspecifications in the BWW models. Those misspecifications were identified as a possible over-engineering of the model: it could include constructs that are not relevant to process modeling, the fact that the BWW evaluation does not take into account the objectives of the modeling grammar during ontological analysis suggesting a need for an individualization of the model, and finally a need to extend the BWW model with enterprise-modeling related constructs (Green & Rosemann, 2000).



Table 1. Ontological analyses – related work – from Green and Rosemann (2000) Study

Type of Grammar Traditional

Wand and Weber (1989) Wand and Weber (1993) Sinha and Vessey (1995) Weber and Zhang (1996) Weber (1997) Green (1997) Parsons and Wand (1997) Opdahl and Henderson-Sellers (1999) Green and Rosemann (2000)

X

Structured

Data-centered

X (DFD)

X (ER) X (ER) X (Relational) X (NIAM) X X

X

O-O

Process

X X (OML) X

Ont. Comp.

Ont. Clar.

Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes

Yes

Yes

These concerns motivate the use of a prototype for completing the ontological evaluation. Since this study will adopt a research methodology that combines ontology and prototyping, the next section will identify the best-adapted language for such a dual evaluation.

Language Selection Based on existing literature, a list of six criteria has been established. The first three — formal basis, executability, and visualization — relate to business processes modeling in general, while the last three — representation of distinct organizations, modeling document exchange, and representation of the three dimensions of workflow — relate more precisely to the context inter-organizational workflows. These two groups are discussed in sequence. The first criterion relates to formality. Curtis et al. (1992) define a formal language as being a language “enactable on a machine”. Therefore, a strictly formal language will have a complete mathematical semantic defining it in order to be understood by the machine. Moreover, a formal language has the advantage of a theoretical framework for analysis and representation (Basu & Blanning, 2000). The second criterion relates to the executability of the language. An executable language is a language that can be simulated. Simulation offers the possibility to support the verification of the formal description with respect to correctness, consistency, completeness, absence of deadlock, and alike (Benyoucef & Keller, 2000). Visualization is another criterion. It is generally accepted that visual information is better understood by humans, and can improve human intuition and understanding about the process (Sutton, Tarr, & Osterweil, 1995). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


277

For this study, three specific criteria have been added to take into account the inter-organizational context. An e-marketplace being an intermediary between multiple buyers and sellers (Choudury, Hartzel, & Konsynski, 1998), the modeling language will have to be able to represent distinct organizations. The modeling of document exchange relates to the actual tendency of linking intraorganizational processes through the exchange of XML documents to form a global B2B inter-organizational process (Skinstad, 2000; Skonnard & Laskey, 2000; RosettaNet, 1999a). Such processes where each partner takes care of a specified part of the process are defined as loosely coupled inter-organizational processes (van der Aalst, 2000). And last but not least, the representation of the three dimensions of workflow corresponds to the foundations on which this work is based, that is to say that workflow technology is the key to creating efficient e-business processes. There is a consensus today on the three dimensions that define a workflow (Leymann & Roller, 2000; van der Aalst, 1998): the business process, representing what is to be done in terms of activities, the IT resource, which will be used in order to automate the tasks, and the organization (Which will perform the activity?) or the cases (When will the task be performed?). Therefore, workflows are often represented using a threedimensional space model called W3 (what, who, which, or when). Five formalisms were candidates for the evaluation: Petri Nets, WfMC, UML, ANSI, and EPC. 1.

The Petri Nets formalism was invented by Karl Adam Petri in the early 60s (Pfleeger, 1998). It was quickly used to model business processes and has now been formalized to model workflows (van der Aalst, 1998). Petri Nets are known for their rigorous semantic.

2.

The WfMC formalism is the only consortium-led language to exist today. To define workflow processes, the WfMC uses a basic meta-model composed of a set of objects (type, activity, transition condition, workflow relevant data, role, and invoked application) that represents simple processes. (WfMC, 1995, 1999).

3.

UML is composed of a set of diagrams of which the activity diagram is used to describe processes. For workflow modeling, these objects could either illustrate the invoked application by an activity or the document flow. Furthermore, different organizations can be modeled using swimlanes in the diagram (Booch, Rumbauch, & Jacobson, 1999). UML enjoys actual popularity and is an object-orientated paradigm.

4.

The ANSI formalism boasts a diagrammatic nature and is used in simulation software such as IGrafix and Process. Swimlanes and large corridor-like partitions of the diagram can be used to represent different organizations



Table 2. Comparison of five workflow modeling languages

Petri Nets WfMC UML ANSI EPC

Formal Basis + − − − −

Executability

Visualization

+ − + + −

+ + + + +

Distinct Organizations + − + + −

Document Exchange + − + + −

W3 − + + − −

and document exchanges between them can be represented using the adequate symbols (Rivard & Talbot, 1998). 5.

Finally, the event-driven process chain (EPC) method (Keller, 1992) is used to make business processes understandable by neophytes (Curran & Ladd, 2000). It is now better known for its use in business process reengineering with the SAP R/3 ERP system.

Table 2 presents a summary of the evaluation. The sign “+” means that the criterion is fully respected, while the “–” means that it is not. A question mark means that it was not possible to determine whether or not the formalism meets a given criterion. The table clearly shows that both the WfMC and the EPC formalisms fail most criteria. As the ANSI formalism is not formal and is unable to represent a particular invoked application, it also has major limitations; which leaves only the Petri Nets and UML along with the classical debate between formal strictness and efficient diagramming in business process modeling. UML was finally chosen to pursue the analysis. This choice is essentially motivated by disciplinary reasons. It is of greater interest to evaluate a language whose strength resides in its representational richness. Although UML possesses a less formal basis than Petri Nets, it allows the representation of the dimensions of workflow that are essential to have an efficient model. Moreover, its actual popularity in information systems renders its evaluation very relevant. Therefore, in the remaining of this chapter we will try to answer the following question: Is UML powerful enough from an ontological and practical point of view for the representation of workflows crossing organizational boundaries?

The UML The grammar that will be used for evaluation and modeling purposes in this chapter is the unified modeling language (UML) in its basic form, as described in the UML User Guide by Booch et al. (1999). To this day, the UML has known several evolutions through different extensions mechanisms. The object constraint language (OCL) for example, which is Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


279

present in the standard specification of the UML, allows the precise definition of constraints on objects by using a formal semantics instead of freely formatted text (Object Management Group, 2001). Furthermore, the Eriksson-Penker Extensions (2000) present a set of extensions for business modeling, which are defined by using the standard extension mechanism present in the UML. Models designed with such extensions are not designed to be transformed into code but can serve as a reference for designing other UML IS diagrams. Also published recently, the UML profile for the Distributed Enterprise Computing (EDOC) is a new standard from the Object Management Group (2002) that aims at defining a standard way to model business interaction onward to executable systems. The EDOC profile defines an Enterprise Collaboration Architecture to define models that are independent from the platform, a Patterns profile that aims at improving reusability of object models by defining standard constructs such as common objects or business concepts and a set of technology dependent models for platform-specific modeling. Finally, as extensions can be freely defined in the UML within the framework of the extensions mechanism, another example can be found with the RosettaNet Consortium UML extensions designed to model their generic B2B processes (RosettaNet, 1999b). The choice of focusing on the UML in its basic form is motivated by two reasons. First, it is a bit premature to validate the usefulness of evolutions of the language if it has not been proven that there exists some form of incompleteness or ambiguity in the first place. Second, recent literature suggest that the evolutions that the language has known are not yet either de facto standards or fully useful for practitioners (Thomas, 2003). In the current study, we will focus on the use of three diagrams: the activity, the state, and the sequence diagrams. This choice is consistent with the previous literature on workflow modeling using activity charts and state charts prior to their inclusion in the UML (Muth, Weissenfels, & Weikum, 1999). It is also motivated by the context of the study. Since WFMS are parameterized by representing graphically the workflow that is to be supported by the system, the associated data model is generally transparent for the analyst. Focusing on the structural aspect of the UML would therefore bring little added value.

Rationales for Language Evaluation A model is, by definition, a simplification of the reality (Booch et al., 1999), that is to say, a description of a real-world extant. Figure 1 represents these concepts. Adequate modeling requires completing a good representation of the reality. As Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Figure 1. Validating language constructs

Reality

REAL W ORLD

Representation

Representation of reality satisfying BW W ontology Formalization - mathematical theory

Based on

Based on

Described by

Real-world extant

Sim ulates Automates

W orkflow System

Language constructs (grammar)

Formal model

W orkflow Simulator

Based on

Based on

presented in Weber (1997), this good representation phenomenon is twofold. First, it means having a valid link between the real world and the models that the users make from it (described as “real-world extant” in Figure 1) and second, it means to have a valid link between this user’s model and the scripts that are derived from it (“formal model” in Figure 1). The first phenomenon of representational goodness, which represents how well individuals can derive a good notion of the real world, is not addressed by the BWW models as it is argued in Weber (1997) that it is not at the core of the IS discipline and already covered by other disciplines. The second phenomenon however, and precisely its deep structure aspects, is argued to be at the core of the discipline. It aims at representing how well some of the features of the script represent characteristics of a user’s model of the real world. The BWW models were derived to analyze such deep structure phenomena. Among these models, the representation model allows the evaluation of grammars used to create scripts. Unfortunately, it is difficult to determine in a BWW evaluation if it is the language used to complete the script that is faulty or the evaluation criteria. For example, Green and Rosemann (2000), in an analysis of the EPC, raised doubts about the validity of all the ontological constructs of the BWW model. Under a classical BWW analysis, the language would have been poorly evaluated while its



281

industrial applications are very numerous. It is therefore important to consider a validation methodology that would complete a BWW evaluation. To evaluate a grammar, we therefore need to find a path from the grammatical constructs to the reality as illustrated in Figure 1. A formalism being a language used to model reality, applying the formalism and testing it in a practical manner is the only way to validate unambiguously the abstraction of the model. UML being only partially formal with State diagrams and executability still being in an early development stage, a prototyping approach is clearly the most adequate and will therefore be used as a second evaluation method. In this research, we have the precisely defined language features of the UML and we will use a model based on its grammar to write the process program, a workflow management system. Going from the model to the system will allow us to identify if the models are clear enough for a successful development of the system. Afterwards, a reverse iteration will be made in order to see if there could be any lack of information between the completed process program and the initially designed models. By combining both the BWW analysis and the prototypical analysis, a convergence between the two sets of results can be expected. Indeed, if a prototypical analysis concludes that there is a lack of information within the models for the development of the system, it would confirm a BWW theoretical conclusion of ontological incompleteness. In our context, the ontological constructs are replaced by the constructs of a process program that aims at automating a business process. Of course, the same language features (or grammatical constructs) are kept for both analyses and are those of the UML. There still exists a possibility of a double bias in the analysis in that both the ontological analysis and the prototypical analysis bring convergent, but biased results. Yet, combining the two analyses is a step towards a more valid and more accepted BWW analysis. It is also a step towards a confirmation of the doubts that Green and Rosemann (2000) raised about a possible over-engineering of the BWW models in their analysis of the EPC. Figure 2 presents our complete language evaluation model, which is the main contribution of this chapter. Using this framework, two analyses need to be conducted. The first analysis is purely theoretical and is an application of the BWW representational model presented earlier. The second analysis is a practical application using a prototype system. Once both analyses are completed, a comparison between the conclusions of the two will be conducted. If convergent results are found, then it will be possible to formulate possible improvements for the UML. If the results are divergent, questions will have to be raised about the validity of the ontological constructs.



Figure 2. Language evaluation model Prototypical Analysis

Ontological Analysis

Require adapted (completeness)

Are compared to (ontological clarity)

Language Features (grammar)

Process Programs

Are defined by writing (clarity)

BWW Ontological Construct

Are compared to (ontological completeness)

CONVERGENCE?

Ontological Validation The methodology for the BWW analysis is similar to previous BWW evaluations presented in Table 1. It consists in using the definitions of all ontological constructs and in finding a possible mapping with the grammatical constructs of the language under evaluation. Afterwards, a forward analysis from the grammatical constructs to the ontological constructs is done to evaluate ontological clarity. The backward analysis will check for ontological completeness. If deficiencies are found, their possible consequences will be identified.

Matching Grammatical and Ontological Constructs The elementary unit in the BWW model is the “thing”. This elementary ontological construct can be associated with the object in our three diagrams. Contrary to the EPC, the activity chart can show the transformations made on objects during activities and therefore solve a case of ontological incompleteness. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


283

An activity in the activity diagram will sometimes involve transformations made on objects. In fact, activities in the activity diagram are formally defined as an “ongoing nonatomic execution within a state machine” (Booch et al., 1999, p. 259). State machines represent accordingly the different possible states for an object. We will therefore interpret these activities as a Property in general for the object. This is relevant with Green and Rosemann’s (2000) analysis of the EPC that interpreted the function in the EPC as a Property in general too. Class and Kind are respectively represented in the UML in the class diagram with the class and the generalization constructs. This diagram represents the static aspects of a system and is therefore not included in the analysis. Consequently, the absence of direct match will not be considered as an ontological incompleteness. States of the thing are represented by the state of the object in the activity diagram or by the state construct in the state diagram. A state machine in the state diagram represents the Conceivable State Space, defined as all the states that a thing may ever assume. A Lawful State Space can be represented in a state diagram using substates. Stable States and Unstable States can respectively be represented by the final state or the initial state in a state diagram. Events are represented as the trigger for a transition in the state diagram. But events can also be represented as an activity in the activity diagram. There is no grammatical differentiation for External and Internal events but the use of the Uses Cases for human-machine interaction diagram or the use of stereotypes could help make the differentiation possible. The Conceivable Event Space can be observed on the state machine of a thing by looking at all transitions triggers. There exists no construct for a poorly-defined event and well-defined events use the same grammatical construct as a normal event. Transformations are represented by an activity in the activity diagram. Lawful transformations are represented by guard conditions on transitions. There is no grammatical construct for Lawful event space. History can be modeled using the shallow history state construct in the state diagram. Acts on cannot be represented in the same way as it is defined in the definitions of the ontological constructs but could eventually be associated to the composition relationship in the class diagram, for example, in a composition relation between a thing “Activity” and a thing “Project”. Coupling relationships between objects (“things”) in the system can be represented using messages in the sequence diagram. In the case of workflow management, it is the coupling between actors, organizational units or organizations (between the swimlanes in the activity diagram) that are most interesting to illustrate cross-organizational workflow. A System can be represented using the sequence diagrams. Indeed, if multiple objects are involved, dividing the system will not eliminate the existing couplings between those objects. It could Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


also be represented using the package construct of the UML. The System composition is represented using the object construct. Once again, the System environment, that is to say external and internal things to the system, cannot be differentiated without a stereotype. The System structure is represented using the message construct in the sequence diagram. Subsystems can be represented using a stereotyped package. Relationships of composition and generalization would show the System decomposition and the Level structure. Unfortunately, the package and the relationships are not part of the three views that originally defined our language.

Results of the Ontological Evaluation The complete ontological evaluation has been transcribed in Table 3. For an analysis of ontological completeness, it seems that several constructs can not find representation in any views: Lawful event space, Acts on, Poorly-defined event. Consequently, from a purely theoretical point of view, for workflow modeling, the UML must be considered as ontologically incomplete. Moreover, many examples show that the UML for workflow modeling is not ontologically clear. Indeed, we face construct overload for the activity construct in the activity diagram that can represent a transformation, a process, a property in general, or an event. Construct overload was also observed for the swimlane of the activity diagram that can represent either a thing (such as an organization) or a hereditary property of the thing (a user of the organization). Finally, overload was also identified for the trigger construct (that can represent either an event or a well-defined event). We also face construct redundancy in the case of the Process ontological construct that can be either represented by a complete activity diagram or by the activity construct in an activity diagram. In the case of the activity diagram, construct excess can also be identified since the branching construct could not find any matching ontological construct. Also, since we are using multiple grammars in the analysis, it is necessary to evaluate ontological overlap between the different views of the UML. Not surprisingly, they mainly concern the activity diagram for which there exist many overlaps with the state diagram. The activity diagram was the last added diagram in the UML and consisted in bringing a “process” view to information systems. Ontological analysis shows that it does not integrate perfectly with the other views. Clearly, the goal of minimizing ontological overlap is not attained here. The consequences of those deficiencies are not negligible for the systems analyst. First, he may not possess all necessary constructs to complete his models. Second, some confusion may arise between different constructs because of the overload and scripts could therefore be interpreted differently from



285

Table 3. BWW representation model analysis for dynamic aspects of UML Ontological construct Thing Property ° In Particular ° In General ° Intrinsic ° Mutual ° Emergent ° Hereditary ° Attributes Class

Activity diagram Object Swimlane Activity Swimlane

State diagram Object

Sequence diagram Object

Class (Cass diagram) Generalization (Class diagram)

Kind State Conceivable State Space State Law Lawful State Space Process Event Conceivable Event Space Transformation Lawful Transformation: ° Stability Condition ° Corrective Action Lawful Event Space History Acts On Coupling: Binding Mutual Property System

State of the object

Activity diagram Activity Activity

State State Machine StateàTransitionàState Substates

Trigger All triggers

Activity Guard conditions on transitions

Shallow history state construct Messages Sequence Diagram Object

System Composition System Environment System Structure Subsystem System Decomposition Level Structure External Event Stable State Unstable State Internal Event Well-Defined Event Poorly-Defined Event

Other views

> Messages

> Final State Initial State > Trigger

Package with >

Package with > Composition Generalization



one analyst to another. Finally, the analyst may be tempted to use only the activity diagram because it covers most of the necessary ontological constructs. But these harsh conclusions for the UML need to be softened for several reasons. First, we need to be cautious about the ontological incompleteness conclusion. Using multiple views, this incompleteness has been minimized to only four constructs that are not necessarily essential to workflow modeling, and this could confirm a conclusion by Green and Rosemann (2000) who raised the question of a possible over-engineering of the BWW model and a need for a contextual individualization of the model. Second, the construct redundancy that has been identified refers to the possibility of having different levels of abstraction for the activity diagram. While this may look confusing at first, adequate stereotyping on different activity diagrams could clearly identify at what level we are. Third, the construct excess refers to the absence of an ontological construct that identifies branching. Intuitively, this is an ontological construct that would definitely be essential to any workflow modeling grammar. Now that conclusions have been raised from the ontological analysis, a prototypical analysis will complete the evaluation.

Prototypical Validation Methodology The prototypical analysis consists in modeling a B2B business process and in automating it using a WFMS. It is precisely the transposition of the model in the system that is analyzed. The assessment of the prototype will determine if the models are clear (if we do not hesitate between constructs in the system) and complete enough (if the model has all the required information) for the successful development of a WFMS. The clarity analysis consists in transposing the model into the WFMS to identify ambiguities. The completeness analysis consists in comparing the process program requirements for adequate execution of the process to the language features we had for modeling the workflow. The research context chosen is the aeronautical industrial sector and, more precisely, the exchange of quality control documents between manufacturers and their numerous sub-contractors. The names of the clients and suppliers are voluntarily omitted for confidentiality reasons. The existing inter-organizational processes lack automation, and it is therefore anticipated that important economies of scale could me made by using a market aggregator that would automate B2B processes in a workflow-oriented manner.



287

We therefore need a development model that: 1) uses UML for modeling, and 2) aims primarily at defining standard B2B processes. The development model for the RosettaNet Consortium that has successfully implemented B2B standard for over 60 companies in the IT and electronics components industry (InternetWeek, 2000) was a natural candidate. The model aims at creating Partner Interface Processes (PIP) that define standard interfaces for developers. A PIP is composed of a new “generic” B2B process, a dictionary of common properties for the industry and of XML document type definitions (RosettaNet, 1999a).

Business Model “As Is” Process Using corporate documentation, a preliminary blueprint of the B2B workflow was drawn and presented to five different experts or managers in aerospace quality control. Two were working for two different large manufacturers while the three others were working for different sub-contractors. Some minor modifications were made to the business blueprint and complementary corporate documentation was sometimes collected during the meetings. With the modifications suggested by the respondents and the supplemental corporate information obtained, a final blueprint of the business process was finally drawn. It is presented in Appendix 1. The process starts with a supplier having produced a given number of items ordered by a manufacturer. There is an optional quality inspection to be made on a randomly chosen item in the shipment if the supplier produces it for the first time or if modifications were made in the manufacturing process. This inspection leads to the writing of a first article inspection report that is kept in the supplier’s documentary vault while another copy is sent with the shipment. The failure of this inspection is not included in the boundaries of our studied process because it involves another process of B2B communication to determine the reasons of non-compliance. For every shipment, the supplier must complete a mandatory inspection that consists again in choosing randomly an item in the shipment and in inspecting it. This leads to the writing of a certificate of conformity, also called a certificate of compliance. Once again, a copy is kept in the supplier’s documentary vault while another copy is joined to the shipment. If the item is found to be noncompliant with the manufacturer’s requirements, a supplier report of nonconformity (RNC) is sent to the manufacturer describing the defect and asking for a study of the non-conformity. If the manufacturer accepts the nonconformity, he sends back the RNC mentioning that the article is accepted “as Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


is”. Otherwise, the RNC is sent back mentioning that the article is rejected. Sometimes, a certificate of acceptance or a certificate of rejection can replace the RNC. Once again, a copy of both of these documents is kept in the vaults of both the supplier and the manufacturer. When the shipment arrives at the manufacturer, quality control documents are inspected. If the supplier has a sufficiently good rating for the manufacturer, inspection at reception can be skipped. Otherwise, another inspection is made and, if it is successful, the received items are placed in the inventory. If the inspection finds a defect, all the received parts are immediately placed in quarantine and the non-conformity is studied. A RNC is filled in and if the item can be accepted “as is”, it is placed in the inventory. The refusal of the item is not included in the boundaries of this process because it involves another complex process of repairing; reworking or modifying the item according to supplemental analyses made and would unnecessarily complicate the case of study.

“To Be” Process The business process analysis phase aims at creating a new generic “to be” process modeled using the UML formalism. Several governmental and quality control agencies impose both the process and the exchanged documents, and therefore little modifications are possible. In fact, it was found that the two manufacturers had very similar processes with their subcontractors and almost similar quality control documents. The redesign of the process includes a third party (the e-hub) and was made using the basic guidelines of business process reengineering (BPR) as presented in Hammer (1990). While the application of those guidelines may be considered as being a non-exhaustive method for workflow analysis and redesign, it is important to remember that the objective of this work is to evaluate an inter-organizational modeling language and not the applicability of BPR methods in an inter-organizational context. The modified business process is presented in Appendix 2.

System Development For the implementation framework phase, the XML document format used consisted of those already present in the WFMS software package. The Dictionaries step was not completed because it is not relevant to this study. A prototypical process model, as presented in Pfleeger (1998), was followed. Every revision to the model was considered as a possible completeness problem or clarity problem.



