The TQM Journal, Volume 20 Issue 4: The International Review Of Organizational Improvement Kansei affective Engineering Design

tqm cover (i).qxd 24/06/2008 13:19 Page 1 ISSN 1754-2731 Volume 20 Number 4 2008 Celebrating 20 years 1989-2008 T...

Author: Jens J. Dahlgaard (Editor)

59 downloads 693 Views 6MB 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

tqm cover (i).qxd

24/06/2008

13:19

Page 1

ISSN 1754-2731

Volume 20 Number 4 2008

Celebrating 20 years 1989-2008

The TQM Journal The international review of organizational improvement formerly The TQM Magazine

Kansei/affective engineering design Guest Editor: Professor Jens J. Dahlgaard

www.emeraldinsight.com

The TQM Journal

ISSN 1754-2731 Volume 20 Number 4 2008

Kansei/affective engineering design Guest Editor Professor Jens J. Dahlgaard

Access this journal online _________________________

283

Editorial advisory board __________________________

284

Guest editorial ___________________________________

285

Perspectives and the new trend of Kansei/affective engineering Mitsuo Nagamachi _____________________________________________

290

Kansei/affective engineering design: a methodology for profound affection and attractive quality creation Jens J. Dahlgaard, Simon Schu¨tte, Ebru Ayas and Su Mi Dahlgaard-Park __________________________________________

299

Customer experience management: influencing on human Kansei to management of technology Shin’ya Nagasawa _____________________________________________

312

Kansei engineering approach for total quality design and continuous innovation Antonio Lanzotti and Pietro Tarantino_____________________________

Access this journal electronically The current and past volumes of this journal are available at:

www.emeraldinsight.com/1754-2731.htm You can also search more than 175 additional Emerald journals in Emerald Management Xtra (www.emeraldinsight.com) See page following contents for full details of what your access includes.

324

CONTENTS

www.emeraldinsight.com/tqm.htm As a subscriber to this journal, you can benefit from instant, electronic access to this title via Emerald Management Xtra. Your access includes a variety of features that increase the value of your journal subscription.

Structured abstracts Emerald structured abstracts provide consistent, clear and informative summaries of the content of the articles, allowing faster evaluation of papers.

How to access this journal electronically

Additional complimentary services available

To benefit from electronic access to this journal, please contact [email protected] A set of login details will then be provided to you. Should you wish to access via IP, please provide these details in your e-mail. Once registration is completed, your institution will have instant access to all articles through the journal’s Table of Contents page at www.emeraldinsight.com/1754-2731.htm More information about the journal is also available at www.emeraldinsight.com/ tqm.htm

Your access includes a variety of features that add to the functionality and value of your journal subscription:

Our liberal institution-wide licence allows everyone within your institution to access your journal electronically, making your subscription more cost-effective. Our web site has been designed to provide you with a comprehensive, simple system that needs only minimum administration. Access is available via IP authentication or username and password. Emerald online training services Visit www.emeraldinsight.com/training and take an Emerald online tour to help you get the most from your subscription.

Key features of Emerald electronic journals Automatic permission to make up to 25 copies of individual articles This facility can be used for training purposes, course notes, seminars etc. This only applies to articles of which Emerald owns copyright. For further details visit www.emeraldinsight.com/ copyright Online publishing and archiving As well as current volumes of the journal, you can also gain access to past volumes on the internet via Emerald Management Xtra. You can browse or search these databases for relevant articles. Key readings This feature provides abstracts of related articles chosen by the journal editor, selected to provide readers with current awareness of interesting articles from other publications in the field. Non-article content Material in our journals such as product information, industry trends, company news, conferences, etc. is available online and can be accessed by users. Reference linking Direct links from the journal article references to abstracts of the most influential articles cited. Where possible, this link is to the full text of the article. E-mail an article Allows users to e-mail links to relevant and interesting articles to another computer for later use, reference or printing purposes.

Xtra resources and collections When you register your journal subscription online you will gain access to additional resources for Authors and Librarians, offering key information and support to subscribers. In addition, our dedicated Research, Teaching and Learning Zones provide specialist ‘‘How to guides’’, case studies and interviews and you can also access Emerald Collections, including book reviews, management interviews and key readings. E-mail alert services These services allow you to be kept up to date with the latest additions to the journal via e-mail, as soon as new material enters the database. Further information about the services available can be found at www.emeraldinsight.com/alerts Emerald Research Connections An online meeting place for the world-wide research community, offering an opportunity for researchers to present their own work and find others to participate in future projects, or simply share ideas. Register yourself or search our database of researchers at www.emeraldinsight.com/connections

Choice of access Electronic access to this journal is available via a number of channels. Our web site www.emeraldinsight.com is the recommended means of electronic access, as it provides fully searchable and value added access to the complete content of the journal. However, you can also access and search the article content of this journal through the following journal delivery services: EBSCOHost Electronic Journals Service ejournals.ebsco.com Informatics J-Gate www.j-gate.informindia.co.in Ingenta www.ingenta.com Minerva Electronic Online Services www.minerva.at OCLC FirstSearch www.oclc.org/firstsearch SilverLinker www.ovid.com SwetsWise www.swetswise.com

Emerald Customer Support For customer support and technical help contact: E-mail [email protected] Web www.emeraldinsight.com/customercharter Tel +44 (0) 1274 785278 Fax +44 (0) 1274 785201

TQM 20,4

284

The TQM Journal Vol. 20 No. 4, 2008 p. 284 # Emerald Group Publishing Limited 1754-2731

EDITORIAL ADVISORY BOARD Dr Ayed Al-Amri Saudi Arabian Quality Council (SAQC), Saudi Arabia Professor Jiju Antony University of Strathclyde, UK Professor Tony Bendell Managing Director of Services Ltd and Visiting Professor at Middlesex University Business School, UK Marcos Bertin President of International Academy for Quality Shirley Y. Coleman Newcastle University, UK Dr Tito Conti PlusGroup, Italy Professor Jens J. Dahlgaard Linko¨ping University, Sweden Professor Su Mi Park Dahlgaard Lund University, Sweden Professor Barrie Dale Manchester Business School, UK Professor John Dalrymple RMIT, Australia Dr John Davies Salford University, UK Professor Kostas Dervitsiotis University of Piraeus, Greece Professor T.N. Goh National University of Singapore, Singapore Dr H. James Harrington James Harrington Institute, USA Dr Philippe Hermel University of Versailles-St Quentin en Yvelines, France Professor Sam Ho Hang Seng School of Commerce, Hong Kong Dr Yasar Jarrar The Executive Office, Government of Dubai, Dubai Dr Stanislav Karapetrovic University of Alberta, Canada Professor Robert Karaszewski University of Nicholas Copernicus, Poland

Professor Bengt Klefsjo Lulea˚ University of Technology, Sweden Dr Graeme Knowles Warwick Manufacturing Group, University of Warwick, USA Professor Yoshio Kondo Kyoto University, Japan Professor Rodney McAdam Ulster Business School, Northern Ireland Professor Christian N. Madu Pace University, USA Dr Robin Mann Director of Centre for Organisational Excellence Research (COER), New Zealand Dr Eitan Naveh Israel Institute of Technology, Israel Professor John Oakland Oakland Consulting plc, UK Dr Fernando F. Padro Monmouth University, USA Dr Hubert K. Rampersad President, TPS International Inc., Florida, USA Professor Zien-Yusuf Rushami Universiti Utara Malaysia, Malaysia Dr Bishnu Sharma University of the Sunshine Coast, Australia Professor Amrik Sohal Monash University, Australia Dr Charles Tennant University of Warwick, UK Dr Manu Vora Business Excellence Inc., ASQ, USA Dr Di Waddell Deakin University, Australia Professor Adrian Wilkinson Griffith Business School, Australia Dr Shirley M.C. Yeung Asia International Open University, Macau, China Professor Klaus Zink University of Kaiserslautern, Germany

Guest editorial The First European Conference on Kansei Engineering The papers selected for this special issue were all except one presented at the 10th International Conference on Quality Management and Organisational Development (QMOD), which took place at Lund University, Campus Helsingborg, Sweden, 18-20 June, 2007. For further information see www.ch.lu.se/qmod. More than 200 participants from 38 countries presented about 160 high quality papers at the QMOD conference, making it a success and marking a milestone in the QMOD history. As the QMOD conference celebrated its ten years jubilee in 2007 we felt that this was the right time to officially introduce the research field of Kansei engineering to QMOD participants by organising the first European Conference on Kansei Engineering integrated into the QMOD conference and in this way contributing to the spread of knowledge about the Kansei engineering methodology and tools to a wider audience. The integrated conference – the First European Conference on Kansei Engineering – which was run as one of five main parallel session tracks became a great success. In total, 32 papers were presented. What was interesting to observe was that the Kansei sessions were very popular and we had to change the planned session rooms so that the Kansei sessions could take place in one of the biggest auditoriums. Many QMOD participants were curious about Kansei engineering and wanted to know more about the subject. As a consequence of this great interest among the participants The TQM Journal decided to publish a special issue on Kansei/affective engineering design in 2008 (issue 4). This is the background of this special edition which I have worked on during several months, together with the authors of the selected papers. The selection base was the 32 Kansei papers presented – grouped into 14 European papers, 14 Japanese papers and four other papers (Taiwan, Australia, Malaysia, and Mexico). One general observation from the conference was that delegates from both academia and business emphasized the need for more Kansei research and also for spreading the applications to new areas. Another observation was that Kansei engineering is in fact widening its application areas from traditional product design to service design areas. Applying Kansei engineering within these new application areas requires strong collaboration from different research disciplines ranging from social to engineering sciences. Kansei engineering toolkits to evaluate and analyze Kansei data are under development by several research groups in Europe, as well as Japan/Asia. Such toolkits were regarded as having the power to attract more interest from companies to implement Kansei engineering in the near future. The selected papers It is natural in this pioneering special issue of The TQM Journal first to present the paper “Perspectives and the new trend of Kansei/affective engineering” written by the pioneer and founder of Kansei engineering, Mitsuo Nagamachi, Japan. Nagamachi declares that Kansei engineering (KE) aims to develop products that people want to have deeply in their mind. The term kansei is a Japanese word and

Guest editorial

285

The TQM Journal Vol. 20 No. 4, 2008 pp. 285-289 q Emerald Group Publishing Limited 1754-2731

TQM 20,4

286

implies psychological feeling and needs in mind. Before purchasing a passenger car, for example, one may have images in mind of “a powerful engine”, “easy operation”, “beautiful and premium exterior”, “cool and relaxed interior” and so on. These words express the kansei, and the consumers really want to have such a vehicle if the manufacturer succeeds in realizing a vehicle fitting to their imaginations. Nagamachi claims however that we have not ever had such a science and technology which can treat psychological feelings and needs (kansei) technologically, but Kansei engineering is able to grasp the consumers’ kansei on a psychological basis, to analyze the kansei using statistical methods, and to transfer the analyzed data into the design domain. In his article Nagamachi presents and discusses the different KE methods and shows various products which have been developed by using Kansei engineering. The second paper in the selection, “Kansei/affective engineering design – a methodology for profound affection and attractive quality creation”, is written by Professor Jens J. Dahlgaard et al. (members from the KE Research group at Linko¨ping University, Sweden). The authors present a model of the KE methodology and illustrate how this model was applied on a simple example which all may understand – design of a new chocolate bar. At the end of the article the authors present and discuss the future of KE. The authors have observed that people today care more and more about whether products and services match and appeal to their feelings, emotions, personal life styles, identities, and even moral/ethical preferences. The most attractive products/services of tomorrow will in the authors’ view be designed to satisfy all dimensions of human needs – manifest as well as latent needs. To be successful companies have to attain a profound understanding of the complexity of different human needs and the power of satisfying these needs. Hence, the research foundation of affective/Kansei engineering should in the authors’ view aim at understanding and balancing the satisfaction of the “Trinity of Human Needs” (Dahlgaard-Park, 2003): (1) physical or biological needs; (2) mental/psychological needs (embracing emotional, intellectual, social and aesthetic needs); and (3) spiritual or ethical needs. By understanding the “Trinity of Human Needs” and by making efforts to build these complex human needs into new products and services Kansei/affective engineering researchers will definitely be able to contribute with quite new research applications which may increase people’s quality of life. By working more explicitly with the “Trinity of Human Needs”, affective/Kansei engineering may also be able to give clear input to understanding and developing a company’s image and product branding which match better to the increased awareness and demands of sustainability. This should in the authors’ view be one of the core applications of “New Kansei/affective engineering”. The authors then suggest a structural model as a possible expanded framework for future Kansei/affective engineering research studies. According to the model profound affection is a result of the following six enabler factors: (1) Sensing experience; (2) emotional experiences (kansei);

(3) (4) (5) (6)

behavioural experiences/action; social experiences/interactions & relations; spiritual experiences/moral, ethics; and intellectual experiences/cognition.

The authors define “Profound Affection” as a very comprehensive state, which is a result of a combination of sensing, intellectual/cognitive, emotional, social, behavioural and spiritual experiences. “Profound Affection” is not only a result of sensing or emotional experiences. Because the Japanese terminology Kansei gives associations of sensing and emotional aspects only and does not embrace other essential aspects such as spiritual, intellectual, social aspect the authors suggest adopting the terminology of affective engineering design, instead of Kansei engineering, when the research aims at understanding the broader scope of “Profound Affection”. The next article, “Customer experience management – influencing on human kansei to management of technology”, is written by Professor Shin’ya Nagasawa, Waseda Business School, Japan. In his article, Nagasawa presents and discusses the concept of customer experience, which has been effectively used as a concrete theory to organically combine people’s kansei or feeling and psychology into the making of products. Furthermore, the concept of customer experience also helps to understand hit products and brand successes. Nagasawa exemplifies customer experience by discussing the following five modules of strategic experience values: (1) sensory experience value; (2) emotional experience value; (3) intellectual experience value; (4) behavioral experience value; and (5) relative experience value. Nagasawa then illustrates these five modules of experience values in his analysis of four different hit products case stories. By using the five modules of experience values to explain hit products there is a strong overlap to the structural model for creating a profound affection presented by Dahlgaard et al. in the previous article. This may not be a coincidence because hit products are in some way the result of a company having been successful in the design of a new product which has resulted in profound affection. The fourth paper selected, “Kansei engineering approach for total quality design and continuous innovation”, is written by Professor Antonio Lanzotti and Pietro Tarantino from University of Naples, Italy. The paper aims at defining a structured process of continuous innovation in the product concept development phase by using a statistical-based Kansei engineering (KE) approach, which consists of the identification of quality elements satisfying both functional and emotional user needs, i.e. the total quality elements. As the approach developed integrates both the Kano methodology for attractive quality creation and Kansei engineering analyses this paper complements

Guest editorial

287

TQM 20,4

288

the previous papers. The proposed approach is exploited through a case study on train interior design developed in a virtual reality (VR) laboratory. The fifth paper selected, “An e-commerce site for gift flower arrangements that fit kansei and social manners”, is written by Professor Keiko Ishihara et al., coming from Hiroshima International University, Japan. With this paper we are “back to Japan” again, and what could be more typical of Japan than flower arrangements? Maybe some of the readers will be inspired to develop a new business inspired by the contents of this article? The authors describe an e-commerce site as a solution for proposing gift flower arrangements suiting the purchaser’s needs according to the integration of results of kansei evaluation and expertise of florists. The proposed system deals with flower arrangements using foams in containers. A purchaser inputs data such as the purpose of the present, relationship with the recipient, and budget. The system receives the data and then retrieves suitable flowers from the database according to the results of a kansei evaluation and social constraints. Then the system displays a list of possible flowers and types of arrangements. The remaining four papers are selected from outside Japan (Canada, New Zealand, the UK and Sweden) and they illustrate well how the Kansei/affective engineering has spread to all parts of the world. The sixth paper in this special edition, “Effect of SmartPhone aesthetic design on users’ emotional reaction: an empirical study”, is written by Parul Nanda et al., from an experience analysts research team, based in Ontario, Canada. The paper investigates emotional reaction of males to varying aesthetic design of the BlackBerry Pearl, a leading wireless SmartPhone solution (BlackBerry Pearl, 2007). Further, the paper empirically evaluates male preferences for the BlackBerry Pearl in different colours and overlay patterns. The term emotion is operationalised to refer to users’ preferences based on instinct rather than intellect. The seventh paper, “Multi-modal visual experience of brand specific automobile design”, is written by Anders Warell, who is Professor at Massey University, Wellington, New Zealand. The paper presents a questionnaire study of brand-specific perceptions of automotive design using subjective rating methods. The purpose of the study was to explore the multiple modalities of the visual product experience of automobile design as perceived by the general public. Furthermore, the experiences were analysed using a framework for Visual Product Experience. Results from the study indicate that there is a correlation/relation between experiential modes, in that respondents tended to rate attributes consistently high or low across modes. This implies that if the aesthetics are not perceived as favourable, neither is the expression of the car. Furthermore, respondents’ assessments of aesthetic appeal and expression are on an average strikingly similar, suggesting that the level of aesthetic appeal correlates with the level of semantic understanding of the design. The general rating of emotional response follows a similar consistent pattern for the two studied cars. The eighth paper selected, “Kansei engineering toolkit for the packaging industry”, is written by Dr Cathy Barnes et al., from the Affective Engineering Research Group, University of Leeds, UK. The paper presents a Kansei Engineering Toolkit for packaging design by using illustrative case studies. The authors present the application of the toolkit to “live” projects to show how each tool supports design development. In this case, relationships are constrained to product appearance alone. It

is found that the suggested Kansei Engineering Toolkit has real value within the packaging development process to inform concept selection decisions based upon actual consumer data. The ninth and the last paper selected, “Affective design of values in primary healthcare”, is written by Ebru Ayas et al. (PhD student from the Kansei Engineering Research at Linko¨pings University, Sweden). The authors have found that whilst considerable research has been devoted to apply affective (kansei) engineering for product design, there are no studies found for servicescape (physical surroundings) design in healthcare services. So the overall aim of the study is to find possible affective design solutions in a servicescape by utilizing Kansei engineering. Another aim with this study is to propose a framework methodology for exploring kansei values and perceived qualities towards waiting areas in health services. The study applied is a qualitative approach for data collection in the Kansei engineering methodology. A data mining technique is used to extract design solutions for a specific feeling. Calm, pleasant and fresh feelings are determined towards creating values for patients for primary healthcare waiting rooms. “Calm” is found as the most desired feeling for creating values that would appeal to human kansei. The core design attributes explored for this feeling are privacy, colours, child play areas and green plants. Good design of lighting, seating arrangements and low sound level are also needed design elements to give a complete design view. Considering the technical qualities giving feelings of safety, functionality and privacy appear important for design of values. Based on interaction qualities, welcoming environment, with caring staff and giving attention to patients are needed. The results for perceived affective qualities from selected waiting areas showed significant differences. With the selection of these nine papers it is my hope that the readers of The TQM Journal will be inspired for further studies of the area of Kansei/affective engineering, which I strongly believe has a lot to contribute to the further development of quality and TQM. As written above, people today care more and more about whether products and services match and appeal to their feelings, emotions, personal life styles, identities, and even moral/ethical preferences. The most attractive products/services of today and tomorrow will be designed to satisfy all dimensions of human needs – manifest as well as latent needs. Hence, quality and quality management must be further developed also to embrace the new dimensions covered within kansei/affective engineering. Jens J. Dahlgaard Division of Quality Technology, Linko¨ping University, Sweden

Guest editorial

289

The current issue and full text archive of this journal is available at www.emeraldinsight.com/1754-2731.htm

TQM 20,4

Perspectives and the new trend of Kansei /affective engineering

290

Hiroshima International University, Hiroshima, Japan

Mitsuo Nagamachi Abstract Purpose – The purpose of this paper is to present and discuss Kansei engineering (KE), which uses a unique ergonomic technology to produce a new product which fits to consumers’ feelings and demands. It is a consumer-oriented product development method based on the consumer mind. It has been applied to realize several new Kansei products so far. Design/methodology/approach – KE is sometimes able to create an invention, but mostly it is powerful to create more comfortable and affective products or services to the customers. KE utilizes psychological methods to grasp the customer’s feelings, and the data obtained by this method are analyzed using multivariate statistical analyses which are transferred to the design domain (design specifications). Findings – It was found that a customer has a hierarchy of values of his/her life. All people, from children to the elderly, want to enhance their quality of life. Having qualified products and services, including service men’s smiles and greetings is an important factor. Originality/value – KE is spreading out in the world at present. The paper (witten by the pioneer and founder of KE) tells “a story” about the methods and procedures to create a new Kansei product and refers to the implications of Kansei/affective engineering. Keywords Customer satisfaction, Consumer behaviour, Product design, Product development, Japan Paper type Conceptual paper

The TQM Journal Vol. 20 No. 4, 2008 pp. 290-298 q Emerald Group Publishing Limited 1754-2731 DOI 10.1108/17542730810881285

Introduction There are two ways of a product development, one is called “product-out” philosophy which implies the manufacturer provides technology and design specifications according to decision making from the manufacturer’s side. Another way is “market-in” philosophy which means consumer-orientation for product development. By this philosophy a manufacturer considers consumer needs and wants and these are transferred to the product function and design as product properties. Nowadays, the consumers desire consumer-oriented products, because they have a lot of goods at home and they want to have goods which are more needed, more attractive, and very sensitive to their personality and their feelings. Kansei engineering aims to develop products that people want to have deeply in their mind. The term kansei is a Japanese word and implies psychological feeling and needs in mind. Before purchase of for example a passenger car one has images in mind of may be “a powerful engine”, “easy operation”, “beautiful and premium exterior, “cool and relaxed interior” and so on. These words express the kansei, and the consumers really want to have such kind of a vehicle if the manufacturer succeeds in realizing a vehicle fitting to their imaginations. However, it is a regret that we have not ever had such a science and technology which can treat psychological feelings and needs (kansei) technologically. Kansei engineering is able to grasp the consumers’ kansei on a psychological basis, to analyze the kansei using statistical methods, and to transfer the analyzed data into

the design domain. Nagamachi introduced Kansei engineering around 1970 at Hiroshima University as a customer-oriented product development method in order to realize products’ best fit to customer needs. He has since been engaged in the development of Kansei engineering over the last 35 years and contributed to the development of several kansei products and Kansei engineering methods (Nagamachi, 1974, 1989, 1995, 1996, 1998, 2002, 2005).

Perspectives of Kansei/affective engineering 291

Methods of Kansei engineering Kansei engineering Type I Kansei engineering Type I (see Figure 1) starts from a company’s decision of product strategy on the design domain as well as on the target (customer type). Then in step 2 kansei words are collected which are related to the product domain. Usually, 30-40 kansei words are collected, adjectives or sentences of feelings related to the product domain, and after that a five-point or seven-point/nine-point SD (semantic differentials) scale is constructed (Osgood et al., 1957). In step 4 product samples are collected, and after that in step 5 items/categories of each sample are identified. Item means a category like size, width, colour, style, function etc., and category implies more detailed features like red, yellow, green, blue for the colour item. The subjects then evaluate in step 6 each product sample on the five-point (or seven-point/nine-point) SD scale sheet and the evaluated data are then analyzed in step 7 using multivariate statistical methods like Principal Component Analysis, Factor Analysis, Regression Analysis, Cluster Analysis, Conjoint Analysis, Quantification Theory Type I, II, III and IV, etc. Most of psychological phenomena are expressed in a qualitative style and Chikio Hayashi devised four non-parametric statistical methods called “Quantification

Figure 1. A flow of the Kansei engineering Type I

TQM 20,4

292

Theory” (Komazawa and Hayashi, 1976), which are feasible to treat qualitative data in a multivariate analysis style. He invented QT-I compatible to Multi-regression Analysis, QT-II compatible to Discriminant Analysis, QT-III compatible to Factor Analysis and QT-IV compatible to Cluster Analysis. Qualitative data are analyzed very easily using Hayashi’s Quantification Theory. Among these analytical methods Quantification Theory Type I is an excellent technique feasible to find relational design rules between the kansei and the design specifications (Komazawa and Hayashi, 1976). Rough Sets Theory Recently, it became clear that Rough Sets Theory may be better to find decision rules fit to the designers’ kansei thinking (Nishino, 2005; Nishino et al., 2005; Nishino et al., 2006a). Rough Sets Theory (founded by Z. Pawlak) is able to deal with ambiguous and uncertain data like the kansei data. Psychological feelings like kansei have in general nonlinear characteristics and since many of the methods used for the statistical analysis are based on the normal distribution, an application of such methods on kansei data may have problems. Rough Sets Theory can deal with both linear- and nonlinear data as well as data from non-normal distributions. Category classification method Another kansei method is the category classification method. By using this method we first set the strategic kansei concept, namely a top concept for the new product to be developed, and then we break down this concept to more concrete and detailed sub concepts. The classification process is continued to more concrete sub concepts until the nth stage. The classification process illustrates a tree structure flowed from the top concept to the nth sub concepts. The finalised diagram shows a whole concept group with a hierarchical structure of the top concept to the bottom sub concepts. The bottom sub concepts will be transferred to ergonomic experimentation in order to decide on the product design specifications. For instance, Mazda settled “Human-Machine Unification” as the top concept, which means a driver’s images of a vehicle in operation as if it was his own body. The second sub concept of this was “self-controlling” and the third sub concept was “feel freely when controlling the gear shift”. An ergonomic experiment was conducted to decide the length of the gear shift, and the data calculated showed that 9.5 cm gave the best feeling of “self-controlling”. This method is very useful and was named “Kansei Category Classification Method”. Applications of Kansei engineering Nagamachi has contributed to the development of more than 20 kansei products and in this section we see some of his product examples with explanations of the development process. Japanese-style refrigerator Japanese used to have one door or two door type of refrigerators. Sharp asked Nagamachi to introduce Kansei engineering in the product development division, and the project team tried to apply it to the design of a new refrigerator. The team visited

monitor houses to take video of wives’ cooking behaviours. Curious pictures of the wives bending their backs when taking out vegetables from the bottom room of the two-door refrigerator were found. Seen from the viewpoint of Ergonomics this bending posture needed about three times as much efforts as in a standing posture. The wives felt that it was hard work (this is a kind of kansei). Hence, the job was to reduce the high working load by changing the design of the refrigerator. By Nagamachi’s suggestion the team proposed Sharp change the vegetable position from the bottom to the top area of the refrigerator (as shown in the right of Figure 2) to realize a more comfortable work. Sharp was successful in redesigning a new refrigerator style, and after that all refrigerators in Japan changed to the new style (freezer to the bottom and vegetables to the top).


Digital cameras Kansei engineering is very useful in the creation of an innovative new attribute/feature to be built into mature products and in this way create new unique products. Sharp has for example been successful in inventing a unique camcorder, called Crystal Liquid View cam, which has been extended to digital cameras. This new attribute/feature is now used worldwide by every body. Shampoo and treatment Milbon, which is a maker of cosmetic goods, invited Nagamachi to introduce Kansei engineering to the company. Following Nagamachi’s suggestion, the company organized a concurrent engineering team. All managers of all divisions joined to the project team from the beginning. The team first learned what Kansei engineering is, and after that the team started a kansei survey by visiting hair salons. They

Figure 2. The ordinary refrigerator (left, freezer on the top) and the Kansei refrigerator (right, freezer on the bottom)

TQM 20,4

294

interviewed 200 ladies on hair care as well as hair problems. These data were analyzed by Quantification Theory Type III to get segmentation of hair kansei and ladies’ hair problems. Zero kansei concepts were decided from the analyzed data as “Soft Touch and Rashly Hair”. The zero-level concept was broken down to 4th stage following a “Category Classification Procedure”, and in the 4th stage sub concepts were transferred to the R&D Department to create about 600 different test materials. By applying the test material on a hair mannequin with human hair the test materials were evaluated concerning kansei fitness to the zero concept, and as a result the test material were reduced in number to just 20. After that the remaining test material was applied on real monitors’ hair and finally just one pair of shampoo and treatment was selected. After that the team continued the development of a container and a perfume. Concerning the container design we followed Kansei engineering Type I. After collecting the kansei words and 62 different containers from different container makers, the team conducted evaluation experiments using a five-point SD scale sheet. After calculation, by using Quantification Theory Type I, the team obtained good results on the relational rules between the kansei and a container’s physical treats (shapes, colours, etc). Kansei engineering system for home and kitchen design Kansei engineering system means a computerized-assisting system to support a designer work or a selection of goods by a customer. The system has several databases, an inference engine and other databases. The customer inputs his/her kansei into the system, and the intelligent system then calculates to find a design candidate. We constructed such systems as FAIMS, HULIS, ViVA etc. HULIS is a kansei home design system. A customer inputs his/her images with words, and the system then analyses the data bases and displays a home design which fits to the customer’s home image. This system is the fundamental system of the ViVA virtual kansei system. Because building a real kitchen is expensive we used a combination of virtual reality technology and Kansei engineering system, which is called Virtual Kansei Engineering (ViVA). Based on 10,000 wives layout data, we constructed a kitchen Kansei engineering system with a virtual space engineering part. A wife sitting in front of a computer inputs her family data, life style data and finally her kansei (her image/associations) related to the new kitchen. Then the kansei kitchen system calculates and shows her the 3-dimensional kitchen design on the screen. All people became surprised at this stage, and if no need of change in design, she is now able to continue into the computerized space to have a virtual experience of the design. The ViVA system became very popular in Japan, and several people came to see it even from overseas. Because the customers visited Matsushita Electric Works’ showroom, the sales cost was almost zero, and the company got a good profit from the Kansei engineering system (ViVA). Figure 3 is a very simple example made by the kansei virtual computer system.