289

To implement this inter-organizational workflow, the Weblogic Process Integrator (WLPI) of BEA SYSTEMS was used. Only one workflow engine was used to support the process. Such a configuration would facilitate the integration with ERP systems, which are prevalent in the aeronautics industry. Also, for smaller firms, the purchase of a WFMS is way beyond budget. A Web-based interface for document management with routing controlled by an outsourced WFMS is much more appropriate to their context. Finally, using an intra-organization-like architecture in an inter-organizational context greatly reduces the risk associated with the prototype development while still being adequate for the study. This system is based on the Weblogic application server that enables the use of Java 2 Enterprise Edition (J2EE) specifications such as Java Server Pages (JSPs), Enterprise Java Beans (EJB) or Java Messaging Services (JMS). At the core of the system is the Weblogic Process Integrator Server, a workflow engine dynamically executing workflows defined in the workflow studio using a flowchart tool. For each activity defined in the workflow, several tasks can be completed such as sending an electronic mail message, launching an application or sending an XML document to a client application. Workflow models, instances and variables are stored in a relational database. For the prototyping context, the WFMS is used as a work coordination tool, which is generally the main task of a WFMS (Chaffey, 1998). Each organization uses a client application that 1) gives reminders to complete an inspection or to fill in a document and 2) asks specific questions on the result of evaluation or of an inspection for adequate workflow routing. Communication between the workflow engine (the electronic intermediary) and the different client applications (the organizations) is made through the exchange of XML documents. The workflow engine predefines the document type definitions used. To define a workflow, WLPI uses a workflow definition meta-model presented in Figure 3. The execution logic is represented using eight grammatical constructs defined as nodes in the meta-model. Each node can invoke different actions such as invoking an application, sending a reminder or sending an XML document. For parameterization, the Workflow studio is used to describe in a flowchart manner the workflow to be automated. It uses eight grammatical constructs that are briefly described in Table 4 with their equivalences in the UML which mainly come from the activity diagram because of the close resemblance between the two grammars. Prior to parameterization, tests were made to verify that the engine could adequately execute a basic workflow and could send simple XML documents to the workflow clients. Afterwards, the models were used to set the WFMS to our context using the translation present in Tables 4-6. The observations made during the system development follow.



Figure 3. Weblogic process integrator workflow meta-model

O rg a n iz a tio n

R o le

T e m p la te

T e m p la te de fin itio n

V a ria b le

Node

User

In sta n c e

Node

V a ria b le

A c tion

Clarity Two cases of ambiguity were identified. The first case concerns the activity construct in the activity diagram. It was ambiguously used as an event construct too because no grammatical construct exists for an event in the activity diagram of the UML while these constructs were distinct in the process program. For example, receiving an XML document is modeled as an activity in our models while it is an event in the process program. Clearly, mistakes could be made while transposing the model in a WFMS because room is left for interpretation. The second ambiguity concerns the swimlanes of the activity diagram. Our process program required making a clear distinction between organizations, roles and users. The swimlanes of our model were not sufficiently precise to make such distinctions. In the case of workflow management, more precision is needed and adequate meta-modeling appears unavoidable to clearly identify the relationships between users, their roles and their organizations. Not surprisingly, such precisions are made in RosettaNet UML extensions (1999b).



291

Table 4. Weblogic process integrator grammatical workflow constructs PROCESS PROGRAM CONSTRUCT

VENDOR DESCRIPTION

UML EQUIVALENT Activity Diagram

Indicates the start of a workflow. Indicates the end of a workflow.

Start Done Task

Decision

Event

Connector

Or

And

State Diagram

Sequence Diagram

Initial State Final State

Defines a task in a workflow.

Activity

Represents a condition in the workflow that evaluates to be true or false. Represents an event that can be triggered either internally or externally by an XML message. Subactions can be performed, and/or workflow variables can be set, as the result of the trigger of the event. Used to connect workflow nodes. The arrow directs you to the subsequent task in the flow. Allows joining of one or more task, decision, or event with an OR condition. Allows joining of one or more task, decision, or event with an AND condition.

Sequential Branch Activity

Trigger

Triggerless Transition ?

Join

Table 5. Non-diagrammatical constructs used in Weblogic process integrator Construct Organization User Role Workflow variable

Representation in UML Swimlane Swimlane Swimlane Object State

Table 6. Some possible actions for tasks Actions Call program Sending an XML document Send an e-mail Assign task to user

Representation in UML Object with stereotype Object with stereotype Object with stereotype Activity



Completeness Only one case of incompleteness was observed. As defined in Booch et al. (1999), “A Stereotype is an extension of the vocabulary of the UML, allowing you to create new kinds of building blocks similar to existing ones but specific to your problem” (p.78). Therefore, for every process program construct that lacked a precise grammatical symbol, we could freely define a new construct to represent it. For our prototype system, we developed stereotypes for each possible action for an activity as represented in Table 6. However, since there does not exist a symbol for the “exclusive or” join, we could not define a stereotype to represent it. This is clearly a completeness deficiency in the activity diagram. Indeed, joining in the activity diagram can only be made on an “AND” basis and not an “EXCLUSIVE OR” basis. This observation was not made during the parameterization of the system but when determining the translation scheme presented of Table 4.

Final Remarks A final observation made during the development of our prototype is the little use we made of the views other than the activity diagram. In fact, this is not very surprising since the flowcharting tool used in WLPI is very similar to the activity diagram. In this analysis, we identified cases of clarity and completeness problem and made an observation on the use of multiple views. We will now compare these results with those of the ontological analysis and discuss convergent or nonconvergent results.

Reconciliation and Discussion In this section, the results obtained from the ontological and from the prototypical evaluation are compared. We will first evaluate completeness issues, followed by clarity problems and by grammatical overlaps.

Completeness From the ontological evaluation, we concluded that the UML was ontologically incomplete because it lacked the Lawful event space, Acts on, and the PoorlyCopyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


293

defined event constructs. Those completeness problems were not observed for the development of the prototype system. In fact, those constructs are fundamentally philosophical and have little to do with workflow modeling. They most probably illustrate a case of contextual over-engineering of the BWW models for a situation as specific as cross-organizational workflows and illustrate the need for a contextual specification of the models. In the prototypical evaluation, we lacked a construct of “Exclusive Or” for joining two activities. This result cannot be compared with the ontological analysis because branching in the UML had no ontological equivalent in the BWW models. But clearly, such a joining is essential in process modeling and this illustrates once again the need for a specification of the BWW models so that it can include branching. It also illustrates the need to add an “Exclusive Or” construct in the activity diagram. Such a construct exists in the OCL specification but it is unclear whether OCL can be used to define constraints on branching between action states in the activity diagram and therefore adds to ambiguity.

Clarity We observed in the ontological evaluation that the activity construct in the activity diagram brought a construct overload problem. This result is convergent with the prototypical analysis in which we faced confusion between the activity and the event construct when transposing our workflow model into the WFMS. Clearly, the activity diagram lacks an event construct and further specifications of the language should aim at including this construct. The construct redundancy problem of the Process ontological construct, which could either be represented by a complete activity diagram or by a single activity in an activity diagram, was not a particular problem for the development of our system. In fact, this result is explainable by the fact that the WFMS imposes indirectly the appropriate level of abstraction of the task as it coordinated the work of individuals. Also, the redundancy for the trigger construct was not an issue as the WFMS used a single event construct in its grammar. Finally, during both ontological and prototypical validation, we faced clarity problems with the swimlanes of the UML activity diagram that could ontologically represent a thing, or a hereditary property of that thing and, in a more practical context, users, roles, and organizations. Further specifications of the UML should aim at defining a more precise semantic for the swimlanes in the activity diagram that could permit the representation of organizational hierarchical levels.



Multiple Views The ontological evaluation revealed that the activity diagram had several overlaps with other views in the UML. During the development of the prototype, little use was made of diagrams other than the activity diagram. These results are clearly convergent. It illustrates once again how the activity diagram integrates poorly with other views of the language. It is difficult here to suggest possible improvements because reducing overlaps could also lower the complete ontological completeness of the language. In fact, further improvements should aim at both reducing overlaps while maintaining the overall ontological coverage, which can be considered as very satisfactory for the UML.

Conclusion To this day, research on inter-organizational workflows has essentially focused on technical aspects of interoperability between WFMS. In fact, very little work has been done in order to define a precise semantic for inter-organizational business modeling. This chapter intended to bridge that gap by finding a solution to this problem from an IS perspective. To provide a framework for this research, we chose to rely on the work of Wand and Weber (1993). This chapter aimed at determining if the ontological validity of available formalisms was sufficient to represent workflows crossing organizational boundaries. A review of several formalisms revealed that the UML fulfils essential representation criteria related to B2B workflows. Moreover, it possesses several extension possibilities that make it a powerful — and popular — language for business modeling. Three contributions can be stressed. First of all, this work presented a more rigorous methodological framework for ontological grammar evaluation than previous studies by combining an analysis using the BWW representation model with a prototypical analysis. Prior research had raised doubts on the validity of the BWW model in workflow-modeling contexts by assuming that the tested grammar had little deficiencies. Clearly, a more rigorous methodological framework was needed in order to discuss the validity of the BWW models. By using the ontology of a WFMS in addition to the BWW ontology, conclusions drawn out of convergent results from both analyses can be considered more rigorous. Moreover, this twofold methodology could provide useful basis for the evaluation of future languages that combine the aim of business and systems analysis.



295

Second, by using this new methodological framework, little ontological deficiencies were identified in the UML. This result could mainly be attributed to the extension capabilities of the grammar. Nevertheless, some room for improvement has been identified and specific enhancements were suggested. The most challenging concerns the overlaps of the activity diagram with other views in the UML. Clearly, this view, which is essential for workflow modeling, does not fit well with the other views of the popular language, and this problem could increase current hesitancy for developers to use a workflow paradigm for IS development. Third, the two analyses confirmed the need for the development of specific ontology for workflow modeling. There is undeniably room for both a universal ontology for the representation of real-world phenomenon such as the BWW models and for more specific contextual ontology, which could also be based on the BWW ontology. In fact, the BWW representation model is probably too fundamental for a precise context such as cross-organizational business process modeling. Indeed, while it was first concluded that the UML had ontological deficiencies, our models were sufficient for the successful development of a prototype system in the aerospace industry. From these contributions, directions for future research can be identified. While this work credited the UML with little deficiencies, some of the aforementioned suggested improvements could boost the already-high quality of its grammar. Optimal combinations of graphical and text constructs in modeling language could be evaluated. Another interesting research direction could be the definition of common extensions for the community of WFMS developers. Indeed, while the extension mechanisms are powerful tools against ontological deficiencies, such extensions need to be defined in a matter that is fully understandable by all business partners. To this day, only extensions to the UML for business modeling at large and not workflow management have been defined. Recent efforts made by the OMG with the EDOC offer interesting possibilities in this area. Such a contextual approach could also be a research track for ontological evaluation models. Further research in this area could aim at defining a particular ontology for workflow modeling based on the BWW models and on the metamodels from several WFMS. There are over 200 WFMS systems available on the market, which would allow a large sample of study. With a precise ontology for workflow, universal extensions could be defined and therefore ease the task of making UML models a possible direct input for WFMS parameterization. This stream of work also has business implications. Accurately modeling information flows has always been a challenge. The models have to be representative of the reality while being easy enough to understand to enable users to validate them. Inter-organizational systems introduce additional challenges for analysts. They have to model information flows across organizations.



In doing so, they face multiple sets of documents, different names, different term definitions, and multiple data dictionaries. Having an ontologically valid formalism does not guarantee that all models will be correct. However, it reduces the risk of errors by eliminating ambiguities, by reducing confusion and by ensuring that all the elements that will have to be implemented in the system have a representation in the model. This reduces the risk of having an inadequate or incomplete system.

References Basu, A. & Blanning, R. (2000). A formal approach to workflow analysis. Information Systems Research, 11(1), 17-36. Benyoucef, M. & Keller, R. (2000). An evaluation of formalisms for negotiations in e-commerce. Proceedings of the Workshop on Distributed Communities on the Web, Quebec City, QC, Canada, (pp. 45-54). Bézivin, J. (1998). Who is afraid of ontologies? OOPSLA ’98 Workshops, Vancouver, Canada. Available at http://www.metamodel.com/oopsla98cdif-workshop/bezivin1 Booch, G., Rumbauch J., & Jacobson, I. (1999). The Unified Modeling Language user guide. Reading, PA: Addison-Wesley. BPMI. (2002). Business process management initiative. Business process modeling notation, working draft V 0.9. Retrieved from http:// www.bpmi.org/bpmn-spec.esp Bunge, M. (1977). Treatise on basic philosophy: Volume 3: Ontology I: The furniture of the World. Dordrecht & Boston: Reidel. Bunge, M. (1979). Treatise on basic philosophy: Volume 4: Ontology II: A world of systems. Dordrecht & Boston: Reidel. Chaffey, D. (1998). Groupware, workflow and intranets: Reengineering the enterprise with collaborative software. Woburn, MA: Digital Press. Choudury, V., Hartzel, K., & Konsynski, B. (1998, December). Uses and consequences of electronic markets: An empirical investigation in the aircraft parts industry. MIS Quarterly, 22(4), 471-507. Curtis, B., Kellner, M., & Over, J. (1992). Process modeling. Communications of the ACM, 35(9), 75-89. Curran, T. & Ladd, A. (2000). SAP R/3 Business Blueprint. Upper Saddle River, NJ: Prentice Hall. Eriksson, H. E. & Penker, M. (2000). Business modeling with UML: Business patterns at work. New York: Wiley Computer Publishing.



297

Gartner Group. (1999a, October 7). E-Business to workflow: Wake up! Available at http://www4.gartner.com/DisplayDocument?ref=g_search& id=300767 Gartner Group. (1999b, December). Why e-business craves workflow technology. Available at http://www4.gartner.com/DisplayDocument?ref=g_ search&id=301217 Green, P. (1996). An ontological analysis of information systems analysis and design (ISAD) grammars in upper CASE tools. Unpublished doctoral dissertation, Department of Commerce, The University of Queensland, Brisbane, Australia. Green, P. (1997). Use of information systems analysis and design (ISAD) grammars in combination in upper CASE tools – An ontological evaluation. Proceedings of the Second CAiSE/IFIP8.1 International Workshop on the Evaluation of Modeling Methods in Systems Analysis and Design (EMMSAD’97) (pp. 1-12), Barcelona, Spain. Green, P. & Rosemann, M. (2000). Integrated process modeling: An ontological evaluation. Information Systems, 25(2), 73-87. Hammer, M. (1990, July-August). Reengineering work: Don’t automate, obliterate. Harvard Business Review, 104-112. InternetWeek. (2000, November 6). RosettaNet sharpens focus: Interview with Jennifer Hamilton, CEO, RosettaNet. InternetWeek [online]. Retrieved from http://www.internetweek.com/interviews/ham110600.htm Keller, G., Nuettgens, M., & Scheer, A. W. (1992). Semantische prozessmodellierung auf der basis “ereignisgesteuerte prozessketten (EPK).” In A. W. Scheer (Ed.), Veroeffentlichungen des Instituts fuer Wirtschaftsinformatik. Saarbruecken: Heft 89. Leymann, F. & Roller, D. (2000). Production workflow: Concepts and techniques. Upper Saddle River, NJ: Prentice Hall. Lucking-Reiley, D. H. & Spulber, D. F. (2001). Business-to-business electronic commerce. Journal of Economic Perspectives, 15(1), 55-68. Muth, P., Weissenfels, J., & Weikum, G. (1999). What workflow technology can do for electronic commerce. Technical report, University of the Saarland, Department of Computer Science, Saarbruecken, Germany. Object Management Group, The. (2001). Unified modeling language specification, version 1.4. Available at http://www.omg.org/cgi-bin/doc?formal/ 01-09-67 Object Management Group, The. (2002). UML profile for enterprise distributed object computing specification, OMG adopted specification. Available at http://www.omg.org/technology/documents/formal/edoc.htm



Opdahl, A. L. & Henderson-Sellers, B. (1999). Grounding the OML metamodel in ontology. Working paper, (pp. 1-59), University of Queensland, Brisbane. Parsons, J. & Wand, Y. (1997). Using object in systems analysis. Communications of the ACM, 40(12), 104-110. Pfleeger, S. L. (1998). Software engineering: Theory and practice. Upper Saddle River, NJ: Prentice Hall. Rivard, S. & Talbot, J. (1998). Le Développement de Systèmes d’Information. Québec: Presses de l’Université du Québec. RosettaNet. (1999a). RosettaNet implementation framework. Available at http://www.rosettanet.org/RosettaNet/Rooms/DisplayPages/ LayoutInitial?Container=com.webridge.entity.Entity[OID[AE9C86B8022CD 411841F00C04F689339 RosettaNet. (1999b). UML extension for e-business partner interface process modeling, version 0.5 draft. Sinha, A. P. & Vessey, I. (1995). End-user data modeling: An ontological evaluation of relational and object-oriented schema diagrams. Working paper, Indiana University, Indiana. Skinstad, R. (2000). Business process integration through XML. Netfish Technologies [online]. Retrieved from http://www.infoloom.com/ gcaconfs/WEB/paris2000/S10-03.HTM Skonnard, A. & Laskey, B. (2000, May). Biztalk server 2000: Architecture and tools for trading partner integration. MSDN Magazine [online]. Retrieved from http://msdn.microsoft.com/msdnmag/issues/0500/biztalk/default. aspx Smith, H. & Fingar P., (2003). Business process management — The third wave. Tampa, FL: Meghan-Kiffer Press. Sutton, S. M., Tarr, P. L., & Osterweil, L. J., (1995). An analysis of process languages. CMPSCI technical report 95-78, Dept. of Computer Science, University of Massachusetts, Amherst. Thomas, D. (2003). UML — Unified or universal modeling language? Journal of Object Technology, 2(1), 7-12. van der Aalst, W. M. P. (1998). The application of Petri Nets to workflow management. The Journal of Circuits, Systems and Computers, 8(1), 2166. van der Aalst, W. M. P. (1999). Interorganizational workflows: An approach based on message sequence charts and Petri Nets. System Analysis — Modelling — Simulation, 34(3), 335-367.



299

van der Aalst, W. M. P. (2000). Loosely coupled interorganizational workflows: Modeling and analyzing workflows crossing organizational boundaries. Information and Management, 37, 67-75. van den Heuvel, W. -J. & Weigand, H. (2000). Cross-organizational workflow integration using contracts. Proceedings of OOPSLA 2000 Workshops, Minneapolis, Minnesota. http://jeffsutherland.org/oopsla2000/ vandenheuvel/vandenheuvel.htm Wand, Y. & Weber, R. (1989). An ontological evaluation of systems analysis and design methods. In E. D. Falkenberg & P. Lindgreen (Eds.), Information systems concepts: An in-depth analysis (pp. 79-107). Amsterdam: North Holland. Wand, Y. & Weber, R., (1990). An ontological model of an information system. IEEE Transactions on Software Engineering, 16(11), 1282-1292. Wand, Y. & Weber, R. (1993). On the ontological expressiveness of information systems analysis and design grammars. Journal of Information Systems, 3(4), 217-237. Weber, R. (1997). Ontological foundations of information systems (Accounting Research Methodology Monograph Series No. 4). Melbourne: Coopers & Lybrand. Weber, R. (2003). Conceptual modelling and ontology: Possibilities and pitfalls. Journal of Database Management, 14(3), 1-20. Weber, R. & Zhang, Y. (1996). An analytical evaluation of NIAM’s grammar for conceptual schema diagrams. Information Systems Journal, 6(2), 147-170. Workflow Management Coalition. (1995, January). The workflow reference model. Available at http://www.wfmc.org/standards/docs/tc003v11.pdf Workflow Management Coalition. (1999, January). Interface 1: Process definition interchange. Available at http://www.wfmc.org/standards/docs/TC1016-P_v11_IF1_Process_definition_Interchange.pdf Zetie, C. (2003). Business process modeling is gaining speed. Giga Information Group, CIO.com [online]. Retrieved from http://www2.cio.com/analyst/ report773.html Zwass, V. (1998). Structure and macro-level impact of electronic commerce: From technological infrastructure to electronic marketplaces. In K. E. Kendall (Ed.), Emerging information technologies. Thousand Oaks, CA: Sage Publications.



Endnote 1

An earlier version of this chapter was published as Dussart, Aymeric, Aubert, Benoit, A., & and Patry, Michel. (2004), An Evaluation of InterOrganizational Workflow Modelling Formalisms, Journal of Database Management, 15(2), April-June 2004, 74-104.



301

Appendices Appendix 1. Final “as is” business blueprint S tart

S UP PL IE R

Is the artic le a newly m anufac tured i tem for the suppl ier?

Item d epo s it

No

Is th e ite m c om plian t w ith Y e s W ri te C erti fic ate of re qu ir em en ts ?

No

E nd

Yes Ship item w ith quali ty c ontrol doc um ents

T em porary i tem deposi t

S tudy o f no nco nfor m ity

D oes the it em sati sfy the Fi rs t Articl e Inspect ion requirem ents?

Com pl ianc e (C . of C.)

No

P lac e item s in inv en to ry

Item de po sit

Do cu m e ntar y Va ult

In s pec tion a nd te s t of ite m

W ri te suppli er report of non c onformi ty and tak e pic tures of item

No

W rite F irst A rticle Inspecti on Report

Do c . V au lt

Ins pe ct Qu ality c on tr ol d oc um e nts w ith item s re c eive d D o w e inspec t it em at rec eption for thi s s upplier?

Y es Ins pec tion a nd te st of ite m

Inventory

Is the item c om plian t with r eq uire m en ts?

Y es

B UY ER

P ro ce ed to Fir s t A rtic le in sp ec tio n of item

Y es

No

Qu ar antine

Plac e item in quarantine,

Stu dy no n- co nfor m ity and w rite s up plier r ep ort of non -c on for m ity A c c ept ite m ,

Is the item at v ariance w ithin w ri te certifi cate suppli er requirem ents but not Y es of acc eptance at varianc e wit h buy er purchase or annotate RN C orders AN D is perform anc e, interchangeability, m aintainability , pla ce ite m weight, safety (...) unaffected?

in in ve ntor y

I s the item at variance w ithi n s upplier requi rements but not at variance w ith buy er purc hase Y es orders AN D i s performanc e, i nterc hangeabi lit y, m aintainabili ty, weight, s afety (...) unaffec ted?

No

No

E nd W rite an d s end Ce rtific a te o f A c ce ptan ce W r ite ce rtifica te o f r eje ction



Appendix 2. Proposed “to be” process MANUFACTURER

e-HUB

SUPPLIER

l:Articles [Not Inspected]

Check if article is newly manufactured [existing production]

[newly] Proceed to FAIR

f:FirstArticleInspectionReport > [COMPLETED]

l:Articles [compliant] [In First Inspection]

[not compliant]

Write FAIR Proceed to inspection [not compliant]

:Message > Reception of e-mail

Write supplier report of non-conformity

Flag Manufacturer for Non-Conformity

l:Articles [In inspection]

[compliant]

l:Articles [non-conform]

c:CertificateOfConformity > [COMPLETED]

Study Non-conformity [acceptable deviation] [unacceptable deviation]

s:SupplierReportofNonConformity > [COMPLETED]

a:AcceptanceReport > [COMPLETED]

Write Certificate of Acceptation

Flag Supplier for Acceptation of non-conformity

Fill Certificate of Rejection

r:RejectionReport > [COMPLETED] Flag Supplier for Rejection of non-conformity

:Message >

Reception of e-mail

:Message >

Wrtie certificate of conformity

l:Articles [Conform]

Ship Articles

Reception of e-mail

Discard Articles

Receive Articles

Inspect QC documents Inspection at reception for supplier?

l:Articles [Received]

[Supplier with inspection]

[Trusted Supplier]

l:Articles [In manufactuer inspection] [Non-compliant]

Inspect Articles [Compliant]

Place articles in quarantine l:Articles Write report of [Quarantined] non-conformity

r:ReportOfNon-conformity [Completed]

Study deviation

Place articles in inventory

Annotate RNC[Acceptable Deviation]

l:Articles [Stored]

[Discard articles]

l:Articles [Discarded]

[Return for modification ] Return article to supplier

l:Articles [Returned]

Discard articles

The new process is organized around the outcome: having a shipment of compliant parts in the manufacturer’s inventory. The original process involved several non-value added activities. In the new process, document-related activities only involve information capturing on a Web-based interface, thus reducing the number of channels to one. All other activities aim at fulfilling the final outcome of the process. The primary control variable of the flow is the state



303

of the article shipment item, which is purely outcome-related. If the outcomes are separated as having 1) a compliant shipment of items sent to the manufacturer and 2) an inspected shipment of items in inventory, both activities are performed by those who will use the outcomes. The new process is to be executed by a centralized WFMS and will coordinate work as if it was done in a single organization (geographical independence). Moreover, all documents are stored in a centralized location. Through aggregation, the system aims at minimizing quality control costs. The decision points are all located where the work is done except for the “study non-conformity” activity. But again, this activity cannot be modified because it has to be completed by the manufacturer. Finally, information capture is now done at the source.

Appendix 3. State machine for the 1:articles object

Not Inspected

articleChecked [is already produced]

articleChecked [is newly produced]

In Inspection

In First Inspection FAIRCompleted [is non-com pliant]

m anufacturersEvaluation Completed [refuses deviation] non-conform inspectionCompleted [is not-compliant]

inspectionCompleted [is com pliant] m anufacturersEvaluation Completed [accepts Conform deviation] articlesSent

FAIRComplete d [is com pliant]

supplierEvaluated [is trusted]

Received supplierEvaluated [is not trusted]

Stored

inspectionCompleted [is compliant] deviationStudied [is acceptable]

Quarant ined

Discarded deviationStudied [is unrepairable]

In m anufacturer inspection inspectionCompleted [is not-compliant]

Returned deviationStudied [is repairable]



Appendix 4. Possible interaction diagram for the new process

M A N.

e-HUB

S UPP .

W rite non-conform ity report

Inspect articles

Flag m anufacturer C onsult report Inspect deviation

report Send decision Flag supplier C onsult report report

C onsult decision and prepare shipm ent

Send articles C onsult quality reports Inspect articles Quality reports


Issues in the Evaluation of System Analysis and Design Techniques

305

Chapter XI

Methodological Issues in the Evaluation of System Analysis and Design Techniques Andrew Gemino, Simon Fraser University, Canada

Abstract This chapter examines methodological issues arising in the comparison of systems analysis and design techniques. An argument is made to establish a foundation of research and more broadly consider the management of scope in analysis and design research. A discussion of why and how we evaluate techniques is provided. A generalized approach combining both deductive and inductive reasoning is presented and a combined grammarbased and cognitive-based approach to comparison is discussed. In addition, concepts from Friedman’s economic methodology are applied in the choice between alternative ontologies that underlie grammar-based comparisons. The chapter concludes with a set of nine questions that researchers should consider when designing and developing research in the evaluation of systems analysis and design techniques. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

306 Gemino

Introduction In his book Ontological Foundations of Information Systems, Weber (1997, p. 30) suggests: The way ahead, I have argued, lies in sustained efforts to develop paradigmatic foundations for the discipline. If we fail to develop such foundations, I believe the IS discipline will remain fragile… we will have squandered our chances (yet again) of coming to a deep understanding of the nature and purpose of information systems. Weber’s words should resonate strongly when we consider two points: 1) that investment in information systems has taken an increasingly large share of the capital invested in western economies (National Science Board, 2003) and 2) in the midst of this investment, information systems courses are being excluded from business school accreditation (Ives, Valacich, Watson, & Zmud, 2002) and from curriculum in MBA training (Avison, 2003). It seems clear that while information systems are recognized as increasingly important, training in the management of information systems is not recognized with the same level of importance. It is in this context of paradigmatic foundations that we consider the methodology of technique evaluation, that is, how we compare techniques in the area of systems analysis and design. The discussion of this topic is divided into three sections. The first section introduces the importance of evaluating the techniques that help define scope. The next section focuses on the type of representations we compare and the two basic functions in modeling: reading and writing. The third section suggests a generalized approach to evaluating analysis techniques, along with some considerations of this general method. While this chapter is concerned primarily with how we evaluate analysis techniques, the lack of paradigmatic foundations that Weber laments overshadow our discussion of generalized methods for comparisons. For this reason, it seems natural to preface the main topic of the chapter with a few remarks about the relationship between systems analysis modeling techniques and the management of scope in information technology projects.



307

Scope and Systems Analysis and Design The Project Management Body of Knowledge (PMBOK) defines nine areas of project management (scope, time, cost, quality, risk, human resources, communication, procurement, and integration) and five project phases (initiating, planning, execution, control, and closing). The use of system analysis modeling techniques impact scope management directly because analysis techniques help analysts and stakeholders define expectations and support the process of requirements definition in the context of planning. Since failures occur when a system does not meet its objectives, analysis techniques are also important in the management of risk and quality. The use of analysis techniques can therefore affect five project management areas (scope, quality, risk, time, and cost). It should also be recognized that scope definition is revisited throughout the project's lifecycle, hence, analysis techniques that help to define scope are relevant throughout the project. The importance of clear requirements is clearly evident in practitioner research into factors affecting project success (Standish Group, 1999). Given the importance of analysis techniques, I believe a fundamental issue underlying the analysis, design, development and implementation of information systems remains the management of scope. While the accounting discipline informs the practice of cost management and operations management has created PERT and Gantt charts for time management, what business school function considers scope? The almost universally accepted IS tradition of training in system analysis and design places the IS discipline as a natural leader in the effective management of scope. Since defining requirements is a recognized skill of importance in IS curriculum, it seems arguable that the management of scope is a discipline that, at least for IS projects, could be exclusive to the IS banner. In searching for paradigmatic foundations, it therefore seems reasonable to consider a research area focused on managing scope. Yet research in the area of systems analysis techniques remains, at least in the casual perusal of top journals in the IS discipline, more peripheral than core. As Weber (1997, p. 8) notes in an aside, “(I count myself among those who are currently viewed as fringe dwellers in the IS discipline).” Many reasons for the relative lack of interest in analysis and design research can be asserted. For example, some might argue the journal limelight is taken by more engaging research topics such as factors affecting the adoption of information technology or the alignment of technology investments with business strategy. These topics are of obvious importance to a large number of organizations and rightly receive a great amount of attention in IS journals. I believe researchers in the area of systems analysis and design would be better served to view the lack of attention to their research not as the result of more engaging topics, but rather a result of


308 Gemino

less compelling research in our area. I believe that in the market place of ideas, increased recognition will only come as a result of improving the quality of the research developed in the area of systems analysis and design. In addition to improved quality, I believe additional journal focus will be provided as we focus more broadly on issues surrounding the management of scope in IS projects. The issue of scope management involves not only the definition of requirements, but also the tools and strategies used to effectively communicate scope to a variety of stakeholders (users, analysts, developers, quality assurance). Scope management includes the processes to control changes and adaptations in projects. And in order to communicate scope, we will need an ability to track it closely, particularly as a projects move through inevitable changes in objectives and requirements. These issues are of a broader concern to managers and the fact that improved scope management may likely positively impact project success rates provides an engaging topic for researchers and practitioners alike. It is for these reasons that I suggest researchers in our area look towards the management of scope as a foundation for their research.

Why Evaluate Scope Modeling Techniques Having made some arguments for the importance of scope management in the IS discipline, we turn our attention to the importance of evaluating techniques for defining scope as represented by IS models. The use of techniques such as data flow diagrams or entity relationship diagrams can improve the clarity of requirements, reduce changes in these requirements and provide more realistic objectives, all of which serve to improve the potential for success in IS projects (Johnson, Boucher, Connors, & Robinson, 2001). Given the importance of techniques in managing scope, it is not surprising a large number of techniques have been proposed (Avison & Fitzgerald, 1995; Chatzoglou & Macaulay, 1996; Oei, van Hemmen, Falkenberg, & Brinkkemper, 1992). The abundance of techniques and a lack of comparative measures have created a need to evaluate alternatives (Johnson, 2002; Wand & Weber, 2002). Many researchers have responded to this need and described useful and engaging comparisons (Batra, Hoffer, & Bostrom, 1990; Bodart, Sim, Patel, & Weber, 2001; Burton-Jones & Meso, 2002; Gove & March, 2003; Green, 1996; Siau & Benbasat, 1997). Workshops such as the Evaluation of Modeling Methods in Systems Analysis and Design associated with CAiSE and the symposiums on Research in Systems Analysis and Design (2002, 2003, 2004) are indications of the growth in research in this area. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


309

But while progress has been made, the foundations for evaluating techniques remain somewhat undefined. Useful frameworks have been developed (Batra et al., 1990; Topi & Ramesh, 2002; Wand & Weber, 2002) and foundations have been proposed (Wand & Weber, 1993; 1995; Weber, 1997), but little consensus has been reached. While some would argue consensus is not necessary for progress, Weber (1997) has argued that some agreement on principles underlying information system modeling would help to focus research in the area. In considering comparisons between scope definition techniques, a clear distinction should be made between the techniques we might compare (such as object oriented analysis, structured analysis, IDEF0, data modeling and many others) and the methodology we use to compare these techniques. It is to the latter that this chapter is addressed. In addition, this chapter narrows the focus by removing considerations of specific techniques (such as measurement instruments, constructs, experimental treatments, ontological analyses) used to develop these comparisons. This is not to suggest that the pursuit of improved techniques for evaluating modeling techniques is without value. To the contrary, it is a necessary pursuit in the development of the area. However, studies of evaluation techniques can shed only a little light on the objectives and assumptions underlying technique evaluations.

Evaluating Alternative Techniques In considering methods for technique evaluation, the journey begins with a simple question: “How can we evaluate alternative modeling techniques?” To understand this question, we must agree on some basic ideas relating to modeling of information systems. If one accepts, as I do, the argument made in Weber (1997, p. 65) that information systems are “representations of histories of things in the real world in terms of the way we have chosen to conceive them” then it is follows that models of information systems are representations of representations of histories. In my experience, statements such as these inevitably lead to conspicuous rolling of the eyes on the part of the reader and the use of mathematical formalism on the part of the writer. While the formalism used by Weber (1997) and Wand and Weber (1993) provides a clear, albeit sometimes challenging path through definitions, I have decided to minimize formalism in this discussion. I willingly accept a larger degree of ambiguity, which I hope to trade off with a somewhat simpler message.


310 Gemino

Four Representations to Consider To begin the discussion, I believe few would argue that a model is a representation and that a model presents information about some things. These statements are consistent with the first proposition in a model of information outlined by Weber (1997, p. 89). In this proposition Weber suggests that, “information is a representation of some thing(s) in the world.” In regards to modeling, this suggests we must recognize the fundamental difference between the thing(s) being modeled and the information about the thing(s) — the model itself. I also believe it is not controversial to suggest that information about a thing can be presented and eventually understood. This statement would be consistent with Weber’s (1997, p. 89) second proposition stating that “things can be known only via information.” This brings us to an important point: Is information presented to a person equal to knowledge? The statement to consider is whether two people viewing the same model, and presented with the same information, will necessarily develop the same cognitive representations of what is being modelled. If the answer to this statement is no, then the researcher accepts that information presented is not necessarily equal to knowledge gained and he or she defines himself or herself as a constructivist (Gemino & Wand, 2003). I would argue that individuals construct knowledge from the combination of information presented and personal, and therefore unique, long-term memory. If the response to the above statement is yes, then the researcher believes information presented is equal to knowledge gained and the researcher can be seen to ascribe to the information-processing model of human cognition (Mayer, 2001). In this view, information is an input that is processed, in a similar way, by a large majority of individuals. This process results in a similar output (knowledge) across many individuals. The choice between a constructivist view and an information processing view is surprisingly important. I believe the information processing view underlies much of the research in the area, particularly where the focus is placed on what is presented rather than how it is understood and processed into knowledge. In my opinion, the constructivist view provides more flexibility in evaluation. Researchers can take into account individual differences, for example in terms of domain and modeling experience, while not precluding the idea that a good representation can be similarly understood by a large number of people. When we argue about things in a model, and not things in cognition, I believe we restrict ourselves to a discussion that assumes everyone viewing the model understands it in the same way. I do not believe this to be true. It is my view that the presentation of information about a thing does not imply knowledge about a thing. Understanding is a process that results in a different type of representation — a cognitive representation of knowledge — that is Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


311

fundamentally different from the model it is derived from. In my view, information presented is not equal to knowledge gained. So in evaluating modeling techniques, Weber’s second proposition can be used to suggest there are really four representations to consider: 1) things in the world, 2) information about things in the world, 3) understanding of things in the world, and 4) understanding of information about things. Note that I do not include modeling techniques or knowledge of these techniques as representations. This is because modeling techniques are the language we use to express information about things. While the language is critical to developing effective communication, the language is not an end in itself but rather a means of expressing understanding. In using the word “understanding”, I separate an individual’s understanding from the more general concept of knowledge (which may be common across individuals). These four representations are shown in Figure 1. The four categories of representations are addressed with different tools. For example, how we define “things in the world” is addressed by ontologies and meta-models. These tools provide a well-defined language for describing things in the world. “Information about things” is created, in our area of research, using analysis and design modeling techniques such as use case models or entity relationship diagramming. These modeling techniques provide a simpler but less complex language for developing “models of things in the world” than ontologies and meta-models. Analysis and design modeling techniques are tools that help us to transforming things in the world into models of things of the world. They are

Figure 1. Four representations to consider in IS modeling

1

2

Things in the world

Information about things in the world

3

Understanding of things in the world

4

Understanding of info. about things


312 Gemino

abstractions of things that should be of interest to stakeholders in a systems project. To compare alternative modeling techniques, it is often considered important to measure the understanding that has been gained from viewing or creating a model. This requires an empirical instrument to evaluate a person’s “understanding of information about things” While this perhaps provides a measure of a person’s understanding of the model, I would suggest another important issue to consider is what a person knows about the real world as a result of viewing the model. This suggests that “understanding information about things” and “understanding things in the world” are not always congruent. In my view, a researcher considering an evaluation of modeling techniques should reflect on all four of these representations. For example, consider an experimental evaluation of two modeling techniques. The first step should be to clearly define differences between the alternatives (Gemino & Wand, 2004). These differences are only apparent when they can be compared against a complete set of modeling constructs, such as those provided by an ontology or meta-model. Once the differences are clearly defined, then an experiment can be developed. Study participants might then be run through an experimental procedure and measurements taken. An important consideration in the choice of instruments will be whether the researcher is measuring a participants’ “understanding of information about things” or “understanding of things in the world”. I would argue that if the objective of modeling techniques is to communicate information about things in the real world, then the latter, understanding of things in the world, is a more appropriate item to measure. Others might argue the real task is for the model viewer to understand the model; hence understanding of information about things would be appropriate. The important issue, however, is to recognize and clearly identify which representation is being considered. This will help to develop clearer and more useful comparisons.

Functions of Modeling There are two basic functions served in any modeling exercise: reading and writing (Wand & Weber, 2002). The two functions suggest two dimensions for evaluating modeling techniques. Norman’s Theory of Action (1986) is useful in understanding these functions (Gemino & Wand, 2004b). The theory is depicted in Figure 2. The theory is best understood by considering two persons who are involved in modeling. The model creator has a goal to communicate his or her understanding by representing his or her understanding in a model. A model viewer has a goal



313

Figure 2. Norman’s theory of action as applied to IS modeling

Design Model

Viewer Model

Designer (model creator)

Stakeholder (model viewer) Model

to understand the domain being represented by interpreting the model. Norman theory clearly distinguishes between a person’s understanding, held in cognition, and the model that is expressed physically. Discrepancies between a person’s understanding of the system and the model used to represent the system leads to issues in both creating and interpreting the diagram. Norman labels these discrepancies the “gulf of execution” and the “gulf of evaluation”. A gulf of execution forms when discrepancies exist between the conception of the domain from a model creator’s viewpoint and the model of the domain (for example a diagram). In this case the model creator is frustrated because the model does not represent his/her view of the domain accurately. These discrepancies may occur due to: 1) constraints on the expressiveness of the technique, 2) lack of skill of model creator, or 3) confusion in the model creator’s conception of the domain. A gulf of evaluation occurs when a discrepancy exists between a model viewer’s cognitive model of the domain and the model representing the domain. In this case, the model viewer is not getting the correct idea because there is a difference between what the model shows and what the viewer understands. The discrepancy may occur when: 1) the user misinterprets the diagram due to lack of experience with the technique, 2) the user develops a different conception of the domain being represented than the one conveyed by the diagram, or 3) there exists ambiguity within the diagram itself. Viewed through Norman’s theory, the modeling process can be seen as an effort to bridge the gulfs of execution and evaluation in order to bring the model Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

314 Gemino

creator’s and model viewer’s conceptions of the system (the design model and the viewer model), closer together. Norman’s theory is useful because it provides a clear separation between the domain represented in a model and the understanding of that domain as represented in the cognitive model of either the designer or the viewer. These insights suggest researchers need to clarify whether their focus is placed on the gulf of execution (where analysts create models) or the gulf of evaluation (where individuals interpret models). More importantly, Norman’s model suggest that a complete evaluation must include considerations of how the modeling technique bridges both gulfs.

A Generalized Approach to Evaluation Having discussed how we approach the comparison of scope defining techniques, our attention can turn more carefully to these comparisons. It is important to note we are not considering why we model, but rather, why we compare alternative techniques for modeling. It is hoped that insights into why we evaluate might help to define not only the research objectives of the area but also the principles underlying these objectives.

Inductive and Deductive Approaches to Questions of Comparisons Gemino and Wand (2003) have suggested that if the objective of comparing alternative modeling techniques is to find which modeling technique performs better, then empirical tests and comparative results fulfill this objective. This is a phenomenon-based approach to comparison based on purely inductive principles, that is, using data to infer performance of techniques. Observations would provide comparative information from which we could infer the preferred analysis techniques when practitioners needed to choose between modeling techniques. While this may be one of the objectives of evaluating alternatives, Gemino and Wand (2003) suggest it should not be the ultimate objective. Instead, the objective should be to understand why these differences occur. Focusing exclusively on results cannot provide explanations of why observed difference exist. To explain why differences exist requires a theory that enables a deductive reasoning directly relating characteristics in a modeling technique to differences in the performance of individuals who use the technique. A deductive approach suggests a reason why performance difference will be observed. We therefore evaluate techniques in order to test theories of how characteristics of



315

modeling techniques affect the eventual understanding of individuals viewing or creating models.

An Argument for Evaluation Given the discussion just mentioned, the evaluation of techniques is perhaps best viewed in the familiar format of an argument. For example, an argument suggesting technique “A” is a better modeling technique than technique “B” might be crafted in the following manner: 1.

Technique A has a higher level of ontological clarity than technique B.

2.

Viewing models that are created using techniques with higher ontological clarity leads to better understanding in individuals .

Therefore: 3.

Individuals viewing models created with technique A will show a higher level of understanding than individuals viewing models created with technique B.

It is useful to consider this argument carefully. The first proposition is a statement that could be verified by comparing techniques A and B to some benchmark set of constructs in an ontology or meta-model. Which ontology or meta-model is used is not important yet (we will return to this question in the next section). Gemino and Wand (2003) have identified this as a grammar-based approach to evaluation. But in and of itself, the first statement does not provide any argument as to which technique will perform better for people. It is simply a statement of fact in regards to mapping of constructs in two techniques to one set of ontological or meta-model constructs. The second proposition differs from the first because it suggests something about individuals viewing models. It is at this point that the area of IS analysis separates from that of computer science. In the IS discipline, the model viewer is an important part of the modeling process. The second proposition is not an objective fact, but rather a statement that is hopefully derived from an underlying theory about how humans perceive characteristics in models. Gemino and Wand (2003) have identified this a cognitive-based approach to evaluation. It is here, in statements such as the second proposition, that I believe the area of scope definition would be best served to focus its attention. Specifically, our research area needs theories to address why a characteristic of a modeling technique, Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

316 Gemino

such as ontological clarity, makes a difference to a model viewer’s understanding. While there are many forms that a statement such as the second proposition can take, none of these statements will ever be objective in the way that the first proposition can be considered objective. Statements like the second proposition can always be questioned and must be backed by theoretical arguments with as much clarity as possible. But even with a cognitive theory, a statement like the second proposition provides little comfort. It is not difficult to consider alternative theories that could argue exactly the opposite expected result from that provided in the second proposition. This is why statements such as the third proposition are so important. Without a testable hypothesis, such as that given in the third proposition, there can be little hope of advancing our understanding of how to make more effective and efficient scope definition techniques. These testable hypotheses give us an opportunity to evaluate alternative theories using empirical data. This recognition is of course nothing new, as Popper (1934/1959) attests to. This argument can therefore be generalized to provide a structure that, I argue, can underlie evaluations of systems analysis and design techniques. This argument structure generally coincides with the approach suggested in Gemino and Wand (2003). The argument is summarized in three propositions: 1.

A consideration of characteristics of techniques being compared.

2.

A deductive statement based on theory that suggests why these characteristics will impact the dependent measure being considered.

3.

A hypothesis that is crafted so that it can be evaluated inductively.

Forming our evaluations in this way provides us with the best of both the inductive and deductive approaches. The empirical results provide us with phenomena that can address issues regarding the performance of alternative techniques. Using theory to drive our comparisons provides an opportunity to build theory and recognize principles for good design of modeling techniques.

Considerations for the Generalized Approach One issue to consider in regards to this general approach relates to the first proposition. What is meant by an objective consideration of characteristics of



317

modeling techniques? Gemino and Wand (2003) have argued earlier that this comparison should be addressed using a benchmark set of constructs. I have used ontology and meta-model, interchangeably, as candidates for these benchmarks because both ontology and meta-models attempt to provide construct definition at a more abstract level than the level provided in analysis techniques. However, a similar issue to the question of “Which theory?” in the second proposition can be raised in the first proposition: “Which ontology or meta-model should be used as a basis for comparison?”. Again, each researcher faces a choice in addressing this question. The first choice in coming to this realization is perhaps the largest; and it is the issue regarding truth. The question to consider is the following: “Is there one true underlying ontology (meta-model) for information systems modeling?”. The researcher answering yes to this question would also likely accept that “all relevant constructs for IS modeling are correctly defined in this true ontology”. But at this point, the researcher would be trapped into a rather difficult problem of induction akin to proving statements such as “All swans are white.” These problems have generally been viewed as intractable. On the other hand, if the researcher answers no to the question above, then the researcher finds himself or herself in a problem of a different sort. If there is no true ontology (meta-model), then how can you choose between alternatives? For example, if two alternative ontologies are well-formed and each provided its own set of clearly defined constructs so there was no logical inconsistency in either definition, how can you choose between them? To address this question, the field of economics provides some insight. In this field, researchers can face comparing alternative economic theories built on different sets of assumptions but that address a similar question (Boland, 1982). Since well-formed economic theories are essentially “closed worlds” that are developed from founding assumptions, they are similar to ontologies and metamodels used in IS research. Some have argued that economic theories might be compared based on the realism of assumptions; however, Friedman (1953) has shown the futility of these types of arguments and instead suggests that, “viewed as a language, theory has no substantive content; it is a set of tautologies. Its function is to serve as a filing system for organizing empirical material and facilitating our understanding of it” (p. 7). This view of theory is similar to our view of ontologies and meta-models as languages that serve as analytical filing systems for constructs. Given this view, the question arises how we can compare alternative theories. Again Friedman (1953, p.7) provides insight:


318 Gemino

The answer to these questions depend partly on logical, partly on factual considerations. The canons of formal logic alone can show whether a particular language is complete and consistent, that is whether propositions in the language are “right” or “wrong”. Factual evidence alone can show whether the categories of the “analytical filing system” have a meaningful empirical counterpart, that is whether they are useful in analyzing a particular class of concrete problems. So the first test of any ontology or meta-model is logical completeness and consistency. This should be a relatively objective exercise. Once an ontology or meta-model has passed this logical test, it can then be used to identify differences among modeling techniques. The impact that these differences have on participants can then be hypothesized using cognitive theory and eventually tested empirically. The ontology (meta-model) that is “better” is the ontology that provides us with differences that lead to “useful” empirical results. In this case useful can be defined as results confirming both the differences identified and their significant impact on participants’ performance. A useful ontology would therefore provide us with a language for identifying differences between techniques. Further, a useful ontology would establish differences that were found to affect the performance of the technique in a significant way across a variety of technique comparisons. In adopting the “usefulness” of the ontology as the main criteria for comparison on ontologies or meta-models, the researcher adopts the conventionalist view. This view suggests that no truths are possible based on induction (inferring from data). So rather than search for truths, we compare ontologies on the basis of agreed upon conventions. Examples might include ease of use, largest number of constructs or any other criteria we see fit to use. We have argued above for the criteria of usefulness. One should note that saying ontologies are useful does not imply that they are true. It is my belief, therefore, that outside of arguing about completeness and consistency, an argument about which ontology is better on any level other than an empirical level in systems analysis and design is an exercise unlikely to bear significant academic fruit.