Figure 3. A simple scene displayed on the screen by the Kansei virtual system

Kansei ergonomics Usually Kansei engineering is able to support the design of a kansei product, and it became successful almost every time. However, Kansei engineering sometimes needs ergonomic sense and technology for creating more comfortable products, especially for making an assisting device for elderly people. Matsushita Electric Works, Ltd and Nagamachi jointly approached to apply the kansei engineering to the production of a new toilet based on the philosophy of a universal design. In this project, we considered three points: (1) As wives usually are very serious about environmental problems, it was decided to reduce water for washing by 80 percent, so that the new toilet only uses 20 percent of water compared with an ordinary toilet. We realized that by devising a new facility which is able to clear up by small amount of water without a water tank. (2) Japan is now the highest aging society in the world and all manufacturers must consider elderly people’s needs. For the elderly to stand up easily, we incorporated arm rests on both sides of the toilet, and the new toilet was tilted three degrees forward to assist elderly people’s standing-up behaviours. (3) Eight different toilets, collected from different makers, were researched from the viewpoint of Kansei engineering. From the data of people’s feelings of sitting comfort of these toilets, Kansei engineering showed the best comfortable three-dimensional curve of a toilet surface. Based on these data we modelled the three-dimensional toilet seat illustrated in Figure 4 (right side).

TQM 20,4

296

Figure 4. The Kansei toilet, TRES (left shows exterior and right horizontal view)

You can see the artistic module of the toilet surface shape which provides affection of comfortable sitting. It was examined whether the three-dimensional surface suggested by Kansei engineering will be the best from the viewpoint of Ergonomics. A body pressure measurement FSA sheet was put on the kansei toilet seat and the subjects sat down on the pressure sheet. The measurements showed that the kansei toilet received high pressure on a limited area, namely only on the middle area of both wings, and then all subjects felt more comfortable when sitting. From the research we found the new ergonomic principle that “the distributed and flattered body pressure” brings the comfortable feeling, not small areas like on just the bone (Nagamachi, 2007). The research of the new toilet, named TRES, on EMG (muscular strength measurement of lower and upper legs) showed that EMG for the elderly in standing up decreased by 90 percent compared with a toilet without the two arm rests and with a flat surface. This is the new research story on Kansei ergonomics. Kansei engineering is very powerful product development, but if it is supported by the ergonomic philosophy, the research outcome will often be more valuable products. Concluding remarks In 1970 Kansei engineering was born at Hiroshima University as a new research branch of ergonomics, and Nagamachi and his colleagues have been engaged in developing the new technology for 35 years. The society has changed over the world. People want to have goods fitting to their needs and emotions (kansei). People seek more and more affective and emotional products nowadays. Kansei engineering has been extended towards more technological and intelligent aspects, but it always stands on a human ware philosophy. It has become wider and deeper in an academic sense, that is, from Kansei engineering to Kansei ergonomics and to Kansei rough set engineering (Nagamachi, 2006; Nagamachi et al., 2007, Nishino, 2005; Nishino et al., 2006a, b, c; Hirata Okamoto, 2007). We always collaborate with the designer group in the final design stage because the designers are invaluably key persons in creating excellent products which the

customers are satisfied with and can enjoy. Kansei engineering/Kansei ergonomics just provides the sensible and sensitive data analyzed by a technology based on the human kansei. The collaboration of the kansei engineer with the designer is needed in the final stage for the success of kansei product development. Finally, a customer has a hierarchy of values of his/her life and they want to have an enjoyable and satisfied life. All people, from children to elderly people, or any person want to enhance their quality of life. They expect to be provided by qualified products and services, including service men’s smiles and greetings. The latter is a quality attribute in the service industry. Companies should provide excellent product quality (product and service quality) which fit customers’ kansei values. Manufacturers should deeply consider producing “Product Quality” which fits to “the Customers’ Kansei Values”. References Hirata Okamoto, R. (2007), “Comparison between statistical and lower/upper approximations rough sets models for beer can design and prototype evaluation”, paper presented at the 10th International Conference of QMOD, Helsingborg, June 18-20, 2007. Komazawa, T. and Hayashi, C. (1976), A Statistical Method for Quantification of Categorical Data on its Applications to Medical Science, North-Holland, New York, NY. Nagamachi, M. (1974), “A study of emotional technology”, Japanese Journal of Ergonomics, Vol. 10 No. 2, pp. 121-30. Nagamachi, M. (1989), Kansei Engineering, Kaibundou, Tokyo. Nagamachi, M. (1995), Introduction to Kansei Engineering, Japan Standard Association, Tokyo. Nagamachi, M. (1996), “Kansei engineering and its applications”, Japanese Journal of Ergonomics, Vol. 32 No. 6, pp. 286-9. Nagamachi, M. (1998), “Kansei engineering: a new consumer-oriented technology for product development”, in Karwowski, W. and Morris, W.S. (Eds), The Occupational Ergonomics Handbook, Wiley, New York, NY, pp. 1835-48. Nagamachi, M. (2002), “Kansei engineering in consumer product design”, Ergonomics in Design, Vol. 10 No. 2, pp. 5-10. Nagamachi, M. (2005), Product Development and Kansei, Kaibundou, Tokyo. Nagamachi, M. (2006), “Kansei engineering and Rough Sets Model”, Rough Sets and Current Trends in Computing: Proceedings of the 5th International Conference, Kobe, Japan, November 6-8, 4259, Springer, New York, NY, pp. 27-37. Nagamachi, M. (2007), “From Kansei engineering to Kansei ergonomics”, Ergonomia, Vol. 4, No. 2. Nishino, T. (2005), “Rough sets and Kansei”, in Nagamachi, M. (Ed.), Product Development and Kansei, Kaibundo, Tokyo. Nishino, T., Nagamachi, M. and Tanaka, H. (2005), “Variable precision Bayesian rough sets model and its application to human evaluation data”, Proceedings of RSFDGrC, LNAI 3641, Springer, New York, NY, pp. 294-303. Nishino, T., Nagamachi, M. and Tanaka, H. (2006a), “Variable precision Bayesian rough sets model and its application to Kansei engineering”, Transactions on Rough Sets V, Springer, New York, NY, pp. 190-206.


TQM 20,4

298

Nishino, T., Nagamachi, M. and Sakawa, M. (2006b), “Acquisition of Kansei decision rules of coffee flavor using rough set method”, Kansei Engineering International, Vol. 5 No. 4, pp. 41-50. Nishino, T., Sakawa, M., Nagamachi, M., Kato, K. and Tanaka, H. (2006c), “A comparative study on approximations of decision class and rule acquisition by rough sets model: an application to the design of children’s shoes”, Kansei Engineering International, Vol. 5 No. 4, pp. 51-60. Osgood, C.E., Suci, G.J. and Tannenbaum, P.H. (1957), The Measurement of Meaning, University of Illinois Press, Champaign, IL. Corresponding author Mitsuo Nagamachi can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints


Kansei /affective engineering design A methodology for profound affection and attractive quality creation

Kansei/affective engineering design 299

Jens J. Dahlgaard, Simon Schu¨tte and Ebru Ayas Department of Management and Engineering, Linko¨ping University, Linko¨ping, Sweden, and

Su Mi Dahlgaard-Park Institute of Service Management, Lund University, Helsingborg, Sweden Abstract Purpose – The purpose of the paper is to present and discuss the Kansei engineering (KE) methodology, and to reflect on the future development of KE. The paper presents a model of the KE methodology and illustrates how this model was applied on a simple example which all may understand – design of a new chocolate bar. Design/methodology/approach – The research methodology is a combination of desk research (literature analysis), data collection, data analysis, reflections and model building. Findings – The paper suggests a structural model as a possible expanded framework for future Kansei/affective engineering research studies. According to the model profound affection is a result of the following six enabler factors: sensing experience; emotional experiences (Kansei ); behavioural experiences/action; social experiences/interactions and relations; spiritual experiences/moral, ethics; intellectual experiences/cognition. Originality/value – The paper defines “Profound affection” as a very comprehensive state, which is a result of a combination of sensing, intellectual/cognitive, emotional, social, behavioural and spiritual experiences. “Profound affection” is not only a result of sensing or emotional experiences. Keywords Product design, Product attributes, Marketing, Chocolate Paper type Research paper

1. Introduction Improving products requires knowledge about how product attributes (properties) affect the consumers. The ambition of any product or service provider is to design products in a way that will evoke a positive impact and make the consumer buy the same product again – in other words creating desire. Creating desire requires profound affection, which is a very comprehensive state resulting from a combination of sensing, intellectual/cognitive, emotional, social, behavioural and spiritual experiences. Kansei engineering, which is a relatively new research field, contains a promising methodology for designing and creating new attractive products and services with a profound affection on the users. The Kansei engineering methodology, which has its roots in Japan from beginning of the 1970s, aims to design and develop products/services that match customers’ emotional, psychological feelings and needs. It is the aim of this paper to introduce the background of Kansei engineering, its methodology, potentiality and


TQM 20,4

300

limitations for designing and building new products with a profound affection, and to suggest a new framework in the form of a structural model, which can be used systematically in understanding the potential enablers of profound affection which may be used for building new innovative and attractive products ( ¼ Attractive Quality Creation). Sections 2 and 3 will introduce the background of Kansei engineering, and its methodology, with a simple example. In section 4, which we have titled “from Kansei/engineering to affective engineering design and attractive quality creation”, we will reflect on the future, and the new suggested framework for understanding the potential enablers of profound affection is presented.

2. From sensitivity to Kansei Emotions and feelings and are often a result of external sensory stimuli provided to the human mind via the human senses. Together with the human mind, this purely objective sensory information is converted to basic emotional reactions (Damaiso, 1996). Outgoing from the evoked emotion a number of appropriate physical reactions may be triggered (Cornelius, 1996). Repeatedly triggered similar emotional responses can also lead to a certain state of mind which can be referred to as mood (Picard, 1997). Emotion and feeling have traditionally found their counterpart in reason. When Baumgarten wrote his Aesthetica in 1750-1758 (Baumgarten, 1961) one of his aims was to create an opposite pole to the field of logic as ratio described in Aristoteles’ Organon (Schweizer, 1973). Also, Kant realized the shortcomings of pure reason and the necessity of creating a counterpart (Kant, 2004). In the field of physiology Damasio (1996) could prove that emotion and reason in fact are hard wired in the human brain. In artificial intelligence Picard (1997) built interaction models describing the relationship between reason and emotion. In another area marketing experts distinguish between feeling appeal, i.e. commercial applying to emotional, subjective impressions of product features and thinking appeal, i.e. commercial applying to logical, objectively verifiable product features (Liu and Stout, 1987). The concept/terminology of Kansei, which is one of numerous Chinese “loan words” assimilated into the daily Japanese as well as the Korean language, can be translated as emotionality, sensuality or sensitivity/sensibility. The terminology is widely used in daily life not only in Japan but also in China and Korea, often as a pairing terminology to intellectuality (Chisei). The position of the word, Kansei (emotionality) in East Asian culture is somewhat remarkable compared to the Western tradition. Generally in the Western tradition rational, intellectual and logical aspects (Chisei) are prioritised and considered to be most important while emotional aspects are either ignored or considered to be something that should be restrained. In East Asian tradition, emotional aspects (Kansei) have been recognised at least equally fundamental and important in all forms of human life, if not considered to be more important than the intellectual aspect. In the context of product development Kansei can be referred to as: [. . .] the impression somebody gets from a certain artefact, environment or situation using all senses of vision, hearing, feeling, smell and taste as well as cognition (Nagamachi, 1989).

3. Kansei/affective engineering Research in Japan during the 1970s on integration of affective values into products was mainly referred to as emotional engineering (Nagamachi (1989). In the 1980s the method spread rapidly in the Japanese car industry and in 1986, Yamamoto, the president of Mazda, coined the technique “Kansei engineering” when he gave lectures about “car culture” at Michigan University. At this lecture he suggested to apply Kansei engineering in car design (Lee, 1992, p. 377). Under this term the methodology spread to Western industry. In Europe many companies had their own methods for evaluation of the affective impact of their products. The academic approach started in different disciplines in the 1990s. Industrial design, mechanical engineering, psychology and ergonomics are some of these disciplines, and many different names evolved such as emotional engineering, emotional design, design for pleasure, Affective engineering, Kansei engineering, etc. Kansei engineering is in Europe often considered as a methodology within the research field of affective engineering. In Japan, on the other hand, Kansei engineering is more or less regarded as a product design philosophy. Some researchers have tried to define the content of the Kansei engineering methodology. Shimizu and Sadoyama (2004) state that:


Kansei engineering is used as a tool for product development and the basic contents are described shortly as follows: Identification of product properties and correlations between those properties and the design characteristics (Shimizu and Sadoyama, 2004).

Figure 1 presents how Kansei engineering works in principle. The psychological feelings (including emotions, moods, impressions, etc.) are collected using suitable methods, e.g. Semantic Differential Method, which is one of the core methods in Kansei engineering. These data are used to build up a Kansei knowledge base (a Kansei engineering system) using statistical methods. This database establishes the links between the feelings and product properties. In this way it is possible to understand how the choice of certain product attributes (properties) may affect the emotional perception of the whole product. The following example may help in understanding the Kansei engineering methodology. A company producing chocolate products provides different chocolate bars for a regional market. They have noticed that the market share of the different products is

Figure 1. Kansei engineering system (KES)

TQM 20,4

302

Figure 2. The working principle of Kansei engineering

different without really knowing why. To understand why they conducted a customer survey using affective engineering methods and found out that few product properties are driving certain impressions. For example eating chocolate bars gives a bad conscious for many people and they feel before and after eating that they should not have done it. However, if chocolate bars give a light and sporty feeling they may not feel as guilty as eating a bar which gives a heavy and sweet feeling. In a next step they surveyed the market for products giving a light and sporty feeling. Comparing these products with each other revealed that light brown chocolate with air bubbles and wafer baking inside will give the desired impression. With this information a new product can be created aiming towards this Kansei ( ¼ light and sporty feeling). The example shows how Kansei engineering may effectively be used for new product creation. Kansei engineering has proven to work fine in such cases but what is happening inside the customer’s mind is often difficult to say. Different functional models have been developed to describe the psychological processes behind the methodology. The most common type of model focuses on practical process mapping of grasping and translating the Kansei into product properties. Figure 2 is an extension of a model developed by Lee et al. (2002). Outgoing from the model Schu¨tte (2005) added an explanation about what in principle happens inside the human brain, and what the outcome is from a psychological point-of-view. A sensory input from one or several of the senses of hearing, sight, smell, taste and touch leads to the building of Kansei. If necessary also other more complex senses can be used such as a sense of balance. At the same time a Chisei (intellectuality, rationality), a type of logical cognitive awareness is built up by the same input, building knowledge through learning processes. In Kansei engineering this process is utilized to study participants interacting with product samples getting emotional experiences. This experience (Kansei) is then expressed verbally or through behaviours. By collecting the way the participants react, researchers can draw conclusions about participants’ and hence customers’ experiences (representation of the Kansei). In the case of the chocolate bar, external stimuli are a variety of chocolate bars given to the participants who test several of them by first looking at the bars, then unpacking them and finally eating them. Consequently, taste and smell of the chocolate are primary sensorial data. Other factors might be the chewing into a Kansei and a corresponding Chisei. The data required for affective development of new products are

therefore both Kansei and Chisei words (logical and technical specifications). The participants’ spontaneous reactions are collected before the customer has time to think about it in detail.

Kansei/affective engineering design

3.1 A model on Kansei engineering methodology In Japanese publications different types of Kansei engineering have been identified and applied in different contexts. Schu¨tte (2005) examined all these types of Kansei engineering and developed model covering the contents of Kansei engineering. The model is presented in the Figure 3.

303

3.2 Choice of domain The domain in this context describes the overall idea behind the product type in general. Choosing the domain includes the definition of the intended target group and user type, market-niche and type, and the product group. Choosing and defining the domain is carried out by using existing products as well as new concepts and not-yet-known design solutions, and using this information a domain description is formulated serving as basis for further evaluation. For the chocolate bar manufacturer the main contents of the domain definitions are shown in Table I. It is seen that the company decided to focus on two different domains with different products. It is expected that healthy chocolate snacks may have a market potential in domain 2 where the target group consists of males and females with a university background living in major cities. 3.3 Span the semantic space The expression semantic space was addressed for the first time by Osgood et al. (1957) who posed that every artefact can be described in a vector space defined by semantic

Figure 3. A model on Kansei engineering

TQM 20,4

304

expressions (words). This is done by collecting a large number of words describing the domain. Suitable sources are pertinent literature, commercials, manuals, specification lists, experts, etc. Experience shows that most words and verbal expressions describing products are adjectives. Depending on the context the number of words identified typically varies between 20 and 800. The raw data collected is three-way data, i.e. the data can be stacked in three dimensions. Each data cell contains a number from 1 to 7 (depending on the measurement scale – other scales may be 1 to 9 or 1 to 5), representing a subject’s judgment of a Kansei word related to a particular concept (product). The raw data are then analysed using mathematical methods such as factor analysis and cluster analysis (Ishihara et al., 1998). Factor analysis reduces the number of data by constructing new latent factors explaining the original data set. In this way it is possible to answer questions of how the different words are related to each other, in which way they affect understanding of a meaning (a latent factor), and how they may support future prototype experiments. From the analysis of the semantic space relevant words are selected representing the different dimensions, and the selected words are linked to product properties. In the chocolate bar example the relevant words were identified using sources like customers’ and manufacturers’ language describing the relevant domains, but also verbal expressions used in ads and commercials. Among others the following words were identified as Kansei words: delicious, delightful, juicy, light, luxurious, nutritious, tasty, sporty, and unusual. 3.4 Span the space of properties The next step is to span the space of product properties relevant to the semantic space. the space of product properties is a collection of selected product properties for further evaluation. The collection of product properties representing the domain is done from different sources such as existing products, customer suggestions, possible technical solutions and design concepts etc. To select properties for further evaluation a Pareto-diagram (compare with Dahlgaard et al., 2002) can assist the decision between important and less important properties. In the chocolate bar example an original set of 14 different product items (properties) were identified and presented to target group customers who rated them by assigning a total of ten points between those items (properties) they personally believed to be the most important ones. The data was used to construct a Pareto-diagram, and the four most important items (properties) were chosen for evaluation: size, shape, colour, and brand. The remaining items (properties) were dismissed. Figure 4 demonstrates how the Pareto chart for the chocolate bar looked like.

Table I. Domain definition for the chocolate bar

Attributes

Domain 1

Domain 2

Gender Edu background Region Products

Male High school Rural areas Traditional chocolate snack

Male and female University Major cities Healthy chocolate snack


Figure 4. Pareto chart for chocolate bar development

3.5 Synthesis In the synthesis step the Semantic Space and the Space of Properties are linked together as displayed in Figure 5. Connections between abstract feelings and technical specifications are established and quantified. For every Kansei word a number of product items (properties) are found affecting the Kansei word. Most common tools for linking the Kansei words with product items are: . category identification (Nagamachi, 2001); . regression analysis/quantification theory type I; . rough sets theory (Mori, 2002); . genetic algorithm (Nishino and Nagamachi, 1999); and . fuzzy sets theory (Shimizu and Jindo, 1995). In the chocolate bar example, linkages between Kansei words and the product items were done using multiple linear regression analysis. For each Kansei word correlation and regression coefficients were determined for estimation of the direction and the strength of the affects.

Figure 5. Synthesis phase

TQM 20,4

306

3.6 Model building and test of validity The final step of validation now remains. This final step is done in order to check if the prediction model is reliable and realistic. If the prediction model fails it is necessary to update the Space of Properties and the Semantic Space and consequently refine the model. A method to simplify presentation of the results is to present the most important results of the model in a resulting table. For the chocolate bar example see Table II. PCC, which stands for partial correlation coefficient, indicates the relative importance of the specific item for the rating of delicious. Apparently, shape is the most important item to design for if a high impression of delicious is required. CS, which stands for category score, is equal to the regression coefficients. In this case the edges should be smooth (CS ¼ 0.5). Sharp edges in fact decrease the impression of deliciousness. An optimal delicious chocolate bar should, according to the results in Table II, possess smooth edges, be light brown in colour, be large and carry the brand of the manufacturer. 4. From Kansei/engineering to affective engineering design and attractive quality creation When considering the future, we are quite convinced that Kansei/affective engineering as a philosophy in designing products and services will get increased interest all over the world. From many fronts we receive clear signals that people today care more and more about whether products and services match and appeal to their feelings, emotions, personal life styles, identities, and even moral/ethical preferences. The most attractive products/services of tomorrow will in our view be designed to satisfy all dimensions of human needs – manifest as well as latent needs. To be successful companies have to attain a profound understanding on the complexity of human’s different needs and the power of satisfying these needs. The research foundation of affective/Kansei engineering should in our views aim at understanding and balancing the satisfaction of the “Trinity of Human Needs” (Dahlgaard-Park and Dahlgaard, 2003): . physical or biological needs; . mental/psychological needs (embracing emotional, intellectual, social and aesthetic needs); and . spiritual or ethical needs. By understanding the “Trinity of Human Needs” and by making efforts to build these complex human needs into new products and services Kansei/affective engineering

Table II. Result for the chocolate bar Kansei word “delicious”

Item

PCC

Category

Size

0.4

Colour

0.4

Shape

0.8

Brand

0.1

Large Small Light brown Dark brown Sharp edges Smooth edges Our brand Competitor brand

CS 0.1 2 0.3 0.5 2 0.1 2 0.4 0.5 0.7 0.3

researchers will definitely be able to contribute with quite new research applications which may increase people’s quality of life. By working more explicitly with the “Trinity of Human Needs” affective/Kansei engineering may also be able to give clear input to understanding and developing a company’s image and product branding which match better to the increased awareness and demands of sustainability. This should in our view be one of the core applications of “New Kansei/affective engineering”. Let us illustrate this core with an example. In 1998 two of the authors of this article were involved in a transformation process in Pioneer Denmark (Dahlgaard-Park and Dahlgaard, 2003a, b). The background of the transformation process was that Pioneer Electronics had competition problems on the world markets especially with the other two Japanese competitors – Sony and Matsushita Electronic Corporation (Panasonic etc.). To start a transformation process aiming at changing the image of Pioneer and its product branding, the president of Pioneer gave a New Year speech to top managers from Pioneer companies all over the world. In this New Year speech he announced Pioneer Electronic Corporation’s new corporate identity, which at the same time was announced as the corporation’s Vision 2005. The vision was presented with a tree metaphor picture as shown in Figure 6. The key message from Pioneer’s president was that all efforts from companies and employees all over the world must now be focused on the goal – customer satisfaction is our ultimate goal. But this ultimate goal can only be achieved if employees all over


Figure 6. Pioneer Electronic Corporation’s new corporate identity (Vision 2005)

TQM 20,4

308

Figure 7. A structural model for building “Profound Affection”

the world in their different jobs participate actively in understanding the different dimensions of customer needs and problems. Having understood the variation, interdependence and depth of customer needs then people can begin to design new products which may be able to “move the heart and touch the soul”. People may gradually understand that attaining such a state requires that the customers will have positive experiences related to sensing, cognition, morality, action (self-realization), and social relations. All these experiences will together determine if the product will “move the heart and touch the soul”. Based on our research and observations from the First European Conference on Kansei/affective engineering, our previous discussions and reflections on the Pioneer metaphor above, and by combining research on human needs (especially the suggested Trinity Model for Human Needs,) we suggest the following structural model as a possible expanded framework for future Kansei/affective engineering research studies (Figure 7). As can be seen in the model profound affection is a result of six enabler factors: X1:

Sensing experience/the five senses. Sensing experiences are often in Kansei engineering related with emotions, but seen from a product development viewpoint it makes sense to regard it as an independent factor which also may be related to other factors of the model in Figure 7. Simple questions to ask could be: Is the product nice to see, nice to hear, nice to touch, nice to smell, and nice to taste?

X2:

Emotional experiences (Kansei). Emotional experience is a result of sensing experience and is more comprehensive than each individual sensing experience.

X3:

Behavioural experiences/action. These experiences reflect user behaviours when interacting with products and services. Questions to ask: Is the product user friendly? Comfortable? Do users feel unity with the product or alienation?

X4:

Social experiences/interactions and relations. Products are here regarded as instruments for building social interaction & relations. Question to ask: Is the product strengthening your social relationships and interactions? Product examples: Mobile telephone, MP3, internet, blocs, etc.

X5:

Spiritual experiences/moral, ethics. Customers of today are now concerned more on spiritual experience including moral and ethics of producing, using and scrapping the products. Questions to ask: Is the product/service produced ethically correct? Is the product dangerous or polluting the environment? How are the products contributing to global warming, etc?

X6:

Intellectual experiences/cognition. These experiences are related to traditional quality attributes belonging to the basic/ must-be quality dimension. Questions to ask: What and how are the functions of the product? Are they functioning logically? Is the product reliable, safe, etc.?

As the model shows “Profound Affection”, where customers’ hearts are moved and their souls are touched, is a very comprehensive state, which is a result of a combination of sensing, intellectual/cognitive, emotional, social, behavioural and spiritual experiences. “Profound Affection” is not only a result of sensing or emotional experiences. Because the Japanese terminology Kansei gives associations of sensing and emotional aspects only and does not embrace other essential aspects such as spiritual, intellectual, social aspect etc., as shown in Figure 7, we suggest adopting the terminology of affective engineering design instead of Kansei engineering when the research aims at understanding the broader scope of “Profound Affection”. As said above, an observation from the QMOD conference is that Kansei engineering is widening its application areas from traditional product design to service design and other areas. When doing that it is a necessity to broaden the traditional scope of Kansei engineering to the new scope of “Profound Affection”. We therefore suggest that Kansei engineering researchers in the future should not be too narrow in their research. Think “New Kansei engineering” which we call “affective engineering design”. Go back to basics, understand human needs, and then try to understand the enablers for moving people’s hearts and touching their souls. In this process there is a need for effective and efficient statistical tools for identifying and understanding the following important detailed relationships behind Figure 7. Profound Affection ¼ (1) X1 þ X2 þ X3 þ X4 þ X5 þ X6 (one factor contributions); (2) þ XiXj þ (interactions 2x2); (3) þ XiXjXk þ (interactions 3x3); (4) þ XiXjXkXl þ (interactions 4x4); (5) þ XiXjXkXlXm þ (interactions 5x5); and (6) þ X1X2X3X4X5X6 (interactions 6x6). We will end this article by shortly commenting on the six enabling factors above. By these comments we also indicate that the relevant contributions to building Profound Affection depend on the context. For some products or services all one- factor


TQM 20,4

310

contributions may be important enablers as well contributions from several of the potential interaction effects, while for other products maybe only a limited number of one-factor contributions may be relevant. However, it is our belief, that the structural model in Figure 7 can be used systematically in building new innovative products or services with a Profound Affection, which will attract the customers ( ¼ Attractive Quality Creation). For that purpose we believe that the model could be used as a check in every phase and gate of major product development processes, from ideation to market launch and post implementation review. References Baumgarten, A.G. (1961), Aesthetica, Georg Olms Verlagsbuchhandlung, Hildesheim. Cornelius, R.R. (1996), The Science of Emotion: Research and Tradition in the Psychology of Emotions, Prentice-Hall, Upper Saddle River, NJ. Dahlgaard, J.J., Kristensen, K. and Kanji, G.K. (2002), Fundamentals of TQM, NelsonThornes, Cheltenham. Dahlgaard-Park, S.M. and Dahlgaard, J.J. (2003a), “Toward a holistic understanding of human motivation: core values – the entrance to people’s commitment?”, The International Journal of AI (Artificial Intelligence) and Society, Vol. 17 No. 2, pp. 150-80. Dahlgaard-Park, S.M. and Dahlgaard, J.J. (2003b), “The human dimension: critical to sustainable quality”, in Conti, T., Kondo, Y. and Watson, G. (Eds), IAQ (International Academy of Quality) Quality into the 21st Century – Perspectives on Quality, Competitiveness & Sustained Performance, Vol. 14, ASQ Press, Milwaukee, WI, pp. 73-103. Damaiso, A.R. (1996), Descartes’ Error: Emotion, Reason and the Human Brain, Papermac, London. Ishihara, S. and Ishihara, K. (1998), “Hierarchical Kansei analysis of beer can using neural network”, in Vink, P., Koningsveld, E.A.P. and Dhondt, S. (Eds), Human Factors in Organizational Design and Management – VI, Elsevier, Amsterdam. Kant, I. (2004), Kritik av det rena fo¨rnuftet, Thales, Stockholm. Lee, S.Y. (1992), Human Factors Engineering for the Future, Pakyoungsa, Seoul. Liu, S.S. and Stout, P.A. (1987), “Effects of message modality and appeal on advertising acceptance”, Psychology & Marketing, Vol. 4 No. 3, pp. 167-87. Mori, N. (2002), “Rough set approach to product design solution for the purposed Kansei”, The Science of Design Bulletin of the Japanese Society of Kansei Engineering, Vol. 48 No. 9, pp. 85-94. Nagamachi, M. (1989), Kansei Engineering, Kaibundo Publishing, Tokyo. Nagamachi, M. (2001), Workshop 2 on Kansei Engineering, International Conference on Affective Human Factors Design, Singapore, 2001. Nishino, T. and Nagamachi, M. (1999), “Internet Kansei engineering system with basic Kansei database and genetic algorithm: TQM and human factors”, Centre for Studies of Humans, Technology and Organization, Linko¨ping. Osgood, C.E. and Suci, C.J. (1957), The Measurement of Meaning, University of Illinois Press, Champaign, IL. Picard, R. (1997), Affective Computing, Massachusetts Institute of Technology, Cambridge, MA. Schu¨tte, S. (2005), “Engineering emotional values in product design – Kansei engineering in development”, Institute of Technology, Linko¨ping University, Linko¨ping.