Conclusions The discussion just mentioned highlights a number of important questions that researchers might consider when developing evaluations of IS scoping techniques. The questions are in no way meant to be exhaustive. Some questions not



319

considered in the following list, for example, are provided in Gemino and Wand (2004). The questions offered here are intended to make explicit some additional considerations relating to technique comparisons. 1.

Does the researcher allow for individual differences in understanding gained from viewing a model?

2.

Which of the four representation(s) will be considered in developing the comparison between models?

3.

Will the focus be placed on issues relating to the gulf of execution and/or on the gulf of evaluation?

4.

In making the comparison, has the researcher clearly identified the differences between modeling techniques? Ontologies and meta-models are suggested for this purpose.

5.

In justifying these differences, has the researcher established the completeness and consistency of the ontology (meta-model)?

6.

In regards to the use of ontology (meta-model), will the researcher be relying on the convention of “usefulness” in regards to the use of the ontology (meta-model) or some other convention?

7.

Has the researcher been able to provide a theoretical link explaining how the differences between techniques result in different cognitive outcomes?

8.

Were the outcomes measured with an appropriate instrument?

9.

Are the results from the empirical procedures significant and meaningful?

It is hoped these questions will help researchers become more aware of the choices they face in evaluating models. This awareness should help to define comparisons more precisely and perhaps begin to build a foundation with a more cumulative tradition. Weber (1997) has argued that the way forward is to develop foundations in our research area. In regards to the evaluation of techniques, I believe this begins with an understanding of the assumptions and principles underlying our techniques. It is my hope that this chapter provides some useful beginning in regards to this discussion. In addition, I hope the reader considers the broader issue of the management of scope as a context for further research into the area of IS analysis and design.


320 Gemino

References Allen, G.& March, S. (2003). Modeling temporal dynamics for business systems. Journal of Database Management, 14(3), 21-36. Avison, D. E. (2003, January). Information systems in the MBA curriculum: An international perspective. Communications of the Association for Information Systems, 11(6), 117-127. Avison D. E. & Fitzgerald, G. (1995). Information systems development: Methodologies, techniques, and tools (2nd ed.). London: McGraw-Hill. Batra, D., Hoffer, J., & Bostrom, R. (1990, February). Comparing representations with relational and EER models. Communications of the ACM, 33(2), 126-139. Bodart, F., Sim, M., Patel, A., & Weber, R. (2001, December). Should optional properties be used in conceptual modelling? A theory and three empirical tests. Information Systems Research, 12(4), 384-405. Boland, L. A. (1982). The foundations of economic method. London: George Allen and Unwin. Burton-Jones, A. & Meso, P. (2002). How good are these UML diagrams? An empirical test of the Wand and Weber good decomposition model. In L. Applegate, R. Galliers, & J. I. DeGross (Eds.), Proceedings of the International Conference on Information Systems 2002, Barcelona, Spain, 15-18 December, 2002 (pp. 101-114). Chatzoglou, P. D. & Macaulay, L. A. (1996). Requirements capture and IS methodologies. Information Systems Journal, 6, 209-225. Friedman, M. (1953). The methodology of positive economics. In Essays in positive economics. Chicago: University of Chicago Press. Gemino, A. & Wand, Y. (2003). Evaluating modeling techniques based on models of learning. Communications of the ACM, 46(10). Gemino, A. & Wand, Y. (2004). Foundations for empirical comparisons of conceptual modeling techniques. Forthcoming in Requirements Engineering Journal, 2004. Green, P. (1996, May). An ontological analysis of information systems analysis design (ISAD) grammars in upper CASE tools. PhD thesis, Department of Commerce, University of Queensland, Australia. Ives, B., Valacich, J., Watson, R., & Zmud, R. (2002). What every business student needs to know about information systems. Communications of the Association for Information Systems, 9(30).



321

Johnson, J., Boucher, K., Connors, K., & Robinson, J. (2001, February/March). The criteria for success. Software Magazine, 21(1), s3-11. Johnson, R. A. (2002). Object-oriented system development: A review of empirical research. Communications of the Association for Information Systems, 8, 65-81. National Science Board, Subcommittee on Science & Engineering. (2003). Science and engineering indicators — 2002. Retrieved September 25, 2003, from http://www.nsf.gov/sbe/srs/seind02/c8/c8s2.htm Oei, J. L. H., van Hemmen, L. J., Falkenberg, E. D., & Brinkkemper, S. (1992, July). The meta model hierarchy: A framework for information systems concepts and techniques. Technical report No. 92-17, Department of Informatics, Faculty of Mathematics and Informatics. Katholieke Universiteir, Nijmegen, (pp. 1-30). PMBOK 2000. (2000). Project management body of knowledge. Project Management Institute. Retrieved September 25, 2003, from http:// www.pmi.org/info/PP_PMBOK2000Excerpts.asp Popper, K. (1934/1959). Logic of scientific discovery. New York: Science Editions. Siau, K. L., Wand, Y., & Benbasat, I. (1997). Information modeling and cognitive biases — An empirical study on modeling experts. Information Systems, 22(2&3), 155-170. Standish Group. (1999). CHAOS: A recipe for success. Retrieved July 15, 2004, from http://www.standishgroup.com/sample_research/PDFpages/ chaos1999.pdf Topi, H. & Ramesh, V. (2002). Human factors research on data modeling: An extended framework and future research directions. Journal of Database Management, 13(2), 3-20. Wand, Y. & Weber, R. (1993). On the ontological expressiveness of information systems analysis and design grammars. Journal of Information Systems, 3, 217-37. Wand, Y. & Weber, R. (1995). On the deep structure of information systems. Journal of Information Systems, 5, 203-223. Wand, Y. & Weber, R. (2002). Research commentary: Information systems and conceptual modeling — A research agenda. Information Systems Research, 13(4), 363-376. Weber, R. (1997). Ontological foundations of information systems (Accounting Research Methodology Monograph No. 4). Melbourne, Australia: Coopers and Lybrand.


322

Wyssusek & Klaus

Chapter XII

Ontological Foundations of Information Systems Analysis and Design: Extending the Scope of the Discussion

Boris Wyssusek, Queensland University of Technology, Australia Helmut Klaus, Queensland University of Technology, Australia

Abstract Ontology has attracted considerable attention in information systems analysis and design (ISAD) research. Ontology is philosophy and bears its own substance and history of debates, which quite often have not been accounted for in information systems research. A more comprehensive consideration of well-known philosophical issues of ontology may help to apprehend precisely the transfer of ontological concepts into ISAD, including insights regarding their limitations and to articulate directions


Ontological Foundations of Information Systems Analysis and Design

323

towards further research. In particular, this requires expanding of the scope of current debates in information systems towards the sociophilosophical aspects of ontology. Only then, it will be possible to determine whether ontology can direct the project of theoretical foundation for ISAD. An outline of the critique of the prevailing rationalistic methodical understanding of information systems development in contemporary IS literature illustrates how the indiscriminating borrowing of philosophical presuppositions has encumbered current understandings. Critical reflection upon these presuppositions can get over persuasions and bring about theorisation.

Introduction In the last two decades, ontology and ontologies have attracted enduring attention in the field of information systems research and practice, especially in the domain of information systems analysis and design (e.g., Checkland, 1981; Boland, 1982; Winograd & Flores, 1986; Wand & Weber, 1988; Floyd, 1992; Hirschheim, Klein, & Lyytinen, 1995; Weber, 1997b; Green & Rosemann, 1999; Milton, Kazmierczak, & Thomas, 2000; Fettke & Loos, 2003; Rosemann, Vessey, & Weber, 2004). The domain of information systems analysis and design is understood to be concerned with the analysis of “real world” systems — the determination of changes that should occur in the “real world” after the introduction or modification of an information system, and finally, based upon the elicited requirements, the design of information systems. Thus, of all domains within information systems research and practice, information systems analysis and design (ISAD) has the most and the strongest ties to the world “out there”. The process of ISAD is embedded in the whole systems development life cycle, that is, a methodical process that covers all activities from the identification of problems and opportunities to the implementation and evaluation of the system (Kendall & Kendall, 1992, pp. 66). In this context, information systems are commonly seen as representational systems, that is, systems that represent facts about the “outside world”. This presupposes knowledge of what there is to be represented, and how to represent it. Consequently, research on ISAD has turned to the philosophical discipline of ontology that is concerned with “being” and “what exists”. It is generally acknowledged that the central activity of analysis and design of information systems is modelling. In analysis, parts of the “real world” are described that should be represented in the information system. Correspondingly, in design certain characteristics of the information system to be developed are Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

324

Wyssusek & Klaus

described in models. In contrast to the arts, where modelling is understood as a highly creative and intuitive activity, modelling in the context of ISAD, conforming to the concept of the overall systems development life-cycle, is generally regarded as a methodical process, that is, following generally agreed upon and prescribed steps. The emphasis on, and the importance of modelling for ISAD is conspicuously demonstrated by the abundance of modelling methods, the continuous efforts of improving them and developing new ones. However, beyond modelling methods themselves, there is hardly any theoretical foundation for ISAD. This deficit has given rise to the interest of the ISAD research community in ontology and ontologies (e.g., Wand & Weber, 1988). Ontology is prestigious as a philosophical discipline, with tradition, famous individuals, a host of literature, and high reputation in the general scientific community. It thus opens the possibility for information systems research to draw on the findings of other well-established disciplines. However, this can be perilous without an awareness of certain limitations. Information systems researchers are seldom philosophers, which makes drawing on an unfamiliar discipline prone to the fallacies of gross misunderstandings, false analogies, simplification, and the like. Especially in philosophy, every concept comes with its own history of debates; precise definitions are rare, and to grasp fully a notion we first need to know and to understand the debates and their history. Nevertheless, engaging in philosophy is neither futile nor too laborious. In fact, it cannot be avoided, since a good part of the answer to the question “why philosophy?” is that the alternative to philosophy is not no philosophy, but bad philosophy. The “unphilosophical” person has an unconscious philosophy, which they apply in their practice — whether of science or politics or daily life (Collier, 1994, p. 17). Unconscious philosophy does not mean that every man is a philosopher by default. It only means that every man has the potential to think and to engage in philosophy. Gramsci (1971) holds that philosophy is not only the business of people called philosophers; on the contrary, language that is given to all humans, always deals with meaning of notions and concepts; every human shares the understanding of a community and strives to act ethically, and lives within a historically given structure of habits, rites and beliefs. Thus, in any intellectual activity and mentation, the human being, irrespective of it being explicit or not, enacts a specific understanding of world. For Gramsci (1971) then, when thinking becomes a conscious activity, thought is immediately connected to an ethical question, namely “Is it better to ‘think’, without having a critical Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


325

awareness, in a disjointed and episodic way?” (p. 324). In refuting the conscious adoption of imposed worldviews, Gramsci’s everyman’s thinking or philosophy cannot be but critical, since only then genuine awareness, action and participation in the world can emerge. This calls to mind that even more, when dealing with philosophical views, circumspection is required, or a critical approach. As the meaning of ontology and ontologies is bound to the respective philosophical doctrines used as horizons of interpretation, not only their immediate consequences for the understanding of ontology and ontologies must be paid attention to, but also to seemingly more distant consequences that are implicated in a particular doctrine, for example, ethical consequences. One common understanding of ontology and ontologies in the ISAD literature is based on the ontology of Mario Bunge (1977, 1979, 1993), which, alas, hardly finds support in contemporary philosophy and sociology. Thus, Bunge’s philosophical doctrine determines a horizon of interpretation that is rather detached from contemporary discourses on the social world. If Bunge’s ideas were followed consistently in unfolding a conceptual foundation for information systems, this foundation would be precluded to relate to current thinking, and would make its communication within the IS community and beyond rather taxing. Moreover, maintaining a dialogue with neighbouring disciplines, such as organisation theory, would be onerous likewise. In other words, the uncritical adoption of Bunge’s ontology restricts understanding of world to a particular doctrine that is solitary. This in turn narrows the perspective to such an extent that limitations as well as ethical consequences of the application of his ontology cannot be questioned. Its proponents may have already forfeited a discriminative concern with that doctrine, while there is little chance that clear-sighted comments from beyond this circle may ever be voiced. Yet, the issue is not supplanting Bunge’s ontology with another one. Rather, critical approaches that are already extant in contemporary information systems research proffer plenty of opportunities to engage in critical reflection on ontology as foundation for ISAD. For example, when following HabermaS (1972), the ontological foundation of information systems analysis and design should not solely be guided by an instrumental-technical or a practical-hermeneutical cognitive interest, but also by an emancipatory interest. In brief, methods have given rise to the interest in ontology in information systems research, and methods are also of interest in philosophy. By critically reconnecting the information systems discourse on method with that of philosophy and the question of language, we expect to make space for discussions on the meaning of ontological foundation of information systems analysis and design.


326

Wyssusek & Klaus

Setting the Scene Since the famous NATO conference in 1968, the information systems research community has been familiar with the notion of software crisis (Naur & Randell, 1969). Plagued by the ever-increasing complexity of hardware and software, the development of reliable, effective, and efficient information systems had become the major challenge for researchers and practitioners alike. It was also on the occasion of this very conference that the notion of software engineering was introduced. “The phrase ‘software engineering’ was deliberately chosen as being provocative, implying the need for software manufacture to be based on the types of theoretical foundations and practical disciplines, that are traditional in the established branches of engineering” (Naur & Randell, 1969, p. 13). Today, it is generally acknowledged that this conference was the birth of the discipline of software engineering, although in its early days, the phrase software engineering was of a metaphorical nature rather than an actual description of the state of affairs. Nevertheless, engineering was identified as the silver bullet to (almost) all the problems encountered when engaging in the development of complex hardware and software systems.

“Methodism” in Software Engineering What is engineering? Definitions of engineering abound, yet share some common clauses: “Creating cost-effective solutions … to practical problems … by applying scientific knowledge … to building things … in the service of mankind” (Shaw, 1990, p. 15). Obviously, the difference between craft or art and engineering rests on the application of scientific knowledge. Analogously to Ryle’s (1949) distinction between “knowing-that” (declarative knowledge) and “knowing-how” (procedural knowledge), scientific knowledge can be characterised as comprised of two different, yet complementary types: knowledge about the subject matter under investigation — the ultimate goal of science — and knowledge about how to achieve this goal. Thus, the latter type of knowledge is about means-ends-relationships, and when codified describes “a specification of steps which must be taken, in a given order, to achieve a given end” (Caws, 1967, p. 339). In general, we refer to such a specification as method, derived from the Greek µετα (meta: along) and οδοσ (odos: way) — following a way. With Descartes (1637), the application of the appropriate method of inquiry became the guarantor for obtaining truth. When engineering makes use of scientific knowledge, it draws on knowledge about the subject matter and on knowledge about means-ends-relationships, that is, methods. However, whereas in the realm of science the goal of applying (or following) a method is to gain knowledge, in the realm of engineering the goal is “building things”. Reaching this Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


327

goal is supposedly ensured — in the tradition of Cartesianism — when using the appropriate method: “The predisposition to believe in the power of methodologies comes from Descartes who proposed that truth is more a matter of proper method than genial insight or divine inspiration” (Hirschheim et al., 1995, p. 21). Even before Descartes had spelt out the dominance of method in the production of knowledge, the obsession of academics with method had already become a target of scathing polemics: Method — no word is more popular in our lectures these days, none more often heard, none gives off a more delightful ring than that term. Everything else, if you use it often enough, will end by nauseating your readers. This is the only thing that never makes them sick. If you leave it out, they think the feast you set before them is disgustingly seasoned and poorly prepared. If you use it often, they will believe that anything you give them is the ambrosial and nectared food of the gods (Turnèbe, 1600; quoted in Ong, 1958, p. 228). The sheer abundance of methods available and applied these days in software engineering means that the metaphor software engineering has been taken seriously and now appears as a description. Methodical software development is the state-of-the-art in the field. Still the question remains: “Have the expectations been met?” Recurring failures in the development of information systems (Standish Group, 1994, 2003; Boustred, 1997), the persistence of the software crisis (Gibbs, 1994), as well as the productivity paradox (Brynjolfsson, 1993; Attewell, 1994; Strassmann, 1997) point towards a negative answer. But why? Ever since the emergence of the discipline software engineering, information systems development has been understood as a “planned, deliberate activity — bounded in time and carried out in a systematic and orderly way” (Bansler & Havn, 2003, p. 51). In other words, it has been understood as a methodical process, to an extent that “the modern concept of method has been so strongly impressed on our thinking about systems development, that the two concepts, information systems development and information systems development method, are completely merged in systems development literature” (Truex, Baskerville, & Travis, 2000, p. 56). In contrast, empirical findings question the very idea of methodical information systems development, since methods are often unsuitable for some individuals (Naur, 1993) and settings (Baskerville, Travis, & Truex, 1992). Similar methods in similar settings yield distinctly different results (Turner, 1987). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

328

Wyssusek & Klaus

Developers may claim adherence to one method while ignoring this method in actual practice (Bansler & Bødker, 1993) (Truex et al., 2000, p. 54). Consequently, Floyd (1992) criticises the discipline’s “view of methods as rules laying down standardized working procedures to be followed without reference to the situation in hand or the specific groups of people involved” (p. 86). This suggests that while research has been preoccupied with method, it has simultaneously neglected (or ignored) the amethodical aspects of information systems development, as explained by Truex et al. (2000): “When the idea of method frames all of our perceptions about systems development, then it becomes very difficult to grasp its non-methodical aspects” (p. 74). In a similar vein, Bansler and Havn (2003) note that these aspects “become marginalized and practically invisible, [for example,] how ISD is subject to human whims, talents and the personal goals of the managers, designers and users involved” (p. 51). Truex et al. (2000), subjecting texts on information systems development to Derrida’s (1978, 1982) deconstructive approach, demonstrate the antagonism between privileged methodical and marginalised amethodical text (Table 1).

Rationalism and Post-Rationalism Methodical texts are a product of classical rationalist thought (Winograd & Flores, 1986, p. 14). Specifically, software development in the tradition of rationalism rests on the following assumptions:

•

There is a given reality “out there” which we come across during software development. By analysing the facts of this reality, we obtain requirements for the software.

•

The essential task of the software developer is — starting from the problem defined in that reality — to find a correct solution in the form of a program system.

•

It is possible to separate the production of software from its use. Software engineering is concerned with the production of software on the basis of fixed requirements.

•

Software production is based on models representing reality. Models should map reality correctly.

•

The whole process is largely independent of individuals. For one and the same problem, different developers should arrive



329

at the same results. The developers should be interchangeable.

•

Communication should be restricted and regulated via fixed interfaces. The division of labour can be worked out on an ad hoc basis. Subject to technical feasibility, any desired parts of the production process can be automated.

•

The developer’s responsibility covers — only — proper construction of the product in accordance with the requirements specification. Any ethical considerations that go beyond this are quite separate from the technical aspects of the work (Floyd, 1992, p. 89).

Table 1. Assumptions and ideals of methodical and amethodical texts (Truex et al., 2000, p. 59) Privileged methodical text

Marginalized amethodical text

1. Information systems development is a managed, controlled process

2. Information systems development is random, opportunistic process driven by accident

idealizing logical decomposition reductionism

idealizing holism creativity

3. Information systems development is a linear, sequential process

4. Information systems development processes are simultaneous, overlapping and there are gaps

idealizing temporal causal chain

idealizing fragmentation parallelism disconnectedness

5. Information systems development is a replicable, universal process

6. Information systems development occurs in completely unique and idiographic forms

idealizing generalization consistency formalisms

idealizing choice change adhocracy

7. Information systems development is a rational, determined, and goal-driven process

8. Information systems development is negotiated, compromised and capricious

idealizing goal predetermination process predetermination human cooperation

idealizing conflict social constructivism human independence


330

Wyssusek & Klaus

This rationalist thought is still guiding information systems research, despite having been confronted with strong criticism (Introna, 1996; Ciborra, 1998), whereas various other disciplines have already adopted post-rationalist or postpositivist thought that had emerged in philosophy during the past century. Of pertinence for ISAD is the prevalence of new thought in the neighbouring domains of organisation and management theory. Here, Burrell and Morgan (1979) have conceptualised sociological paradigms. In following them, the ontological and epistemological assumptions of post-positivist social-constructivist thought (applied to ISAD) have been rendered by Hirschheim et al. (1995): The epistemology is that of anti-positivism reflecting the belief that the search for causal, empirical explanations for social phenomena is misguided and should be replaced by the will and need to make sense of oneself and the situation. The ontology is that of nominalism (constructivism) in that reality is not a given, immutable “out there” but is socially constructed. It is the product of the human mind. Object systems emerge as part of the ongoing reality construction and the act of bounding the scope of object systems and defining requirements contributes to the ongoing process of sense making. The paradigm, social relativism, focuses on understanding social phenomena and is primarily involved in explaining the social world from the viewpoint of the organizational agents who directly take part in the social process of reality construction (p. 75). There is obviously a nexus between theory and practice of software development; for example, Hirschheim et al. (1995) argued that in the history of information systems development, methodologies and the respective approaches have always been informed by (socio-)philosophical foundations. This means for the proposal of new theoretical foundations for systems analysis and design (methods), that it must be informed by the outcome of past discourses; otherwise it would run the risk of repeating the past in an unreflected manner. Every effort towards the development of theoretical and, especially, ontological foundations of information systems analysis and design must therefore enter into a dialogue with past discourses. The pertinence of such a dialogue is exemplified in the following.



331

Modelling With or Without Philosophical Presuppositions? In a critique of a proposal by Hirschheim et al. (1995) for a less dogmatic approach to ISAD, Weber (1997a) denies the link between data modelling approaches and philosophical assumptions: From an ontological perspective, I argue that whether we adopt a realist position or an [sic] a subjectivist position has little bearing on the data modelling grammars we might choose to design and construct. If we are realists, we believe that the world exists independently of the knowing subject. Whether we can discover this world, however, is another matter. If we are subjectivists, we believe that the knowing subject constructs his or her world. The world does not exist independently of the knowing subject. In both cases, however, we use data models to represent the knowing subject’s world. We must choose modelling objects and rules that allow us to represent either an objective world or a subjective world. Whether we have an objective world or a subjective world has little bearing on the modelling objects and rules we select to incorporate in our data modelling grammars (p. 306). Weber’s denial of the nexus between presuppositions and practice of ISAD is ill-founded. First, Weber does (at first sight) not realise that Hirschheim et al. (1995) do not write about modelling grammars, but about methodologies (not in a philosophical sense!). Hence, their argument has a much wider scope, incorporating the philosophical presuppositions that underlie the documented practice of data modelling. Second, the ongoing efforts of Weber and others towards the development of an ontological foundation of modelling grammars (e.g., Wand, 1994; Wand & Wang, 1996; Wand & Weber, 1988, 1989, 1990, 1992, 1994, 1995; Weber, 1997a, 1997b; Weber & Colomb, 1998)—which evolved into the so-called Bunge-Wand-Weber ontology—are actually proof to the opposite of the intended statement. In fact, the entire Bunge-Wand-Weber approach is based on philosophical presuppositions that determine the very objects and rules to be selected for incorporation in the respective modelling grammar. If it were the case that philosophical presuppositions had no bearing on the selection process of methods, then why bother at all with ontological foundation of modelling grammars and methods? For Weber (1997a), the nexus between presuppositions and doing still requires empirical verification:


332

Wyssusek & Klaus

In practice, however, Hirschheim et al. (1995) may be right. Empirically we need to investigate whether an association exists between the type of data modelling approach that designers use and the philosophical assumptions they make. They [i.e., Hirschheim et al.] have provided us with a rich basis for theorizing about potential links between philosophical assumptions and data models. Moreover, they have highlighted the need for empirical research to be undertaken to determine whether putative links exist in data modelling practice (p. 308).

The Criticism of the Myth of Formal Method Another potentially enlightening critical discourse in information systems research does not target the development process but the very methods themselves. The use of formal methods is probably one of the most prominent results of the understanding of software development as an engineering process. There has been a flurry of papers advocating the use of “formal methods” in the software industry…. Academicians, with and without industrial experience, apologetically missionarize for formal methodism, under various degrees of radicalism. Sometimes, they even berate industrial software engineers for not using such supposedly formal methods, warning them of imminent disasters if these methods are not adopted, as they are perceived to be a key solution to the chronic software crisis (or plague, rather). In some cases, even practicing engineers are found preaching the gospel of formal methods to their fellow members of the industrial congregation (Le Charlier & Flener, 1998, p. 1). Formal methods and their application is a highly contested terrain: “Formal methods are controversial. Their advocates claim they can revolutionize development. Their detractors think they are impossibly difficult” (Hall, 1990, p. 11). The question whether formal methods in development are useful or detrimental detracts firstly from the circumstance that they are only conceivable within a particular constellation of power/knowledge (Westrup, 1999). Secondly, this instrumental view of methods obscures that they are not founded in themselves and in their utility, but as De Millo, Lipton, and Perlis (1979), Le Charlier and Flener (1998), and Balzer, Goldman, and Wile (1978) have argued, that formal methods are determined through a realm that can only be called informal. First, because formal methods are always constructed by means of necessarily informal methods. Otherwise, we would end up with an infinite regress. Second, Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


333

the application of formal methods takes place in an informal world. Third, the meanings of the constructs of formal methods and of the results of their application only exist in realm of the informal world. Formal semantics does not provide any meaning when not related to semantics of the informal world. Meanings are concepts that cannot be formalised; we can only formalise the representation of concepts (Le Charlier & Flener, 1998, p. 12). Nevertheless, formal methods are valuable instruments in the realms of information systems analysis and design (e.g., Hall, 1990; Bowen & Hinchey, 1995). However, a broader perspective must be taken for recognising their scope and limitations, and to avoid their mystification. For instance, even mathematical proofs — the very incarnation of the application of formal methods — can only be understood within a social, hence informal context (De Millo et al., 1979). The legacy of proofs that eventually have been proved wrong substantiates this claim. Consequently, when seeking to develop or to improve formal methods their always given informal and social aspects have to be taken into consideration. Similarly, it might be surmised that an understanding of informal and social aspects of a formal method has a more positive impact on its application than would have an improvement of the formalism of that method.

Ontology, Ontologies, and Information Systems Analysis and Design During the last two decades, ontology has gained considerable attention in the field of information systems research and development. Yet, it must be acknowledged that there is not a single concept of ontology. The plurality of meanings of the word suggests that a profound scepticism is appropriate if information systems analysis and design should be founded on a concept of ontology. Instead of delving into the etymology of the word “ontology”, it might suffice to note for the present purpose that, although ancient Greek philosophers have of course dealt with ontological problems, it was only in the 17th century that the word “ontology” was introduced to denote a branch of philosophy (contemplative sciences) (Goclenius, 1613, p. 16). An encyclopaedic definition would be: The word “ontology” is used to refer to philosophical investigation of existence, or being. Such investigation may be directed towards the concept of being, asking what “being” means, or what it is for something to exist; it may also (or instead) be concerned with the question “What exists?”, or “What general sorts of things are there?” (Craig, 1998). Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

334

Wyssusek & Klaus

This definition suggests that both questions — “What does it mean for something to exist?” and “What exists?” — could be answered independently, or that the question of meaning can be avoided altogether. Already Kant (1929) had criticised the notion that ontology could “supply, in systematic doctrinal form, synthetic a priori knowledge of things in general” (p. B303). He held that neither realism nor idealism as ontological stances could provide this fundamental knowledge. Thoughts without content are empty, intuitions without concepts are blind. It is, therefore, just as necessary to make our concepts sensible, that is, to add the object to them in intuition, as to make our intuitions intelligible, that is, to bring them under concepts. These two powers or capacities cannot exchange their functions (Kant, 1929, p. B75). Moreover, Kant reversed the classical view of epistemology, an insight known as the Copernican turn in philosophy. Instead of understanding knowledge as conforming to objects, we have to understand the objects as conforming to the conditions of our knowing. Thus, human knowledge is limited to appearances; we are not able to know of the things-in-themselves, as it was claimed earlier for ontology. In brief, according to Kant, all ontology is epistemic bound, or ontology without epistemology, that is, without considering the possibilities of knowing, is without any merit. The concept of ontology has been further radicalised in the last century by Heidegger. In response to Kant (1929), to whom the “scandal to philosophy” was that no proof has yet been given of the “existence of things outside us” (p. Bxl), Heidegger (1962) argues that the scandal is “not that this proof has yet to be given, but that such proofs are expected and attempted again and again” (p. 249). According to Heidegger, the question of earlier ontology concerning the nature of reality of the external world poses a pseudo-problem. For Heidegger (1962), the task of ontology is to deal with the question of meaning of Being and not to lay out principled knowledge of world: Basically, all ontology, no matter how rich and firmly compacted a system of categories it has at its disposal, remains blind and perverted from its own aim, if it has not first adequately clarified the meaning of Being, and conceived this clarification as its fundamental task (p. 31).



335

In the classical understanding dating back to ancient Greek philosophy, ontology was understood as a discipline within or next to metaphysics. There was no plural of the word ontology, as there still are no “biologies” although different scientists have a different understanding of biology. The use of the word ontology in its plural form, points however to the idea of the connectedness of world and language, which has been a philosophical issue at least since the early 19th century, when Wilhelm von Humboldt (1836) had discovered that people of different languages construct their world differently. The significance of language for the knowledge of world was further radicalised by Nietzsche (1954), arguing that our entire understanding and knowledge of the world is bound to our language, and denying any objective meaning or any actual reference of language to world, posited that this knowledge is of essentially metaphorical nature. Even when reference between knowledge and world was conjectured, as in the logical positivism of the Vienna Circle, this had been influenced by the early Wittgenstein (1922), who had proclaimed: “The limits of my language mean the limits of my world” (para. 5.6). In biology, with von Uexküll (1934) the concept took hold that different species live in different worlds since their modes of cognition are structured differently. Based on von Uexküll’s findings, Cassirer (1967) developed the idea that the world humans consciously live in is essentially a symbolic world. The structure of this world does not so much depend on the “outside” world rather than on the ways humans interact socially by means of symbolisation. Men live neither in a world of things or a world of acting, but rather within language that has been given to them; language is neither an instrument of exchanging messages, nor of performing operations within the mind — it is rather constitutive of world within a particular community (Sapir, 1949; Whorf, 1956). It is warranted to say that almost the entire philosophy of the 20th century was concerned with the relationship between language and cognition. Accordingly, researchers concerned with linguistic representation of knowledge about world have called their constructs ontologies (e.g., Gruber, 1993; Guarino, 1995). This has however been done in complete awareness of the differences between ontologies they construct and ontology in philosophy, as for example shown by Gruber (1993): The word “ontology” seems to generate a lot of controversy in discussions about AI.… In the context of knowledge sharing, I use the term ontology to mean a specification of a conceptualization. That is, an ontology is a description … of the concepts and relationships that can exist for an agent or a community of agents. This definition is consistent with the usage of ontology as set-ofconcept-definitions, but more general. And it is certainly a different sense of the word than its use in philosophy.… An ontology is a specification used for making ontological commitments.… Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

336

Wyssusek & Klaus

Practically, an ontological commitment is an agreement to use a vocabulary … in a way that is consistent (but not complete) with respect to the theory specified by an ontology (para. 2-4). Hence, ontologies in artificial intelligence are linguistic conventions that do not tell us anything about the outside world. Yet, in philosophy we have come to use ontologies in a similar vein (e.g., Quine, 1961). Within linguistic relativity, the play of language structures (symbolic) worlds. Humans speak and write to objectify experiences, thereby making it possible to communicate these experiences even if they lie in the past. Objectification is also the way of projecting potential future experiences and communicating these projections (Berger & Luckmann, 1966). Thus, language can be seen as a means of representation of experiences. Yet, this is only possible with a caveat: The meaning of linguistic expressions is not fixed; symbols are multivalent. In short, the same expression may convey different meanings. Quite often objectification has been confused with objective meaning of linguistic expression. Moreover, linguistic expressions can also not be conflated with their meaning of conceptualisation. This has also been clarified by Gruber (1993), who understands ontology as a description or a specification of conceptualisations available to an (artificial) agent. And we may remind that an artificial agent does not conceptualise — artificial agents only command over a linguistic structure, not over conceptualisations; the concepts always remain in the realm of the human mind. When developing an ontological foundation for information systems analysis and design, differentiation between the two distinct meanings of ontology depicted above is paramount. Regarding the Bunge-Wand-Weber (BWW) ontology, this distinction is can be made obvious in this way: on the one hand, there are all the constructs that are descriptions or specifications of conceptualisations (e.g., Wand & Weber, 1990a, p. 64). Thus, these constructs are parts of an ontology — according to Gruber’s definition. On the other hand, there is the claim that these constructs are the “things” that make up the world (Wand & Weber, 1988, pp. 213). This claim belongs to the realm of philosophical ontology. In the light of the preceding elaborations, this claim is highly questionable, since it is based on a positivist philosophical position that not too many would subscribe to these days (for some critical self-reflection see, Wand et al., 1995; Weber, 1997b, pp. 174). According to this position, an ontological model is regarded as a representation (mapping) of the “true” reality, that is, given perceptions. This representational notion of “model” presupposes a direct relationship between the model (the representation) and the model source (the original). A model is “good” or “true” if it corresponds with reality — the essence of the correspondence theory of truth. Accordingly, for the development of an ontological foundation for ISAD it is decisive to “find” the true objects and relationships in the world. Yet nobody Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


337

knows how to do this, which is not a problem at all. The “scandal” is that there are still people around who try to figure it out. In conclusion, the effort geared towards the development of an ontological foundation for ISAD will only be fruitful if the idea of an ontological foundation in the classical philosophical (or metaphysical) sense is being abandoned. The symbolic world, structured by means of language and symbolic interaction, is actually a plurality of worlds. Therefore, we recommend that adherence to the ontological, epistemological, methodological, and anthropological positions held by Bunge is not required. Rather, it seems to be more promising to understand “ontological foundation” in the sense of Gruber. The constructs provided by an ontology do not need to be grounded in some metaphysics. They might serve us well enough, if we understand them as descriptions or specifications of conceptualisations. If we commit to an ontology, we commit ourselves to a vocabulary and a grammar that might or might not be useful to be used when we speak about the world. We will know its adequacy, if we try to express our conceptualisations by means of this “ontology”. If it does not serve our purpose, we do not have to change our worldview — changing the vocabulary and the grammar will suffice. The notion of linguistic relativism will help us to understand why people understand an “ontology” differently, or, in other words, why they attribute different meanings to one and the same “ontology”. And, it will also direct us to a question that especially needs our attention: How do we develop conceptualisations that are at least compatible in such a way that we are able to communicate by means of an “ontology” we have subscribed to?

Summary and Conclusion The persistence of the software crisis — budget overruns, exceeded time frames, not meeting users’ requirements, and total failure of information systems development projects — provides a host of motivations for the reconsideration of the state-of-the-art in information systems development. In order to overcome these highly undesirable results, it is widely believed that the development of more rigorous theoretical foundations for information systems development is the key to success. One approach that has gained considerable attention during the last two decades is the development of ontological foundations for information systems analysis and design. Quite in contrast to the general lack of interest in philosophical issues of information systems, philosophical ontology has become an inspiring source for information systems research and practice. Understanding information systems as essentially representational systems, that is, as systems that represent knowledge about certain domains, ontology is largely being used for the Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

338

Wyssusek & Klaus

identification or the definition of that what constitutes these domains. Thus, information systems research is more interested in the ontical question of “What exists?” rather than in the ontological question of “What does it mean to exist?” — a preference and a limitation with considerable consequences. It was the linguistic turn in philosophy that brought the language dependency of all knowledge to our attention. Among others, von Humboldt, Nietzsche, Wittgenstein, Cassirer, Sapir, and Whorf argue that our access to the world is bound to language and that there is no way to transcend our knowledge beyond the means provided by language. With the linguistic turn, classical ontology was overcome. Ontology is now seen as being bound to language and the question of existence has been turned into a question of ontological commitment (e.g., Quine, 1961). This understanding of ontology is accompanied by the acceptance of the existence of multiple ontologies and the proximity of philosophical ontology to philosophy of language and interpretation. The inadequacy of everyday language for formal logic was the starting point for the development of formal languages, which in turn provide the basis for the development of formal ontologies. However, formal languages (and formal ontologies) do not help to overcome the problem of subjectivity and linguistic relativism; they rather sidestep it. The constructs of formal languages have per se no meaning in our world. Formal semantics do not tell us anything about “our” world, that is, a world of informal languages. The “real world” meaning of formal language constructs is derived from symbolic interaction between humans and is always relative to particular communities. It is only within such communities that objectifications by means of language develop a stable yet not fixed meaning that enables members of the respective community to communicate effectively. If we intend to say something about the world by means of formal ontologies, we have to develop a common language practice that eventually will lead to the desired stable meanings. So far, this problem has hardly ever been addressed, nor have its socio-political implications been articulated. The restricted scope of discourses and debates on the ontological foundation of information systems analysis and design is partly due to the negligence of already well-documented debates within the information systems research community. The contentions surrounding the methodical versus the amethodical understanding of software development and the myth of formal methods have shown that the limitations of these debates have to be transcended. Focussing entirely on formal aspects of the ontological foundation of ISAD may implicate an error of the third kind, that is, finding the right answers to the wrong questions (e.g., Mitroff, 1998, pp. 13-32). For example: Why bother with the evaluation of modelling grammars? If we are convinced of a formal ontology, then why not develop a modelling grammar from scratch, based on this very “ontology”?



339

Research on the evaluation of modelling grammars in the context of ISAD has provided proof of the usefulness of formal ontologies for a certain purpose (e.g., Green & Rosemann, 1999; Milton et al., 2000; Fettke & Loos, 2003). However, such a proof only shows that one axiomatic system conforms to another axiomatic system, or that it does not. The proof that the “axiomatic reference system” — the “ontology” — is suitable if we want to express something about “the” world cannot be established. Nor can it be established that the analytical approach toward the development of the “axiomatic reference system”, exemplified by the Bunge-Wand-Weber ontology, is superior to any other approach. This chapter has tried to establish that when dealing with fundamental issues of theory and practice it is advisable to create an awareness of the potential and limitations of our knowing and doing. This entails considering marginalised positions in a critical discussion of approaches towards information systems analysis and design. The status quo of information systems research in general, and of research in information systems analysis and design in particular, needs to be reflected upon in this sense — to be able to ask the right questions. These questions then extend the scope of, for example, the discussion of ontological foundations of information systems analysis and design, and open the possibility to move beyond the status quo.

Acknowledgments The authors acknowledge the helpful comments of two anonymous reviewers.

References Balzer, R., Goldman, N., & Wile, D. (1978). Informality in program specifications. IEEE Transactions on Software Engineering, 4(2), 94-102. Bansler, J., & Bødker, K. (1993). A reappraisal of structured analysis: Design in an organizational context. ACM Transactions on Information Systems, 11(2), 165-193. Bansler, J. P., & Havn, E. C. (2003). Improvisation in action: Making sense of IS development in organizations. In G. Goldkuhl, M. Lind, & P. J. Ågerfalk (Eds.), Proceedings of Action in Language, Organizations and Information Systems (ALOIS), Linköping, Sweden, (pp. 51-64).


340

Wyssusek & Klaus

Baskerville, R., Travis, J., & Truex, D. P. (1992). Systems without method: The impact of new technologies on information systems development projects. In K. E. Kendall, K. Lyytinen, & J. I. DeGross (Eds.), The impact of computer supported technologies in information systems development (pp. 241-260). Amsterdam: North-Holland. Berger, P. L., & Luckmann, T. (1991). The social construction of reality. London: Penguin. Boland, R. J. Jr. (1982). The process of system design: A phenomenological approach. In Proceedings of the International Conference on Information Systems 1982, (pp. 31-45). Boustred, C. (1997). ATSI seminars. In K. Liu (2000), Semiotics in information systems engineering (p. 3). Cambridge: Cambridge University Press. Bowen, J. B., & Hinchey, M. G. (1995). Seven more myths of formal methods. IEEE Software, 12(3), 34-41. Brynjolfsson, E. (1993). The productivity paradox of information technology. Communications of the ACM, 36(12), 67-77. Bunge, M. (1977). Treatise on basic philosophy: Volume 3: Ontology I: The furniture of the world. Boston: Reidel. Bunge, M. (1979). Treatise on basic philosophy: Volume 3: Ontology II: A world of systems. Boston: Reidel. Bunge, M. (1993). Realism and antirealism in social sciences. Theory and Decision, 35, 207-235. Burrell, G., & Morgan, G. (1979). Sociological paradigms and organizational analysis — Elements of the sociology of corporate life. London: Heinemann. Cassirer, E. (1967). An essay on man. An introduction to a philosophy of human culture (18th ed.). New Haven: Yale University Press. Caws, P. (1967). Scientific method. In P. Edwards (Ed.), The encyclopedia of philosophy, volume 7 (pp. 339-343). New York: Macmillan. Checkland, P. (1981). Systems thinking, systems practice. Chichester: Wiley. Ciborra, C. U. (1998). Crisis and foundations: An inquiry into the nature and limits of models and methods in the information systems discipline. Journal of Strategic Information Systems, 7(1), 5-16. Collier, A. (1994). Critical realism: An introduction to the philosophy of ROY BHASKAR. London: Verso. Craig, E. (1998). Ontology. In E. Craig (Ed.), Routledge encyclopedia of philosophy, volume 7, Nihilism — Quantum mechanics, interpretation of. London: Routledge.



341

De Millo, R. A., Lipton, R. J., & Perlis, A. J. (1979). Social processes and proofs of theorems and programs. Communications of the ACM, 22(5), 271-280. Derrida, J. (1978). Structure, sign and play in the discourse of human sciences. In J. Derrida (Ed.), Writing and difference (pp. 278-293). Chicago: University of Chicago Press. Derrida, J. (1982). Différance. Chicago: University of Chicago Press. Descartes, R. (1637). Discours de la méthode. Pour bien conduire sa raison, et chercher la verite dans les sciences. Leyden: Maire. Fettke, P., & Loos, P. (2003). Ontological evaluation of reference models using the Bunge-Wand-Weber model. Proceedings of the Americas Conference on Information Systems 2003 (pp. 2944-2955). Floyd, C. (1992). Software development as reality construction. In C. Floyd, H. Züllinghoven, R. Budde & R. Keil-Slawik (Eds.), Software development and reality construction (pp. 86-100). Berlin: Springer. Gibbs, W. W. (1994). Software’s chronic crisis. Scientific American, 271(3), 86-95. Goclenius, R. (1613). Lexicon philosophicum, quo tanquam clave philosophiæ fores aperiuntur. Frankfurt: Becker. Gramsci, A. (1971). Selections from the prison notebooks of ANTONIO G RAMSCI. New York: International Publishers. Green, P., & Rosemann, M. (1999). An ontological analysis of integrated process modelling. In M. Jarke & A. Oberweis (Eds.), Proceedings of the 11th International Conference on Advanced Information Systems Engineering (CAiSE ’99), (pp. 225-240). Berlin: Springer. Gruber, T. R. (1993). What is an ontology? Retrieved from http:// www.ksl.stanford.edu/kst/what-is-anontology.html Guarino, N. (1995). Formal ontology, conceptual analysis and knowledge representation. International Journal of Human-Computer Studies, 43(5/6), 625-640. Habermas, J. (1972). Knowledge and Human Interest. Boston: Beacon Press. Hall, A. (1990). Seven myths of formal methods. IEEE Software, 7(5), 11-19. Heidegger, M. (1962). Being and Time. (J. Macquarrie & E. Robinson, Trans.). New York: Harper and Row. Hirschheim, R., Klein, H. K., & Lyytinen, K. (1995). Information systems development and data modeling: Conceptual and philosophical foundations. Cambridge: Cambridge University Press. Introna, L. D. (1996). Notes on ateleological information systems development. Information Technology & People, 9(4), 20-39.