¨ sthetik als Philosophie der sinnlichen Erkenntnis, Schwabe & Co., Verlag, Schweizer, H.R. (1973), A Basel. Shimizu, Y. and Jindo, T. (1995), “A fuzzy logic analysis method for evaluating human sensitivities”, International Journal of Industrial Ergonomics, Vol. 15, pp. 39-47. Shimizu, Y. and Sadoyama, T. (2004), “On-demand production system of apparel on basis of Kansei engineering”, International Journal of Clothing Science and Technology, Vol. 16 Nos 1/2, pp. 32-42. Further reading Alikalfa, E. (2006), “Designing quality feeling in a reach truck – a Kansei engineering approach”, Institute of Technology, Linko¨ping University, Linko¨ping, p. 147. Choi, K. and Jun, C. (2007), “A systematic approach to the Kansei factors of tactile sense regarding the surface roughness”, Applied Ergonomics, Vol. 38 No. 1, pp. 53-63. Hanako, K. (2004), “Information retrieval by taste impression with an integrated metadata extraction mechanisms”, Information Processing Society of Japan (IPSJ), Vol. 71, pp. 145-52. Kanda, T. (2005), “Evaluation of human meal Kansei using AHP”, Networking, Sensing and Control, 2005: Proceedings: 2005 IEEE, March, pp. 431-6. Kano, N. (2001), “Life cycle and creation of attractive quality”, Proceedings of the 4th QMOD Conference, Linko¨pings University, Sweden. Makoto, K. (2001), “Perception of deliciousness and Kansei biosensing technologies, the sense of smell”, Foods & Food Ingredients Journal of Japan, Vol. 193, pp. 47-56. Nagai, H. (2002), “Application of Kansei engineering for new production development for beverages”, Foods and Food Ingredients Journal of Japan, Vol. 202. Nagamachi, M. (1995), “Kansei engineering: a new ergonomic consumer-oriented technology for product development”, International Journal of Industrial Ergonomics, Vol. 15, pp. 3-11. Nagamachi, M. (2002), “Kansei engineering as a powerful consumer-oriented technology for product development”, Applied Ergonomics, Vol. 33 No. 3, pp. 289-94. Nishio, C. (1994), “Marketing models by neural network”, Operations Research, Vol. 39, pp. 203-8. Penfield, W. and Rasmussen, T. (1950), The Cerebral Cortex of Man: A Clinical Study of Localization of Function, Macmillan, New York, NY. Schu¨tte, S.T.W. (2002), “Designing feeling into products-integrating Kansei engineering methodology in product development”, Institute of Technology, Linko¨ping University, Linko¨ping, p. 184. Schu¨tte, S. and Eklund, J. (2004), “Concepts, methods and tools in Kansei engineering”, Theoretical Issues in Ergonomics Science, Vol. 5, pp. 214-32. Spector, P.E. (1992), Summated Rating Scale Construction: An Introduction, Sage, Newbury Park, CA. Corresponding author Jens J. Dahlgaard can be contacted at: [email protected]




TQM 20,4

312

Customer experience management Influencing on human Kansei to management of technology Shin’ya Nagasawa Waseda Business School, Graduate School of Commerce, Waseda University, Tokyo, Japan Abstract Purpose – The purpose of this paper is to explain the relationships and the meaning of the customer experience management approach, which involves manufacturing and fabrication influenced by human kansei with respect to the management of technology (MOT). Design/methodology/approach – Four cases of experience value creation from earlier work are presented. An interview was held with the product manager of each product or CEO of each company. According to the interview, the paper analyses experience values of four cases based on the five modules. Findings – As a result of analyzing INAX “SATIS”, NISSAN “X-TRAIL”, Canvas Bag by “Ichizawa Hampu” and Albirex Niigata from the viewpoint of the creation of customer experiences, it was found that each of them has high standards for all values of SENSE, FEEL, THINK, ACT and RELATE, meaning that they are like an ensemble of customer experiences. They create not only functional benefit but also customer experiences by the MOT approach. Originality/value – This paper explains the relationships and the meaning of the customer experience management approach, which involves manufacturing and fabrication influenced by human kansei with respect to the Management of Technology (MOT) and will be of interest to those involved in that field. Keywords Product development, General management, Critical success factors, Project management, Japan Paper type General review


1. Introduction From the viewpoint of management of technology (MOT), the importance of management of new product development can be illustrated as follows. The production rules in the twentieth century manufacturing industry placed stress on “how to make” something. The production rules in the twenty-first century stress “what to make”. The manufacture of attractive goods and products is most important for all divisions of a corporation. Management of new product development plays a central role and is influenced by such multi-directional views as the combined viewpoints of technological aspects and non-technological aspects. The problem of manufacturing industries in Japan, as elsewhere, is that high technology products will not necessarily lead to high sales volumes. This situation does not arise from the technology itself but from the management philosophy or methodology, and therefore education and research on management of technology (MOT) should be emphasized. Management of new product development involves a commitment to all aspects of developing a new

product to make sure that the developed products will become successful (Nagasawa, 2004). There are certain rules and common attributes among cases of hit product development. The author tried to identify a set of general success factors behind project management by interviewing product and project managers, who had succeeded in developing hit products, and by scrutinizing the gifts, talents and abilities required to develop hit products (Nagasawa and Kino, 2004, etc.). The methodology for developing a new hit product involves “the Seven Tools for New Product Planning”. The tools are widely used and contribute to producing hit products in various fields (Ofuji et al., 2000, etc.). At the same time the author emphasizes based on long experience that products should not only have functional elements and quality, but also something more such as design that moves, touches and impresses people’s kansei (a word representing feeling, taste, emotion, etc.) and psychology (Amasaka and Nagasawa, 2000; Nagasawa, 2002, 2003, etc.). Nevertheless, up to the present, the author could not develop any theory and methodology for including such features and attributes in new products, even if he scrutinized and tried to do so. The concept of customer experience, which has recently attracted much attention, has been effectively used as a concrete theory to organically combine people’s kansei or feeling and psychology into the making of products. Furthermore, the concept of customer experience also helps us to understand hit products and brand successes which success cannot be explained by traditional marketing theory. It is the aim of this article to describe and discuss how the philosophy of experiential marketing and customer experience management, which is manufacturing and fabrication that influences the connection with kansei, is related to the management of technology (MOT). 2. Design of intangible events for tangible product There are many aspects of products and their management that are considered in MOT, but the most important concern is the customer. It is frequently said that “the product did not sell well even though it was technically good”. This kind of product is a technical product but it does not constitute consumer goods. If customers do not accept advanced technological and high quality products, these cannot be considered as consumer goods. So the most important requirement for such “goods” is to design an intangible event that creates a smooth relationship between the tangible product and the user. In this discussion, the emphasis is placed on the point of contact and the interface or point of touch between the product and the user. Furthermore, it is important to provide some “meeting” point to ensure that the user encounters the product. The design of the meeting point between the user and a tangible product does not only mean determining the shape or styling and the form of the product, but also the provision of a product that makes the user feel good or become happy when using it. A tangible product designed with full consideration of its users should sell well. The intangible event means to combine a product with the concept of goods as two sides of the same coin. This involves designing an intangible event, which ensures that the user meets the tangible product. In other words, the most important aspect is how useful the designed product will be for the people who buy it (Iwakura et al., 2005).

Customer experience management 313

TQM 20,4

Designers and producers are used to having a strong sense of satisfaction and confidence in designing a technologically advanced product. However, customers do not understand the high technology involved and do not perceive whether the technology contributes to or has a relationship to their happiness. This tendency becomes more significant the more advanced the technology is. This can be explained by experiential marketing and customer experience management.

314 3. Customer experience and strategic experience value modules Experiences are private intangible events that occur in response to some stimulation (e.g. as provided by marketing efforts before and after purchase). Experiences involve the entire living being. They often result from direct observation and/or participation in events – whether they are real, dreamlike, or virtual. Schmitt provided a brief description of the five types of customer experiences that form the basis of the Experiential Marketing Framework (Schmitt, 1999). “Customer experience” indicates the value of something that impresses and appeals to the sense of kansei, involving the senses and impressions of the user such as those the customer actually feels directly and is impressed with when coming into contact with a company and a brand. “Customer experience” is not an incidental value but an essential and intrinsic value where the products and services are understood from the customer’s point-of-view as those provided by the company and the brand. The objective of marketing, which creates “Customer experience” (“Experiential marketing”), is not to provide products and services as tangible objects to customers, but to take the aspect of consuming in the context of the customers’ lifestyles and to interpret their consumption by appealing to their senses and feelings in the process. Schmitt classifies experience values into five modules as a strategic basis of marketing activities. The five modules of the strategic experience values, which are shown in Table I, will be discussed in this section and further exemplified in section 4, with four different hit products case stories. 3.1 SENSE (sensitive customer experience) SENSE marketing appeals to the senses. The objective is to create sensory experiences through sight, sound, touch, taste and smell. SENSE marketing may be used to differentiate companies and products, to motivate customers, and to add value to products. As we will see, SENSE marketing requires an understanding of how to achieve sensory impact (Schmitt, 1999).

Table I. Strategic experiential modules provided by Bernd H. Schmitt

Module

Contents of customer experience

SENSE FEEL THINK ACT RELATE

Sensory experience value that appeals to the five senses Emotional experience value that appeals to feelings and moods Intellectual experience value that appeals to creativity and cognitive functions Behavioral experience value that appeals to physical behavior and lifestyle Relative experience value that appeals to confirmative groups and cultural groups

Source: Nagasawa (2006)

It is a sensory experience value to provide an exciting stimulus by appealing directly to the five senses of the consumer such as the visual sense, auditory sense, sense of touch and taste, and sense of smell. In the case of a car, for instance, a Jaguar provides a sensitive experiential value that appeals to the aesthetic sense. On the other hand, a Porsche has a sensitive experiential value that stimulates excitement. 3.2 FEEL (affective customer experience) FEEL marketing appeals to customers’ inner feelings and emotions, with the objective of creating affective experiences that range from mildly positive moods linked to a brand (e.g. for a non-involving, non-durable grocery brand or service or industrial product) to strong emotions of joy and pride (e.g. for a consumer durable, technology, or social marketing campaign). As we will see, most affection occurs during consumption. Therefore standard emotional advertising is often inappropriate because it does not target feelings during consumption. What is needed for FEEL marketing to work is a close understanding of what stimuli can trigger certain emotions as well as the willingness of the consumer to engage in perspective taking and empathy (Schmitt, 1999). FEEL is about emotional experience values that appeal to the inner feelings and moods of the customers. The relaxed feeling which we experience when drinking a cup of coffee at Starbucks Coffee, the enthusiasm we experience in enjoying a ride at Disneyland, and so on are emotional experience values. 3.3 THINK (creative/cognitive customer experience) THINK marketing appeals to the intellect with the objective of creating cognitive, problem-solving experiences that engage customers’ creativity. THINK appeals to engage customers’ convergent and divergent thinking through surprise, intrigue, and provocation. THINK campaigns are common for new technology products. But THINK marketing is restricted mainly to high-tech products. THINK marketing has also been used in product design, retailing, and in communications in many other industries (Schmitt, 1999). THINK is about intellectual experience values that appeal to the intellect of the customers through cognitive and problem solving experiences. Edible oils “Healthy Acona” or the drinking tea “Healthya Green Tea” (produced by Kao) with the efficacy to help contain the accumulation of body fat give intellectual experience values to customers by appealing to a health-oriented lifestyle. 3.4 ACT (physical customer experience) ACT marketing aims to affect bodily experiences, lifestyles, and interactions. ACT marketing enriches customers’ lives by enhancing their physical experiences, showing them alternative ways of doing things (e.g. in business-to-business and industrial markets), alternative lifestyles, and interactions. Analytical, rational approaches to behaviour change related to ACT are often motivational, inspirational, and spontaneous in nature and brought about by role models, e.g. movie stars or famous athletes (Schmitt, 1999). ACT is about behavioural experience values that appeal to physical behaviours, lifestyles and the mutual relationships between people. The “iPod” produced by Apple


TQM 20,4

316

and the “Mini Cooper” produced by BMW satisfy customers’ need for self-identity due to the segmentation of lifestyles. 3.5 RELATE (social-identity customer experience) RELATE marketing contains aspects of SENSE, FEEL, THINK, and ACT marketing. However, RELATE marketing expands beyond the individual’s personal, private feelings, and thus adding to “individual experiences” and relating the individual to his or her ideal self, to other people or to cultures (Schmitt, 1999). RELATE is about relative experience values that appeals to individual self-realization. Harley Davidson, the symbol of a free spirit, stimulates owners to put tattoos on their arms and body. 4. Case studies of customer experience creation in Japan This section presents four cases of experience value creation from Nagasawa (2005). We had an interview with the product manager of each product or CEO of each company. According to the interview, we analyse experience values of four cases based on the five modules. 4.1 INAX “SATIS” “SATIS” produced by INAX shown in Figure 1 is a series of sensational sanitary ceramic products used in toilets, bathrooms and powder rooms. It is a toilet facility without a water tank and the design is impressive with regard to customers’ kansei. When it was put on the market, it caused a sensation and became a very influential product. “SATIS” changed the perspective of people’s understanding of things. The concept of a toilet was replaced by that of a hospitality space. The conventional concept of a

Figure 1. INAX “SATIS”

toilet was recreated with something that has a different value for customers and that provided a unique added value. We had an interview with Mr Nobuho Miyawaki, director of the Space Design Centre, INAX and the product manager of “SATIS.” According to the interview, we analyse experience values of “SATIS” based on the five modules. Table II shows the systematical analysis of INAX “SATIS”, by using the Strategic Experiential Modules in Table I, as a concept that creates value for customers, as to whether it gives any impression to the customers, or the kind of customer value that it creates. Of course, we should discuss interaction values because there would be a lot of interactions among five modules. Especially SENSE seems to have interactions with several other experience values. Though interactions are important, it is quite difficult to analyse them because there are many combinations among five modules. As a matter of convenience we tried to distinguish each customer value among the five modules.


4.2 Nissan “X-TRAIL” Nissan has sold the “X-TRAIL” (Figure 2) well since November 2000 when the car was launched. They maintained the top position in domestic sales of SUVs (Sports Utility Vehicles) for five years in a row up to 2005. They have sold a total of 160,000 X-TRAILs in 170 countries and the vehicle has maintained the top place in its class in all of these countries. The “X-TRAIL” was developed based on “the Seven Tools for New Product Planning”. “Seven Tools for New Product Planning” proposed by Ofuji et al. (2000, etc.) is a set of tools to create hit products appealing to the Kansei of consumers. The seven tools consist of an interview survey, a questionnaire survey, positioning analysis, idea generation, idea evaluation and selection analysis, conjoint analysis and quality tables. Module

Customer experience included in INAX “SATIS”

SENSE

Design making customers feel that it provides an aesthetic space Toilet space as a new circumstance in daily living It makes customers perceive clean conditions A sense of relief is aroused Appraised as an ideal toilet space The lack of a tank has extended the toilet space Sufficient coordination that is full of imagination Tankless washing beyond expectations High functionality appealing to customers senses Intelligent toilet space Toilet behavior is changed due to full automation The toilet can be offered without hesitation to guests and friends Eco-design based on social responsibility Appeals to customers due to branding Building a new social category

FEEL THINK

ACT RELATE


Table II. Customer experience of INAX “SATIS” as a strategic experiential module

TQM 20,4

318

Figure 2. Nissan “X-TRAIL”

We had an interview with Mr Masahiro Toi, Chief Product Specialist of NISSAN and the product manager of “X-TRAIL.” According to the interview, we analyse experience values of X-TRAIL” based on the five modules. Table III shows the result of analysing the different kinds of customer experiences of NISSAN’s “X-TRAIL”, which has a price of 2 million JPY. Customers take into consideration that paying the high price should enable them to realize the experience values shown in the table. 4.3 Canvas bag produced by a small Kyoto company “Ichizawa Hampu” “Ichizawa Hampu,” a veteran company of canvas products in Kyoto, is famous for producing and selling canvas bags (Figure 3) throughout Japan. Ichizawa Hampu is a famous Japanese brand spoken of as the “Japanese Louis Vuitton.” The company has one shop and catalog sales with a constant 3,000 to 4,000 orders. Based on interviews with the fourth owner and Chief Executive Officer, Shinzaburo Ichizawa, and his wife Emi Ichizawa, member of the Board of Directors, we analyzed the secrets of the Ichizawa Hampu brand from the viewpoint of customer experience. As shown in Table IV, Ichizawa Hampu creates customer experiences by emphasizing customers among the young generation with consideration of the customers’ lifestyles. It is a veteran company with a traditional industry in Kyoto, symbolized by natural sailcloth adhering to the craftsman’s art. The source of customer experience value creation of Ichizawa Hampu is summarized as the three points of the development of their products:

Table III. Customer experiences of Nissan “X-TRAIL”

Module

Customer experience included in Nissan “X-TRAIL”

SENSE FEEL THINK ACT RELATE

Rectangular design CM of a falling person stimulates the psychology that a person is playing a sport Astonishment and thought of the washable interior of the car Events such as “X-TRAIL JAM” converts people to outdoor sports “X-TRAIL” Building a flexible fan club for “X-TRAIL” in outdoor sports



Figure 3. Canvas bags by Ichizawa Hampu

Module

Customer experience included in canvas bag by “Ichizawa Hampu”

SENSE

Brand recognition using visual labels Visual sense of the canvas Careful finish with a good supple texture The long life of the products creates attachment to the brand The labels bring out a sense of nostalgia in people Careful craftsmanship alludes to the spirit of study Lifestyle of taking good care of things Customer loyalty created through a repair and re-upholstery service

FEEL THINK ACT RELATE


(1) their attitude of attention to careful craftsmanship; (2) new discovery of traditional natural textiles from sailcloth products; and (3) feedback on the needs of the customers. This means that the development strength of the company and the customer experience are inextricably linked. 4.4 Football J1 team “Albirex Niigata” The Football J1 team “Albirex Niigata” has an incomparable drawing power. It offers a spectacle and it is even impressive that every match can gather more than 40,000 spectators who wear an orange uniform with the team’s colours as shown in Figure 4. Niigata is said to be not pro-sport. There is no company willing to be a sponsor of Niigata because they have no star players. From the viewpoint of customer experience we analyzed the phenomena of Albirex Niigata, which is considered to be a “miracle”. In the analysis we held discussions at the Waseda Business School and used various

Table IV. Customer experiences included in Ichizawa Hampu’s canvas bag

TQM 20,4

320

Figure 4. Emblem of football J1 “Albirex Niigata”

interviews that Mr Hiromu Ikeda, CEO of Albirex Niigata, had given, and which illustrated the efforts to create customer experiences in Albirex Niigata. From the viewpoint of customer experience Albirex Niigata very effectively succeeded in combining the enthusiastic space of 40,000 fans and the experience of appealing to their love of their home town. The success of Albirex Niigata showed that the frame of reference of customer experience is effective not only for developing products but also for the sports business and the entertainment industry. Albirex Niigata is not a general product, but the Albirex Niigata reference provides a means to understand how to plan and create intangible events. The frame of reference of the customer experience value of Albirex Niigata is illustrated in Table V.

5. Relationship between functional benefits and customer experiences In conclusion, let us summarize the new relationship of customer experience with conventionally functional benefits and the role customer experience plays. It is useful to analyze the relationship between the functional benefits and customer experiences to discuss what can be provided and what can be created from the viewpoint of creating value for the customers. Module

Customer experience included in “Albirex Niigata”

SENSE

Enthusiastic experience of 40,000 fans that cannot be encountered anywhere else Visual experience of the orange team colours Passionate experience of attachment to their own handmade team without any support from large companies and without any star players Experience of turning the negative image of Niigata into a positive image for the city A festival experience once every two weeks Experience of a feeling of unity with loud cheering by the supporters Experience of the unity of the supporters in their own home city of Niigata

FEEL THINK ACT Table V. Customer experiences of “Albirex Niigata”

RELATE

Source: Nagasawa (2005, p. 222)

From the viewpoint of functional benefits, for instance, a toilet that provides improvements in functionality and usability of the toilet space has customer value. In addition, the depth and breadth of the choice by providing various options that improve customers’ selection based on their own taste, by unifying designs that improve the image of the toilet can be understood in the same way as a result of value creation. From the viewpoint of customer experience, the total space including the toilet, creates value for the customer, and the toilet space itself also creates value for the customer. That is, providing a space and atmosphere in a toilet different from conventional ones, creates customer value by appealing to the customer’s psychology and kansei to totally and radically change the recognition of what a toilet is. This means to create totally new customer values that influence customers’ lifestyles. A comparison between customers’ values created by functional benefit and customer experience values shows that these values have a complementary relationship even though there is some overlap. The functional benefits of a toilet room provides customer values through the improvement of the functional and beneficial aspects, but customer experiences creates customer value by improving the psychological aspects of the customers’ kansei, which functional benefits cannot provide. In other words, the functional benefits give physical and materialistic satisfaction and the customer experiences give psychological and kansei satisfaction. Figure 5 illustrates the relative relationship between functional benefits and customer experience. These are interpreted as having both their own field and being in a complementary relationship. It should be noted that supporting the complementary relationship can be understood from the MOT approach. That is, when the MOT approach realizes an innovative technology (development of direct valve washing and so on) it provides functional benefits (achievement of a smaller space due to the absence of a tank). The customer experience (change in the image of a toilet space) influences the customer’s mind by an innovative technology and creates value to influence kansei. The complementarity between functional benefits and customer experience creates a totally different and innovative value for the customers (a toilet as a hospitality space).


Figure 5. Image of complementary relationship between the functional benefit and the customer experience

TQM 20,4

322

6. Conclusion The concept of customer experience, which has recently attracted much attention, has been effectively used as a concrete theory to organically combine people’s kansei or feeling and psychology into the making of products. Furthermore, it also helps us to understand hit products and brand goods the success of which cannot be explained by traditional marketing theory. The objective of this paper was to explain the relationships and the meaning of the Customer Experience Management approach, which involves manufacturing and fabrication influenced by human kansei with respect to the Management of Technology (MOT). As a result of analyzing INAX “SATIS”, NISSAN “X-TRAIL”, Canvas Bag by “Ichizawa Hampu” and Albirex Niigata from the viewpoint of the creation of customer experiences, it was found that each of them has high standards for all values of SENSE, FEEL, THINK, ACT and RELATE, meaning that they are like an ensemble of customer experiences. They create not only functional benefit, but also customer experiences by the MOT approach.

References Amasaka, K. and Nagasawa, S. (2000), Basics and Applications of Sensory Evaluation: for the Kansei Engineering in Automobile Development, Japan Standards Association, Tokyo (in Japanese). Iwakura, S., Iwatani, M. and Nagasawa, S. (2005), Strategic Design Management in Honda: Destructive Creation and Evolution of Brand, Nihon Keizai Shimbunsha, Tokyo (in Japanese) and Korean Human Books, Seoul (translated in Korean). Nagasawa, S. (Ed.) (2002), Product Development Relating Kansei: Its Methodology and Practice, Japan Publishing Service, Tokyo (in Japanese). Nagasawa, S. (2003), Practice of Kansei Product Development: Featuring Kansei as Product Elements, Japan Publishing Service, Tokoyo (in Japanese). Nagasawa, S. (Ed.) (2004), Live MOT: Messages of Product Managers, Nikkagiren Shuppansha, Tokyo (in Japanese). Nagasawa, S. (Ed.) (2005), Value Creation through Customer Experience That Enables to Develop Hit Products: Manufacturing and Fabrication That Influence Kansei, Nikkagiren Shuppansha, Tokyo (in Japanese). Nagasawa, S. (Ed.) (2006), Creating Customer Experience in Long Standing Companies: Design Management of Glance of Customer, Doyukai, Tokyo (in Japanese). Nagasawa, S. and Kino, R. (2004), Identity of Nissan and Honda: Product Managers Who Pursue Product Development, Doyukai, Tokyo (in Japanese). Ofuji, T., Okamoto, S., Konno, T., Nagasawa, S. and Maruyama, K. (2000) in Kanda, N. (Ed.), Easy-To-Understand Version of the Seven Tools for New Product Planning Series to Create Hit Products, Vol. 2, Nikkagiren Shuppansha, Tokyo (in Japanese) and Taiwan Productivity Center, Taipei (translated in Chinese, 2003). Schmitt, B.H. (1999), Experiential Marketing: How to Get Customers to SENSE, FEEL, THINK, ACT, and RELATE to Your Company and Brands, Free Press, New York, NY.

Further reading Nagasawa, S. and Tsai, P.-J. (2007), Marketability of Environment-Conscious Products: Applications of “The Seven Tools for New Product Planning”, Koyo Shobo, Kyoto. Pine, B. II, Joseph, B. and Gilmore, J.H. (1999), The Experience Economy, Harvard Business School Press, Watertown, MA. Schmitt, B.H. (2003), Customer Experience Management: A Revolutionary Approach to Connecting With Your Customer, John Wiley & Sons, Indianapolis, IN. Schmitt, B.H. and Simonson, A. (1997), Marketing Aesthetics: The Strategic Management of Brands, Identity, and Image, Free Press, New York, NY. Corresponding author Shin’ya Nagasawa can be contacted at: [email protected]




TQM 20,4

Kansei engineering approach for total quality design and continuous innovation

324

Antonio Lanzotti and Pietro Tarantino Department of Aerospace Engineering, University of Naples Federico II, Napoli, Italy Abstract Purpose – This paper aims at defining a structured process of continuous innovation in the product concept development phase by a statistical-based Kansei engineering (KE) approach. It consists in the identification of quality elements satisfying both functional and emotional user needs, i.e. the total quality elements. Design/methodology/approach – The approach is developed integrating results from Kano and KE analysis. Three statistical methods considered to be suitable for KE study, are used: supersaturated design for concept configuration, ordinal logistic regression for data analysis, and EVA method for quality evaluation of the optimal concept. These methods are compared with the most used ones in KE regarding their efficacy, efficiency and easiness of use. An innovative procedure to exhibit concepts in a KE session is also presented. It uses the abstraction and association idea principles to elicit users’ grade of agreement for a particular Kansei word. Findings – The proposed approach is fully exploited through a case study on train interior design, developed in a virtual reality (VR) laboratory. The evaluation of comfort improvements obtained by means of a new handle and handrail design is carried on with expert users in VR. A consistent increase of a quality index, by using the defined approach, was obtained. Originality/value – This work aims at contributing to the conception of new product solutions, which are appealing and saleable. The availability of virtual reality technologies and software capable to manage complex statistical analyses, will concretely aid designers and engineers in the ideation of high-emotional-quality products, which can be helpful for innovative enterprises to maintain and even increase their market position. Keywords Product development, Total quality management, Customer satisfaction, Design and development Paper type Research paper

1. Introduction After being underestimated for many years, methodologies that help designers to take into account emotional variables are now viewed with increasing interest. Some of the developed methodologies are part of Emotional Design. This is succinctly defined as a design philosophy that focuses on the emotions’ influence on the way humans interact with objects (Norman, 2004). Among these methodologies, Kansei engineering (KE) is finding a very considerable interest of product design teams (Nagamachi, 1995; Nagamachi and Matsubara, 1997; Schu¨tte and Eklund, 2005). The TQM Journal Vol. 20 No. 4, 2008 pp. 324-337 q Emerald Group Publishing Limited 1754-2731 DOI 10.1108/17542730810881311

This work was financially supported by the project (PRIN) “Statistical design of continuous product innovation” funded by the Italian Ministry of University and Research. The authors wish to thank Professor Giuseppe Di Gironimo, ing. Giovanna Matrone of the VR-lab CdCRT Test and the FIREMA S.p.A. for cooperation in the concept generation and evaluation phases.