342

Wyssusek & Klaus

Kant, I. (1929). Critique of pure reason (N. Kemp Smith, Trans.). London: Macmillan. Kendall, K. E., & Kendall, J. E. (1992). Systems analysis and design. Englewood Cliffs, NJ: Prentice-Hall. Le Charlier, B., & Flener, P. (1998). Specifications are necessarily informal or the ultimate myths of formal methods. Research paper RP-98-001, University of Namur. Milton, S., Kazmierczak, E., & Thomas, L. (2000). Ontological foundations of data modeling in information systems. Proceedings of the Americas Conference on Information Systems 2000, (pp. 1537-1543). Mitroff, I. (1998). Smart thinking for crazy times. The art of solving the right problems. San Francisco: Berrett-Koehler. Naur, P. (1993). Understanding TURING ’s universal machine: Personal style in program description. The Computer Journal, 36(4), 351-372. Naur, P., & Randell, B. (1969). Software engineering: A report on a conference sponsored by the NATO Science Committee, Garmisch, 7-11 October, 1968. Brussels: NATO Scientific Affairs Division. Nietzsche, F. (1954). On truth and lie in an extra-moral sense. In W. Kaufmann (Trans., Ed.), The portable Nietzsche (pp. 42-47). New York: The Viking Press. Ong, W. J. (1958). RAMUS, method, and the decay of dialogue: From the art of discourse to the art of reason. Cambridge: Harvard University Press. Quine, W. V. O. (1961). From a logical point of view: 9 logico-philosophical essays (2nd ed.). Cambridge: Harvard University Press. Rosemann, M., Vessey, I., & Weber, R. (2004). Alignment in enterprise systems implementations: The role of ontological distance. Proceedings of the International Conference on Information Systems 2004 (pp. 439-448). Ryle, G. (1949). The concept of mind. Chicago: University of Chicago Press. Sapir, E. (1949). Selected writings of EDWARD SAPIR in language, culture and personality. Berkeley: University of California Press. Shaw, M. (1990). Prospects for an engineering discipline of software. IEEE Software, 7(11), 15-24. Standish Group. (1994). The CHAOS Report. Retrieved from http:// www.standishgroup.com/chaos/toc.php Standish Group. (2003). CHAOS Chronicles v3. Retrieved from http:// www.standishgroup.com/chaos/toc.php



343

Truex, D. P., Baskerville, R., & Travis, J. (2000). Amethodical systems development: The deferred meaning of systems development methods, Accounting, Management & Information Technologies, 10(1), 53-79. Turnèbe, A. (1600). De methodo libellus. In A. Turnèbe, Libelli de vino, calore, et methodo nunc primum editi. Folio 2. Paris: Morel. Turner, J. (1987). Understanding the elements of system design. In R. J. Boland & R. A. Hirschheim (Eds.), Critical issues in information systems research (pp. 97-111). Chichester: Wiley. von Humboldt, W. (1836). Über die Verschiedenheit des menschlichen Sprachbaues und ihren Einfluss auf die geistige Entwicklung des Menschengeschlechts. Berlin: Königliche Akademie der Wissenschaften. von Uexküll, J. (1934). Streifzüge durch die Umwelten von Tieren und Menschen. Ein Bilderbuch unsichtbarer Welten. Berlin: Springer. Wand, Y. (1996). Ontology as a foundation for meta-modelling and method engineering. Information and Software Technology, 38, 281-287. Wand, Y., Monarchi, D. E., Parsons, J., & Woo, C. (1995). Theoretical foundations for conceptual modeling in information systems development. Decision Support Systems, 15, 285-304. Wand, Y., & Wang, R. Y. (1996). Anchoring data quality dimensions in ontological foundations. Communications of the ACM, 39(11), 86-95. Wand, Y., & Weber, R. (1988). An ontological analysis of some fundamental information system concepts. Proceedings of the International Conference on Information Systems 2004 (pp. 213-224). Wand, Y., & Weber, R. (1989). A model of systems decomposition. Proceedings of the International Conference on Information Systems 1989 (pp. 41-51). Wand, Y., & Weber, R. (1990). An ontological model of an information system. IEEE Transactions on Software Engineering, 16(11), 1282-1292. Wand, Y., & Weber, R. (1990a). Toward a theory of the deep structure of information systems. Proceedings of the International Conference on Information Systems 1990 (pp. 61-71). Wand, Y., & Weber, R. (1992). On the ontological completeness of information systems analysis and design grammars. Working paper 92-MIS-003, Faculty of Commerce and Business Administration, University of British Columbia, Vancouver. Wand, Y., & Weber, R. (1995). On the deep structure of information systems. Information Systems Journal, 5(3), 203-223.


344

Wyssusek & Klaus

Weber, R. (1997a). The link between data modeling approaches and philosophical assumptions: A critique. Proceedings of the Americas Conference on Information Systems 1997 (pp. 306-308). Weber, R. (1997b). Ontological foundations of information systems. Coopers & Lybrand research methodology monograph no. 4. Melbourne: Coopers & Lybrand. Weber, R., & Colomb, R. M. (1998). Completeness and quality of an ontology in an information system. In N. Guarino (Ed.), Formal ontology in information systems (pp. 207-217). Amsterdam: IOS Press. Westrup, C. (1999). Knowledge, legitimacy and progress? Requirements as inscriptions in information systems development. Information Systems Journal, 9, 35-54. Whorf, B. L. (1956). Language, thought, and reality. Selected writings of BENJAMIN LEE WHORF. Cambridge: MIT Press. Winograd, T., & Flores, C. F. (1986). Understanding computers and cognition. A new foundation for design. Norwood: Ablex. Wittgenstein, L. (1922). Tractatus Logico-Philosophicus. London: Routledge and Kegan Paul.


Applications of a Unified Foundational Ontology 345

Chapter XIII

Some Applications of a Unified Foundational Ontology in Business Modeling Giancarlo Guizzardi, University of Twente, The Netherlands Gerd Wagner, Brandenburg University of Technology, Cottbus, Germany

Abstract Foundational ontologies provide the basic concepts upon which any domain-specific ontology is built. This chapter presents a new foundational ontology, UFO, and shows how it can be used as a guideline in business modeling and for evaluating business modeling methods. UFO is derived from a synthesis of two other foundational ontologies, GFO/GOL and OntoClean/DOLCE. While their main areas of application are natural sciences and linguistics/cognitive engineering, respectively, the main purpose of UFO is to provide a foundation for conceptual modeling, including business modeling. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.

346 Guizzardi & Wagner

Introduction A foundational ontology, sometimes also called “upper level ontology”, defines a range of top-level domain-independent ontological categories, which form a general foundation for more elaborated domain-specific ontologies. A well-known example of a foundational ontology is the Bunge-Wand-Weber (BWW) ontology proposed by Wand and Weber in a series of articles (e.g., Wand & Weber, 1990, 1995) on the basis of the original metaphysical theory developed by Bunge (1977, 1979). As has been shown in a large number of recent works (e.g., Green & Rosemann, 2000; Evermann & Wand, 2001; Guizzardi, Herre, & Wagner, 2002a, b; Opdahl & Henderson-Sellers, 2002), foundational ontologies can be used to evaluate conceptual modeling languages and to develop guidelines for their use. Business modeling can be viewed as the main application domain of conceptual modeling languages and methods. In the model-driven architecture approach of the Object Management Group (OMG), a business model is called a “computationindependent model” because it must not be expressed in terms of IT concepts, but solely in terms of business language. The business domain, since it contains so many different kinds of things, poses many challenges to foundational ontologies. A unified foundational ontology represents a synthesis of a selection of foundational ontologies. Our main goal in making such a synthesis is to obtain a foundational ontology that is tailored towards applications in conceptual modeling. For this purpose we have to capture the ontological categories underlying natural language and human cognition that are also reflected in conceptual modeling languages such as ER diagrams or UML class diagrams. In Gangemi, Guarino, Masalo, Oltramari, and Schneider, (2002) this approach is called “descriptive ontology” as opposed to “prescriptive ontology”, which claims to be “realistic” and robust against the state of the art in scientific knowledge. For UFO 0.2, the second1 (still experimental) version of our unified foundational ontology (UFO), we combine the following two ontologies: 1) the general formal ontology (GFO), which is underlying the general ontological language (GOL) developed by the OntoMed research group at the University of Leipzig, Germany; (see www.ontomed.de and Degen, Heller, Herre, & Smith, 2001); 2) the OntoClean ontology (Welty & Guarino, 2001) and the descriptive ontology for linguistic and cognitive engineering (DOLCE), developed by the ISTC-CNRLOA research group in Italy, as part of WonderWeb Project (see http:// wonderweb.semanticweb.org/). Existing foundational ontologies, notably SUO, OntoClean-DOLCE, GFO-GOL, and even BWW, all have severe limitations in their ability to capture the basic concepts of conceptual modeling languages. For instance, Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


1.

SUO, OntoClean-DOLCE, and BWW do not make a clear distinction between entities and sets, which is needed to capture the characteristic difference between entity type and datatype.

2.

SUO, GFO-GOL, and BWW do not include an ontology of entity type categories, which is needed to capture the categories of role types, phase types, and mixin types.

3.

SUO, GFO-GOL, and BWW do not pay much attention to the sphere of intentional and social things with the core category of agents, which is needed to capture the characteristics of business processes.

UFO does not have these (and some other) shortcomings of SUO, OntoCleanDOLCE, GFO-GOL, and BWW. Our choice to use OntoClean-DOLCE and GFO-GOL as its basis rests on the fact that these two ontologies offer more constructs that are relevant to conceptual modeling than the other foundational ontologies. Specifically, OntoClean-DOLCE include an ontology of entity type categories and an account of agents, while GFO-GOL includes the fundamental distinction between entities and sets. We have obtained our synthesis by: 1) selecting categories from the union of both category sets; 2) renaming certain terms in order to create a more “natural” language; and 3) adding some additional categories based on relevance for conceptual modeling according to our experience. Using the acronyms “BWW”, “owl”, “UML”, “ISO”, and “BSBR”, we also make references to BWW, the Web ontology language OWL (W3C, 2004), the Unified Modeling Language (UML), the terminology standard ISO1087-1:2000 (ISO, 2000), and to the Business Rules Team submission to the OMG Business Semantics for Business Rules RFP (Chapin, Hall, Ross, Morgan, & Baisley, 2004). For making a distinction between terms used differently in different vocabularies, we use the XML namespace prefix syntax and write, for example, “BWW:thing” and “owl:Thing” for distinguishing between the concepts termed “thing” in BWW and in OWL. We present UFO 0.2 both as a MOF/UML model (OMG, 2004) and as a vocabulary in semi-structured English, similar to the BSBR Structured English of Chapin et al. (2004). MOF/UML is a fragment of the UML class modeling language that is recommended by the OMG as a language for defining modeling languages; in other words, MOF/UML is a meta-modeling language. There are two reasons for using MOF/UML for defining a foundational ontology: first, it allows to express it graphically in the form of a UML class diagram; second, it facilitates the communication of the foundational ontology by making it accessible to the large (and still growing) language community of people familiar with the UML.



An alternative, and more flexible, mode of expression for defining a modeling language such as UFO consists of using semi-structured English to specify the vocabulary of the modeling language. Our UFO vocabulary has three kinds of entries marked up with different font styles:

•

term : A term in this font style denotes being of a type and is used to refer to things of that type; for example, the term individual in the phrase “individual that is wholly present whenever it is present” stands for a thing of type “individual” (i.e., it stands for an individual).

•

name : This is a name of an individual or a type; when abc is a type term referring to things of that type, abc is a name referring to the type itself.

•

term1 relationship phrase term2 : This is a name of a binary relationship type.

A vocabulary entry may contain, additionally,

•

“Corresponding terms” (or “corresponding relationship type expressions”): terms (or relationship type expressions) that are roughly equivalent.

• •

Examples. Constraints: logical statements that have to hold in any given ontology based on UFO.

When there is a primary source for a definition, we append it in brackets, like in “...[based on GFO]”. UFO is divided into three incrementally layered compliance sets: 1) UFO-A defines the core of UFO, excluding terms related to perdurants and terms related to the spheres of intentional and social things; 2) UFO-B defines, as an increment to UFO-A, terms related to perdurants; and 3) UFO-C defines, as an increment to UFO-B, terms related to the spheres of intentional and social things, including linguistic things. This division reflects a certain stratification of our “world”. It also reflects different degrees of scientific consensus: there is more consensus about the ontology of endurants than about the ontology of perdurants, and there is more consensus about the ontology of perdurants than about the ontology of intentional and social things. We hope that this division into different compliance sets will facilitate both the further evolution of UFO and the adoption of UFO in business modeling and ontology engineering. In the next section, we present UFO-A 0.2, while UFOCopyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


B 0.2 and UFO-C 0.2 are presented in the two subsequent sections, respectively. The next section illustrates how UFO can be used to evaluate some business modeling methods, and a final section concludes the chapter.

UFO-A: The Core of the Unified Foundational Ontology Things, Sets, Entities, Individuals, and Types We first present the upper part of UFO-A 0.2 as a MOF/UML model in Figure 1. Notice the fundamental distinction made between sets and entities as things that are not sets (called “urelements” in GFO). In structured English, the upper part of UFO 0.2 can be introduced as follows.

•

thing : This is anything perceivable or conceivable [ISO:object]. Corresponding terms: GFO:entity; DOLCE:entity, owl:Thing; BSBR:thing.

•

set: This is a thing that has other things as members (in the sense of set theory).

•

thing is member of set : This is the name of a formal relationship type that is irreflexive, asymmetric and intransitive.

•

member : This is the role name that refers to the first argument of the thing is member of set relationship type.

•

set is subset of set : This is the name of a formal relationship type that is reflexive, asymmetric and transitive. Constraint: For all t:thing; s1, s2 : set — if t is member of s1 and s1 is subset of s2, then t is member of s2

•

entity :This is a thing that is not a set; neither the set-theoretic membership relation nor the subset relation can unfold the internal structure of an entity [GFO:urelement].

•

entity type : This is an entity that has an extension (being a set of entities that are instances of it) and an intension, which includes an applicability criterion for determining if an entity is an instance of it; and which is captured by means of an axiomatic specification, that is, a set of axioms that may involve a number of other entity types representing its essential features. An entity type is a space-time independent pattern of features, which can be realized in a number of different individuals [based on



GFO:universal]. Corresponding terms: UML:class; DOLCE:universal; owl:Class; BSBR: “generic thing”.

•

entity is instance of entity type: This is the name of a formal relationship type (called classification).

•

instance: This is a role name that refers to the first argument of the entity is instance of entity type relationship type.

•

set is extension of entity type: This is the name of a formal relationship type. Constraint: For all o:entity, t:entity type, s:set — if o is instance of t and s is extension of t, then o is member of s.

•

extension : This is a role name that refers to the first argument of the set is extension of entity type relationship type.

•

entity type is subtype of entity type: This is the name of a formal relationship type that is irreflexive, asymmetric and transitive (also called generalization). Constraint: For all t1, t 2 : entity type; s1, s2 : set — if t1 is subtype of t2 and s1 is extension of t1 and s2 is extension of t2, then s1 is subset of s2.

•

subtype: This is a role name that refers to the first argument of the entity type is subtype of entity type relationship type.

•

individual: This is an entity that is not an entity type. An entity type that classifies individuals is called individual type. Corresponding terms: GFO:individual; DOLCE: particular.

•

thing is part of individual: This is name of a formal relationship type that is reflexive, asymmetric and transitive (also called aggregation).

•

part: This is a role name that refers to the first argument of the thing is part of individual relationship type.

•

entity type is classification type of entity type: This is the name of a formal relationship type where the first argument is a higher-order entity type whose instances form a subtype partition of the second argument (also called higher-order classification). Examples: BiologicalSpecies is classification type of Animal; PassengerAircraftType is classification type of PassengerAircraft. Constraint: For all t1, t2, t 3: entity type — if t3 is classification type of t1 and t2 is instance of t3, then t2 is subtype of t1.

•

classification type: This is a role name that refers to the first argument of the entity type is classification type of entity type relationship type. Corresponding names: GFO: “higher-order universal”; BSBR: “categorization type”; UML: powertype.

•

entity type is classified by entity type: This is the name of a formal relationship type that is the inverse of the entity type is classification type



of entity type relationship type. Corresponding relationship type expressions: BSBR: “type has categorization-scheme”.

Different Kinds of Types In UFO, we make a fundamental distinction between datatypes, which are sets, and entity types, which are not sets, but whose extensions are sets. Based on Wiggins (2001), van Leeuwen (1991), Gupta (1980), and Hirsch (1982), we distinguish between several different kinds of entity types, as shown in Figure 2. These distinctions are elaborated in Guizzardi, Wagner, and van Sinderen (2004a), in which we present a philosophically and psychologically well-founded theory of types for conceptual modeling. In Guizzardi, Wagner, Guarino, and van Sinderen (2004b), this theory is used to propose: 1) a profile for UML whose elements represent finer-grained distinctions between different kinds of types and 2) a set of constraints defining the admissible relations between these elements. One should refer to Guizzardi et al. (2004a, 2004b) for: a) an in-depth discussion of the theory underlying these categories as well as the constraints on their relations; b) a formal characterization of the profile; and c) the application of the profile to propose an ontological design pattern that addresses a recurrent problem in the practice of conceptual modeling. In structured English, the different kinds of types are defined as follows.

Figure 1. The upper part of UFO-A 0.2 as a MOF/UML model * A b str ac t Thing

Thing

Part *

M em ber

* Ins t anc e

S ubs et

En tity

Set

1.. *

1. .* Ex t ens ion

*

*

{ c om pl ete }

* In d iv i d u al

*

1

1. .*

En tity T yp e

D a tatyp e

C las s if ic ati onT y pe

Subty pe * *

1



•

datatype: This is a set whose members are data values. Examples: Integer; String.

•

sortal type: This is an entity type that carries a criterion for determining the individuation, persistence, and identity of its instances. An identity criterion supports the judgment whether two instances are the same. Every instance in a conceptual model must have an identity and, hence, must be an instance of sortal type.

•

base type: This is a sortal type that is rigid (all its instances are necessarily its instances) and that supplies an identity criterion for its instances [OntoClean:type]. Examples: Mountain; Person. Corresponding terms: BWW: “natural kind”.

•

phase type: This is a sortal type that is anti-rigid (its instances could possibly also not be instances of it without loosing their identity) and that is an element of a subtype partition of a base type [OntoClean:“phased sortal”]. Examples: Town and Metropolis are phase subtypes of City; Baby, Teenager, and Adult are phase subtypes of Person.

•

role type: This is a sortal type that is anti-rigid and for which there is a relationship type such that it is the subtype of a base type formed by all instances participating in the relationship type [OntoClean:role]. Examples: DestinationCity as role subtype of City; Student as role subtype of Person.

•

mixin type: This is an entity type that is not a sortal type and can be partitioned into disjoint subtypes, which are sortal types (typically role

Figure 2. Different kinds of types in UFO-A 0.2 Se t

Ex t ens ion

En t it y T yp e

So r tal T yp e

B as eT y p e

1. .*

1

D at atyp e

M ixi n T yp e

R o le T yp e

Ph a se T yp e



types) with different identity criteria. Since a mixin is a non-sortal it cannot have direct instances [OntoClean:non-sortal]. Examples: Object; Part; Customer; Product

•

relationship type: This is a type whose instances are (material or formal) relationships .

Notice that role types and phase types cannot supply an identity criterion for their instances. For this reason, they must be derived from suitable base type from which they inherit their identity criterion. The theory of types, which is part of UFO-A, provides a foundation for a number of modeling primitives that, albeit often used, are commonly defined in an ad hoc manner in the practice of conceptual modeling. In particular, this theory can be considered as an extension of the BWW account of types. In Evermann and Wand (2001), it is proposed that a UML class should be used to represent a BWW: “natural kind” (i.e., it should be equivalent to the “functional schema” of a BWW: “natural kind”). As discussed in Guizzardi et al. (2004a), the concept of a natural kind corresponds to the UFO concept of a base type, that is, a natural kind is a rigid entity type that provides an identity criterion for its instances. It has been argued, however, (e.g., Welty & Guarino, 2001; Gupta, 1980; Wiggins, 2001; van Leeuwen, 1991; Guizzardi et al., 2004a, 2004b), that, in addition to this concept, several other type concepts are needed in descriptive ontologies and in conceptual modeling.

Different Kinds of Individuals We distinguish between a number of different kinds of individuals, as shown in Figure 3. The fundamental distinction between endurants and perdurants corresponds to the colloquial distinction between “objects” and “processes”. In structured English, the different kinds of individuals considered in UFO are explained as follows.

•

endurant: This is an individual that is wholly present whenever it is present, that is, it does not have temporal parts, and that persists in time while keeping its identity [DOLCE]. Examples: a house; a person; the moon; a hole; the redness of a certain apple; an amount of sand. Corresponding terms: GFO:3D-individual.

•

perdurant: This is an individual that is composed of temporal parts; whenever a perdurant is present, it is not the case that all its temporal parts



Figure 3. Different kinds of individuals in UFO-A 0.2 In d i vid u al

{ di sj oi nt}

Per d u ran t

En d u r an t

1.. * Am ount of M atter {dis j oi nt, c om p lete }

{ d is jo int}

Su b stan ce I n d ivi d u al

1 bears

? i nheres in 1. .* M o m en t I n d ivi d u al

Ph ysi cal Ob j ect

{ d is jo int}

R el ato r

1. . * In tr in si c M o m en t

Qu al i ty

are present [DOLCE]. Examples: a storm; a heart attack; a conversation; the Second World War; a business process.

•

substance individual: This is an endurant that consists of matter (i.e., is “tangible” or concrete), possesses spatio-temporal properties, and can exist by itself; that is, it does not existentially depend on other endurants, except possibly on some of its parts) [based on GFO:substance]. Examples: a house; a person; the moon; an amount of sand. Corresponding terms: BWW:thing

•

moment individual: This is an endurant that cannot exist by itself; that is, it depends on other endurants, which are not among its parts [based on GFO:moment]. Examples: the redness of a certain apple; a belief of George Bush; a flight connection between two cities.

•

endurant bears moment individual: This is the name of a formal relationship type [based on GFO: “substance bears moment”].

•

physical object: This is a substance individual that satisfies a condition of unity and for which certain parts can change without affecting its identity. Examples: a house; a person; the moon.

•

amount of matter: This is a substance individual that does not satisfy a condition of unity; typically referred to by means of mass nouns. An amount of matter is mereologically invariant, that is, it cannot change any of its parts without changing its identity [DOLCE]. Examples: a liter of water; a piece of gold; a pile of sand.



•

intrinsic moment: This is a moment individual that is existentially dependent on one single individual. Examples: the redness of a certain apple; a belief of George Bush.

•

intrinsic moment inheres in endurant: This is the name of a formal relationship type [GFO].

•

quality: This is an intrinsic moment that inheres in exactly one endurant and can be mapped to a value (DOLCE:quale) in a quality dimension (Gärdenfors, 2000). Corresponding terms: GFO:quality; DOLCE:quality; BWW: “intrinsic property”. Examples: the color (height, weight) of a physical object; an electric charge. Constraint: For all e1, e2 : endurant; q:quality — if q inheres in e1 and q inheres in e2, then e1 is equal to e2. Examples: the redness of a certain apple.

•

relator: This is a moment individual that is existentially dependent on more than one individual. Relators provide the basis for material relationships (Guizzardi, Herre, & Wagner, 2002b) [GFO:relator]. Corresponding terms: BWW: “mutual property”, UML:link, owl:. Examples: a particular employment (Susan is employed by IBM); a particular flight connection (LH403 flies from Berlin to Munich).

The notion of relators is supported in several works in the philosophical literature (see, e.g., Smith & Mulligan, 1983, 1986). The concept of relators plays an important role in: 1.

distinguishing material relationship types, such as “person is married to person” and “person studies at university”, from formal relationship types, such as “number is greater than number” and “day is part-of month”);

2.

answering questions of the sort: What does it mean to say that John is married to Mary? Why is it true to say that Bill works for Company X but not for Company Y?

Putting all UFO-A terms and relationship-type expressions together in one UML/MOF diagram results in Figure 9 (see Appendix A).



Some Applications of UFO-A to Business Modeling Problems Modeling Customers Most business information systems include a “business object class” Customer for representing the customers of the business. In Figure 4, the role type Customer is defined as a supertype of Person and Corporation. This model is deemed ontologically incorrect for two reasons: first, not all persons are customers, that is, it is not the case that the extension of Person is necessarily included in the extension of Customer. Moreover, an instance of Person is not necessarily (in the modal sense) a Customer. Both arguments are also valid for Organization. In a series of papers (e.g., Steimann, 2000), Steimann discusses the difficulties in specifying supertypes for roles that can be filled by instances of disjoint types.2 As a conclusion, he claims that the solution to this problem lies in separating the hierarchies of role type and base type (named natural type in the article) — a solution, which strongly impacts the meta-model of all major conceptual modeling languages. By using the theory of types underlying UFOA we can show that this claim is not warranted and we are able to propose a design pattern that can be used as an ontologically correct solution to this recurrent problem (Guizzardi et al, 2004b). In this example, Customer has in its extension individuals that obey different identity criteria, that is, it is not the case that there is a single identity criterion, which applies both for Persons and Corporations. Customer is hence a mixin type (a non-sortal). Since every instance in the model must have an identity, thus, every instance of Customer must be an instance of one of its subtypes (forming a partition) that carries an identity criterion. For example, we can define the sortals PersonalCustomer and CorporateCustomer as subtypes of Customer (Figure 5). These sortals, in turn, carry the (incompatible) identity criteria supplied by the base types Person and Corporation, respectively.

Product Modeling In many business information systems, both individual products and product types have to be represented. In a prototypical case, the product individual type, whose instances are identified with the help of serial numbers, is classified by the corresponding product model type, which is a second order classification type, whose instances are subtypes of the product individual type. Figure 6 shows this situation for the case of cars and car models.



Figure 4. An ontologically incorrect model of the entity type Customer Customer

Person

Corporation

Figure 5. An ontologically-correct model of the Customer entity type according to UFO 0.2 «mixinty pe» Customer «base ty pe» Person

«base ty pe» Corporation

«role ty pe» PersonalCustomer

«role ty pe» CorporateCustomer

Figure 6. UML product modeling with UFO-based stereotypes «isClassif iedBy »

Car Color

*

VW Passat

1

«classif icationty pe» CarModel

«instance»

BMW Z3 «instance»

4711 : Car Color = black

In a proposal for the ontological foundations of the Resource-Event-Agent (REA) model (Geert & McCarthy, 2000, p. 13), the authors argue about the importance of the distinction between individual types and classification types accounted here:



Economic Resources like (especially) inventory have an instance/ type definition problem that must be solved in the REA ontology (or in any information system)… cars in an automobile dealership would be modeled with instances (a car with a given engine#) …with classes of cars (1975 Corvette) as type-images.

UFO-B: The Ontology of Perdurants A complete treatment of an ontology of perdurants requires an ontology of temporal entities (GFO:chronoids) (Degen et al., 2001). In this section, instead, we restrict our attention to the most basic perdurant categories for defining UFO-B 0.2 as a foundation for defining some intentional and social entities later. In the sequel we discuss the following basic kinds of perdurants shown in Figure 7: (atomic and complex) events and states.

•

state: This is a perdurant that is homeomeric, that is, each of its temporal parts belongs to the same state type as the whole [based on DOLCE].

•

event: This is a perdurant that is related to exactly two states (its pre-state and its post-state). An event is related to the states before and after it has happened.

•

atomic event: This is an event that happens instantaneously, that is, an event without duration, relative to an underlying time granularity [based on BWW:event and GFO:change]. Examples: an explosion; a message reception.

Figure 7. The perdurant categories of UFO-B 0.2 Perdurant (from UFO-A)

PreState State

Event 1

1

2..*

*

PostState

*

Atomic Event

Complex Event *



•

complex event: This is an event that is composed of other events by means of event composition operators. Examples: a parallel occurrence of two explosions; an absence of a message reception (within some time window); a storm; a heart attack; a football game; a conversation; a birthday party; the Second World War; a Web shop purchase.

•

process: This is a complex event that is a sequence of two or more (possibly parallel occurrences of) atomic events. Examples: a storm; a heart attack; a football game; a conversation; a birthday party; the second World War; a Web shop purchase.

• •

state is pre-state of event: This is a name of a formal relationship type. state is post-state of event: This is the name of a formal relationship type.

UFO-C: The Ontology of Intentional, Social, and Linguistic Things The “objective” perdurant categories (atomic and complex) event and state defined in UFO-B are essential concepts for process modeling, but they are not sufficient for business process modeling, where intentional and social concepts such as action, activity, and communication are needed. The following account of intentional and social things is at an early stage of development and therefore rather incomplete. Nevertheless, we think that it gives an impression of the range of ontological categories that is needed to explain business process modeling.

•

physical agent: This is a physical object that creates action events affecting other physical objects, that perceives events, possibly created by other physical agents, and to which we can ascribe a mental state. Examples: a dog; a human; a robot.

•

action event: This is an event that is created through the action of a physical agent.

•

non-action event: This is an event that is not created through an action of a physical agent.

•

physical agent creates action event: This is the name of a formal relationship type.

•

physical agent perceives event: This is the name of a formal relationship type.



•

non-agentive object: This is a physical object that is not a physical agent. Examples: a chair; a mountain .

•

mental moment: This is an intrinsic moment that is existentially dependent on a particular agent, being an inseparable part of its mental state. Examples: a thought; a perception; a belief; a desire; an individual goal. Constraint: For all mm : mental moment; e:endurant — if mm inheres in e, then e is physical agent.

•

communicating physical agent: This is a physical agent that communicates with other communicating physical agents. Examples: a dog; a human; a communication-enabled robot.

•

institutional agent: This is an institutional fact (Searle, 1995) that is an aggregate consisting of communicating agents (its internal agents), which share a collective mental state, and that acts, perceives and communicates through them. Examples: a business unit; a voluntary association.

•

agent: This is an endurant that is either a physical agent or an institutional agent.

•

communicating agent: This is an agent that communicates with other communicating agents.

•

social moment: This is a moment individual that is existentially dependent on more than one communicating agent. Examples: a commitment; a joint intention.

Figure 8. The categories of the UFO-C 0.2 agent ontology PhysicalObject (from UFO-A) Moment Individual (from UFO-A)

perceiv es

{disjoint}

Relational Moment

Intrinsic Moment *

inheres in

*

creates

1 Physical Agent

3 bears *

2..*

*

Communicative ActionEvent *

Communicating PhysicalAgent {disjoint}

Commitment

*

1..*

Perception

SocialMoment

Non-Action Event

ActionEvent

{disjoint} Belief

Event (from UFO-B)

{disjoint}

Non-Agentive Object

MentalMoment

*

Institutional Agent

Sender

1 Receiv er

Communicating Agent *

1..*

InternalAgent

*



Agents may interact with their inanimate environment, or they may interact with each other, which involves some form of communication; in the latter case, we speak of social interaction. We consider a business process as a special kind of a social interaction process. Unlike physical or chemical processes, social interaction processes are based on communication acts that may create commitments and are governed by norms. We distinguish between an interaction process type and an interaction process individual, while in the literature the term business process is used ambiguously both at the type and at the instance level.

•

interaction process: This is a process that includes at least one perception event and one action event perceived and performed by agents that participate in it. Examples: someone turning on the light in the office when it becomes dark outside; a football game; a conversation; a birthday party; the second World War; a Web shop purchase.

•

social interaction process: This is an interaction process that includes at least one communicative action event. Examples: a football game; a conversation; a birthday party; the second World War; a Web shop purchase.

•

business process: This is a social interaction process that occurs in the context of a business system and serves a purpose of that system. Examples: a football game; a Web shop purchase.

Using UFO to Evaluate Business Modeling Methods In the following subsections, we briefly present some preliminary results in order to exemplify how UFO can be used to evaluate business modeling methods.

The Enterprise Ontology The enterprise ontology, which was developed in a project led by the AI Applications Institute at the University of Edinburgh (see Uschold, King, Moralee, & Zorgios, 1998). Based on a simple upper-level ontology (“metaontology”) consisting of the three modeling concepts entity, relationship, and actor, it provides definitions for nearly 100 terms, both in natural language and in the formalism of Ontolingua. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


For simplicity, the distinction between an individual and an entity type is avoided. An agent (called actor) is defined as a special entity that can play an actor role in certain relationships (such as in performs Activity, has Capability, etc.). There is no independent concept of an event in the enterprise ontology: events are defined as “a kind of activity”. Remarkably, the authors consider also events that take place as a result of natural necessity (such as “water flowing down a hill”) as activities of “inanimate actors” (such as gravity). The following points highlight some shortcomings of the enterprise ontology: 1) For conceptual modeling, it is essential to distinguish between individuals and entity types; 2) It seems to be questionable to view natural forces that cause certain events to happen, such as gravity, as actors/agents; in UFO agents have a mental state and are able to act (create action events), perceive and possibly to communicate; and 3) Events must not be subsumed under activities. Rather, they should be first-class citizens of the meta-model. Unlike events, activities are always associated with an agent (their performer).

The Eriksson-Penker Business Extensions: Subsuming Agents Under Resources In Eriksson and Penker (1999), an approach to business modeling with UML based on four primary concepts is proposed: resources, processes, goals, and rules. In this proposal, there is no specific treatment of agents. They are subsumed, together with “material”, “products”, and “information” under the concept of resources. This unfortunate subsumption of human agents under the traditional “resource” metaphor, which is common in many business modeling methods, prevents a proper treatment of many important agent-related concepts (such as commitments, authorization, communication and interaction).

The REA (Resource-Event-Agent) Model The REA framework, whose ontological foundations are defined in Geert and McCarthy (2000), is based on a notion of an “economic exchange”. An economic exchange comprises a pair of economic events: an inflow and an outflow event. Economic agents participate in economic events and resources are affected (e.g., produced, used, acquired) by these events. In UFO, an economic event is a type of complex action event and resource is a type of substance individual (resources can be physical objects or amounts of matter). In REA, an individual is an (economic) agent by virtue of its participation in an economic event, while in UFO an agent is an individual to which we can ascribe a mental state.



Despite considering both individuals and entity types, the authors do not elaborate on the different sorts of entity types, which are necessary for conceptual enterprise modeling. An example of lack of ontological clarity is found when the authors mix the notions of event and commitments. For instance, in Figure 5 of the aforementioned paper, commitment and economic event are collapsed in one single typeimage. Additionally, the relationships partner and reserves (defined to hold between agent/commitment and resource/commitment, respectively) are considered as subtypes of participation and stock-flow (defined between agent/ economic event and resource/economic event). In our framework, whilst an economic event is a complex action event, a commitment is a social moment. Examples of other types of social moment defined in REA are accountability, responsibility, assignment, and custody. Despite recognizing the importance of part-whole relations in the enterprise domain (for example to model the relation between a resource and its parts), the treatment offered is insufficient. The authors only briefly mention a relation of composition that, together with other relations such as substitutes (meaning that a resource can substitute another), is subsumed under the relation of linkage between resources. No axiomatization for composition is provided. In a companion paper (Guizzardi, Herre, & Wagner, 2002b), we provide a formal characterization for parthood and discuss different types of this relation, which are important for conceptual modeling.

Conclusions The unified foundational ontology UFO 0.2 presented in this chapter should be viewed as an attempt to assemble a foundational ontology for conceptual modeling on the basis of other, already well established and philosophically justified foundational ontologies. We have stratified UFO into three ontological layers in order to distinguish its core, UFO-A, from the perdurant extension layer UFO-B and from the agent extension layer UFO-C. Although there is not much consensus yet in the literature regarding the ontology of agents, such an ontology is needed for building the foundation of conceptual business process modeling. UFO-C 0.2 is a first attempt to construct these foundations. We hope that we can validate and further improve it by investigating its applicability to business modeling problems.



References Bunge, M. (1977). Treatise on basic philosophy. Vol. 3. Ontology I. The furniture of the world. New York: Reidel. Bunge, M. (1979). Treatise on basic philosophy. Vol. 4. Ontology II. A world of systems. New York: Reidel. Chapin, D., Hall, J., Ross, R., Morgan, T, & Baisley, D. (2004). Business semantics of business rules (BSBR). Initial submission to OMG BEI RFP br/2003-06-03, 12 January 2004. Available from http://www.omg.org/cgibin/doc?bei/04-01-04. Degen, W., Heller, B., Herre, H., & Smith, B. (2001). GOL: Towards an axiomatized upper level ontology. In B. Smith & N. Guarino (Eds.), Proceedings of FOIS’01, Ogunquit, Maine, October 2001. ACM Press. (pp. 34-46). Eriksson, H. E. & Penker, M. (1999). Business Modeling with UML: Business Patterns at Work. New York: John Wiley & Sons. Evermann, J. & Wand, Y. (2001). Towards ontologically based semantics for UML constructs. In H. S. Kunii, S. Jajodia, & A. Solvberg (Eds.), Proceedings of ER 2001, (pp. 354-367). Berlin: Springer-Verlag. Gangemi, A., Guarino N., Masolo C., Oltramari, A., & Schneider L. (2002). Sweetening ontologies with DOLCE. Proceedings of EKAW 2002, Siguenza, Spain. Gärdenfors, P. (2000). Conceptual Spaces: The Geometry of Thought. Boston: MIT Press. (pp. 166-181). Geert, G. & McCarthy, W. E. (2000). The ontological foundation of REA enterprise information systems. Retrieved from http://www.msu.edu/ user/mccarth4/rea-ontology Green, P. F. & Rosemann, M. (2000). Integrated process modelling: An ontological evaluation. Information Systems, 25(2), 73-87. Green, P. F. & Rosemann, M. (2002). Usefulness of the BWW ontological models as a “core” theory of information systems. Proceedings Information Systems Foundations: Building the Theoretical Base, Canberra, 2002, (pp. 147-164). Guizzardi, G., Herre, H., & Wagner G. (2002a). On the general ontological foundations of conceptual modeling. Lecture notes in Computer Science, Vol. 2503, 65-78. Berlin: Springer-Verlag. Guizzardi, G., Herre, H., & Wagner, G. (2002b), Towards ontological foundations for UML conceptual models. Lecture notes in Computer Science, Vol. 2519, 1100-1117. Berlin: Springer-Verlag. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Guizzardi, G., Wagner G., Guarino, N., & van Sinderen, M. (2004b). An ontologically well-founded profile for UML conceptual models. Lecture notes in Artificial Intelligence, Vol. 3084, 112-126. Berlin: SpringerVerlag. Guizzardi, G., Wagner, G., & van Sinderen, M. (2004a). A formal theory of conceptual modeling universals. Proceedings of the International Workshop on Philosophy and Informatics (WSPI), Cologne, Germany (pp. 7380). Gupta, A. (1980). The logic of common nouns: An investigation in quantified modal logic. New Haven, CT: Yale University Press. Hirsch, E. (1982). The concept of identity. New York, Oxford: Oxford University Press. ISO. (2000). ISO 1087-1. Terminology work — Vocabulary — Part 1: Theory and application. Retrieved from http//:webstore.ansi.org Object Management Group (OMG). (2003). Meta object facility (MOF) 2.0 core specification, version 2.0. Retrieved from http://www.omg.org/docs/ptc/ 03-10-04.pdf Opdahl, A. L. & Henderson-Sellers, B. (2002). Ontological evaluation of the UML using the Bunge-Wand-Weber model. Software and Systems Journal, 1(1), 43-67. Searle, J. R. (1995). The construction of social reality. New York: Free Press. Smith, B. & Mulligan, K. (1983). Framework for formal ontology. Topoi, 3, 7385. Smith, B. & Mulligan, K. (1986). A relational theory of the act. Topoi, 5(2), 115130. Steimann, F. (2000). On the representation of roles in object-oriented and conceptual modeling. Data & Knowledge Engineering, 35(1), 83-106. van Belle, J. P. (1999). Moving towards generic enterprise information models: From Pacioli to CyC. Proceedings of the 10th Australasian Conference on Information Systems, Wellington. (pp. 1084-1094). van Leeuwen, J. (1991). Individuals and sortal concepts: An essay in logical descriptive metaphysics. PhD thesis, University of Amsterdam. Wand, Y. & Weber, R. (1990). Mario Bunge’s ontology as a formal foundation for information systems concepts. In P. Weingartner & G. J. W. Dorn (Eds.), Studies on Mario Bunge’s treatise. Atlanta: Rodopi. Wand, Y. & Weber, R. (1995). On the deep structure of information systems. Information Systems Journal, 5, 203-223. Welty, C & Guarino, N. (2001). Supporting ontological analysis of taxonomic relationships. Data and Knowledge Engineering, 39(1), 51-74. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Wiggins, D. (2001). Sameness and substance renewed. Cambridge, MA: Cambridge University Press. World Wide Web Committee (W3C). (2004). Web ontology language (OWL). Retrieved from http://www.w3.org/2001/sw/WebOnt

Endnotes 1

UFO 0.2 differs from UFO 0.1, which has been presented at the EMOIINTEROP Workshop at CAiSE’04, by adding the categories of datatype, process, and business process.

2

This problem is also mentioned in (van Belle, 1999, p. 1089): “How would one model the customer entity conceptually? The Customer as a supertype of Organisation and Person? The Customer as a subtype of Organisation and Person? The Customer as a relationship between or Organisation and (Organization or Person)?”



Appendix Figure 9. UFO-A 0.2 as a MOF/UML model * A b str act T h in g

T h in g

Part *

M ember

* I ns tanc e

Subs et

En ti ty

Se t

1. .*

1.. * Ex tens ion

*

*

* *

1. .* Am ount of M atte r { c om pl et e}

Clas s ific at ionT y pe

Subt y pe

En d u r an t

So r ta lT y p e

1

M i xin T y p e

1 bears

? inheres in 1.. *

Su b s tan ce In d i v id u al

D atat yp e

1.. *

En ti ty Type

I n d i v id u al

Pe r d u r a n t

*

1

*

{c o m p lete}

B a se T yp e

R o l eT yp e

Ph as eT y p e

M o m en t In d i v i d u al

P h ys ical O b jec t

1. .*

R e lat o r

I n tri n si c M o m en t

Qu al ity


368 About the Authors

About the Authors

Peter F. Green is a professor of electronic commerce and business information systems cluster leader in the UQ Business School at the University of Queensland, Australia. He has qualifications in computer science and accounting, and a PhD in commerce (information systems) from the University of Queensland. Dr. Green is a chartered accountant and a member of the Australian Computer Society. He has worked during his career as the systems support manager at the South-East Queensland Electricity Board (SEQEB), for a chartered accountancy firm, and for a Queensland government department. He has researched, presented, and published widely on systems analysis and design, conceptual modeling, information systems auditing, and e-commerce. Dr. Green’s publications have appeared in such internationally refereed journals as Information Systems, Journal of Database Management, and the Australian Journal of Information Systems. Michael Rosemann is a professor at the School of Information Systems and coleader of the Business Process Management Group at the Faculty of Information Technology, Queensland University of Technology, Brisbane, Australia. He received his masters of business administration (1992) and his PhD degree (1995) from the University of Münster, Germany. Dr. Rosemann’s areas of interest are conceptual modeling, business process management, and enterprise systems. He is Chief Investigator of a number of research projects funded by the Australian Research Council and industry partners such as SAP AG. Besides 75 refereed journal and conference papers, he is the author of two books and the editor of two books. He is a member of the editorial board of six journals. ****


About the Authors 369

Benoit A. Aubert is currently a professor and director of the Research Group in Information Systems (GReSI) at HEC Montreal, where he holds a professorship in governance and information technology. He is also a fellow at the CIRANO (Center for Interuniversity Research and Analysis on Organizations), where he works on issues in governance, outsourcing, and integrated risk management. His main research areas are outsourcing, outsourcing risk management, and new organization forms (virtual, network, alliances, etc.). Jörg Becker is the acting director of the Institute of Information Systems and chair of Information Management at the University of Muenster, Germany. He studied at the University of Saarland, Germany and the University of Michigan, Ann Arbor and holds PhD in business administration from the University of Saarland. Dr. Becker is one of two editors in chief of Information Systems and E-Business Management and serves furthermore on the editorial boards of Information and Management, Wirtschaftsinformatik (German), and Information Management and Consulting (German). His research interests include information management, information modelling, data management, logistics, retail information systems, and strategic IT-management consulting. Dr. Becker authored numerous books and journal and conference articles. He also serves regularly as program committee member of various conferences such as the European Conference on Information Systems. Dov Dori is head of the Information Systems Engineering Area with the faculty of industrial engineering and management, Technion, Israel Institute of Technology, and research affiliate at MIT, Cambridge, MA. Between 1999-2001, he was a visiting faculty member at MIT’s Engineering Systems Division and Sloan School of Management. Dr. Dori received his bachelor of science degree in industrial engineering and management from the Technion in 1975, his master of science degree in operations research from Tel Aviv University in 1981, and his PhD in computer science from Weizmann Institute of Science, Rehovot, Israel, in 1988. Dr. Dori is on the editorial board of the International Journal of Web Engineering Technologies (IJWET) and International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI). He was associate editor of IEEE Transaction on Pattern Analysis and Machine Intelligence (T-PAMI) and the International Journal of Document Analysis and Recognition (IJDAR). Dr. Dori is co-editor of three books, author of more than 90 journal papers and book chapters, and 70 conference publications. He is a fellow of the International Association for Pattern Recognition (IAPR), a senior member of IEEE, and a member of IEEE Computer Society and ACM.