However, a complete and a further diffusion of KE methodologies among researchers and companies seems, at the moment, constrained by two limitations. First, traditional methodologies attempt to incorporate declared, tangible and functional user needs only, and KE try to do the same but with emotional and intangible users’ needs. These approaches seem to be alternative. Second, the KE approach is still lacking a solid scientific basis. This work has the scope of turning out the validity and usefulness of a KE integrated approach, to be used in the product concept development phase, and its benefits in improving the perceived “total quality” of future products. Henceforth, we will define “total quality product” as a product that satisfies both functional and emotional user needs, and “total quality elements” as the corresponding product attributes (also called design features). The proposed approach (fully described in section 2) integrates the traditional methodologies used in the product concept design with KE principles and statistical methods (briefly described in section 3) such as supersaturated design, ordinal logistic regression and EVA method. An innovative procedure to exhibit concepts in KE sessions is also presented. It uses the abstraction and association idea principles to elicit users’ grade of agreement for a particular kansei word. The proposed approach is fully exploited through a case study on train interior design, developed in a virtual reality environment (described in section 4). The last part of the article is reserved for conclusions and suggestions for possible future works. 2. The KE approach for identifying “total quality” elements The term concept design is used to describe the early phase of the product development process, i.e. the phase where a product concept is created (Ulrich and Eppinger, 2000). A procedure to assess a product’s functional quality in concept design can be schematized into five phases (Di Gironimo et al., 2006): (1) identification of quality elements, i.e. the definition of elements satisfying the declared/functional users’ needs; (2) classification of the identified quality elements, i.e. the identification of those elements with a high impact on user needs; (3) generation of the product concept, i.e. several design solutions, representing different combinations of quality elements, are built usually using a CAD system; (4) quality evaluation, i.e. the quality level for the generated concept and associated elements is quantitatively measured during experimental session in virtual environment; and (5) definition of the optimal or winning concept, i.e. the concept with the highest quality index and the better-evaluated elements, is further developed. Several methods have been developed to support each phase of the above illustrated procedure (King and Sivaloganathan, 1999). Less methodologies exist for achieving a product’s emotional quality. KE is one of these methodologies that has seen a remarkable diffusion in Japan initially and in Europe subsequently. The success of KE is mainly due to its systematic procedure by which it is possible to determine the “quantitative” relationships between users’ emotions and feelings of and product

Kansei engineering

325

TQM 20,4

326

elements (Nagamachi, 1995). An efficient procedure to assess a product’s emotional quality by KE can be schematized into five phase (Schutte and Eklund, 2005): (1) exploration of the semantic dimension, i.e. the identification of words and phrases describing the emotional bond between users and the product under study; (2) exploration of the physical properties dimension, i.e. the identification of important product elements and the selection of a products concept that represents adequately these elements; (3) synthesis, i.e. the collection of users’ impression of the chosen product concept (phase 2) according to the identified words (phase 1); (4) analysis of the collected data (phase 3) for predicting how strong the different product elements are related to the users’ emotional response; and (5) definition of the new product development strategy according to the results of the analysis made in phase 4. In general, design teams choose one of the two above-mentioned procedures to achieve the concept design phase, which contributes to increase the conflict between emotional and functional quality elements. 2.1. Statistical-based KE approach for total quality design In this section a new approach for identifying total quality elements in the concept design phase is proposed. The approach aims at defining a structured process of continuous innovation starting from both functional and emotional user needs. This approach can be divided into two phases. The first phase aims at the exploration and identification of user needs. Innovation is represented by parallel identification of declared-tangible-functional quality elements and emotional-kansei quality elements. Consequently, the first phase is divided into the following three sub-phases: (1.1) Identification of M-O-A quality elements. In this phase, the quality elements satisfying the declared and conscious user needs are identified by the traditional methods of marketing research such as direct interview, focus group, critical incident technique, etc. (Griffin and Hauser, 1993). These elements are then classified by the Kano model (CQM, 2003) as Must-be (M), One-dimensional (O), Attractive (A). The must-be elements are not crucial for this phase, whereas for one-dimensional and attractive quality elements different design solutions will be generated in order to maximize users’ satisfaction. (1.2) Identification of emotional quality elements. In this phase, a KE study is conducted at an abstract level. Sketches and real prototypes are used in place of CAD prototypes for the chosen product concepts. The elicited emotional quality elements will be henceforth called kansei elements. (1.3) Concept generation according to total quality elements. In this phase, one-dimensional, attractive and kansei elements will be used for realizing a set of virtual prototypes using a CAD system.

The second phase aims at selecting the optimal concept, at validating the choice through a confirmatory step and at establishing a continuous step-by-step innovation process. It can be divided into three sub-phases: (2.1) Concept evaluation. In this phase, the concepts generated at phase (1.3) are evaluated in an immersive virtual reality environment. The collected data are then statistically analyzed.

Kansei engineering

327

(2.2) Optimal concept selection and confirmatory session. In this phase, by basing on the analysis results, the design team can choose the optimal concept, i.e. the total quality product concept. A confirmatory session is conducted to establish if the results fit both the prior designer hypothesis and the users’ session responses. (2.3) Innovation. In this phase, the collected information and the collaborative process between designers and users should be updated until a satisfactory improvement result is obtained. This objective can be reached by subdividing the complex innovation process in short steps of design research supported by statistical analyses. A graphical scheme of the proposed approach is illustrated in Figure 1. 3. Statistical methods supporting a KE study In order to give a concrete support to designers, KE needs integration with quality and statistical tools (Nagamachi and Matsubara, 1997). Some of the works in KE area already use such methods for kansei words identification (factor analysis, affinity diagram, textual data analysis), product elements selection (Pareto Diagram), concepts generation and configuration (Fractional Factorial Design) and results analysis (Quantification theory type I, Principal Component Analysis, Rough Set Analysis, etc). Frequently, these tools are used individually and with few observations regarding their appropriateness on the specific application context. In this section, we review three statistical and quality methods that are considered to be suitable for a KE study, especially in a concept design phase: supersaturated design (SSD) for concept configuration, ordinal logistic regression (OLR) for data analysis, and Erto-Vanacore (EVA) method for quality evaluation of the “winning concept”. These methods are

Figure 1. Logical flow of the proposed approach for identifying total quality elements in the concept design phase

TQM 20,4

328

compared with the most used ones in KE regarding their efficacy, efficiency and easiness of use. 3.1 Supersaturated design In a KE study the product concepts should to be chosen among those that equally represents the combination of the alternative products’ elements. For this aim factorial design are often employed (Ellekjaer and Bisgaard, 1998). Because of needs to consider several products’ elements without increasing the number of generated concept and consequently users’ fatigue, fraction factorial design are used instead of full factorial designs (Gustafsson et al., 1999). A supersaturated design is a special class of fractional factorial design useful when there are many factors to be investigated and expensive or time-consuming experimental runs (Wu and Hamada, 2000). In fact, with such designs it is possible to study k . n 2 1 factors with only n runs. For instance, by using a supersaturated design as that proposed by Lin (Lin, 1993a), we can study more than k ¼ 8 factors with only six runs, i.e. eight runs less than 2824 IV ¼ 16 fractional factorial design, six runs less than L12 orthogonal array and three runs less than p-efficient designs. As pointed out by Wang et al. (1995) supersaturated designs provide good plans for very early stages of experimental investigation (as in the case of concept design phase) involving many factors and they can be used for gaining some additional objective and quantitative information in respect to the only expert knowledge (designer hypotheses). 3.2 Ordinal logistic regression The declared purpose of KE is to measure how strong the different design elements are related to users’ kansei. Different statistical methods have been used in these phases (see Figure 1) but only two of them are widely used. The first and most recognized method is Hayashi’s Quantification theory type I (Tanaka, 1979). It is a variant of the linear multiple regression analysis that uses dummy variables to handle explanatory variables with nominal scale value. The second method extensively performed is Principal Component Analysis (Morrison, 2005). It is a method that reduces data dimensionality without loss much of information by performing a covariance analysis between factors. An alternative method to perform data analysis in KE context is ordinal logistic regression (OLR) (Barone et al., 2007). When response variable is the users’ agreement for concepts, as in a KE study, the rating scale is ordinal, e.g. it is measured on a scale ranging from 1 to 5 (7 or 9), with 5 (7 or 9) being “most satisfied”. Ordinal Logistic Regression is a modification of logistic regression model. For an easy and complete discussion about logistic regression see (Lawson and Montgomery, 2006). The way to interpret the results from an ordinal logistic regression analysis will be clarified in the case study. 3.3 EVA method Once that product concepts have been generated according to the individualized total quality elements, and users’ agreement for that concept collected, designers and engineers needs to analyze data for selecting the best quality or “winning” concept.

One of the most used quantitative method for assessing this objective is the Analysis of Mean (Ott, 1967) (ANOM). An alternative effective measurement instrument to evaluate a specific quality level for a product concept is the method EVA proposed in (Erto and Vanacore, 2002). It can be useful to define a quantitative quality index for a product concept. The individual contribution of physical quality elements is determined by a stochastic approach, different according to their Kano classification (CQM, 2003). An ordinary global index of quality for product concept is defined as the product of must-be quality index and one-dimensional quality index (E Q ¼ Qm · Qo ). The main advantage of this method is its quantitative nature that allows design team to make a comparison with the global quality index after a design modification as shown in the proposed case study. 4. Case study: train interior design In some market sectors, such as mobile phones and automobile, where companies brand and style is well established, designers tend to pay very high attention to emotional variables. They work essentially with their own creativity and feelings, following a more or less defined mental model. On the contrary, for other products as trains, poor effort is put on the integration of emotional variables with the traditional design paradigms. This was essentially due to economic reasons. After privatization, the railway industry has begun to look more frequently to users’ requirements and the way to improve the quality of their trip. The aim of this study is to prove how the proposed approach for concept design can improve the users’ perceived total quality for train interior design. For the sake of clarity, this section follows the structure of the approach presented in section 2. 4.1 Preliminary study – traditional concept design approach In collaboration with FIREMA Trasporti S.p.A., an Italian railway industry, a study was conducted to investigate the passengers’ preferences for regional train interior design. A traditional concept design approach (Di Gironimo et al., 2006), based on the identification of M-O-A quality elements, was used. At the end of the process, partially conducted at the VR-lab of CdCRC Test (Competence Center for the Qualification of the Transportation Systems funded by the Campania Region) in Caserta (Italy), a concept with a quality index of Qo Qm ¼ 2:78 ðQm ¼ 0:42; Qo ¼ 6:67Þ was selected. Extensive details of this study can be found in Di Gironimo et al. (2007). 4.2 Identification of emotional quality elements – KE analysis In total, 20 regular travellers participated in the survey. By scanning several sources of information (magazines, manuals, web pages of train manufacturers, etc.) 39 words, describing the emotional bond between travellers and the train interior, were identified. These words were reduced to a more manageable number by using both Factor Analysis (Morrison, 2005) and Affinity Diagram (Tague, 2004). Data for factor analysis were collected using the responses given by travellers to 13 existing train interiors on a five grade Likert scale. The affinity process was performed by the authors together with members of the Firema S.p.A CEO. Both methods gave very similar results. The final chosen kansei words were: comfort, originality, mobility, versatile, simple. The next step was the collection of physical elements of the train interior. Inspiration material was collected from internet and at the end of this search process 65

Kansei engineering

329

TQM 20,4

330

elements were identified. These elements were merged into four groups according to their affinity: passive and active safety elements, general elements, attractive elements, information and communication elements. In order to understand which elements had a high impact on travellers, on-line interviews were carried out. Initially the travellers were asked to select ten elements from the whole list. By a Pareto diagram (Ishikawa, 1990) it was possible to establish the relative importance of the above-mentioned groups. Subsequently, the same travellers were asked to select one element from the first group, three elements from the second group, and one element from the third group. The fourth group was considered not important. For each element, two alternatives were chosen in respect to the Italian railway-industry norms (Table I). A supersaturated design was constructed following the Lin’s approach. The generated design presented ten columns and six runs. The elements assignment to the design columns is crucial for the following analysis and it depends from the needs of the design problem under study. In this case, the chosen assignment led to a saturated design (Table II). The remaining columns may be used to estimate the interaction effects among elements, but in the product concept phase this analysis is not strictly needed. By starting from mood boards technique (McDonagh et al., 2002), an innovative procedure to exhibit concepts in KE sessions was used. The procedure is based on the abstraction and association idea principles to elicit respondents’ grade of agreement for a particular kansei word. A sketch on train interior design constituted an image base from which to construct the alternative concepts. Each element’s alternative was manually drawn on the image base sketch. Finally, the different product concepts were created by adding the drawn element’s alternative on a placard. An example of the alternatives’ images for Handle and Handrails elements is shown in Figure 2. Six alternative product concepts were shown to 30 respondents for evaluation. Most of the respondents were students in Industrial Design at the II University of Naples and Levels

Table I. Description of the chosen elements and levels

Table II. The saturated design generated for the synthesis phase

Elements

Description

A B C D E

Video surveillance system Wide space Handles and handrails Large windows Recyclable materials

0

1

No Min dimension: 650 mm Handle on seat Min dimension: 1m2 45 percent of recyclablility

Yes Max dimension: 800 mm Handle on seat & handrail Max dimension: 1.5 m2 80 percent of recyclablility

Run

A

E

C

B

1 2 3 4 5 6

1 0 0 1 1 0

1 0 0 0 1 1

1 1 0 0 0 1

1 0 1 0 0 1

D 1 1 0 1 0 0

1 1 0 0 1 0

1 1 1 0 0 0

1 0 0 1 0 1

1 0 1 0 1 0

1 0 1 1 0 0

therefore they had a natural attitude toward the presented abstract images. Both the order of appearance of the product concept and the kansei words were opportunely randomized. The authors spent some minutes to explain the way how the alternative product concepts had to be understood and the meaning of each kansei word. The respondents were asked to give their impression about each product concept on a five-grade Likert scale. It was pointed out that the first impression had to be the most important. No time constraints were considered. The collected data were analyzed by ordinal logistic regression. A separate regression model was created for each kansei word. In each model, the score given by respondents to each kansei word was used as the response variable, whereas the quality elements were used as explanatory variables. The data were analysed by MINITABw Release 14.1 software. An example of MINITAB output, for the kansei word “Comfort” is presented in Table III. The p-value for Pearson and Deviance chi-square tests was greater than the chosen significance level (0.05), giving no concern about model fitting. The p-value of the Log-likelihood G-test was less than 0.05, therefore, at least one explanatory variable was related to the response variable “Comfort”. By observing both the p-values for quality elements and 95 percent confidence interval built around odds ratio, it was possible to conclude that quality elements A (Video Surveillance System), B (Wide Space) and C (Handles and Handrails) had a statistical influence over the kansei word “Comfort”. Then, by observing the logistic coefficient, it was possible

Kansei engineering

331

Figure 2. An example of alternative images for Handle and Handrails elements

Comfort Goodness of fit tests Pearson Deviance Log-likelihood (G) Quality elements A B C D E

x2 22.852 23.451 47.861 Coeff. 1.012 1.258 1.488 20.347 0.310

df 15 15 5 SE coeff. 0.434 0.436 0.425 0.418 0.418

p-value 0.087 0.075 0.000 p-value 0.020 0.004 0.000 0.407 0.462

Odds ratio 2.75 3.52 4.43 0.71 1.36

95 percent Confidence interval Lower Upper 1.17 6.44 1.49 8.28 1.93 10.18 0.31 1.60 0.60 3.09

Table III. 3 Ordinal logistic regression output for the kansei word “Comfort”

TQM 20,4

332

to conclude that quality element C had a slightly greater impact over the kansei word “Comfort” in comparison with the others. A qualitative synthesis for the analysis with the other kansei words is presented in Table IV. 4.3 Concept generation according to total quality elements The optimal concept, identified in the preliminary study according to the Kano model, was considered as the concept base to integrate with the result of KE analysis. For the sake of the approach presentation, only the kansei word “Comfort” and its highest related element (Handles and Handrails) will be considered in this article. By using principles of usability, the shape and the position within the train of Handles and Handrails were studied. For each design factor, three-design solution were proposed (Table V). The different design solutions were developed according to a 32 full factorial design, and designed in CAD trough pro/ENGINEERw. An example of the created design solution for the design factor shape is presented in Figure 3. The design solutions were then added to the concept base, generating nine alternative train interior concept designs. 4.4 Concept evaluation The generated train interior concept designs were evaluated in an immersive virtual reality environment. It enabled an efficient analysis of the proposed concept. In fact, virtual prototypes are able to simulate specific characteristics of the concepts saving time and resources compared with physical prototypes (Ottosson, 2002). This advantage is particular evident for a big-size product such as a railway coach. Another three advantages are particularly evident (Lee et al., 2004):

Table IV. Relationship between quality elements and kansei words

Comfort Originality Mobility Versatile Simple

Wide space

Handles and handrails

a

a

b

Large windows

Recyclable materials

a

Note: a Medium relation; b Strong relation

Design factors Shape Table V. Design factors and relative solutions for handles and handrails

Video surveillance system

Position

Levels 1 2 3 1 2 3

Handles

Handrails

Smooth 1 Wavy Smooth 2 On all seats Alternate on seats Alternate on seats

Smooth Wavy Helicoidal Alternate on seats Absent Alternate on seats

Kansei engineering

333

Figure 3. An example of the design solution for the design factor “shape”

(1) Flexibility: design elements can be rapidly set by the experimenter. (2) Adaptability: new and not yet built design elements can be tested. (3) Credibility: in a virtual reality environment users have the impression to interact with a real prototype. Moreover, it can be used for anticipating aesthetic, ergonomic and usability verifications already in the concept development phase (Wilson, 1999). All necessary experimental conditions in order to ensure reproducibility were satisfied (Barone and Lanzotti, 2002). In particular, a sufficient visual realism and a virtual concept conforming to the real one were realized. A total of 20 people took part in the evaluation carried out at the CRdC test. Two evaluation sessions were performed. Initially, respondents were asked to express their satisfaction of the concepts on a nine-grade Likert scale. An operator drove participants into the virtual space. The same navigation experience was submitted to all users. A picture of the evaluation session into the virtual reality lab is shown in Figure 4.

Figure 4. Concept evaluation phase at the VR-lab of CdCRT test and identified optimal concept

TQM 20,4

334

4.5 Optimal concept selection and confirmatory session The collected data were analyzed by observing both the main effects plot for shape and position factor (Figure 5) and the distance between the mean concept score and the grand mean. The analysis identified concept no. 4 as the optimal one. In fact, this concept presented the biggest positive distance form the grand mean and it was generated from the combination of level 2 for shape factor and level 1 for position factor. By comparing the score given by respondents to the new optimal concept (7.6) with that determined in the preliminary study (5.5), it was possible to attest a consistent improvement in the perceived quality of the train interior concept. To confirm this improvement, the same respondents were asked to classify handles and handrails quality element according to the Kano model. Finally, some additional questions were asked for applying the EVA method. The handles and handrails elements were classified as a one-dimensional and the resulted quality index, according to the EVA method, was Qo ¼ 2:6. Accordingly, the total quality index for new optimal concept was QoQm ¼ 3.89 [Qm ¼ 0.42;Qo ¼ 9.27(6.67 þ 2.6)]. In comparison with the preliminary study an improvement of 40 percent in the quality index was obtained. 4.6 Innovation By using the proposed procedure, it was possible to identify a new design element by which to improve users’ sense of Comfort and their global quality perception. If the design objectives have been met with success, the acquired information will be used in the next phase of product development. Otherwise, the procedure is reiterated, step-by-step, through the introduction of new innovative elements.

Figure 5. Main effect plots for the shape and position factors

5. Conclusion In this work, an integrated approach for conceptual design has been presented. It consists in the integration of the Kano-based concept design approach with the Kansei engineering methodology, with the aim of improving the “total quality” of a product concept. The approach tries to overcome the two main limitations in the design approaches hitherto adopted: first, the conflict between declared and tangible user needs on the one hand and latent and emotional user needs on the other hand, second, the lack of quantitative and objectives methods for supporting a KE analysis. There were three statistical and quality methods presented as suitable for the proposed approach: (1) supersaturated design, used in the Kansei engineering phase for constructing product concepts without increasing experimental time and costs; (2) ordinal logistic regression, useful in the respondents data analysis for measuring the strength of relationship between different product elements and users’ Kansei; and (3) EVA method used for defining a quantitative quality index for optimal product concept. The adoption of statistical methods, together with the introduction of a new method for concept exhibition and the use of virtual reality, allow designers to concretely support their design actions with objective information on subjective feelings and emotions. In particular, use of virtual reality technologies already in the concept development phase, allows designers to anticipate aesthetic, ergonomic and usability verifications and at the same time reduces time and cost for the arrangement of physical prototypes. The proposed integrated approach was exploited in a project on train interior design where a consistent improvement of a concept quality index was obtained. Future works are still needed in the KE context for addressing the following issues: . to reduce slightly the time needed for carrying out a KE process; . to develop new appropriate methods of analysis for broadening its use under several assumptions and different industrial situations; and . to aid the methodology with informatics database and inferential engines for disseminating its use among companies. The introduced approach will provide designers with more tools both for stimulating user feelings and for correctly analyzing users’ responses. References Barone, S. and Lanzotti, A. (2002), “Quality engineering approach to improve comfort of a new vehicle in virtual environment”, Proceedings of the American Statistical Association; Spring Research Conference, Statistical Computing Section, Alexandria, VA. Barone, S., Lombardo, A. and Tarantino, P. (2007), “A weighted logistic regression for conjoint analysis and Kansei engineering”, Quality and Reliability Engineering International, Vol. 23, pp. 689-706. Center for Quality Management (2003), “Kano’s methods for understanding customer-defined quality”, The Center for Quality Management Journal, Vol. 2 No. 4, pp. 3-36.

Kansei engineering

335

TQM 20,4

336

Di Gironimo, G., Lanzotti, A. and Vanacore, A. (2006), “Concept design for quality in virtual environment”, Computer & Graphics, Vol. 30 No. 6, pp. 1011-9. Di Gironimo, G., Lanzotti, A., Matrone, G., Papa, S. and Tarantino, P. (2007), “Concept design for quality: case study on the interiors of a railway”, Proceedings of IPMM 2007, June 25-29, Salerno, Italy. Ellekjaer, M.R. and Bisgaard, S. (1998), “The use of experimental design in the development of new products”, International Journal of Quality Science, Vol. 3 No. 3, pp. 254-74. Erto, P. and Vanacore, A. (2002), “A probabilistic approach to measure hotel service quality”, Total Quality Management, Vol. 13 No. 2, pp. 165-74. Griffin, A. and Hauser, J.R. (1993), “The voice of customer”, Marketing Science, Vol. 12 No. 1, pp. 1-26. Gustafsson, A., Ekdahl, F. and Bergman, B. (1999), “Conjoint analysis: a useful tool in the design process”, Total Quality Management, Vol. 10 No. 3, pp. 327-43. Ishikawa, K. (1990), Introduction to Quality Control, 3A Corporation, Tokyo. King, A.M. and Sivaloganathan, S. (1999), “Development of a methodology for concept selection in flexible design strategies”, Journal of Engineering Design, Vol. 10 No. 4, pp. 329-49. Lawson, C. and Montgomery, D.C. (2006), “Logistic regression analysis of customer satisfaction data”, Quality and Reliability Engineering International, Vol. 22 No. 8, pp. 971-84. Lee, S., Chen, T., Kim, J.K., Sungho, H.G.J. and Pan, Z. (2004), “Affective property evaluation of virtual product designs”, IEEE Virtual Reality, March, pp. 27-31. Lin, D.K.J. (1993a), “A new class of supersaturated designs”, Technometrics, Vol. 35 No. 1, pp. 28-31. McDonagh, D., Bruseberg, A. and Haslam, C. (2002), “Visual product evaluation: exploring users’ emotional relationship with products”, Applied Ergonomics, Vol. 33, pp. 231-40. Morrison, D.F. (2005), Multivariate Statistical Methods, 4th ed., Brooks/Cole, Belmont, CA. Nagamachi, M. (1995), “Kansei engineering: a new ergonomic consumer-oriented technology for product development”, International Journal of Industrial Ergonomics, Vol. 15, pp. 3-11. Nagamachi, M. and Matsubara, Y. (1997), “Hybrid kansei engineering system and design support”, International Journal of Industrial Ergonomics, Vol. 19, pp. 81-92. Norman, D.A. (2004), Emotional Design: Why We Love (or Hate) Everyday Things, Basic Books, New York, NY. Ott, E.R. (1967), “Analysis of means: a graphical procedure”, Industrial Quality Control, Vol. 24, pp. 101-9. Ottosson, S. (2002), “Virtual reality in the product development process”, Journal of Engineering Design, Vol. 13 No. 2, pp. 159-72. Schu¨tte, S. and Eklund, J. (2005), “Design of rocker switches for work-vehicles: an application of kansei engineering”, Applied Ergonomics, Vol. 36, pp. 557-67. Tague, N.R. (2004), The Quality Toolbox, 2nd ed., ASQ Quality Press, Milwaukee, WI. Tanaka, Y. (1979), “Review of the methods of quantification”, Environmental Health Perspectives, Vol. 32, pp. 113-23. Ulrich, K.T. and Eppinger, S.D. (2000), Product Design and Development, 2nd ed., McGraw-Hill, New York, NY. Wang, P.C., Wold, N.K. and Lin, D.K.J. (1995), “Comments on Lin (1993)”, Technometrics, Vol. 37 No. 3, pp. 358-9.

Wilson, J.R. (1999), “Virtual environments applications and applied ergonomics”, Applied Ergonomics, Vol. 30, pp. 3-9. Wu, C.F.J. and Hamada, M. (2000), Experiments: Planning, Analysis, and Parameter Design Optimization, Wiley, New York, NY. Further reading Agresti, A. (2002), Categorical Data Analysis, 2nd ed., Wiley, New York, NY. Box, G.E.P., Hunter, J.S. and Hunter, W.G. (2005), Statistics for Experimenters: Design, Innovation and Discovery, 2nd ed., John Wiley & Sons, Hoboken, NJ. Burdea, G.C. and Coiffet, P. (2003), Virtual Reality Technology, 2nd ed., John Wiley & Sons, Hoboken, NJ. Hosmer, D.W. and Lemeshow, S. (2000), Applied Logistic Regression, 2nd ed., Wiley, New York, NY. Karwowski, W. and Marras, W.S. (2003), Occupational Ergonomics: Design and Management of Work Systems, CRC Press, London. O’Shaughnessy, J. and O’Shaughnessy, N.J. (2003), The Marketing Power of Emotion, Oxford University Press, New York, NY. Schu¨tte, S. (2005), “Engineering emotional values in product design – Kansei engineering in development”, Linko¨ping Studies in Science and Technology, Dissertation 951. About the authors Antonio Lanzotti is Senior Professor in Design and Methods of Industrial Engineering at the University of Naples Federico II. He is the corresponding author and can be contacted at: [email protected] Pietro Tarantino is a PhD student in TQM at the University of Naples Federico II.


Kansei engineering

337


TQM 20,4

An e-commerce site for gift flower arrangements that fit kansei and social manners

338

Keiko Ishihara, Ryo Nakagawa, Shigekazu Ishihara and Mitsuo Nagamachi School of Psychological Science, Hiroshima International University, Higashi-Hiroshima, Japan Abstract Purpose – Gift flowers should be chosen to depict a message with the sender’s kansei and are bound by nature of flowers and social manners, to maintain social relationship between the sender and the recipient. Few buyers, but most florists, have expert knowledge of the flowering time, scent, price, and nature of each flower, and are experts in arranging flowers that meet a given purpose. The purpose of this paper is to incorporate handling constraints into the inference process of a kansei engineering system. Design/methodology/approach – The paper collected the expert knowledge concerning nature of flowers, composing flower arrangements and social manners on gifts from specialists of flower arrangements including a florist and special books. At the same time, kansei evaluation experiments on the kinds of flowers and colors were conducted. The expertise and the results of kansei experiments were organized into a flower database and inference rules for choice of a main flower, arrangement shapes and combination flowers. The rules were implemented as server-side programs. Users input information about the recipient, purpose of the arrangement and purchase information using a web browser. The system outputs a solution; a list of main flowers, combination flowers, greens and the shape of arrangement. Findings – Traditional kansei engineering studies revealed the relationships between design elements and kansei with developing new analyzing methods. Different constraints come into the actual product design and manufacturing should be integrated with findings obtained from the kansei evaluation to successfully utilize kansei engineering for product development. Practical implications – The inference rules will be able to tell the reasons for choosing the mainand combination flowers and arrangement shapes to satisfy the customers. Originality/value – The proposed system suggests the original arrangement of flowers unlike most online florists selling ready-made arrangements. The paper shows a solution to incorporate different constraints underlying in a real production process into the inference process based on the result of kansei analysis. Keywords Electronic commerce, Social values, Culture, Botany, Horticulture, Japan Paper type Research paper

1. Introduction We give people flowers on special occasions. For example, we give vases of field flowers to family members on their birthdays or passionate flowers to our lovers. Flowers are chosen to depict a message with the sender’s kansei (Japanese word that The TQM Journal Vol. 20 No. 4, 2008 pp. 338-347 q Emerald Group Publishing Limited 1754-2731 DOI 10.1108/17542730810881320

The authors wish to thank Ms Satoe Yamaoka, who graduated from Hiroshima International University in March 2005, for participating in the development of their system. This work was supported by the Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for Scientific Research (A) No. 15200016.

means sense, feeling, or emotion) and are bound by social manners. For example, flowers with strong smell may not be good for hospital visit, cheap flowers may be out of a festive gathering, passionate flowers presented for a boss might make him be embarrassed, and there are much more local customs (see, e.g. Iwashita and Ito, 2005 for Japanese gift manners). Few buyers, but most florists, have expert knowledge of the flowering time, scent, price, and nature of each flower, and are experts in arranging flowers that meet a given purpose. Typically, a florist infers an acceptable solution after obtaining information from the customer. This study sought to incorporate handling constraints into the inference process of a kansei engineering system. Traditional kansei engineering studies revealed the relationships between design elements and kansei with developing new analyzing methods (for example, Nagamachi, 1991, 1995; Ishihara et al., 1995, 1999). Different constraints come into the actual product design and manufacturing should be integrated with findings obtained from the kansei evaluation to successfully utilize kansei engineering for product development. In the early kansei engineering studies, for example FAIMS (Nagamachi, 1991), on garment design, the researchers tried to incorporate the design constraints into the inference process. We describe our e-commerce site in this paper as a solution for proposing gift flower arrangements suiting the purchaser’s needs according to the integration of results of kansei evaluation and expertise of florists. The proposed system deals with flower arrangements using foams in containers. That kind of flower arrangements become appreciated in Japan, because the recipients have no need for preparing vases nor keeping in case of potted plants (Iwashita and Ito, 2005). A purchaser inputs data such as the purpose of the present, relationship with the recipient, and budget. The system receives the data and then retrieves suitable flowers from the database according to the results of a kansei evaluation and social constraints. Then the system displays a list of possible flowers and types of arrangements.

Flower arrangements that fit kansei 339

2. Flowers, colors, and kansei 2.1 Types of flowers In a flower arrangement, flowers are classified as form, line, mass, or filler according to their shape and size, and the flower to be attached (see Figure 1). A different arrangement composition role is given to flowers in each classification (The arrangement style is different by culture; in Japan, for example Shiraishi, 1993). A form flower is comparatively large with a clear shape and has a strong individuality. It becomes the center-piece of the arrangement. A line flower has flowers attached along the stem and is used to establish the framework of the straight lines and curves in the arrangement. A mass flower is a gathering of many petals or small flowers that form a round shape. They are often used as the main flowers to produce familiarity. A filler flower has many small flowers and leaves attached to the branches. It is used to bury the space and give an appearance of solidity and unity.

Figure 1. Types of flowers

TQM 20,4

340

2.2 Relationship between flowers and kansei We conducted a flower evaluation test using photographs of 25 types of flowers in which only a single flower was shown against a white background. We used mostly red flowers; if the particular type of flower was not red, then we used white or yellow flowers. A photographwas displayed on the monitor of a personal computer. The subjects evaluated each flower by answering the degree of kansei evoked on five-point SD scale for 57 kansei word pairs, such as “light – not light” and “casual – not casual.” The subjects consisted of ten men and women between 21 and 50 years of age who were familiar with flowers. A principal component analysis was performed using the mean value obtained from the evaluation results of each sample flower. The first and the second principal components (PCs) with large eigenvalues were used in our interpretation of the results. The kansei words that produced large positive first PC loadings were “thick”, “spicy”, “showy”, “tropical”, and “vivid”. The kansei words that produced large negative first PC loadings were “cool”, “neat”, “fresh”, “light”, and “soft”. Therefore, this PC showed the relative “lightness/heaviness” of a given flower. The kansei words with large positive second PC loadings were “childish”, “casual”, “pop”, “happy”, and “natural”. The kansei words with large negative second PC loadings were “luxury”, “adult”, “graceful”, “mysterious”, and “chic”. Thus, this PC showed the relative refinement of a given flower. Figure 2 shows the PC score of each flower. Since form flowers have strong meanings, those with unique meanings are located at the edge of the corresponding kansei space. For example, the anthurium on the right edge is “heavy”; at the upper right, the sunflower is “heavy” and “not refined”. The lily and narcissus at the left edge are “light”; the calla at the bottom left is “light” and “refined”. Mass flowers were distributed at various positions of the meaning space. In particular, the rose and gerbera daisy were located at a refined pole. A good correspondence was obtained between the classification according to the flower kansei evaluation and that found in flower arrangements. The flowers and kansei were associated by classifying the PC scores in a hierarchical cluster analysis. We used the reliability presumption method with a multi-scale and bootstrap (Shimodaira, 2002) to determine the cluster. In many cases of kansei evaluation, the sample (evaluation data) set is not big because it is difficult to sample the population many times. The bootstrapping method was designed to assume the variation and reliability of the population from small set of samples (Efron, 1985). The method of presuming the reliability of clustering through bootstrapping was developed by Felsenstein (1985). The compensated method was the multiscale bootstrap proposed by Shimodaira (2002). We used the correlation coefficient as the similarity and the group average method (unweighted pair group method with arithmetic mean) to make the clusters. A total of eight clusters were adopted that had high values of appearance probability. In each cluster, the kansei word, with a high evaluation value that corresponded well to the plot of the PC loads and was often used in actual arrangements, was selected as a representative of the cluster. 2.3 Relationship between colors and kansei Round samples, 5 cm in diameter, were created using colored paper corresponding to 50 colors described in the flower database of our system. Each color sample was placed


Figure 2. Principal component score of each flower

on gray background paper and evaluated. The same set of 57 kansei word pairs that was used to evaluate the flowers was also used to evaluate the colors. The color PCs that showed a relative lightness/heaviness and refinement were included with the flower results. The PC score of each color and the corresponding kansei word were also classified into eight clusters. 3. Flower arrangement expertise A general procedure was used to select the main flower, determine the final shape of the arrangement, and then to choose other flowers with suitable colors and shapes. 3.1 Development of the flower database The name, type, color, unit price, ease of scattering, and flowering time of each flower were included in the database.

TQM 20,4

342

3.2 Shape of the arrangement The system considers five American-style shapes: round, triangular, diamond, fan, and spray. Each type has three sizes: large, medium, and small. An appropriateness degree is assigned to each shape using an integer value of [ 2 10, 10] according to the purpose of the arrangements. The type of flower included is also restricted by the shape of the arrangement. A round arrangement is made by putting flowers together to form a ball. It consists mostly of mass flowers with few line flowers. A triangle shape is made from the straight forms of line flowers while filler flowers are used to make little triangles inside the arrangement to give a solid appearance. The height of a diamond arrangement is suppressed and the flowers extend sideways. Mass, filler and form flowers are mainly used with few line flowers. A fan arrangement is a large design that projects the stalks radially from the center of a round arrangement. Line flowers are necessary to make this shape. It is usually a very large arrangement that cannot be carried by hand. A spray-shaped arrangement looks as if a bouquet is being laid down on the container. Mass and form flowers are mainly used for this type of arrangement. 4. Inference rules 4.1 Selection of the main flower The rules used to select the main flower are given priority in the following order and applied according to the buyer’s specifications: (1) If a flower illustrated on the input screen is particularly desired it is assumed to be the main flower. (2) If a special occasion is selected for the arrangement, such as “hospital visit” or “Buddhist memorial service”, a specific atmosphere (kansei ) is set. Here, “lively” and “neat” would be chosen, respectively. (3) A flower with a flowering time corresponding to the delivery date is chosen among the flowers that have the desired atmosphere for the recipient. (4) A flower with a flowering time corresponding to the delivery date is chosen using the correspondence rule between the interpersonal relationship and the kansei. For example, if the recipient is a superior, a “neat” flower is chosen. (5) A flower with a flowering time corresponding to the delivery date is chosen using the correspondence rule between the recipient’s age and gender and the kansei. For example, a “pop” flower is chosen for girls aged 0-20 years. 4.2 Selection of arrangement shape The shape and size of the arrangement are determined based on the purpose, main flower type, buyer’s requests, and amount of money to be spent. The degree of appropriateness of each arrangement type is determined using the following constraint rules: (1) Usage constraint. A degree of appropriateness [ 2 10, 10] is assigned to each arrangement type by considering the location, height, attractiveness, and ease of carrying. For example, for a party center piece, a diamond arrangement was set to 10 degrees of appropriateness because it is not very tall and looks attractive. A large arrangement that stands out was set to a high degree of appropriateness for a concert.

(2) Main flower type constraint. If the main is a line flower, a linear arrangement was set to a high degree of appropriateness while a small round arrangement was set to a low degree of appropriateness. (3) Buyer’s constraints. If the buyer is concerned about the ease of carrying, small round, small triangle, and spray arrangements were assigned a degree of appropriateness of 10, while larger arrangements were set to a lower degree. (4) Budget constraints. If the budget is less than a certain amount, a low degree of appropriateness was set to large arrangements.


The type of arrangement with the highest degree of appropriateness is adopted by the system (Figure 3). 4.3 Selection of flower combinations Selection rules that reflect human relationships and the desired usage are applied to all the flowers in the database. (1) Price. If the recipient is a superior, relative, senior associate, junior fellow, or someone of whom the buyer is a fan, flowers that cost more than 250 yen are used to make an attractive arrangement. (2) Nature. If the buyer designates the “ease of carrying” as important, flowers that are not easily scattered are used. (3) Flowering time. Flowers whose flowering time includes the delivery date are used. The flowers to be combined are determined among the remaining flowers by applying one of the color rules below: (1) Flowers with the same hue as the main flower and neighboring tones are combined. (2) Flowers with the same tone as the main flower and neighboring hues are chosen. (3) Flowers with the same tone as the main flower and opposite hues are used. (4) If the usage is “Buddhist memorial service”, flowers with pale or light tones are combined.

Figure 3. Shapes of arrangements

TQM 20,4

344

4.4 Selection of the greens Commonly used types of greens are added to the shape of the arrangement. 5. Implementation of the system The purchaser inputs data through a web browser. The system receives the input data from scripts running on a server, and then displays the results of the inference on the user’s browser window. The system components are shown in Figure 4. 5.1 Input and output items Items in three groups are required to be input: (1) information about the recipient (age category, gender, relationship, atmosphere, particular flower) is used to determine the flowers in the arrangement; (2) the purpose of the arrangement and shape specification are used to choose the type and size of the arrangement; and (3) the budget, delivery date, contact address, and destination if different from the contact address are incorporated into the constraints for the arrangement size and types of flowers. Output items consist of a list of flowers in the resulting arrangement, including greens and the shape and size of the arrangement. The flowers are described by their name, flowering time, color name, tone, and hue in PCCS, as well as their type and unit price. 5.2 Algorithm When the buyer inputs data through the Web browser and sends them to the server, the system determines the main flower, type of arrangement, combined flowers, and greens added, in this order, according to the expertise and inference rules described above. After that, the system processes the output data for display on the web browser. We do not consider the wrapping process, which is performed by an actual florist. Snapshots of the screen are shown in Figure 5.

Figure 4. Components of our system


Figure 5. Snapshot of the screen. Left: input, Right: output

5.3 Application examples Example 1. A pop arrangement for a birthday present: . Input. The buyer wanted a “pop” arrangement for a female friend on her 23rd birthday, September 15, delivered directly to her house. The type of arrangement was requested to have a general spreading shape, and there was no special height requirement. The budget was 2,000 yen. . Output. A spray-shaped arrangement composed of an orange garbera daisy as a main flower with brown-orange-yellow combination flowers was chosen (Figure 6, Table I). Garbera is a “pop” autumn flower, and the orange color is also considered “pop”. Example 2. An adult arrangement for a farewell party: . Input. The buyer wanted to carry an adult arrangement to a farewell party on March 10 for a superior who was a 51-year-old man. The budget was 3,000 yen.

Figure 6. Image of the output arrangement (example 1)

TQM 20,4

Role

Flower

Color

PCCS code

Main Combination

Garbera Garbera

Orange Orange Yellow Orange Yellow Brown Orange Yellow Yellow Yellow Green Green Green

v5 v7 v8, lt8 v5, b6, v7 b8 dk4 v7 v8 v8 lt8 v12 v10 v10

346

Chrysanthemum (small) Chrysanthemum (spray) Cosmea

Table I. Contents of the output arrangement (example 1)

Aster Scabiosa Eucalyptus Doudan-azalea Asparagus-pera

Leaves

.

Table II. Contents of the output arrangement (example 2)

Figure 7. Image of the output arrangement (example 2)

Output. A medium-sized round arrangement was chosen, with a fringed purple tulip as a main flower. All the flowers had a purple tone gradation. Both the fringed tulip and the purple color correspond to “adult.” Expensive flowers were used by the selection rule to make an attractive arrangement (see Table II). Easy-to-scatter flowers were excluded. A rough image of the arrangement is shown in Figure 7.

Role

Flower

Color

PCCS code

Main Combination

Tulip (fringe) Tulip (lily) Tulip (double) Anemone Rose (standard) Rose (spray) Gentian Galax Asparagus (million) Ivy

Purple Purple Purple Purple Purple Purple Purple Green Green Green

lt20 lt20 lt20 lt22, v22 lt20, lt22 lt20, lt22 v20 v11 v10 v14

Leaves

6. Conclusions We proposed an inference system to integrate kansei engineering based on flower evaluation tests and constraints of gift flower arrangements. The output arrangements were approved by the assessment participants, who had experience presenting flowers. This study provides a means to develop an e-commerce site that will propose products to satisfy the buyers’ needs, while most current online shops merely allow buyers to choose ready-made goods. References Efron, B. (1985), “Bootstrap confidence intervals for a class of parametric problems”, Biometrika, Vol. 72, pp. 45-58. Felsenstein, J. (1985), “Confidence limits on phylogenies: an approach using bootstrap”, Evolution, Vol. 39, pp. 783-91. Ishihara, S., Ishihara, K. and Nagamachi, M. (1999), “Analysis of individual differences in Kansei evaluation data based on cluster analysis”, KANSEI Engineering International, Vol. 1 No. 1, pp. 49-58. Ishihara, S., Ishihara, K., Nagamachi, M. and Matsubara, Y. (1995) Vol. 1995, “arboART: ART-based hierarchical clustering and its application to questionnaire data analysis”, Proceedings of the 1995 IEEE International Conference on Neural Network, Perth, pp. 532-7. Iwashita, N. and Ito, M. (2005), Kurashi no Ehon – Okurikata no mana- to kotsu (Picture book for the daily life – manners and tips of the presents), Gakushu-kenkyu-sha, Tokyo (in Japanese). Nagamachi, M. (1991), “An image technology expert system and its application to design consultation”, International Journal of Human-Computer Interaction, Vol. 3, pp. 267-79. Shimodaira, H. (2002), “An approximately unbiased test of phylogenic tree selection”, Systematic Biology, Vol. 51, pp. 492-508. Shiraishi, S. (1993), Flower arrangement – hana no atsukaikata kara bu-ke made, Vo-gu kiso siri-zu (From handling flowers to bouquets, Vogue basic series), Vogue Japan, Tokyo (in Japanese). Further reading Nagamachi, M. (1995), “Kansei engineering: a new ergonomic customer-oriented technology for product development”, International Journal of Industrial Ergonomics, Vol. 15 No. 1, pp. 3-11. Soudo-Shuppan (2002), Hana-zukan [Kiribana], Soudo Hana-zukan siri-zu, zouho kaitei-ban, rev ed., (Picture book of flowers [Cut flowers], Soudo Picture Book of Flowers Series), Soudo-Shuppan, Tokyo (in Japanese). Corresponding author Keiko Ishihara can be contacted at: [email protected]




TQM 20,4

Effect of smartphone aesthetic design on users’ emotional reaction

348

An empirical study Parul Nanda Research in Motion, Mississauga, Canada, and

Jeff Bos, Kem-Laurin Kramer, Catharine Hay and Jennifer Ignacz Research in Motion, Waterloo, Canada Abstract Purpose – This paper discusses the impact of aesthetic design of smartphones on users’ emotional reactions and preferences towards the product. To this end, the paper presents a study that explores emotional reaction of males to varying aesthetic design of the BlackBerry and empirically evaluates their preferences for the BlackBerry in different colours and overlay patterns. The paper then presents the statistical results of the study in an innovative graphical representation. Design/methodology/approach – A quantitative and qualitative research design was used, including three types of data-collection instruments (direct observations, rating scales, and interviews) to investigate if males have a stronger positive emotional reaction for visually treated BlackBerry Pearl devices over the original treatment (piano black) of the BlackBerry Pearl. A one-way analysis of variance (ANOVA) was carried out with an independent within subjects variable “Pattern” with ten different levels (i.e. ten different visual treatments). Findings – The study indicates that varying the aesthetic design of the BlackBerry Pearl has an impact on emotional reaction of males. However, it was found that males in this population sample prefer the original, piano black treatment of the BlackBerry Pearl over the visually treated versions of the smartphone. Participants reported significantly higher scores for the original treatment of the smartphone, piano black (mean ¼ 5.5) than for other visual treatments such as skittles (mean ¼ 2.8). Originality/value – The paper gives an insight the mobile phone industry and the effect that phones have had on people, who see them as a fashion accessory, as well as a communicating tool. Keywords Mobile communication systems, Clothing and accessories, Product design, Consumer behaviour Paper type Research paper


1. Introduction Smartphones have gained popularity as a communication tool providing users with “smart” functionalities of both mobile phones and Personal Digital Assistants (PDA). These devices have become an integral part of the everyday life of their users; they are are not merely computational devices but also personal expressions of the users’ lifestyle (Castells, 2006). Thus, providing only smart functionalities (Va¨na¨nen-Vaino-Mattila and Ruuska, 2000) through usable interfaces (Monk et al., 2002) does not translate into a competitive advantage for smartphone manfacturers (Lindstrom, 2005). In order to build a brand image, it is essential for companies to design smartphones that engage users in an emotional experience (Desmet et al., 2001; Redstro¨m, 2001; Gobe´, 2001). Emotional reactions are different from cognitive processes. The emotional system makes judgments and quickly helps in

determining which things in the environment are dangerous or safe, or good or bad (Norman, 2004). The cognitive system interprets and makes sense of the world (Norman, 2004). Emotional and cognitive information processing systems are independent of and interdependent on each other and affect human choices and preferences. Despite the reciprocity between the two systems, leading wireless companies until recently have addressed only users’ cognitive needs by providing feature-laden smartphones, but have either marginalized or completely ignored users’ emotional needs (Norman, 2004). In a recent study conducted by J.D. Power and Associates (2006), consumers rated physical design (24 percent) as the most important factor over other factors such as operation (22 percent) and features (20 percent) in their overall satisfaction with wireless handsets. Thus, designing for the optimal experience requires treating users holistically by understanding both emotional and cognitive needs (Sanders and Dandavate, 1999; Csikszentmihalyi, 1991). One way to strike a balance between these two needs is to focus on the visceral design of the device (Norman, 2004). Visceral design engages the human emotional information processing system in which humans react to visual and other sensory aspects of a product that can be perceived before significant interaction occurs (Norman, 2004). Some manifestations of visceral design for smartphones have been achieved by varying the aesthetics of the industrial design of the devices (BlackBerry Pearl, 2007; Yun et al., 2003; Hallna¨s and Redstro¨m, 2002; see Figure 1). Research indicates that better aesthetics lead to users’ perception of increased usability (Tractinsky, 1997). Aesthetically pleasing designs also help to build a strong brand sense by influencing the most important human sensory organ, the eyes and consequently extending users’ emotional attachment to the device (Vincent, 2005; Lindstrom, 2005).

Effect of smartphone design 349

Figure 1. Examples of varying colours of the BlackBerry Pearl

TQM 20,4

350

Mobile phones in different colours and applique´s have provided users with personalized options of these devices and transformed them from a functional tool to a fashion accessory (Research and Markets, 2005; Nokia, 2005; see Figure 2). Industry trends reveal that such fashionable mobile phones are common among females. However, the current trends do not give much insight into male preferences for such phones. This paper investigates emotional reaction of males to varying aesthetic design of the BlackBerry Pearl, a leading wireless smartphone solution (BlackBerry Pearl, 2007). Further, the paper empirically evaluates male preferences for the BlackBerry Pearl in different colours and overlay patterns. For the purpose of this paper, the term emotion is operationalized to refer to users’ preferences based on instinct rather than intellect (i.e. the observer loves or hates the design at first glance), and does not necessarily address the effect of design change on emotional state or mood (i.e. the design makes the observer feel happy or sad). 2. Research methodology This section describes the empirical study that was conducted to explore the emotional reaction of males to varying the aesthetic design of the BlackBerry Pearl. The study was intended to investigate if males have a stronger positive emotional reaction for visually treated Pearl devices over the original treatment (piano black) of the Pearl. 2.1 Design A quantitative and qualitative research design was used, including three types of data-collection instruments (direct observations, rating scales, and interviews). The design was used to help cross-validate the different types of findings.

Figure 2. Examples of varying applique´s of Nokia phones – L’Amour Collection

2.2 Participants In total, 50 males, aged 19 to 43, from a leading smartphone manufacturing company participated in this study. All participants were experienced smartphone users. 2.3 Procedure The industrial design of the BlackBerry Pearl was modified by changing the faceplates with respect to colour, image overlay, pattern overlay, and jeweled overlay. The participants were shown ten different visually treated images of the smartphone for four seconds each. These visual treatments were selected based on colours and patterns that are currently popular in mobile phones and other small wireless devices. Immediately after looking at the image, participants were asked to assign a value for each visual treatment on a seven-point scale for their emotional reaction to the visual aesthetics of the device (e.g. 1 for “Hate it” and 7 for “Love it”). The users were also interviewed at the end of the session.


3. General results A One-Way Analysis of Variance (ANOVA) was carried out with an independent within subjects variable “Pattern” with ten different levels (i.e. ten different visual treatments). The results indicate a significant difference for the main effect of Pattern (F(9, 441) ¼ 28.148, p , 0.05). Participants reported significantly higher scores for the original treatment of the smartphone, Piano Black (Mean ¼ 5.5) than for other visual treatments such as Skittles (Mean ¼ 2.8). Multiple comparisons using post-hoc Tukey tests for Pattern ( p , 0.05) also indicated participants rated Piano Black significantly higher than Light Blue (Mean ¼ 4.7), Green Swirl (Mean ¼ 4.4), Fractal Shell (Mean ¼ 3.7), Orange Flower (Mean ¼ 3.7), Fuchsia (Mean ¼ 3.5), and Gold Bling (Mean ¼ 3.1) (Figure 3).

Figure 3. ANOVA results for the main effect of PATTERN

TQM 20,4

352

4. Discussion The study indicates that varying the aesthetic design of the BlackBerry Pearl has an impact on emotional reaction of males. However, it was found that males in this population sample prefer the original, Piano Black treatment of the Pearl over the visually treated versions of the smartphone. Research indicates that current graphical representations of Tukey’s method for multiple comparisons have several limitations (Hsu and Peruggia, 1994). This paper presents an innovative way of representing the Tukey results in the form of a matrix. Figure 4 indicates the preferred patterns identified according to post-hoc Tukey multiple comparisons. Ratings for patterns listed across the top of the matrix are compared to ratings for patterns listed across the left side of the matrix. Patterns that are rated significantly higher according to the Tukey analysis are graphically represented in the matrix. Results indicate the Piano Black pattern was preferred to all other patterns except Red and Dark Blue. 5. Conclusion and future work In this paper, a significant effect of Pattern on emotional reaction was found for male smartphone users. The participants preferred the Piano Black, Dark Blue, and Red colour treatments to the other patterns on the device form factor. The findings here confirm earlier empirical studies on the impact of aesthetics on user perception of mobile phones. More research is required to investigate the impact of specific colour combinations, treatments, and textures on user emotional reaction to smartphones. The research conducted to generate the data was limited due to time and resource constraints when planning the study: . Owing to concerns regarding participant confidentiality and intellectual property, the population sampled only included employees of a leading smartphone manufacturing company. This data sample limits the ability to generalize the research findings to the larger population of male smartphone users. . A limited sample of patterns and textures was selected to gather feedback regarding the emotional reaction of smartphone users. The colour palette selected did not represent every colour in the colour wheel. Further, several patterns contained multiple colours. . The decision to use a single bipolar rating scale limited the scope of the analysis to describe one single dimension of the smartphone user emotional reaction. A Rough Set Analysis approach (Du¨ntsch and Gediga, 2000 could yield additional data regarding user preferences for colour and patterns by identifying dependencies between smartphone attributes, reducing attribute descriptions, analyzing attribute significance, and generating decision rules describing smartphone user preferences. The proposed mitigation for the identified limitations for future research would be to involve representative participants from the general smartphone user population and to systematically present a more comprehensive range of colour patterns and textures. A conventional Kansei engineering approach could then be applied involving multiple smartphone form features to generate Kansei words and decision rules to help


Figure 4. Results matrix

TQM 20,4

354

designers identify the aesthetic preferences of smartphone owners (Fuqian et al., 2007). The Kansei approach can provide the advantage of identifying detailed design recommendations based on specific design criteria that are meaningful to the user population. The Kansei approach can also be used to identify rating criteria and complex relationships between design elements which may not be easily identified using other statistical methods.

References BlackBerry Pearl (2007), available at: www.blackberrypearl.com Castells, M., Fernandez-Ardevol, M., Qiu, J.L. and Sey, A. (2006), Mobile Communication and Society: A Global Perspective, The MIT Press, Cambridge, MA. Csikszentmihalyi, M. (1991), Flow: The Psychology of Optimal Experience, Harper, New York, NY. Desmet, P.M.A., Overbeeke, C.J. and Tax, S.J.E.T. (2001), “Designing products with added emotional value: development and application of an approach for research through design”, The Design Journal, Vol. 4, pp. 32-47. Du¨ntsch, I. and Gediga, G. (2000), Rough Set Data Analysis: A Road to Non-invasive Knowledge Discovery, Methodos Publishers, Bangor. Fuqian, S., Shouqian, S. and Jian, X. (2007), “Association rule mining of kansei knowledge using rough set”, Proceedings of the 2nd International Conference of Fuzzy Information and Engineering (ICFIE), Springer, New York, NY, pp. 949-58. Gobe´, M. (2001), Emotional Branding: The New Paradigm for Connecting Brands to People, Allworth Press, New York, NY. Hallna¨s, L. and Redstro¨m, J. (2002), “From use to presence: on the expressions and aesthetics of everyday computational things”, ToCHI, Vol. 9 No. 2, pp. 106-24. Hsu, J.C. and Peruggia, M. (1994), “Graphical representations of Tukey’s multiple comparison method”, Journal of Computational and Graphical Statistics, Vol. 3 No. 2, pp. 143-61. J.D. Power and Associates (2006), US Wireless Mobile Phone Evaluation Study, available at: www.jdpower.com/corporate/news/releases/pdf/2006251.pdf Lindstrom, M. (2005), BRAND Sense: Build Powerful Brands through Touch, Taste, Smell, Sight, and Sound, Free Press, New York, NY. Monk, A., Hassenzahl, M., Blythe, M. and Reed, D. (2002), “Funology: designing enjoyment”, Proceedings of Conference on Extended Abstracts on Human Factors in Computer Systems, Minneapolis, Minnesota, pp. 924-5. Nokia (2005), available at: www.nokia.com/NOKIA_COM_1/About_Nokia/Press/Press_Events/ zz_lamour/lamour_collection_press_release.pdf Norman, D. (2004), Emotional Design: Why We Love (Or Hate) Everyday Things, Basic Books, New York, NY. Redstro¨m, J. (2001), “Designing everyday computational things”, doctoral dissertation, Go¨teborg University, Go¨teborg. Research and Markets (2005), Fashion and Style in the Mobile Handset Industry, available at: www.researchandmarkets.com/reportinfo.asp?report_id ¼ 302177 Sanders, E.B.-N. and Dandavate, U. (1999), “Design for experiencing: new tools”, in Overbeeke, C.J. and Hekkert, P. (Eds), Proceedings of the 1st International Conference on Design and Emotion, Delft University of Technology, pp. 87-91.