Alexander Dreiling received his master’s degree in information systems from the University of Münster, Germany, in 2001. After one year of industry experience in the United States, he returned to the University of Münster for a PhD program in information systems, which he still works on. In February 2004, he started his current position as a research fellow at Queensland University of Technology, Australia. Dr. Dreiling’s academic career so far has led to about 20 conference and journal publications, best paper award nominations at international conferences, and the best track award on IRMA 2004 in his function as a co-track chair. Aymeric Dussart holds both a master’s degree in economics from University of Montreal and a master’s degree in management information systems from HEC Montreal. His research interests focus on the areas of workflow management and business process integration. He presently works for a management consulting firm in Montreal specializing in technology management for major Canadian clients in the information technologies and telecommunications fields. He lectured at HEC Montreal, INSEAD/CEDEP, and MIT/Sloan School of Management. Jörg Evermann received his PhD in management information systems from the University of British Columbia. His research interests are in the area of system analysis. He is particularly interested in the philosophical foundations of IS development, including the use and evaluation of ontologies and system modeling languages. Before receiving his PhD, Dr. Evermann was involved in a number of ERP implementation projects with KPMG Consulting. He is currently a lecturer at the School of Information Management at the Victoria University of Wellington. Peter Fettke received a master’s degree in information systems (DiplomWirtschaftsinformatiker) from the University of Münster, Germany. Since 2002, he is a research assistant for the chair of information systems and business administration (Prof. Peter Loos), Johannes Gutenberg-University Mainz, Germany. Currently he is working on his PhD thesis. His research interests include information systems analysis and design, especially the use of conceptual modeling and component-based system paradigm. He has published articles on reference modeling, conceptual modeling, and component-based engineering in both national and international journals and at conferences. Andrew Gemino is an assistant professor in management information systems with the faculty of business administration at Simon Fraser University. His



degrees include a master of arts degree in economics, a master’s of business administration and a PhD in management information systems. He has published articles in academic journals, including the Communications of the ACM, European Journal of Information Systems, and Requirements Engineering Journal. Beyond his academic interest, he is a co-founder of a software company involved in the development of commercial software used by professional sports teams in the NHL and NBA, as well as automated employee scheduling for sports and entertainment companies. Asunción Gómez-Pérez is the director of the Ontology Group at Politécnica University of Madrid since 1995 and an associate professor at the Computer Science School also at Politécnica University of Madrid. She has a bachelor of arts degree in computer science (1990), a master of science degree in knowledge engineering (1991), and a PhD in computer science (1993) all from Politécnica University of Madrid. She also has a master’s degree in business administration (1994) from Pontificia University of Comillas. She was visiting professor (19941995) at the Knowledge Systems Laboratory at Stanford University. She was the executive director (1995-1998) of the Artificial Intelligence Laboratory at the Computer Science School at Politécnica University of Madrid and is currently a research advisor of the same lab. Her research activities include: interoperability between different kinds of ontology development tools; methodologies and tools for building and merging ontologies; ontological reengineering; ontology evaluation and ontology evolution, as well as uses of ontologies in applications related with semantic webs; and e-commerce and knowledge management. Dr. GómezPérez also acts as consultant of the European Commission on the FP6 program on the topic of knowledge technologies. She has published more than 70 papers on the aforementioned issues and has led several national and international projects related to ontologies funded by various institutions and/or companies. Giancarlo Guizzardi is a research assistant at the Architecture and Services of Network Applications Research Group in the University of Twente, The Netherlands. He is currently in the very last stage of his PhD research, which focus on the application of Foundational Ontologies for the design of philosophically and cognitively well-founded methodological tools for conceptual modeling. His other research interests include: software engineering (in particular domain engineering and software reuse) and the design of domain-specific visual languages. Brian Henderson-Sellers is director of the Centre for Object Technology Applications and Research and a professor of information systems at University of Technology, Sydney (UTS). He is the author of 11 books on object technology Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


and is well known for his work in OO methodologies (MOSES, COMMA, OPEN, OOSPICE) and in OO metrics. Dr. Henderson-Sellers has been a regional editor of object-oriented systems, a member of the editorial board of Object Magazine/Component Strategies and an object expert for many years, and is currently on the editorial board of Journal of Object Technology and Software and Systems Modeling. He was the founder of the Object-Oriented Special Interest Group of the Australian Computer Society (NSW Branch) and chairman of the Computerworld Object Developers’ Awards committee for ObjectWorld 94 and 95 in Sydney. He is a frequent, invited speaker at international OT conferences. In 1999, he was voted number three in the Who’s Who of Object Technology (Handbook of Object Technology, CRC Press, Appendix N). He is currently a member of the review panel for the OMG’s software process engineering model (SPEM) standards initiative and is a member of the UML 2.0 review team. In July 2001, Henderson-Sellers was awarded a doctor of science degree from the University of London for his research contributions in object-oriented methodologies. Roland Holten holds the chair for information systems engineering with the faculty of economics and business administration at Goethe University in Frankfurt, Germany. He habilitated in information systems with the economics faculty at the University of Muenster in Germany, and holds two master’s degrees in business administration and computer science and a PhD in information systems. His work has been published in peer-reviewed journals such as Information Systems, Information Systems and e-Business Management (ISeB), and WIRTSCHAFTSINFORMATIK. Dr. Holten is a member of the editorial board of the journal ISeB and works in the areas of information warehousing, IS modeling, and IS integration. Ed Kazmierczak received his PhD from the University of Tasmania, Australia, in 1992 in the areas of logic and formal methods. His post-doctoral work at the Laboratory for the Foundations of Computer Science at the University of Edinburgh, United Kingdom, was in the areas of modeling and refining functional programs. Following Edinburgh, he joined the Software Verification Research Centre at the University of Queensland, Australia, working in the areas of modeling and refining software for safety critical systems. In 1996, he joined the University of Melbourne, Australia, where he has been the coordinator for software engineering, teaches software engineering, and specializes in software modeling and analysis. Helmut Klaus obtained a master’s degree in information technology and a PhD in 2004 from Queensland University of Technology, Australia. Dr. Klaus is Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


currently a senior research assistant with the Research Centre for Information Technology Innovation at Queensland University of Technology. His research interests include information systems in government, information systems expertise, social and political theory, and philosophy. Peter Loos received a master’s degree in business administration from the University of Saarland in Germany. He completed his Ph. thesis in 1991 and his habilitation thesis in 1997. From 1987 to 1992, he was manager of the software engineering department at the IDS Scheer AG. Since 1999, he has been a professor for information systems at Chemnitz University and at Mainz University, both in Germany. His research interests include information systems applications in industry and financial service provider, conceptual modeling, and system development. He has published numerous articles on ERP systems, information modeling, and development methodologies in both national and international journals and at conferences. Adolfo Lozano-Tello has been a teaching/research assistant professor of computer languages and systems since 1995 at Extremadura University in Spain. He received his bachelor of science degree in computer science at Granada University, Spain, in 1993. He has worked at PROMAINEX (1993-1995) as a computer consultant and a programming team leader. He was the director of the Computer Science Area in Minimally Invasive Surgery Centre in Cáceres, Spain, in 1995. He received his PhD in 2002 in computer science from Extremadura University, with a special prize for having an extraordinary thesis. His PhD research addressed the use of software components and ontologies in applications. His research interests include reusable software components rating, ontology design, and software component and ontology features identification. He has published more than 20 papers on the aforementioned issues in software engineering and knowledge engineering and belongs to several projects related to these topics. At present, he is also the director of the Computer Science Service at Extremadura University. Simon Milton received his PhD from the University of Tasmania, Australia, in 2000 in the area of ontological analysis of data modeling languages. In his thesis, he undertook the first comprehensive ontological analysis of data modeling languages using a common-sense ontology (Chisholm’s ontology). His primary focus remains studying the ontological foundations of data modeling languages — helping data models to Œmap the world. More recently, Dr. Milton has been studying the role of theories of agency in information systems design methodologies and associated ontological commitments. He has been in the department of information systems at the University of Melbourne, Australia, since 2000. Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


Jasmina Nuredini is a consultant with Simsion Associates in Australia. She comes from a financial and data modeling background, with solid experience in data management, process modeling, and enterprise (corporate) data modeling. Nuredini has also held research positions at Deakin University, the University of Melbourne, and Monash University, all in Australia. Through her academic and professional experience, she has developed a keen interest in exploring new ideas and relating them to industry practices, particularly in data and process modeling. Her research results have been published at several Australian and international conferences. Andreas L. Opdahl is a professor of information systems development at the department of information science, University of Bergen, Norway. He is the author, co-author, or co-editor of around 50 journal articles, book chapters, refereed archival conference papers, and books on requirements engineering, multi-perspective enterprise and IS modeling, software performance engineering, and other areas. He is a member of IFIP WG8.1 on design and evaluation of information systems. He serves regularly as a reviewer for premier international journals and on the program committees of renowned international conferences and workshops. He can be contacted at: Department of Information Science and Media Studies, University of Bergen, P.O. Box 7800, N-5020 Bergen, Norway, or at [email protected] or http://www.ifi.uib.no/staff/ andreas. Michel Patry holds a master’s degree in business science from HEC Montreal and a PhD in economics from the University of British Columbia. Dr. Patry is currently a full professor at the Institut d’économie appliquée of HEC Montreal and the associate-director of academic affairs and strategic planning of HEC. He is also a fellow at the CIRANO (Center for Interuniversity Research and Analysis on Organizations). He was vice president of CIRANO from 1998 to 2001. A specialist of the economics of the organization, Dr. Patry’s recent work covers the areas of outsourcing and delegated management, the economics of information technology, the analysis of regulation and contracts, and the impact of regulation on productivity. Iris Reinhartz-Berger received her PhD in 2003 from the faculty of industrial engineering and management, information systems, Technion, Israel Institute of Technology. Both her master of science degree and her PhD were on improving OPM. During her master of science degree work, she formalized conversion rules from object-process language, a subset of English, to Java. Her PhD dissertation dealt with developing Web applications with object-oriented approaches and OPM. Currently, she is a faculty member in the department of Copyright © 2005, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of Idea Group Inc. is prohibited.


management information systems, University of Haifa, Israel, working on improving information systems development processes in various modeling languages (including UML, OPM, XP, etc.). Graeme Shanks is a professor in the School of Business Systems at Monash University, Australia. He has also held academic positions at the University of Melbourne and Chisholm Institute of Technology, both in Australia. Before becoming an academic, he worked for a number of private and government organizations as a programmer, systems analyst, and project leader. His teaching and research interests include information quality, conceptual modeling, identity management, implementation and impact of enterprise systems, and decision support systems. He has published the results of his research in more than 100 refereed journal and conference papers. He is on the editorial boards of the Journal of Data Warehousing, the Journal of Database Management, International Journal of Data Warehousing and Mining, and the Journal of Knowledge Management Research and Practice. He is also a member of the College of Experts of the Australian Research Council, a member of IFIP WG8.3 Decision Support Systems, and a fellow of the Australian Computer Society. Gerd Wagner is a professor of Internet technology at the Brandenburg University of Technology at Cottbus, Germany. His research interests include agentoriented modeling and agent-based simulation, foundational ontologies, (business) rule technologies, and the semantic web. He is an active member of the European research network REWERSE.net and of the international standardization effort RuleML.org. Yair Wand is a CANFOR professor of MIS in the management information systems division of the faculty of commerce and business administration at the University of British Columbia. He holds a doctor of science degree in operations research (Technion, Israel) and a master of science degree in physics (The Weizmann Institute, Israel). Dr. Wand is interested in information systems modeling, theoretical foundations and methods for information systems analysis and design, database design, enterprise modeling, and business process modeling. He is a member of the editorial boards of Journal of the Association of Information Systems, Requirements Engineering Journal, Database Management, and the Journal of Data Semantics and Applied Ontology. Ron Weber is the dean of the faculty of information technology at Monash University, Australia. His primary research interests are in the areas of ontology and conceptual modeling. He is a past president of the Association for Informa-



tion Systems and the Accounting and Finance Association of Australia and New Zealand. In 2000, Dr. Weber was co-chair of the International Conference on Information Systems. From 2002-2004, he was editor-in-chief of MIS Quarterly. He is a fellow of the Association for Information Systems, the Institute of Chartered Accountants in Australia, CPA Australia, the Australian Computer Society, and the Academy of the Social Sciences in Australia. Boris Wyssusek obtained a master’s degree in computer science, a master’s degree in business administration, and a PhD in information systems, all from Technical University in Berlin, Germany. He is currently a postdoctoral research fellow with the Research Centre for Information Technology Innovation (CITI) at Queensland University of Technology, Brisbane, Australia. Dr. Wyssusek’s research interests include information systems, organizations, and methodology.


Index 377

Index

A ad hoc workflows 272 adequacy 64 administrative workflows 272 analytic hierarchy process (AHP) 252 ANSI 277 ARIS 11 “as is” process 288

B BPEL4WS 5 BPML 5 Bunge-Wand-Weber (BWW) ontological models 4, 336 business modeling 345 business object class 356 business process modelling 2 business process reengineering (BPR 288 business to business (B2B) 271 BWW 346

C CASE tools 4 Chisholm’s ontology 5, 218 classes 232 cognition 33 cognitive behaviour patterns 42 cognitive-based approach 315

common-sense realism 219 composite 31 compound concept 237 comprehension 39 concept 224 conceptual comparison 238 conceptual evaluation 234 conceptual model 82, 252 construct 32, 220 construct excess 4 construct of an ontological model 58 construct overload 4 construct redundancy 4 conventionalist view 318 CYC ontology 250

D DAML+OIL 250 data flow diagrams 308 data models 218 database design 30 deep understanding 35 design model 82 design Techniques 305 DOLCE 345 domain 83 domain ontology 252 domain-specific ontology 223


378 Index

E ebXML 5 end-user understanding 28 enterprise modelling 2 enterprise ontology 361 enterprise resource planning (ERP) 5 enterprise theory 189 entity relationship diagrams 308 entity-relationship (ER) 4 EPC 277 epistemological question 177 ERM 60 event-driven process chains (EPC) 16 Excess 61, 64

F FLogic 250 focused ontology 21 formal method 332 foundational ontology 345 Fregean theory of language 76 functional data model 219

G GFO-GOL 347 GFO/GOL 345 goals 362 grammar 61, 274 grammar-based approach 315 grammatical constructs 274 gulf of evaluation 313 gulf of execution 313

I inadequacy 64 incompleteness 61 individual 231, 353 information modeling 57, 176 information systems (IS) 322, 83 information systems analysis and design (ISAD) 322 information systems development 174 inter-organizational workflow 271 interpretation mapping 61

interpretivism 177 IT 175

K knowledge 310

L language convergence 273 language critique approach 174 linguistic analysis 187 LISP 250 logical data flow diagramming (LDFD) 4 LOOM 250

M managerial analysis 200 mapping 87, 238 meta-model 12 meta-model-based methods 176 method 234, 326 methodism 326 metrics 266 MIS 188 modelling rules 85

N Nijssen Information Analysis Method (NIAM) 4

O object constraint language (OCL) 106 object modelling technique 219 object-role 29 OCML 250 OIL 250 OMG’s model-driven architecture 106 OntoClean 345 OntoClean-DOLCE 347 ontolingua 250 ontolingua server 251 ontological clarity 4 ontological completeness 4 ontological model 58 ontological normalization 62


Index 379

ontological overlap 275 ontological question 177 ontology 85, 174, 249, 312, 322, 333 ontology-based applications 249 ontology-driven method engineering 174 OntoMetric method 249 OntoMetric tool 253 OPEN modeling language (OML) 4 overload 32, 61 OWL 250, 347

P part-whole 5 petri nets 277 post-rationalism 328 problem-solving 39 processes 362 product modeling 357 production workflows 272 project escalation 175 project failure 175 purpos 21

Q quality 57

R rationalism 328 RDF 250 REA (resource-event-agent) model 362 real world systems 323 realistic ontology 218 redundancy 61 reference model 5, 56, 64 reference ontology (RO) 221, 252 relation 233 representation 33 representation mapping 61 requirements engineering 179 resources 362 rules 362

scripts 58, 274 semantic data model 219 semantic field 237 semiotic 182 SHOE 250 software engineering 326 specific reality 223 speech act theory 76 stereotype 292 SUO 346 surface-level 39 system analysis 305

T template 105 “to be” process 288

U UFO 345 UFO-A 348 UFO-B 348 UFO-C 348 unified modeling language (UML) 60, 82, 219, 271, 278

V verbal protocol technique 42

W Web service 18 WfMC 277 workflow 277 workflow management system (WFMS) 272 WSCI 5

X XML 250 XOL 250

S scientific realism 106 scope management 307


Instant access to the latest offerings of Idea Group, Inc. in the fields of I NFORMATION SCIENCE , T ECHNOLOGY AND MANAGEMENT!

InfoSci-Online Database BOOK CHAPTERS JOURNAL AR TICLES C ONFERENCE PROCEEDINGS C ASE STUDIES

“

The Bottom Line: With easy to use access to solid, current and in-demand information, InfoSci-Online, reasonably priced, is recommended for academic libraries.

The InfoSci-Online database is the most comprehensive collection of full-text literature published by Idea Group, Inc. in:

”

- Excerpted with permission from Library Journal, July 2003 Issue, Page 140

n n n n n n n n n

Distance Learning Knowledge Management Global Information Technology Data Mining & Warehousing E-Commerce & E-Government IT Engineering & Modeling Human Side of IT Multimedia Networking IT Virtual Organizations

BENEFITS n Instant Access n Full-Text n Affordable n Continuously Updated n Advanced Searching Capabilities

Start exploring at www.infosci-online.com

Recommend to your Library Today! Complimentary 30-Day Trial Access Available! A product of:

Information Science Publishing* Enhancing knowledge through information science

*A company of Idea Group, Inc. www.idea-group.com

New Releases from Idea Group Reference

Idea Group REFERENCE

The Premier Reference Source for Information Science and Technology Research ENCYCLOPEDIA OF

ENCYCLOPEDIA OF

DATA WAREHOUSING AND MINING

INFORMATION SCIENCE AND TECHNOLOGY AVAILABLE NOW!

Edited by: John Wang, Montclair State University, USA Two-Volume Set • April 2005 • 1700 pp ISBN: 1-59140-557-2; US $495.00 h/c Pre-Publication Price: US $425.00* *Pre-pub price is good through one month after the publication date

Provides a comprehensive, critical and descriptive examination of concepts, issues, trends, and challenges in this rapidly expanding field of data warehousing and mining A single source of knowledge and latest discoveries in the field, consisting of more than 350 contributors from 32 countries

Five-Volume Set • January 2005 • 3807 pp ISBN: 1-59140-553-X; US $1125.00 h/c

ENCYCLOPEDIA OF

DATABASE TECHNOLOGIES AND APPLICATIONS

Offers in-depth coverage of evolutions, theories, methodologies, functionalities, and applications of DWM in such interdisciplinary industries as healthcare informatics, artificial intelligence, financial modeling, and applied statistics Supplies over 1,300 terms and definitions, and more than 3,200 references

DISTANCE LEARNING

April 2005 • 650 pp ISBN: 1-59140-560-2; US $275.00 h/c Pre-Publication Price: US $235.00* *Pre-publication price good through one month after publication date

Four-Volume Set • April 2005 • 2500+ pp ISBN: 1-59140-555-6; US $995.00 h/c Pre-Pub Price: US $850.00* *Pre-pub price is good through one month after the publication date

MULTIMEDIA TECHNOLOGY AND NETWORKING

ENCYCLOPEDIA OF

ENCYCLOPEDIA OF

More than 450 international contributors provide extensive coverage of topics such as workforce training, accessing education, digital divide, and the evolution of distance and online education into a multibillion dollar enterprise Offers over 3,000 terms and definitions and more than 6,000 references in the field of distance learning Excellent source of comprehensive knowledge and literature on the topic of distance learning programs Provides the most comprehensive coverage of the issues, concepts, trends, and technologies of distance learning

April 2005 • 650 pp ISBN: 1-59140-561-0; US $275.00 h/c Pre-Publication Price: US $235.00* *Pre-pub price is good through one month after publication date

www.idea-group-ref.com

Idea Group Reference is pleased to offer complimentary access to the electronic version for the life of edition when your library purchases a print copy of an encyclopedia For a complete catalog of our new & upcoming encyclopedias, please contact: 701 E. Chocolate Ave., Suite 200 • Hershey PA 17033, USA • 1-866-342-6657 (toll free) • [email protected]

IT Solutions Series Humanizing Information Technology: Advice from Experts Authored by: Shannon Schelin, PhD, North Carolina State University, USA G. David Garson, PhD, North Carolina State University, USA

With the alarming rate of information technology changes over the past two decades, it is not unexpected that there is an evolution of the human side of IT that has forced many organizations to rethink their strategies in dealing with the human side of IT. People, just like computers, are main components of any information systems. And just as successful organizations must be willing to upgrade their equipment and facilities, they must also be alert to changing their viewpoints on various aspects of human behavior. New and emerging technologies result in human behavior responses, which must be addressed with a view toward developing better theories about people and IT. This book brings out a variety of views expressed by practitioners from corporate and public settings offer their experiences in dealing with the human byproduct of IT.

ISBN 1-59140-245-X (s/c) • US$29.95 • eISBN 1-59140-246-8 • 186 pages • Copyright © 2004

Information Technology Security: Advice from Experts Edited by: Lawrence Oliva, PhD, Intelligent Decisions LLC, USA

As the value of the information portfolio has increased, IT security has changed from a product focus to a business management process. Today, IT security is not just about controlling internal access to data and systems but managing a portfolio of services including wireless networks, cyberterrorism protection and business continuity planning in case of disaster. With this new perspective, the role of IT executives has changed from protecting against external threats to building trusted security infrastructures linked to business processes driving financial returns. As technology continues to expand in complexity, databases increase in value, and as information privacy liability broadens exponentially, security processes developed during the last century will not work. IT leaders must prepare their organizations for previously unimagined situations. IT security has become both a necessary service and a business revenue opportunity. Balancing both perspectives requires a business portfolio approach to managing investment with income, user access with control, and trust with authentication. This book is a collection of interviews of corporate IT security practitioners offering various viewpoint on successes and failures in managing IT security in organizations.

ISBN 1-59140-247-6 (s/c) • US$29.95 • eISBN 1-59140-248-4 • 182 pages • Copyright © 2004

Managing Data Mining: Advice from Experts Edited by: Stephan Kudyba, PhD, New Jersey Institute of Technology, USA Foreword by Dr. Jim Goodnight, SAS Inc, USA

Managing Data Mining: Advice from Experts is a collection of leading business applications in the data mining and multivariate modeling spectrum provided by experts in the field at leading US corporations. Each contributor provides valued insights as to the importance quantitative modeling provides in helping their corresponding organizations manage risk, increase productivity and drive profits in the market in which they operate. Additionally, the expert contributors address other important areas which are involved in the utilization of data mining and multivariate modeling that include various aspects in the data management spectrum (e.g. data collection, cleansing and general organization).


E-Commerce Security: Advice from Experts Edited by: Mehdi Khosrow-Pour, D.B.A., Information Resources Management Association, USA

The e-commerce revolution has allowed many organizations around the world to become more effective and efficient in managing their resources. Through the use of e-commerce many businesses can now cut the cost of doing business with their customers in a speed that could only be imagined a decade ago. However, doing business on the Internet has opened up business to additional vulnerabilities and misuse. It has been estimated that the cost of misuse and criminal activities related to e-commerce now exceeds 10 billion dollars per year, and many experts predict that this number will increase in the future. This book provides insight and practical knowledge obtained from industry leaders regarding the overall successful management of e-commerce practices and solutions.


Its Easy to Order! Order online at www.cybertech-pub.com, www.idea-group.com or call 717/533-8845 x10 Mon-Fri 8:30 am-5:00 pm (est) or fax 24 hours a day 717/533-8661

CyberTech Publishing Hershey • London • Melbourne • Singapore

Excellent additions to your library!

Business Systems Analysis with Ontologies

Ontologies Based Business Integration

Ontologies-Based Business Integration

Business Analysis with QuickBooks

Reference Modeling for Business Systems Analysis

Business Information Systems: Analysis, Design & Practice

Reference Modeling for Business Systems Analysis

Creating Agile Business Systems with Reusable Knowledge

Integrated Business Processes with ERP Systems

Mastering Business Analysis with Crystal Reports 9

Mastering Business Analysis with Crystal Reports 9

Business Rules and Information Systems: Aligning IT with Business Goals

Systems Analysis and Design with UML

Systems Analysis

Introduction to Bio-Ontologies

Business Database Systems

Business and Competitive Analysis

Business Ethics and Organizational Values: A Systems Theoretical Analysis

Business Information Systems: Analysis, Design & Practice, 5th Edition

Doing Business with China

Doing Business with Slovakia

Doing Business with Poland

Doing Business with Estonia

Doing Business with Lithuania

Doing Business with Slovenia

Doing Business with China

Doing Business with Kazakhstan

Doing Business with Jordan

Doing Business with Ukraine

Hydrocarbon Migration Systems Analysis

Equilibrium Analysis Complex Systems

Business Systems Analysis with Ontologies

Ontologies Based Business Integration

Ontologies-Based Business Integration

Business Analysis with QuickBooks

Reference Modeling for Business Systems Analysis

Business Information Systems: Analysis, Design & Practice

Reference Modeling for Business Systems Analysis

Creating Agile Business Systems with Reusable Knowledge

Integrated Business Processes with ERP Systems

Mastering Business Analysis with Crystal Reports 9

Mastering Business Analysis with Crystal Reports 9

Business Rules and Information Systems: Aligning IT with Business Goals

Systems Analysis and Design with UML

Systems Analysis

Introduction to Bio-Ontologies

Business Database Systems

Business and Competitive Analysis

Business Ethics and Organizational Values: A Systems Theoretical Analysis

Business Information Systems: Analysis, Design & Practice, 5th Edition

Doing Business with China

Doing Business with Slovakia

Doing Business with Poland

Doing Business with Estonia

Doing Business with Lithuania

Doing Business with Slovenia

Doing Business with China

Doing Business with Kazakhstan

Doing Business with Jordan

Doing Business with Ukraine

Hydrocarbon Migration Systems Analysis

Equilibrium Analysis Complex Systems

Recommend Documents