Tractinsky, N. (1997), “Aesthetics and apparent usability: empirically assessing cultural and methodological issues”, Proceedings of the CHI 97, Los Angeles, CA, April 18-23, pp. 115-22. Va¨na¨nen-Vaino-Mattila, K. and Ruuska, S. (2000), “Designing mobile phones and communicators for consumers needs at Nokia”, in Bergman, E. (Ed.), Information Appliances and Beyond: Interaction Design for Consumer Products, Morgan Kaufmann, San Mateo, CA, pp. 169-204. Vincent, J. (2005), “Emotional attachment to mobile phones: an extraordinary relationship”, in Hamill, L. and Lasen, A. (Eds), Mobile World: Past Present and Future, Springer, New York, NY. Yun, M.H., Han, S.H., Hong, S.W. and Kim, J. (2003), “Incorporating user satisfaction into the look-and-feel of mobile phone design”, Ergonomics, Vol. 46 Nos 13-14, pp. 1423-40. Corresponding author Parul Nanda can be contacted at: [email protected]




TQM 20,4

Multi-modal visual experience of brand-specific automobile design

356

Affect – The Centre for Affective Design Research, Institute of Design for Industry and Environment, Massey University, Wellington, New Zealand

Anders Warell

Abstract Purpose – This paper presents a questionnaire study of brand-specific perceptions of automotive design using subjective rating methods. The purpose of the paper is to explore the multiple modalities of the visual product experience of automobile design as perceived by the general public. Furthermore, the experiences were analysed using a framework for visual product experience (VPE). Design/methodology/approach – Respondents were asked to assess the design of two car models at an international car show in relation to brand perceptions and visually perceived attributes using, among other tools, visual analogue scales. Analysis was done using a qualitative technique. Findings – Results from the study indicate that there is a correlation/relation between experiential modes, in that respondents tended to rate attributes consistently high or low across modes. This implies that if the aesthetics are not perceived as favourable, neither is the expression of the car. Furthermore, respondents’ assessments of aesthetic appeal and expression are on an average strikingly similar, suggesting that the level of aesthetic appeal correlates with the level of semantic understanding of the design. The general rating of emotional response follows a similar consistent pattern for the two studied cars. Originality/value – Study approach as a way to gain insights into subjective perceptions of products based on appreciation and interpretation of visual product form. VPE framework recognising, mapping and clarifying the multiple modes of the visual experience. Keywords Automotive industry, Brand image, Marketing strategy, Customer satisfaction Paper type Research paper

Introduction It is widely recognised that the role of visual design for creating pleasurable and appealing product experiences is ever increasing. In fact, Mano and Oliver (1993, p. 452) report that the dimension of “hedonic” or “aesthetic” performance, which includes the valuation of products for their intrinsically pleasing properties, is one of the two major dimensions of product relevance; the other being the notion of instrumental or utilitarian performance. Consequently, the understanding users’ experience of product form has become widely recognised as essential to ensure product success. This is true especially for saturated and mature market segments such as the automotive industry, where advantages to be gained over competitors in terms of, e.g. price or product reliability are likely to be marginal (Jordan et al., 1996). Cornet and Krieger (2005) have shown that aesthetic and identity related factors such as “exterior styling”, “interior styling”, The TQM Journal Vol. 20 No. 4, 2008 pp. 356-371 q Emerald Group Publishing Limited 1754-2731 DOI 10.1108/17542730810881348

The author would like to acknowledge the work done by Johanna Pa˚hlstorp and Diana Wang in collating and editing the responses from the questionnaire study. The support provided by Affect, the Centre for Affective Design Research, is also gratefully acknowledged.

“trendy” and “makes me feel attractive” are among the ten most important purchasing criteria for mid-sized sedans in Japan, Germany and the USA. Aspects such as aesthetic appeal, emotional response, brand impression and expression are heavily influenced by product appearance, and thus of main concern for automotive manufacturers today. The study of user needs has traditionally focussed on requirements related to utilitarian, functional attributes such as performance or features through approaches such as Quality Function Deployment and User Centred Design. However, work in a number of fields has acknowledged and explored the role of hedonic and experiential aspects of products for pleasure or displeasure in product use. For example, in marketing research, Hirschman and Holbrook (1982, p. 92) defined hedonic consumption as “those facets of consumer behaviour that relate to the multisensory, fantasy and emotive aspects of product usage experience”. While including hedonic aspects such as emotional arousal and product symbolism, they moved on from the object-focussed, instrumental or utilitarian stream of marketing studies into the multi-sensory subjective experience, focussing on feelings and meaning. Mano and Oliver (1993, p. 451) emphasise the interrelationship between product satisfaction and product-elicited emotions, acknowledging the importance of the subjective, experiential response for product satisfaction. In the field of affective ergonomics, Jordan (1998, p. 25) argued for the importance of making products not only usable, but also pleasurable to use. Jordan found that properties of products that are salient in terms of influencing the level of pleasure/displeasure with a product include aesthetics, apart from traditional aspects such as features, usability, performance and reliability. Authors in a number of fields, from psychology (e.g. Norman, 2004) to design have offered a variety of viewpoints which provide insights into the complex nature of product experience and its relation to, e.g. meaning (Vihma, 1995), formal aesthetics (Muller, 2001; Warell, 2001), emotions (Desmet, 2002) and brand perception (Karjalainen, 2004). This paper proposes a method for the study of user perceptions of visual product appearance using subjective rating methods. A questionnaire study was employed to obtain qualitative and quantitative data, which was analysed using the VPE framework. The framework of visual product experience The visual product experience (VPE) framework is an emerging model for studies of perceived experiences of product design. The term “perceived” as used in this study refers to the visually based aesthetic experience, including the pleasurable hedonic aspects, the meaningful interpretation, and the emotional response to that experience. The perceived experience is different to the experience of utilitarian values, which is related to engaging in product interaction for functional, utilitarian, and task-oriented objectives. The VPE framework offers a theoretical model and a methodical approach which acknowledges, maps and relates the various modes of this complex multi-modal visual product experience. The term “multi-modal” here refers to visual stimuli giving rise to a variety of experiences related to the perception, understanding and judgement of the product. While the VPE framework focuses on the visual experience; the general framework is universally applicable and valid for any sensory experience considered. Thus, the framework would in principle apply for the senses of, e.g. smell, hearing and touch.

Visual experience of automobile design 357

TQM 20,4

358

Figure 1. The visual product experience (VPE) model, showing the core modes of sensory, cognitive and affective experiences (centre)

According to Hekkert (2006), the concept of experience can be defined as “the entire set of effects that is elicited by the interaction between a user and a product, including the degree to which all our senses are gratified (aesthetic experience), the meanings we attach to the product (experience of meaning), and the feelings and emotions that are elicited (emotional experience)”. Building on the work of Hekkert, the experience is in the VPE model seen as being composed of a number of modes, including a sensory, a cognitive and an affective experience (see Figure 1). Affect as used in this context refers to any emotion or subjectively experience feeling, or the involvement of such processes (Schu¨tte, 2005, p. 15). Furthermore, Vihma (1995) states that products are perceived in two ways; as presentations of themselves, and as representations of something else. While Vihma makes a distinction between the aesthetic and the semiotic part of the experience, another view represented by, e.g. Osgood et al. (1957, p. 28), includes the meaning of the form in the notion of aesthetics. In the VPE model, this division is made explicit as two dimensions of the visual experience as a way to enable a more distinctive analysis of product perception. The dimension of presentation is the “pleasurable” and “hedonic” part of the experience (Hirschman and Holbrook, 1982) and has been treated by, e.g. Muller (2001), Warell (2001), and Akner-Koler (2007). The dimension of representation is the “meaningful” part of the experience. Distinctly different from presentation, representation is dependent on semiotic interpretation (Pierce, 1931-1966), which has strong socio-cultural implications for product perception. With its origins in the 1950s (Hirschman and Holbrook, 1982) and the notion of “symbolic consumption” (Gardner and Levy, 1955; Levy, 1959, 1963), this side of product perception has more recently been treated by many authors, e.g. Krippendorff and Butter (1984), Vihma (1995), Mono¨ (1997), Karjalainen (2004), Warell et al. (2006). In the VPE model, the presentational and representational qualities of the visual product appearance are seen as being “two sides of the same coin”. In perception, the two dimensions are intimately intertwined and impossible to distinguish at a phenomenological level, however, they are fundamental for understanding the various aspects of product design that need to be considered when designing an appropriate, appealing and meaningful product. Each of the two dimensions in turn consists of three modes, acknowledging and reflecting a visual experience based on sensory, cognitive and affective modes of perception (Hekkert, 2006). The modes are briefly described in Table I (see also Warell et al., 2006; Warell, 2006, 2007). In the VPE model, the impression and recognition modes map against the sensory part of the experience. In the same manner, appreciation – comprehension and emotion – association map against the cognitive

VPE dimension

VPE mode

Description

Presentation: The hedonic, pleasure-based experience of the product

Impression

Perceiving and distinguishing the product amongst other products through characteristic design elements and/or overall form and product gestalt Approval of formal aesthetic values, acknowledgement of aesthetic appeal of design elements, compositions and structure in visual product appearance Experience of feelings elicited through an appraisal based on concerns and appearance of design elements and/or overall form and product gestalt Perceiving and identifying a product or design element through meaning making based on resemblance and similarity to previously encountered visual references Grasping the nature, significance, or meaning of product specufuc characteristics such as properties, performance and mode-of-use through visual references of a product or design element Creation of meaning and value concepts through the linking of socio-culturally based notions or ideas, such as brand and cultural references, with a product or design element

Appreciation

Emotion

Representation: The meaningful, semiotically-based experience of the product

Recognition

Comprehension

Association

and affective experiential modes, respectively. Furthermore, the VPE model acknowledges the highly complex interaction and the temporal ambiguity between all modes. Together, this multi-modal approach provides a complete picture of all possible types of experiences based on the visual perception of product design. Measuring visual experience of product form Karlsson et al. (2003, p. 1409) recognise the importance of being able to measure the perceived experience of a product. From a manufacturer’s point-of-view, reasons include that a pleasant product is easier to sell than an unpleasant one. Furthermore, Karlsson et al. state that there is a risk that utilitarian values, which are traditionally easier to measure, gain a disproportionate focus compared to “emotional” user needs. Despite this fact, however, there are few standardised methods available to measure perceived values. Karlsson et al. (2003, p. 1412ff) provide an overview of methods available for assessment of perceived experience, including Semantic Environment Description or SMB (Ku¨ller, 1975), Kansei engineering or KE (Nagamachi, 1995, 1999), and Product Semantic Analysis or PSA (Wikstro¨m, 2002). Common for all approaches is the use of semantic differential scales; polar rating scales with anchors (endpoints) of differentiating meaning, and a continuous or discrete ordinal scale between these anchors. Since the semantic differential measures connotative meaning more readily than denotative aspects of meaning, semantic differential scales are ideally suited for the measurement of aesthetic aspects (Osgood et al., 1957, p. 290).


Table I. Dimensions and modes of the visual product experience (VPE) framework

TQM 20,4

360

While these methods have slightly different origins they are designed for the same basic objective; i.e. to assess and measure the perception of designed products or environments. Also, they exhibit slightly varying methodological approaches. The method employed to study the visual product experience in this research shares some of the characteristics of the other methods but is also distinct in some ways: . VPE is a generic method in the sense that it is not constrained to use for any specific type of product or environment, in contrast to Semantic Environment Description, which is designed for the assessment of architectural environments. . Similar to PSA, VPE utilises a combination of qualitative and quantitative data collection methods, while the other methods use only quantitative data collection methods. . By providing a categorisation between six experiential modes, the VPE framework recognises that product perception is a multi-modal experience and that the experience of each of these modes is distinctively different. Consequently, each mode is individually assessed and utilises distinct sets of words to acknowledge the individual nature of each mode. Semantic Environment Description and Kansei engineering utilise a variety of adjectives, however the factors are not defined in terms of modality. Product Semantic Analysis focuses on one mode only (the comprehension mode in VPE). . VPE utilises unipolar Visual Analogue Scales, which provide a qualification of the perception which is easily converted to quantitative data for analysis. This is similar to Product Semantic Analysis. Semantic Environment Description uses unipolar seven point Likert scales while Kansei engineering uses a variety of scales. . In VPE, the respondent assesses one adjective at a time. As for Product Semantic Analysis, the interpretation of each adjective is focussed using associative descriptors for each adjective. Semantic Environment Description utilises eight factors which include a number of adjectives that are positively and negatively correlated to the factors. . As for PSA, VPE uses adjectives that are customisable for each specific study. SMB uses a standardised set of factors originally designed for studies of physical architectural environments, which however has found applications in the study of automotive interiors (Laike, 1999; Karlsson et al., 2003). . In VPE as for PSA, interpretation and analysis of semantic word scale assessments is done using qualitative techniques, while the other methods utilise a variety of statistical analysis tools. Laike (1999) states a set of four criteria when adopting the Semantic Environment Description method to measure the impression of vehicle interiors. The criteria were (Karlsson et al., 2003, p. 1412) that: (1) the method should be easy to administrate; (2) have a high reliability and validity; (3) be easily adaptable for cross-cultural comparisons; and (4) provide useful information for design, development and marketing. These criteria are general in nature and are seen as relevant for the VPE method as well.

Design of semantic differential scales The design of semantic differential scales to measure perceptions of the product experience is an important factor for validity of results. A variety of semantic differential scales is found in literature; two of the most common being the Likert type scale (as used by Osgood et al.) and the Visual Analogue Scale (VAS). According to Gould et al. (2002, p. 706), Visual Analogue Scales “is a measurement instrument that tries to measure a characteristic or attitude that is believed to range across a continuum of values and cannot easily be directly measured”. Both types of scales can be either unipolar (i.e. measure a single attribute for a given concept) or bipolar (i.e. measure two opposed attributes for a given concept). Semantic differentials featuring an adjective and its antonym (e.g. happy-sad) were used by Osgood et al. (1957). The problem with this type of scale is that not all adjectives have a singularly defined antonym, and thus may not be interpreted as expected (Wikstro¨m, 2002; Schu¨tte, 2005). As a means to avoid this problem, Ku¨ller (1975) employed a seven-step Likert scale with anchor points titled “slightly” and “very”. Other researchers (e.g. Wikstro¨m, 2002; Schu¨tte, 2005) have employed similar approaches. Specifically, Wikstro¨m (2002, p. 43ff) uses three steps in the design of semantic word scales: (1) Step 1: Identifying adjectives. Through focus group interviews, adjectives are identified and related to desired and undesired expressions of the product, forming the basis for the semantic word scale. Scenario methods and a simplified version of KJ analysis are used to systematically identify adjectives. (2) Step 2: Categorising adjectives. Wikstro¨m collated words that were given the same or a similar meaning by the focus group. Some words were changed into antonyms in order to achieve a word scale that was not only composed of adjectives with positive connotations. In order to align the interpretation of each adjective by respondents, each selected adjective was presented together with an associative word. (3) Step 3. Design of assessment tool (visual analogue scales). Wikstro¨m uses a VAS where respondents assess the expression word (an adjective) on a scale 100 mm in length, ranging from “maximally” (at 0 mm) to “the opposite” (at 100 mm). The mid-point of the scale is indicated by a “0” (zero, at 50 mm). According to Wikstro¨m (2002, p. 47), an assessment of more than 70 or less than 30 indicates that the product has a strong expression of the assessed adjective (or its opposite). Values from 45 to 55 indicate the inexistence of the expression, while other values indicate a weak expression. In this study, the overall approach of Wikstro¨m for the design of semantic word scales was used. However, due to the specific objectives and the nature of the study, there are a few noticeable differences. These are specifically related to the fact that a representative user group was not available beforehand, and therefore the identification and categorisation of adjectives could not be informed by users, e.g. through focus groups. Instead, the identification of adjectives is based on brand specific promotional and marketing material. Adjectives relating to both expressive and formal aesthetics properties used on web site presentations of the two cars were collated and grouped into words with the same or similar meaning. Similar methods of identifying expressive adjectives have been used by, e.g. Karlsson and Wikstro¨m (1999) and Opperud (2000) with good results.


TQM 20,4

362

However, the expressive mode studied by Wikstro¨m is only one of the modes studied in VPE. Of the six VPE modes, the modes studied in this research included comprehension (semantic expressions), appreciation (aesthetic appeal), and emotions (evoked feelings). Thus, adjectives were sought that related to the distinctively different nature of these three modes. For the comprehension mode, this meant collating adjectives that were relevant for the specific products studied. Naturally, the expressions for Saab would be different from the expressions for Volvo, as identified from brand specific material. For the appreciation mode, an identical group of adjectives were used for both cars in order to objectively assess formal aesthetic appeal. The same approach was used for emotions, which featured five sets of bipolar emotional pairs. This approach taken for the appreciation and emotion modes is theoretically sound as they both belong to the hedonic and pleasure related dimension of presentation, which is unrelated to brand-specific subjective interpretation (the representation mode). Furthermore, a slightly different design of VAS, similar to Schu¨tte (2005, p. 73) and Ku¨ller (1975) was adopted. The VAS used in this study uses unipolar analogue scales measuring 100 mm in length, where respondents measure each adjective on a scale with the left anchor titled “Not at all” (at 0 mm), and right anchor titled “To a great extent” (at 100 mm). This type of scale avoids the risk of ambiguity when interpreting the opposite of each adjective, which may enter as a factor in Wikstro¨m’s (2002) and Osgood’s et al. (1957) scales. However, it does not pick up if respondents perceive the expression to be opposite. Instead, such opposed adjectives have to be deliberately included through separate VAS scales. Also, Schu¨tte (2005, p. 73) notes the problem that “the extremes are sometimes considered to be indefinite which in turn means that the distances are not considered to be completely equal”. A further complication with Wikstro¨m’s scales is that it is not known what meaning respondents assign to the term “the opposite”, which, furthermore, will be different from subject to subject. This complexity is avoided with the approach taken in this study. Methodology The purpose of the study was to explore visual aesthetic appeal, evoked emotional response and meaning of product form as interpreted by the respondent group. The study was carried out during two days at an international automotive exhibition[1]. Two selected cars, the Saab Aero-X concept car, and the Volvo C30 production model[2], were studied through a questionnaire study with attending exhibition audience. The questionnaire employed a combination of qualitative and quantitative questions, including categorical multi choice questions, open-ended questions and a combination of these, as well as Visual Analogue Scales (VAS). Respondents were approached at each respective stand and asked to answer the questions of the questionnaire in the presence of the selected cars. If required, respondents were guided through the questionnaire by the interviewer, who clarified the meaning if any uncertainties occurred. The average time spent per respondent was around 15 minutes. In all, 18 respondents were interviewed; ten at the Saab exhibition stand, and eight at the Volvo stand. The questionnaire was divided into five sections according to the following: (1) Part 1 covered questions related to the respondent and their relation to cars in general, such as respondent’s interest and knowledge about cars, age, country of origin, and car purchase decision making. Answers were given using

(2)

(3)

(4)

(5)

five-point Likert scales and multi choice tick-box replies, with the opportunity to provide open-ended qualitative responses where applicable. Part 2 covered questions relating to the specific brand of car studied, such as brand impression, distinctiveness and typicality of the car for the brand and in relation to other brands, brand-specific features, and desirability for driving or owning. Answers were given through a combination of five-point Likert scales, multi choice tick-box replies and/or as open-ended qualitative responses. Part 3 assessed a respondents’ appreciation of the “formal aesthetics” of the product, i.e. how appealing the design is to the visual sense. Visual analogue scales (VAS), featuring a line 100 mm in length, were used for respondents to answer the question “What do you think about the car’s looks?”. The left VAS anchor was labelled “Not at all”, while the right anchor was labelled “To a great extent”. Respondents indicated their assessment by marking on the VAS line the degree to which they appreciated a range of nine visual formal attributes describing the appearance. Part 4 focused on emotional response evoked by the studied model. In line with approaches proposed by, e.g. Desmet et al. (2003), respondents answered the question “How does this car make you feel?”. Five visual analogue scales each featuring a set of diametrically opposed emotional states were utilised to measure emotional response. Each bipolar VAS featured a neutral mid-point indicated with a “0”, and one emotional state at each anchor. Part 5 studied expressions elicited from the visual appearance of the car. In the questionnaire, respondents were given a set of eight expressive adjective terms and asked to respond to the question “How is this car to you?”. Respondents indicated on visual analogue scales, ranging from “Not at all” at the left anchor to “To a great extent” at the right anchor, to what extent the product is perceived to express these properties. The expressive terms were further defined by associative descriptors intended to align the interpretation of the terms across respondents.

Findings In the following, the results are presented according to the structure of the study questionnaire. Part 1: Respondent details The respondent groups for the two brands exhibit similar characteristics with respect to their relation to cars, although the ten surveyed Saab subjects indicated a slightly higher interest in cars than the Volvo group (an average of 4.2[3] and 3.6 of a maximum of 5, respectively). The perceived importance of the choice of car was rated equally high at 4.8[4]. Part 2: Brand associations and aspirations Respondents’ general appreciation of the two brands showed some interesting findings. When asked how “distinctive/clear” each brand is, respondents rated the two brands very similarly at 3.8[5] for Saab and 3.9 for Volvo. When asked to qualify responses, comments for Saab as a brand included that the “design is easily


TQM 20,4

364

recognizable” and that it is a “high level brand”. For Volvo, comments included “immediately characteristic”, “big”, “used to be square”, and “coherence”. However, one respondent commented that Volvo as a brand “doesn’t surprise anymore”. Shifting focus to the specific models surveyed, respondents were asked how “typical” they found the Saab Aero X and the Volvo C30 for each respective brand. Interestingly, the rating was identical for the two cars; 4.1[6]. When asked how typical the two cars were in relation to other brands, the Saab Aero X was perceived as more typically Saab at 3.7[7], while Volvo C30 at 3.0 was seen as a car that could come from any other brand. When enquired how strongly they would like to see themselves driving or owning the selected cars, the rating for the Saab Aero X was 4.2[8], compared to 3.8 for Volvo C30. Part 3: Appreciation (aesthetic appeal) The assessments of aesthetic appeal of the studied cars are presented in Figures 2 and 3. The Saab Aero X was overall given very high positive scores (Figure 2). The range for most attributes are fairly focused, with the mean consistently towards the top end. The assessments of the formal attributes “Balanced/Proportional” and “Simple/Clean” have a greater range. The attribute “Ugly/Non-appealing”, which at first may be regarded as the opposite of “Beautiful/Stunning”, is clearly assessed in its own right, yielding a significantly greater spread than any other attribute. The assessments for Volvo C30 (Figure 3) follow a similar pattern. However, ratings are not as positive as for the Saab Aero X and the range is overall greater. In contrast to Saab Aero X, the ratings for “Ugly/Non-appealing” exhibit the smallest range, indicating a more consistent experience of this attribute.

Figure 2. Assessment of appreciation (aesthetic appeal) of Saab Aero X, according to nine standardised adjective pairs


Figure 3. Assessment of appreciation (aesthetic appeal) of Volvo C30, according to nine standardised adjective pairs

Part 4: Evoked emotional response In this part of the study, subjects were asked to respond to the question “How does this car make you feel?”. As seen in Figure 4, mean values of Saab Aero X show a clear positive trend. The range for all assessments is quite limited, especially for the emotions “Delighted” and “Attracted”. The emotion “Surprised” exhibits the greatest spread. The assessments of Volvo C30 indicate a more mixed response (Figure 5). Ratings are on the positive side for all emotions except for a negative tendency for “Bored”. The spread is significant for all emotions, especially for the pair “Disappointed-Satisfied”, which exhibits the full range of assessments. Part 5: Comprehension (elicited expressions) Subjects were asked to assess their comprehension of the studied cars in terms of expressions elicited by the visual form, in response to the question “How is this car to you?”. Figure 6 shows the assessments against specific expressions, taken from Saab Aero X-specific promotional material. Overall, expressions are perceived as quite strong, thus indicating a positive relation between desired expressions (from the brand

Figure 4. Assessment of emotional response evoked by the Saab Aero X, against five sets of opposed emotional pairs

TQM 20,4

366

standpoint) and experienced expressions (as perceived by respondents). Only one expression, “Discreet” is assessed significantly lower. Ranges are most confined for the expression “Different”, and most distributed for “Discreet” and “Progressive”. Similarly, the assessment of expressions of Volvo C30 is based on a set of eight Volvo specific adjectives, taken from C30-specific marketing material. All assessments indicate a positive correlation between desired and perceived expressions, the strongest expression being “Stylish” and the weakest “Extreme”, although differences in mean values are small (Figure 7). All expressions show a significant spread; the expression “Exciting” exhibiting the greatest and “Premium” the smallest spread. Discussion Results In the following, the findings will be discussed in relation to the VPE framework (Figure 1). Overall, the range of ratings given by respondents is greater for Volvo C30 than for Saab Aero X across the comparative modes of appreciation and emotion. This

Figure 5. Assessment of emotional response evoked by the Volvo C30, against five sets of opposed emotional pairs

Figure 6. Assessment of expressions elicited by the Saab Aero X, according to eight brand-specific adjectives


Figure 7. Assessment of expressions elicited by the Volvo C30, according to eight brand-specific adjectives

indicates a more mixed response to the design of the C30 than for the Aero X. Moreover, mean ratings are overall lower for Volvo C30 than for Saab Aero X, indicating a higher level of appreciation and a more positive emotional response for the Aero X than the C30. These findings are interesting and may be related to the quite different nature of the two vehicles. While the C30 is a production model, the Aero X is a concept study. As concept cars do not adhere to the same cost and manufacturing constraints as production cars, they offer an opportunity for manufacturers to explore aesthetic solutions that would not have been feasible for production models. Consequently, concept vehicles typically have a stronger expression, a more clearly defined identity, and a bolder formal visual language with respect to, e.g. stance, proportion and differentiation. This fact may be reflected in the higher ratings for the Saab Aero X. The economic reality of manufacturing also tend to force formal solutions to adopt a more bland expression and the utilisation of more mainstream visual solutions, which may be a factor for the greater range in ratings for the Volvo C30 – respondents are more likely to have seen similar solutions before, and find it less appealing and interesting. With respect to brand impression, previous research (Fjellner and Stridsman-Dahlstro¨m, 2006; Warell et al., 2006) has proposed that the associative mode (including notions of brand core values) has a dominating role over comprehension and recognition for product identity. This study suggests a positive relation between the different modes with respect to how respondents rate their visual experience. In both study groups, respondents tended to give either low or high average ratings with respect to the appreciation (aesthetic appeal), emotion and comprehension (expression) modes. However, there was no evident connection to brand awareness; in fact, for the Saab study, two of the respondents giving some of the highest and lowest ratings with respect to these modes claimed they did “not know the Saab brand”. It would be interesting to study whether knowledge of the brand would correlate to higher ratings in these modes, which, if so, would support the findings of previous research.

TQM 20,4

368

Furthermore, for both cars, respondents’ assessments of formal aesthetics and elicited expressions show a striking similarity in terms of average ratings. For Saab Aero X, averages are 80.6 and 79.0; for Volvo C30, 67.1 and 66.6, respectively. The general rating of emotional response follows a similar consistent pattern for the two studied cars. This observation is interesting as it raises the question whether a high level of aesthetic appeal correlates with a high level of semantic understanding (in this case, in the form of elicited expressions). These attributes relate to the appreciation and comprehension modes, respectively, which, according to the VPE framework, both are cognitively based experiences. A fascinating theoretical proposition stemming from this observation is that good product semantics (i.e. a perceived understanding of properties expressed by the product form) cannot be achieved without a high level of aesthetic resolution, e.g. through order and clarity. Moreover, previous research (Warell et al., 2006) suggests that the comprehension mode is essential for brand identification through the use of visual references that support a coherent interpretation of typical brand characteristics and core values. In relation to the current study, this suggests that brand-specific communication is supported by aesthetic appeal and expression – provided visual references employed are perceived as genuine (coherent with brand values) and authentic for the brand (recognisable). Future studies will need to further investigate these issues, as well as the issue whether the presentational (the “pleasurable”, aesthetically related part) or representative dimensions (the “meaningful”, identity related part) of the visual appreciation are generally more highly assessed or significant. Study approach Warell (2001, p. 46) states that the intention with using semantic word scales in combination with VAS was not to provide data for statistical analysis, but rather as a way to determine whether a product has the expressed property or not. The purpose of Wikstro¨m’s study was to measure expressions of products for use in early assessments of design concepts in the product development process. The approach of the method employed in this study has similar characteristics. However, VPE as an approach is unique in the sense that it offers both a theoretical framework for understanding the phenomenon of the multi-modal visual experience, and a methodical approach for empirical study of respondents’ perceptions. While other methods employ either a random set of unstructured adjectives (Kansei engineering), a factor reduced collection of terms for a certain type of environment (Semantic Environment Description), or specific adjectives for one experiential mode only (Product Semantic Analysis), VPE relies on the definition of adjectives that are customised for each specific study. Furthermore, it distinguishes between the six experiential modes as they are considered significantly different and distinct types of experiences. Other approaches do not recognise the variety of experiential modes, which limits the range of the acquired data and the implications for design work. As for the other approaches, the VPE method is based on semantic transformation; the interpretation of perceived experiences into verbal concepts. This approach has specific challenges, which may be circumnavigated by using non-verbal techniques such as visual stimuli in order to avoid subjective interpretation. The issue of interpretation is an added challenge when it comes to multi-cultural studies, which exhibit specific market based interpretations. In the study, a non-verbal technique was

employed in addition to the VAS assessments. Respondents were encouraged to indicate on provided image material what elements of the product form gave rise to certain appreciations, expressions and emotional response. However, as this material was not complete across respondents, it was not included in this paper. As in Wikstro¨m’s (2002) study, the qualitative technique used for the analysis of VAS results is based on each separate adjective independently within each VPE dimension. The benefit of qualitative analysis is that it provides an opportunity for deeper understanding of results. However, it requires that the analysis is done by individuals with intimate knowledge of the material. The drawback of this approach is that it is not known whether adjectives correlate positively or negatively to some common factor, e.g. for aesthetic appeal, adjectives such as “harmonious”, “beautiful” and “sleek” may be positively correlated to a factor. The qualitative approach also limits the ability to draw conclusions relating to correlations between dimensions; for example whether “aesthetic appeal” is correlated to “elicited expression”. Applying quantitative tools for analysis would, sample size permitting, possibly allow for a more complete understanding of study data, and is planned for future studies. Future development of the VPE method will include studies of correlations between adjectives used in the various modes, as well as whether respondents are actually able to discern between the various “origins” of the experience defined by the modes of the VPE framework (impression, appreciation, emotion, recognition, comprehension, and association) through the data gathering methods used. The relation between perceptions and specific form elements will also be studied in more detail. Conclusions The paper contributes to design research in several ways. First, the utilised framework for visual product experience presents a holistic view of the possible modes of visual product design experience – aesthetically as well as meaning-related, whether based on sensory, cognitive or affective experiential modes. Second, the method used in the study recognises, distinguishes between, and addresses the various possible experiential modes. By avoiding “fuzziness”, it contributes to a high degree of clarity to the respondent, as well as a high level of transparency in analysing and interpreting results. For design practice, the approach taken is seen has having a significant potential for operative as well as strategic design management and tasks related to branding, marketing and identity development, due to the fact that multi-modal perceptions are clarified and distinguished. Notes 1. The 2006 Mondial de L’Automobile, Paris, September 30-October 15, 2006. 2. While the Volvo C30 was internationally launched at the Paris exhibition, the Saab Aero X concept car had previously been presented at the 76th International Geneva Auto Show, 2-12 March, 2006. 3. Based on answers from 9 of 10 respondents. 4. Based on answers from 9 of 10 Saab respondents. 5. Based on answers from 8 of 10 respondents. 6. Based on answers from 9 of 10 Saab respondents. 7. Based on answers from 7 of 10 respondents.


TQM 20,4

370

8. Based on answers from 9 of 10 respondents. 9. In the questionnaires, the visual analogue scales used did not feature any value indications. Respondents were asked to qualitatively assess each attribute on an unmarked line with a length of 100 mm.

References Akner-Koler, C. (2007), “Form and formlessness”, Chalmers University of Technology, Gothenburg. Cornet, A. and Krieger, A. (2005), Customer-Driven Innovation Management, available at: www. autoassembly.mckinsey.com.edn Desmet, P.M.A. (2002), “Designing emotions”, doctoral thesis, Technical University of Delft, Delft. Desmet, P.M.A., Hekkert, P. and Hillen, M.G. (2003), “Values and emotions; an empirical investigation in the relationship between emotional responses to products and human values”, Proceedings of the 5th European Academy of Design conference, Barcelona. Fjellner, C. and Stridsman-Dahlstro¨m, J. (2006), “Understanding the identity conveyed by visual product design: a study of perception and interpretation of the brand essence of Saab automobile”, MSc thesis, Chalmers University of Technology, Gothenburg. Gardner, B. and Levy, S.J. (1955), “The product and the brand”, Harvard Business Review, Vol. 33, March-April, pp. 33-9. Gould, D.J., Kelly, D., Goldstone, L. and Gammon, L. (2002), “Examining the validity of pressure ulcer risk assessment scales: developing and using illustrated patient simulations to collect the data”, Journal of Clinical Nursing, Vol. 10, pp. 697-706. Hekkert, P. (2006), “Design aesthetics: principles of pleasure in product design”, available at: http://studiolab.io.tudelft.nl/hekkert/publications (accessed 5 April 2006). Hirschman, E.C. and Holbrook, M.B. (1982), “Hedonic consumption; emerging concepts, methods and propositions”, Journal of Marketing, Vol. 46, Summer, pp. 92-101. Jordan, P.W. (1998), “Human factors for pleasure in product use”, Applied Ergonomics, Vol. 29 No. 1, pp. 25-33. Jordan, P.W., Thomas, B. and McClelland, I.L. (1996), “Issues for usability evaluation in industry: seminar discussions”, in Jordan, P.W. and McClelland, I.L. (Eds), Usability Evaluation in Industry, Taylor & Francis, London, pp. 231-43. Karjalainen, T.-M. (2004), Semantic Transformation in Design, University of Industrial Arts, Helsinki. Karlsson, B., Aronsson, N. and Svensson, K. (2003), “Using semantic environment description as a tool to evaluate car interiors”, Ergonomics, Vol. 46 Nos 13/14, pp. 1408-22. Karlsson, M.A. and Wikstro¨m, L. (1999), “Beyond aesthetics! Competitor advantage by a holistic approach to product design”, paper presented at the 6th EIASM International Product Development Management Conference, Cambridge, July 5-6. Krippendorff, K. and Butter, R. (1984), “Product semantics: exploring the symbolic qualities of form”, The Journal of the Industrial Designers Society of America, Spring. Ku¨ller, R. (1975), Semantisk miljo¨beskrivning (SMB), Psykologifo¨rlaget, Stockholm. Laike, T. (1999), Att ma¨ta upplevelse av bilinterio¨rer, Lund University, Lund. Levy, S.J. (1959), “Symbols for sale”, Harvard Business Review, Vol. 37, July-August, pp. 117-19. Levy, S.J. (1963), “Symbolism and life style”, in Greyser, S.A. (Ed.), Toward Scientific Marketing, American Marketing Association, Chicago, IL.

Mano, H. and Oliver, R.L. (1993), “Assessing the dimensionality and structure of the consumption experience: evaluation, feeling, and satisfaction”, The Journal of Consumer Research, Vol. 20 No. 3, pp. 451-66. Mono¨, R. (1997), Design for Product Understanding, Liber AB, Stockholm. Muller, W. (2001), Order and Meaning in Design, Lemma Publishers, Utrecht. Nagamachi, M. (1995), “Kansei engineering: a new ergonomic consumer-oriented technology for product development”, International Journal of Industrial Ergonomics, Vol. 15, pp. 3-11. Nagamachi, M. (1999), “Kansei engineering and its applications in automotive design”, SAE paper No. 1999-01-1265. Norman, D. (2004), Emotional Design: Why We Love (Or Hate) Everyday Things, Basic Books, New York, NY. Opperud, A. (2000), “Is the beauty in the eye of the beholder?” (“En semiotisk studie av mobiltelefondesign”), Chalmers University of Technology, Gothenburg. Osgood, C.E., Suci, G.J. and Tannenbaum, P.H. (1957), The Measurement of Meaning, University of Illinois Press, Urbana, IL. Pierce, C.S. (1931) in Hartshorne, C., Weiss, P. and Burks, A.W. (Eds), Collected Papers of Charles Sanders Pierce, Vols 1-8, Harvard University Press, Cambridge, MA. Schu¨tte, S. (2005), “Engineering emotional values in product design – Kansei engineering in development”, doctoral thesis, Linko¨ping University, Linko¨ping. Vihma, S. (1995), “Products as representations: a semiotic and aesthetic study of design products”, dissertation, University of Art and Design, Helsinki. Warell, A. (2001), “Design syntactics to a functional approach to visual product form. Theory, models, and methods”, dissertation, Chalmers University of Technology, Gothenburg. Warell, A. (2006), “Identity recognition in product design: an approach for design management”, paper presented at the 13th International Product Development Management Conference, Politecnico di Milano, Milan, June 11-13. Warell, A. (2007), “Visual experience of brand-specific automobile design: Studying appreciation, emotion and comprehension using the VPE framework”, paper presented at the 10th QMOD Conference: Quality Management and Organizational Development and 1st European Conference on Kansei/Affective Engineering, Lund University, Helsingborg, 18-20 June. Warell, A., Fjellner, C. and Stridsman-Dahlstro¨m, J. (2006), “Visual product identity: understanding identity perceptions conveyed by visual product design”, Proceedings of the 5th International Conference on Design and Emotion, Chalmers University of Technology, Gothenburg, September 27-29. Wikstro¨m, L. (2002), “Produktens budskap: metoder fo¨r va¨rdering av produkters semantiska funktioner ur ett anva¨ndarperspektiv”, doctoral dissertation, Chalmers University of Technology, Gothenburg. Corresponding author Anders Warell can be contacted at: [email protected]




TQM 20,4

Kansei engineering toolkit for the packaging industry Cathy Barnes, Tom Childs, Brian Henson and Stephen Lillford

372

Affective Engineering Research Group, School of Mechanical Engineering, University of Leeds, Leeds, UK Abstract Purpose – The purpose of this paper is to describe the Kansei engineering toolkit that has been developed to provide a set of tools and techniques to support better packaging design. Design/methodology/approach – The toolkit has its foundations in Kansei engineering but the work has extended the scope and increased reliability of results by: including structured linkages to designers; replacing “highest level Kansei” from Kansei type 1 with brand values; introducing a more structured process for the elicitation of type 2 selection of pack physical properties; reducing the complexity of the semantic differential survey used to elicit consumer perceptions; and structuring a process for selection of the Kansei words. Findings – The work has shown that the proposed toolkit is able to support the design of packaging by illustrating the process with industrial case studies. Research limitations/implications – Kansei engineering and the techniques presented in this toolkit are inevitably simplifications of the real situation, since many more variables affect the consumers purchase decision than is tested in this process. There is still a need to test the insights gained by the toolkit into a wider investigation. Practical implications – This paper offers the packaging industry a robust and repeatable method to develop better packaging. Originality/value – The paper presents an overall description of the Kansei engineering toolkit for packaging design and is a structured process that provides quantitative results for the relationship between branding, consumer perception and design variables. Keywords Packaging, Brand management, Product design, Linguistics, Semantics Paper type Case study

1. Introduction Products can be considered important “touchpoints” by which a company interacts with its consumer. Packaging, media promotion and customer support are other touchpoints that the business communicates and interacts with their consumers to create an overall “experience”. In competitive markets where the consumer has greater power, due to extensive product choice, business goals have shifted to focus on increasing the level of affinity the consumer has with the values of the business. The aim of this is to prolong the interaction (i.e. repeat purchase) and to ultimately become a brand that the consumer will endorse to others. For business to have such brands, they


The work reported in this paper was carried out as part of Knowledge Transfer Partnership No. 6179, to “Apply Kansei engineering to the European Packaging Industry” in conjunction with Faraday Packaging Partnership and PIRA International Ltd. The authors would also like to thank all the participating company members of the Faraday Packaging Partnership who had significant input into the definition of this process. Special thanks must go to the client companies who provided the case studies reported here.

must have a defined and unique personality that relates to consumer values and lifestyle choices, these are described in the “brand essence” which to be successful, must be communicated uniformly across all touchpoints. Packaging, in the context of Fast Moving Consumer Goods (FMCG), has a number of functions. It must protect the product from contaminants, it should contain the product in a dose suitable for purchase and use and finally, it must communicate essential information about the contents to the consumer. In addition, the fourth function of packaging is to market the product at point of sale. This means that the packaging has a fundamental role to play in the transmission of the brand personality to the consumer and in the enticement to purchase. The semiotic perspective of design focuses on viewing products as signs capable of communication (Vihma, 1995). If products are to be considered as signs that are interpreted by users, it is useful to consider consumer response to both product and packaging appearance as one stage in an integrated process of communication. The framework from (Crilly, 2004) breaks down the path between the producer (business and design team) and the consumer into “design”, “product”, “senses” and “response”. This illustrates how effective communication through product design requires an understanding of manipulating the physical characteristics of the product to enhance the sensory input and how this sensation is interpreted by the consumer. As we have seen, this mechanism is arguably of equal (or perhaps in certain circumstances, more) importance for the design of the packaging for consumer products. We have leveraged work from the disciplines of design, marketing psychology and linguistics to develop a decision support toolkit which assists the design of better packaging designs. The toolkit has its roots in Kansei engineering (Nagamachi, 1995), a consumer centred product development process and uses the framework in Figure 1

Kansei engineering toolkit 373

Figure 1. Framework for design as a process for communication

TQM 20,4

374

to help understand the relation between product physical properties and consumers’ response. However, as the original Kansei engineering did not explicitly consider brand values, which we know are critical to successful packaging designs; we report some new methods to extend the scope of the toolkit. We have also supplemented the technique with linguistic expertise to improve the selection process for the adjectives used in the consumer survey which has improved the robustness and repeatability of the results. This paper presents the Kansei engineering toolkit for packaging design by using illustrative case studies. We present the application of the toolkit to “live” projects to show how each tool supports design development. In this case, relationships are constrained to product appearance alone, but other projects have been conducted to explore relationships with other senses (Henson, 2003). This shows that this Kansei engineering toolkit has real value within the packaging development process to inform concept selection decisions based upon actual consumer data. 2. Toolkit for packaging design Traditionally, industrial designers use inspiration and cultural observation to translate feelings into product properties, but these methods pose problems which range from misinterpreting the design brief, to poor support in decision making and a lack of rationale. This in turn can contribute to the poor market performance of many new products. When an ad hoc approach is also used to develop the design and form of the packaging this adds to the overall potential lack of appeal. Thus, this toolkit aims to provide a structured and rational approach to the translation of consumer insight into appealing packaging. It supports the collection of consumer focused attributes and provides methods for empirically quantifying the relationships between the design and the required consumer affect. The process can be used in many ways from defining the final packaging design to screening concepts. The possible implications for the product development process are discussed later. 2.1 Kansei engineering We analysed the different Kansei methods proposed (Nagamachi, 1995) and then focused on the translation of Kansei types I and II because of their applicability and proven success in industry case studies. Kansei engineering type 1 – category classification. In Kansei engineering type I, a product strategy and a market segment is identified through behavioural, lifestyle and values surveys. Insights are turned into descriptions of the intended feeling the user will get from interacting with the product, these are tested for affinity with consumers. Descriptions which are most important to the consumers are used for the product strategy (Nagamachi, 2002). Each feeling description is divided into factors (termed “sub-concepts on a lower Kansei level”) these have been identified as contributing to the feelings. Typically, an expert team expands on each factor until they can be related to physical design properties. The tree diagram in Figure 2 illustrates the principle for the Kansei word “acceleration” for an automobile (Schutte, 2005). However, Schutte noticed that some Kansei words could not be linked to physical properties through team discussion because they can have several interpretations. For

example, “Soft” could be expressed against the construction of the seat, the aesthetic appearance or to the weak construction of the automobile. Kansei engineering type 2 – engineering system. Kansei engineering type 2 uses statistical methods and a set of products defined through a design of experiments array (Nagamachi, 1995b). These link product physical properties with consumers’ feelings (their Kansei ). The output of Kansei engineering type 2 is a computer-based engineering system comprising a database of semantics, design properties and their relations obtained through a semantic differential style consumer survey. This system, when developed, can be used to find personalised solutions to consumers based on their pre selection of Kansei words, or it can give designer feedback on consumers’ interpretation based on their construction of the physical properties (Schu¨tte, 2005). In this variant, Kansei words can be collected from experts, consumers or trade press. These words should be selected to be representative of the consumers’ feelings about a product (Nagamachi, 1995). The approach requires a wide range of existing designs for successful application, and makes the assumption that these can be deconstructed into a set of lower level properties, see Figure 3. The consumer survey involves participants evaluating each product against each semantic (Kansei word) using a semantic differential questionnaire (Osgood and Suci, 1969). Then the underlying meaning and semantic space can be calculated using factor analytic techniques (Nagamachi, 1995b) and multivariate statistics can help to identify correlations between product properties and consumer perceptions.


2.2 Our methodology The Kansei engineering toolkit for packaging design is shown in Figure 4 and comprises a set of exercises and supporting software to facilitate a pack development team to create and evaluate new packaging concepts. We have divided the traditional Kansei process into six tools which are compatible with the different stages within an industrial development process. The tools have been developed to be used individually or as a holistic process to provide insights and guidance. The toolkit proposed in this paper does not represent a direct translation of the Kansei engineering approach. We have built upon the key aspects and supplemented its functionality to address issues specific to packaging design. . We have included structured linkages to designers. Understanding the relationships between pack appearance and consumer response is important

Figure 2. Example of Kansei words connection to physical properties of a automobile through Kansei engineering type 1

TQM 20,4

376

Figure 3. Deconstruction of a set of watches into lower level properties

Figure 4. Kansei engineering toolkit

.

.

.

.

as visual appeal is often dominant (Posner and Nissen, 1976). The process of deconstructing the appearance of a product or pack into lower level properties is often difficult and the principle of Gestalt shows that appearances can be perceived in different ways by different people (Lidwell, 2003). Through our work with industry, design managers have expressed concerns about the validity of the assumption that assembling a set of “optimal” attributes will result in an appealing whole pack. Thus, this toolkit includes processes to ensure that the designer and the consumer are central to the process whilst reducing the chances of a suboptimal assemblage of attributes. We have replaced the type 1 “highest level Kansei” with brand values. Early in the process we include consideration of the brand essence document which contributes to the word selection. This ensures that later evaluations of concepts will ensure congruency with the brand personality. We have introduced a more structured process for the elicitation of type 2 selection of pack physical properties. The visual appearance of packs is extremely varied. They can come in all shapes and sizes from regular cartons through to the complex surfaces prevalent in moulded polymer bottles. Thus, we have developed a filtering process whereby the range of geometry under consideration is reduced whilst ensuring research objectives are met. We have reduced the complexity of the semantic differential survey used to elicit consumer perceptions. In a standard Kansei type 2 experiment the length of the survey and its repetitive tasks can be extremely demanding on participants and lead to questions about the repeatability and robustness of the results. To improve this we have introduced an approach which allows us to reduce each survey session to 30 minutes. This is even more important as industry partners often require us to include additional more traditional “market research” type questions after the semantic scales. However, this approach has provided useful data to reinforce the Kansei findings. We have added in a structured process for selection of the Kansei words. To increase the robustness and repeatability of the results we have collaborated with Linguistic experts to develop a process to easily gather, filter and select the most appropriate words for use in a Kansei study (Delin et al., 2007). The aim of the method is to ensure that adjectives with appropriate levels of precision and recall are presented to participants. We have developed a collaborative process where words are collected from facilitated exercises with consumers. However, we have also defined a more objective approach which inputs seed words into the British National Corpus and uses computational linguistic techniques to extend, refine and prune the list. To ensure that the final list is easily understood with few misinterpretations, we have developed a manual checklist based upon well understood linguistic and grammatical principles that we use to evaluate the final list before presentation in a semantic study.

2.3 Implications for product and pack development Analysis has shown (Childs et al., 2006) that major brand owners and packaging manufacturers have similar product development processes and these are in line with theoretical engineering design processes. However, it found that the boundaries and


TQM 20,4

378

thus the decisions made at the transitions between stages are not common. Figure 5 shows a common process defined in (Childs et al., 2006), which demonstrates the process but fail to show the iteration within and between stages. The work identified general agreement on the product development process, but highlighted some differences in the tasks carried out (or knowledge required) by the different companies at the different stages of pack development depending on the strategic goal of the development process. These goals can be categorised as: . creating a new market, with a step change innovation; . targeting a gap in the market; . extending an existing brand, with a new product; and . refreshment, of an existing brand and product. The toolkit supports different approaches to product development and insight requirements through its individual “tool” construction; a tool can be selected and used depending on each project requirement. Primarily the tools are applied early in the product development process during the TARGET, IDEATE and DEVELOP phases shown in Figure 5. More detail on each tool is explained through the cases studies in the next section building upon the framework presented in Figure 4. 3. Natural evaluative language generation It is very important that the right set of adjectives is used for the consumer survey to ensure that the Kansei process is robust and repeatable and gives relevant results. The adjectives must accurately describe the product and its desired brand identity and also reflect the judgements that participants might want to make. Unsuitable adjective choices can result in a range of problems, including: . misinterpretation of an experimental question resulting either in a flat response distribution or a “double peak” that indicates two interpretations; . adjectives with similar meanings can artificially weight a particular response; and . confusion from unfamiliar or ambiguous adjectives. Kansei engineering suggests that adjectives are selected by talking to consumers or by searching relevant literature (Nagamachi, 1995). These rely upon experience to ensure that words are not missed that represent key design parameters. This issue is especially relevant when designing packaging. The influence of brand is a key to the design process and no method currently exists to ensure that this is represented in the adjective set.

Figure 5. Common product development process among industrial members

3.1 Tool 1: exploration of the semantics Figure 6 shows how this tool supports the generation of an extensive list of seed words developed from three sources: (1) Product functional benefits: ensures that packaging under consideration is congruous with contents and product purpose. (2) Brand values: extracting relevant words from brand documents means that the evaluation also tests for harmony and fit with the values that the brand elicits. (3) Packaging formats: we found seed words from this category rarely elicit any useful results. However it is still important that this is considered for a complete analysis.


Figure 6. Tools 1 and 3

TQM 20,4

380

The next stage is to extend the seed words list by identifying alternative synonyms using a series of accepted linguistic techniques (Delin et al., 2007). Then, a corpus based technique, specially developed for this toolkit, automatically generates all possible words from modern English usage relating to the seed words. This is based on the British National Corpus (BNC), a 100 million word collection of texts, representing a comprehensive picture of British English (Aston and Burnard, 1998). This approach ensures that the selected adjectives are widely used in everyday language. It can result in over 1,000 adjectives related to the desired qualities but this increases the probability of covering all parameters of interest. 3.2 Tool 3: refining the semantic space Kansei engineering uses Principle Component Analysis (PCA) to select the adjectives to present to consumers. However, this requires an expensive (time and money) survey and can result in missing important product and brand qualities. Our approach, exemplified in tool 3 and shown in Figure 6, uses linguistically informed guidelines to categorise the adjectives. These guidelines help to identify suitable candidate adjectives and thus eliminate inappropriate candidates. Some examples are: . Guideline V: remove adjectives requiring additional context to be understood. . Guideline VI: remove comparative adjectives. . Guideline VII: remove non-gradable adjectives. . Guideline XIII: remove adjectives that relate to a prolonged experience. After the application of the guidelines there are usually too many candidate adjectives remaining. We link the adjectives back to the original seed words and with the use of manual checklist, select a set of 10-20 words representing all product, brand and pack qualities. 3.3 Case study: natural evaluative language generation for a cleaner bottle We illustrate our description of the Natural Evaluative Language Generation contained within tools 1 and 3 by means of extracts from an industrial case study, in which we evaluated different packaging concepts for a new cleaning product. Tool 1: exploration of the semantics. We identified a set of appropriate seed words for input into the system. In this situation, packaging attributes did not result in any suitable adjectives and so will not be discussed any further within this case study. These seed words were manually extended using the linguistic process described in section 3.1 to about 70 words, see Table I. Each word was then input into the BNC as Brand equity BE1: Delight BE2: High-standard

Table I. Extending seed words

BEn

Product benefits Enjoyment, surprise, joy, content, pleasure, delighted etc. Best, exceptional, exclusive, extraordinary, unique . . . etc.

PB1: Skin kindness

Kind, caring, smooth, moisturising etc

PB2: Perfume

Scent, odour, aroma . . . etc.

PBn

described in section 3.1 to automatically identify other adjectives that occur in natural language and a list was produced of each frequent word from the British National Corpus. Tool 3: refining the semantic space. The next stage searches the relevant adjective set to find those suitable for the consumer study, as described in section 3.2. In this case study there were 100 adjectives which fulfilled all guidelines. Thus, the adjectives were arranged into a matrix, a sample of which is shown in Table II to facilitate the selection of appropriate adjectives for the consumer survey. Adjectives that have multiple roots test more than one concept and can reduce the number of questions required. The client must be confident that all the relevant brand, product and pack attributes are covered by the adjectives and this matrix can be used to ensure this occurs. The final adjective list for the consumer survey is highlighted in Table II, and included words such as tender, conventional, fun, luxurious, everyday, slender, cosy and bold. 4. Concept range generation It is vital that suitable concepts span all design possibilities and are appropriately presented to survey participants. Kansei engineering type 1 can be used to develop product concepts based on the original “Kansei” or feeling. However, this technique was found unsuitable as decomposing a brand to feelings was highly subjective. Because of the complexity of packaging, deconstructing into items and categories of attributes (as in Kansei engineering type 2) frequently results in many categories. Therefore qualitative research techniques are used to define the concept range by providing preliminary insights through a more open structure than the semantic differential survey. Our approach allows the toolkit to be used for incremental change of existing packs or to radically redesign the pack starting from a “blue sky” approach. 4.1 Tool 2: design classification For incremental design, a large and diverse range of packs should be collected. However, designing a totally new pack concept requires a more fundamental approach. Tool 2 supports the collation of a set of visual stimuli to be used as a starting point (Figure 7). Through the use of typically abstract pictures (from image libraries such as Gettyimages) the team creates “Kansei Boards” by “decoding” what elements in the stimuli create associations to the target brand qualities. This information is used as the inspiration behind a creative sketching and brainstorming activity to develop a large number of potential pack concepts. 4.2 Tool 4: concept definition and selection This tool takes the outputs of tool 2, either a set of existing packs or a set of original concepts and delivers two processes: (1) Concept reduction process: A semantic mapping process supports a representative selection of consumers to physically place pack prototypes on a two-dimensional map, with axes defined by key product and brand attributes (e.g. clean and fresh or spirited and traditional). Participants find it easy to justify their concept placements which lead to insights into correlations between design attributes and affects. A reduced range of concepts can be


Moisturising

U

U

Smoothness

Balanced

U

Skin kindness

Product:

U

U

U

U

U

Comfortable

Happy

U U

U

Delight

Love

Brand:

Showy

Everyday

Slender

U

U

U

U

U

U

U

U

U

U

U

U

U

U

U

U

U

U

U

U

Bold

Cosy

Luxurious

Mid coverage Fun

U

U

Friendly

U

U

Simple

U

Traditional

U

U

Natural

U

U

Romantic

Least coverage

U

Advanced

U

U

Amusing

382

Conventional

Tender

Table II. Matrix of relevant adjective set against original seed words

Highest coverage

U

Spiritual

U

U

Casual

TQM 20,4



selected to test based on the hypothesis that dispersed locations on the map represent a diverse range of concepts and attributes properties. (2) Preliminary identification of important pack attributes: This uses a triadic sorting exercise to define the important attributes necessary for multivariate regression techniques. It is based on the Repertory Grid Technique (Kelly, 1955). Participants are asked to select three packs from the set and identify a likeness between two packs and define the contrast with the third. For example, a consumer might say, “Packs 4 and 9 are similar because they have a circular footprint and pack 14 is different as the footprint is square”. Having completed this exercise a number of times, a large number of pack attributes can be recorded that are “obvious” to the consumer. These can be prioritised according to frequency of observation or other design objectives.

TQM 20,4

384

4.3 Case study: sample range generation for an alcoholic bottle design We illustrate our description of the sample range generation contained within tools 2 and 4 by means of extracts from an industrial case study, in which we evaluated different packaging concepts for a modern bottle for a new alcoholic premixed drink. Tool 2: design classification. The project addressed the affect evaluation of an existing prototype so Kansei Board and concept generation was not carried out. Instead, glass bottles were collected from a wide range of markets and brands. Overall, 20 unique clear and brown bottles of volume appropriate to the product were collected and their labelling removed to avoid familiarity biases. Green bottles were not relevant to the product so were not included. Tool 4: concept definition and selection – concept reduction process. The company selected the two axes on the semantic map as modern and American. A representative selection of consumers were given the set of glass bottles and asked to place them on the two-dimensional grid. Participants were able to clearly distinguish between the suitability of the designs and afterwards could justify why designs were placed in each location. Figure 8 shows one result of the semantic mapping research experiment for the alcoholic premixed drink. Tool 4: concept definition and selection – identification of important pack attributes. Figure 9 shows a three bottles selected from the set presented to the participants of the triadic sorting exercise. The explanation of the selection was: Bottles A & B are both clear and have convex collars, contrasting with bottle C which is brown and has an inverted collar.

This exercise results in decomposition of the concepts into design attributes (even ones that are hard to describe). Participants can often relate the attribute to their preference and with further questioning rationalise their choices. For example, participants felt the collar was a feature which contributed to the bottle’s modern style. Results from the two exercises were used to inform the creation of a smaller range of bottles of priority attributes to the project domain. 5. Tools 5 and 6 – consumer survey and interpretation A self report semantic differential survey and statistical techniques similar to those in traditional Kansei is used to understand the underlying relationships, and shown in Figure 10. Participant fatigue and the measurement of the affect that this has on participants’ evaluations is a vital issue to the validity of the research. The changes introduced in this toolkit through the earlier tools result in a shorter survey length; this is

Figure 8. Semantic mapping exercise


Figure 9. Bottles being compared


TQM 20,4

386

accompanied by techniques in tool 5 to uniquely order questionnaires for each participant to reduce fatigue affects. 5.1 Tool 5: affective consumer survey A statistically significant cohort of consumers is recruited and the pack concepts and adjective lists are prepared. A bespoke programme is used to facilitate data presentation, collection and analysis. It randomises order and polarity of the word pairs helping to reduce survey biases. Bipolar adjectives (e.g. cosy – not cosy) are used across a seven-point Likert scale (strongly positive – neutral – strongly negative) to increase accuracy of responses and to eliminate misinterpretations from using antonyms. We also include additional questions to gather further insight and to allow us to better interpret the PCA results. 5.2 Tool 6: affect interpretation This tool, described in Figure 10, shows how to analyse the data gathered through the survey in tool 5 and is very similar to a typical Kansei analysis. PCA reduces the data into orthogonal components and multivariate regression is used where possible to investigate which attributes contribute to eliciting each semantic response. Pack concept results are plotted in semantic space and correlated with additional data for interpretation and design rule definition. 5.3 Case study: interpretation of data from glass bottles for adult sauce We illustrate the setup and evaluation of the consumer survey through a research project to assess the suitability of candidate glass bottle shapes to communicate a brand and “adult” image of a company’s new flavoured sauce. Tool 5: affective consumer survey. The survey tested a prototype design against seven other glass bottles selected from existing products with similar volume and glass density. Figure 11 shows three of the designs. A total of ten adjectives were selected which closely represented the target packaging domain; 60 consumers took part in the

Figure 11. Three of the test bottles

study; they were presented with one bottle at a time and gave their adjective ratings and preference score using the bespoke software. Tool 6: affect interpretation. Principal Component Analysis showed that participants’ use of the adjectives could be simplified using three factors: modernness, adultness and uniqueness and the packaging concept were plotted against these as shown in Figure 12. The study found that the prototype bottle performed well against the range of evaluations, but bottle 2 was the most suitable to product and brand even though it scored low against Unique. Recommendations to increase the consumer perception of the prototype were made by assessing common features in well-liked bottles. One such observation was that long necked bottles were seen as appealing. This provided a basis for further development and explained why consumers did not prefer the prototype.


6. Summary This paper has presented an overview of techniques to support and improve the well-established Kansei engineering process for use in the packaging industry. It has also shown how the philosophy of Kansei engineering can be used innovatively to add new insights to design decision-making. Although the case studies presented in this paper are packaging related, the toolkit has its roots in Kansei engineering and so would provide insights for products as well. Other applications of Kansei engineering have been developed in the packaging industry and could be considered for the next development phase. For example, the use of artificial intelligence techniques to analyse coffee cans, milk cartons and beer packaging (Ishihara et al., 1997) showed how consumers clustered different packs by label design styles. Kansei engineering and the techniques presented in this toolkit are inevitably simplifications of the real situation, since many more variables affect the consumers purchase decision than is tested in this process. There is still a need to test the insights gained by the toolkit into a wider investigation.

Figure 12. Semantic map of sauce bottles

TQM 20,4

388

References Aston, G. and Burnard, L. (1998), Handbook: Exploring the British National Corpus with SARA, Edinburgh University Press, Edinburgh. Childs, T., Agouridas, V., Barnes, C.J. and Henson, B. (2006), “Controlled appeal product design: a life cycle role for affective (Kansei) engineering: engage network work package 2”, available at: www.engage-design.org Crilly, N. (2004), “Seeing things: consumer response to the visual domain in product design”, Journal of Design Studies, Vol. 25 No. 6, pp. 547-77. Delin, J., Sharoff, S., Barnes, C.J. and Lillford, S.P. (2007), “Linguistic support for concept selection decisions”, Artificial Intelligence for Engineering Design, Analysis and Manufacturing, Vol. 21 No. 2, pp. 123-35. Henson, B. (2003), “Sensual surfaces: engaging consumers through surface textures”, paper presented at the DPPI’03, Pittsburgh, PA, June 23-26, 2003. Kelly, G. (1955), The Psychology of Personal Constructs, Vols 1 & 2, Routledge, London. Lidwell, W. (2003), Universal Principles of Design, Rockport, Gloucester, MA. Nagamachi, M. (1995), “Kansei engineering: a new ergonomic consumer-oriented technology for product development”, International Journal of Industrial Ergonomics, Vol. 15, pp. 3-11. Nagamachi, M. (2002), “Kansei engineering as a powerful consumer-oriented technology for product development”, Applied Ergonomics, Vol. 33, pp. 189-94. Osgood, C.E. and Suci, G.J. (1969), “Factor analysis of meaning”, in Osgood, C.E. and Snider, J.G. (Eds), Semantic Differential Technique – a Source Book, Aldine Publishing Company, Chicago, IL, pp. 42-55. Posner, M.I. and Nissen, M.J. (1976), “Visual dominance: an information-processing account of its origins and significance”, Psychological Review, Vol. 83, pp. 157-71. ¨ Schutte, S. (2005), “Engineering emotional values in product design”, dissertation, No. SE-58183, Linkoping Universitet, Linkoping. Vihma, S. (1995), “Products as representations: semiotic and aesthetic study of design products”, University of Art & Design, Helsinki. Further reading Ishihara, S., Ishihara, K., Tsuchiya, T., Nagamachi, M. and Matsubara, Y. (1997), “Neural network approach to kansei analysis on canned coffee design”, Proceedings of the 13th Triennial Conference of the International Ergonomics Association, Vol. 2, pp. 211-13. Nagamachi, M. (1995), An Account of Kansei Engineering, Japan Standards Association, Tokyo, Chapters 3 & 4. Corresponding author Cathy Barnes can be contacted at: [email protected]



Affective design of waiting areas in primary healthcare

Design of waiting areas

Ebru Ayas Division of Quality and Human Systems Engineering, Linko¨ping University, Linko¨ping, Sweden

389

Jo¨rgen Eklund Division of Quality and Human Systems Engineering, Linko¨ping University, Linko¨ping, Sweden, and Division of Industrial Ergonomics, STH, Royal Institute of Technology, Huddinge, Sweden, and

Shigekazu Ishihara School of Psychological Science, Hiroshima International University, Hiroshima, Japan Abstract Purpose – This paper seeks to deal with affective design of waiting areas (servicescapes) and has twofold aims. The first, is to explore affective values for waiting areas. The second, is to identify interactions between physical design attributes and affective values. Design/methodology/approach – This study included a free association method for data collection, applying Kansei engineering methodology to extract design solutions relating to specific ¨ stergo¨tland County, Sweden. In feelings. The study was undertaken at six primary health centres in O total, 88 participants (60 patients and 28 staff) were interviewed. Findings – The selected waiting areas show significant differences for their perceived affective qualities. The most desired feeling for creating affective values is found to be “calm”. The core design attributes contributing to this feeling are privacy, colours, child play-areas and green plants. Good design of lighting, seating arrangements and a low sound level are also important design attributes to give a more complete design solution. Research limitations/implications – The study provides useful insights for understanding affective needs in servicescapes, and it provides design suggestions. The results have not been analysed separately for gender or different age groups. Practical implications – The paper proposes a framework model to be applied when dealing with affective values in servicescapes. Originality/value – This paper makes an original contribution to understand affective values towards the physical environment in servicescape design. It offers a methodology to study complex environments with many alternative design solutions using limited resources. Moreover, this study uses a combination of a free association method and Rough Sets theory in affective design. Keywords Rooms, Design, Affective psychology, Health services sector, Community health centres Paper type Research paper

The authors would like to thank Mattias Elg for his feedback and discussions on the study and psychologist, Ms Rebecka Lundgren, who graduated from Linko¨ping University in 2007, for her work during interviews. Special thanks are due to Ms Maria Rasch and Ms Marie Lindstro¨m from O¨stergo¨tland County Council, Sweden for their fruitful discussions during the planning of the study.


TQM 20,4

390

1. Introduction Design of “quality” for products is moving beyond functionality and usability to satisfying people’s subjective quality needs and values (Childs et al., 2006). In recent years affective design has been applied increasingly in product development to fulfil subjective needs and preferences of customers and users. The resulting emotions are referred to as “Kansei” in Japanese. “Kansei” is defined as “individual’s psychological feeling and image resulting from a series of information processes from a certain artefact, environment, or situation” (Nagamachi et al., 1999). Kansei engineering was proposed as a methodology for affective design of products in the early 1970s (Nagamachi, 1995). This methodology aims at translating human psychological processes, such as feelings and emotions, into appropriate product design elements, such as size, shape, and surface characteristics. However, measuring the Kansei is not easy and will always build on a subjective basis, since the measurement methods are depending on the reactions of the humans (Schu¨tte et al., 2007). The main challenge of this methodology originates from difficulties in mapping Kansei to perceptual design elements (Jiao et al., 2006). It is considered that identifying the essential subjective qualities by understanding affective (Kansei ) values could contribute to better product design. Accordingly, affective values obtained through personal experience indicate the value of something that impresses and appeals on the Kansei (Nagasawa, 2006). Values are interactive relativistic preference experiences that refer to the evaluation of some object/goods, service, and environment by some subject (consumer or other customer) (Holbrook, 1998). Although considerable efforts has been devoted to applying Kansei engineering as a methodology in product design, the authors have not found studies concerning physical environment design in healthcare services. Looking at the healthcare services understanding patients’ needs in waiting areas is a growing concern (Arneill and Devlin, 2003; Leather et al., 2003). However, exploring Kansei values and needs for design of waiting areas in primary healthcare are not given the same emphasis. Waiting areas are referred as “servicescapes” where a part of the service is delivered, perceived and where the staff and patients interact (Bitner, 1992). Managing the psychological experience of customers’ waiting experience in a waiting area by reducing the perceived waiting time may be as effective as reducing the waiting time itself (Katz et al., 1991). Besides, waiting areas are also part of the perceived service quality and belong to the tangibles quality dimension of services (Parasuraman et al., 1998). An extensive review by Dijkstra (2006) points out the possible interactions between different environmental stimuli, and it is likely that the effects of different environmental stimuli will reinforce or weaken another. Exploring the interaction effects of design attributes for an environment in relation to affective judgements have not been captured fully by the methods suggested in the literature (Osgood et al., 1957; Ku¨ller, 1980; Russel et al., 1981). Improving the interaction experience of physical design attributes would help to create customer value for psychological aspects of Kansei (Nagasawa, 2006). Therefore, there is a need to distinguish between functional and affective values for an object or an environment and to show how the two interact. In this research, the

Kansei Engineering methodology is proposed to examine the interaction between servicescape design attributes that may affect affective values. The overall aim of the study is to understand waiting areas from an affective design perspective. The aim of the present study is twofold. The first is to explore affective values for waiting areas. The second is to identify interactions between physical design attributes and affective values. Specifically, three main research questions are investigated: RQ1. Are there differences regarding perceived affective values between waiting areas? RQ2. What are the affective values desired by the patients/staff when experiencing waiting areas? RQ3. How do waiting area design attributes interact in creating affective values? The plan of the paper is as follows. In section 2, a framework model is presented for data collection and analysis. Next, empirical research is explained in section 3. In section 4, results of the study are presented. Last, the paper is concluded with theoretical and practical implications for design of waiting areas, limitations of the study and implications for Kansei engineering. 2. The proposed framework model The traditional way of applying Kansei engineering is done through collecting adjectives (Kansei words) to construct the semantic space for the product of concern. The Kansei words are then reduced to a feasible number with the help of Factor Analysis (Nagamachi, 1995; Schu¨tte, 2005). Selected words are thereafter prepared for subjective evaluation on a Semantic Differential scale (Osgood et al., 1957). The questions in such surveys are normally highly structured and employ a forced-choice design in a multiple set of words, which may have an effect on the assessment of the experiences or needs for the product. So also the results of the Kansei engineering study may be influenced by the data collection methodology. Therefore, in this study a “free association technique” (Freud, 1913) is proposed for data collection. This technique has been used in qualitative research mainly for conducting psychoanalysis (Parker, 2004). Using this type of data collection may yield more useful insights into what participants actually think (Cozby, 1989). Besides, this way of collecting data may help to understand the affective reactions of the individual. A framework model is proposed and Figure 1 illustrates the application steps of the study using the free association method: (1) Step 1: The domain is decided. It defines the context of the application for a particular type of product in a Kansei engineering study. (2) Step 2: The participants are identified. This step is related to the selection of domain. (3) Step 3: Data collection is performed through in depth interviews, where free association is used as a method. First, participants talk about important feelings for them (Kansei words). Thereafter, the participants relate their feelings to design attributes.


391

TQM 20,4

392

Figure 1. Flowchart for the proposed Kansei engineering model

(4) Step 4: The semantic space of feelings (Kansei words) and the space of design attributes collected in the previous step are recorded. (5) Step 5: Based on the data recorded, the relationship of feeling statements to design attributes are verified for each participant. (6) Step 6: An MS Excel database (called a Kansei database) is constructed to store the feeling and design attribute responses. (7) Step 7: Through statistical methods (e.g. x2 Independency Tests), significance of common Kansei words is tested for all participants. The common feelings identified in this step are assumed to indicate Kansei values. (8) Step 8: Data mining methods are used to discover useful design patterns associated with the Kansei values (see step 7). “Rough Sets” (RS) method is a applied to extract decision rules between the Kansei values and design attribute categories from the data set. (9) Step 9: Data extraction quality is checked for validation. There are different validation algorithms in RS theory that may be used due to data nature and the nature of the aim (minimization or maximization) of the study. (10) Step 10: Decision rules are extracted by algorithms that are used in RS (e.g. minimal covering rule (LEM2) algorithm, Michalski Qm measure, m-method, dual beta-lower and upper approximations (Nishino et al., 2006). 3. Conducting the study Selection of waiting areas Primary healthcare centres for the present study were selected based on two criteria of qualitative research: pragmatism and representativeness (Hackley, 2003). The study ¨ stergo¨tland was undertaken representing socially and regionally distinct areas in O County, Sweden. The six waiting areas coded as A, B, C, D, E and F were chosen from 35 primary healthcare centres considering such as regional characteristics, proximity to city centres and social characteristics of the patient population. In Table I the characteristics of the healthcare centres and the design characteristics of each waiting area were defined. Pictures of each waiting area (A, B, C, D, E and F) are also presented in Plates 1, 2, 3, 4, 5 and 6 respectively. Data collection The main interview questions used in the study are presented in Table II. Accordingly, the same questions were also applied to the staff. The responses were documented by note-taking. In the last section, gender and ages of the participants were assessed. Approval for conducting the study was obtained from the county council and from the health care centre managers. Face-to-face interviews were conducted in the waiting areas of the selected healthcare centres. The participants were encouraged to reply extensively. Patients received information handout about the study on their arrival to the healthcare centre and were asked for voluntary participation by the receptionists. The patients were interviewed while they waited to see a physician or to have laboratory tests. If a person preferred not to participate, information was given to the next patient.


393

TQM 20,4

394

Table I. Characteristics of waiting areas

Waiting area

Regional characteristics

Social characteristics

A

City centre

Multi-cultural, newly built

B

City centre

Multi-cultural

C

Town

Families, newly built

D

Town

Families, industry

E

Suburb

Multi-cultural

F

Suburb

Multi-cultural

Design characteristics of the waiting areas Sofa, art, information and reading material, not a separate waiting area, big windows, light and soft colours (e.g. orange, pink) on the walls, built stairs Small sitting groups, information and reading material, art, light colours Sofa, private seating, child play area, aquarium, central lighting, information and reading material Chairs for private seating, art, TV, pink and sand colour on the walls, light colour on the floor material, central lighting, information and reading material Wooden seating, sand colours on the walls and dark vinyl material on the floor, drinking facilities, information billboard, central lighting, reading material Sofa, child play area, art, light colours on the walls and matching colour on the floor with the sofa, central lighting, information and reading material

Plate 1. Waiting area A (city centre)

Data analysis Table III presents an overview of data collection and analysis methods. The responses from the first interview question were classified according to perceived Affective qualities for the physical environment according to Russell et al. (1981). Correspondence analysis (CA) was applied to investigate the “perceived affective values” for the selected waiting areas. CA (Greenacre, 1984) is seen as a generalization


395

Plate 2. Waiting area B (city centre)

Plate 3. Waiting area C (town)

of Principal Component Analysis (PCA) where the variables to be analyzed are categorical instead of quantitative (Abdi and Valentin, 2007). For the second interview question, the feeling statements (gathered from patients and staff) were grouped. Next, x2 analysis was applied to test whether there were significant differences in the participants’ feeling statements. The hypotheses were: H0.

All categories of feelings are equally important.

H1.

Some categories of feelings are more important.

The data from patients and staff were classified under three service quality dimensions for further analysis: Technical quality, Interaction quality (Brady and Cronin, 2001)

TQM 20,4

396

Plate 4. Waiting area D (town)

Plate 5. Waiting area E (suburb)

and Affective quality dimensions (Russell et al., 1981). According to each quality category frequencies of feeling statements were calculated and a Pareto analysis was applied as a supportive tools for prioritizing feeling statements under each quality dimension. Based on the x2 and Pareto analysis results, commonly desired feelings of the participants were derived. The third interview question was investigated based on the commonly desired feelings from the second interview question. The interactions between design attributes for the commonly desired feelings were analyzed by the RS method (Pawlak, 1992). The RS approach provides advantages such as identifying relationships that would not be found by using statistical methods, allowing qualitative and quantitative data, and finding minimal sets of data (Pawlak, 1992).


397

Plate 6. Waiting area F (suburb)

Participants

Interview questions

Research question

Patients and staff

1. How do you feel in the waiting area? 2. What important feelings would you like to get from a waiting area? 3. Try to think how do you relate your important feelings to design features in a waiting area?

(1) (2)

Patients

Type

Analysis

Interview question 1 Correspondence analysis Interview question 2 x2 independency tests

Interview question 3 Rough sets analysis

(3)

Investigation

Data classification

Perceived affective values for the selected waiting areas Commonly desired feelings from waiting areas for all patients and staff Comparison of patient and staff responses Extracting the design attribute interactions for calm feeling

Affective qualities (Russel and Pratt, 1981) Affective qualities (Russel and Pratt, 1981) Service quality dimensions (Brady and Cronin, 2001) Design attributes (Leather et al. (2003)) and Bitner (1992))

The SPSS version 15.0 (SPSS, 2006) was used to apply CA and x2 analyses. ROSE software (Predki et al., 1998) was used for the RS method. 4. Results Participants In total, 88 participants were interviewed from the six primary healthcare centres, i.e. 60 patients and 28 staff working in different positions: managers, doctors, receptionists

Table II. Interview questions used in the study and their relation to the research questions

Table III. Overview of the data collection and analysis approaches

TQM 20,4

398

and nurses. The mean ages for patients and for staff were 56 and 48 years respectively. More than half of the participants were women (61 percent). In earlier studies, it is found that men and women do not differ in their level of positive affectivity (Watson, 2000). Therefore, responses were not categorized according to gender. Drop-out rates were judged to be below 15 percent for patients and 20 percent for staff. One of the reasons for drop-out was lack of time. Correspondence analysis of perceived affective values Table IV shows the classification of responses according to the overall perceived affective qualities (n ¼ 239) gathered from patients and staff for the selected waiting areas in the study. The first column displays the classification of waiting areas according to affective qualities. The second column shows Affective qualities attributed to environments. The frequency of responses for each quality dimension is shown in the third column. The last column shows examples of verbal expressions i.e. feelings related to Affective qualities. Accordingly, CA was applied to the classification of responses by affective quality dimensions for each waiting area. According to the results from the CA, there is a significant relationship between the health centres and attributed affective qualities from the participants (x2 (30) ¼ 93.67; p ¼ 0.00, p , 0.001). The results showed that two dimensions were sufficient to explain the variability underlying the classification. The first principal axis explains 66 percent and the second principal axis explains 20 percent of the variability. For the first dimension Distressing Quality, Relaxing Quality and Unpleasant Quality are dominant and account for 80 percent of the variability. For the second dimension Gloomy Quality and Pleasant Quality together account for 83 percent of the variability. The two axes together explain 86 percent of the variability for all affective qualities (there were no observations obtained on the Sleepy quality dimension). With a two dimensional solution each quality dimension contributed with 53 percent-98 percent of the total inertia (eigenvalues). A plot of category quantifications on the two retained dimensions is presented in Figure 2. The first dimension (vertical) appears to distinguish between negative and positive affective qualities. Positive and negative qualities are located in different quadrants, 2-3 and 1-4 respectively. In the first and fourth quadrant Gloomy, Unpleasant and Distressing Qualities are located. In the second quadrant Relaxing Quality, Exciting Quality, Arousing Quality and Pleasant Quality are located together. Corresponding waiting areas

Table IV. Classification of responses according to affective quality dimensions

A,C,D,F – A,C,D,E,F A,B,D,E,F A,B,C,D,E,F C,D A,B,C,D,E,F A,B,C,E,F

Affective qualities

Frequencies

Arousing quality Sleepy quality Relaxing quality Distressing quality Exciting quality Gloomy quality Pleasant quality Unpleasant quality Total

12 0 44 40 20 4 83 36 239

Examples of verbal expressions E.g. fresh, inviting, cheering – E.g. calming, cosy, quiet E.g. unsecure, worried, stressful E.g. interesting, exciting E.g. cold, not personal, negative E.g. nice, home-feeling, home-comfort E.g. hopeless, boring, monotonous


399

Figure 2. Mapping affective qualities versus waiting areas

The second dimension (horizontal) distinguishes Pleasant and Distressing Quality independently from Affective Quality. Pleasant Quality is positioned alone in the third quadrant and Distressing Quality is positioned alone in the fourth quadrant. Waiting area E is perceived as Distressing Quality (servicescapes that are both pleasant and non-arousing would be called distressing, see Zeithaml and Bitner, 2003) in the fourth quadrant. On the other hand waiting area C appears to be perceived with positive qualities such as Relaxing Quality and Exciting Quality in the second quadrant. Waiting area A, B and E are located near Unpleasant and Gloomy Quality (unpleasant and sleepy) representing negative Affective Qualities, while the rest of the waiting areas C, D and F are associated with positive qualities. Commonly desired feelings for waiting areas x2 analysis was applied to the data gathered for important feelings. As a result the calm, welcome and safety-security feelings appeared as significant for patients and staff respectively (x2 (26) ¼ 218; x2 (21) ¼ 49.3, p ¼ 0.00, p , 0.001). There was a difference between patients and staff in relation to the order of feelings. Safety-security feeling was desired by the staff prior to welcome feeling. Table V shows the frequencies of desired feelings and related type of qualities from application of a breakdown structure within Kansei engineering type I (Nagamachi, 1995). Quality is chosen to represent all the qualities as a core Kansei. The relative amount of responses (n ¼ 192) for the waiting areas were classified under the three headings, Affective Qualities, Technical Quality and Interaction Quality. A large amount of patients and staff responses were related to Affective Qualities (54 percent), while fewer were related to Technical Quality (28 percent) and Interaction Quality with the staff (18 percent). In Figures 3-6, feelings were classified by Pareto diagrams under each quality dimension from Table V. The Pleasant Quality dimension is related to the feelings pleasant, comfortable, warm, home feeling and cosines (Figure 3). It can be seen that

TQM 20,4

Kansei levels Level 0 Feeling of quality

Level 1

Level 2

Feelings for affective qualities (54 percent)

Relaxing quality (47 percent) Calm (90 percent)

400 Feelings for technical quality (28 percent) Table V. Breakdown of quality dimensions according to the Kansei engineering type I approach

Level 3

Pleasant quality (46 percent) Pleasant (38 percent) Arousing quality (7 percent) Fresh (57 percent) Security-safety (42 percent)

Functionality (25 percent) Privacy (21 percent) Feelings for interaction Welcome (57 percent) quality (18 percent) Caring staff (12 percent) Staff give attention (11 percent)

Figure 3. Pleasant qualities

pleasant, comfortable and warm together represent more than 80 percent of all responses. Calm and relaxing constitute the Relaxing Quality dimension (Figure 4) while calm represents more than 80 percent of responses. Fresh feeling together with bracing appeared important for the Arousing Quality dimension (Figure 5). Security-safety, functionality and privacy are the important issues that represent more than 80 percent of the responses for the Technical Quality dimension (Figure 6). According to the results from the qualitative responses; security-safety feeling in general is related to cleanliness, order of furniture and environment, hygienic design material, reception’s placement, alarm buttons, how staff notice and give attention to


401

Figure 4. Relaxing qualities

Figure 5. Arousing qualities

the patients, size of waiting area, warm colours, soft shapes, stable design of furniture and giving a professional impression with the environment and the staff. Functionality is in general related to lighting, low sound level, reduced noise, layout with open, airy and spacious design, easy access to emergency exit, big windows to provide daylight and also service design attributes such as providing queue numbers. Welcoming environments including staff that cares and gives attention to patients are the important characteristics accounting for 80 percent of the responses for Interaction Qualities (Figure 7).

TQM 20,4

402

Figure 6. Technical qualities

Figure 7. Interaction qualities

Experimental results on design attribute associations for calm feeling Table VI presents the design attributes and categories used to investigate calm feeling. Classified attribute categories are presented in Table VI from patients’ statements (n ¼ 33). There were 15 design attribute categories stated during the interviews. Attributes were coded with signs (a1, a2, . . .) in the third column. The classification of collected design attributes were done under the categories of functionality, facility, interior appearance and activity (Leather et al., 2003; Bitner, 1992). Further, data from each waiting area were binary coded as decision attributes (0 ¼ non-existence of attribute; 1 ¼ existence of attribute). The RS model (Pawlak, 1992) was applied to identify relationships between calm and the design attribute categories.


Feeling

Design attributes

Attribute categories

Sign

Decision (D1)

Calm

Functionality

Privacy, lighting, noise level, layout design Child play-area, drink, interaction quality, sitting places Colour, art, order, plants Entertainment, music, reading material

a1, a2, a3,a4

Waiting area A

a5, a6, a7, a8 a9, a10, a11, a12

Waiting Waiting Waiting Waiting

a13, a14, a15

Waiting area F

Facility Interior appearance Activity

area area area area

B C D E

403 Table VI. Design attributes and categories

First, approximations of data for decision (D1) classes and classification accuracies were calculated. The results vary between 0.7-1. Table VII shows that for all design attribute categories, the overall quality of classification is equal to 0.93 (the ratio of correctly classified objects) and the value for core attributes is 0.15. Core attributes for calm feeling resulted in the subset of design attributes privacy, child play area, colour and plants, presented also in Table VII. In addition to selecting core design attributes, a manual selection procedure was conducted to see how other design attributes would contribute to the overall quality of design. The predicted loss on quality of classification found if such an attribute were removed. Lighting, sitting places, sound level are the design attributes also to be considered for design of calm feeling. The rest of the attributes were not considered as important with values less than 0.03. Application of minimal covering rule (LEM2 algorithm) resulted in 16 decision rules of association between design attributes. The decision rules were then examined according to their relative strength to explain the data and with their syntax to identification of important design attributes. Further, the strong decision rules (n ¼ 9) were selected to express the feeling calm and presented in Table VIII. Considering the results from, e.g. waiting area “F”; decision rule number 9 shows that “a2a3a10” are interacting design attributes. This rule represents that lighting design, reduced noise level and provision of art interact to generate calm feeling. Covering index values show the ratio of main attribute category compositions for calm. Covering index is 1 for this rule which represent that the power of the rule is sufficient to design calm feeling. Interaction of different design attributes to design calm specifically in each waiting area can be seen from the Table. Thus, particular design solutions also need to be explored for waiting areas.

Quality of classification Core

For all attribute categories For core attribute categories a1 a5 a9 a12

0.93 0.15 Privacy Child play-area Colour Plants

Table VII. Significance of attributes and core attributes for calm

TQM 20,4

Waiting Decision area rule no. Decision rule A

C

1 2 3 4 5

a1a8 a2a8a10a12 a7a8 a2a12a13 a3a8a9a12a15

D E

6 7 8

a1a12a13 a2a8a10 a1a2a3a7a8a12a15

F

9

a2a3a10

B

404

Table VIII. Selected strong decision rules for calm

Represented design attribute categories Privacy, sitting places Lighting, sitting places, art, plants Interaction quality, sitting places Lighting, plants, entertainment Noise level, sitting places, colour, plants, reading material Privacy, plants, entertainment Lightning, sitting places, art Privacy, lighting, noise level, interaction quality, sitting places, plants, reading material Lighting, noise level, art

Covering index 0.40 0.40 0.33 0.33 0.33 0.33 0.44 1 1

5. Discussion and conclusions This study made an attempt to explore servicescapes from an affective design perspective. There are several factors that may have affected the results in this research. Foremost, the framework proposed in this paper should be developed further in order to better handle studies of complex contexts and environments with extremely many design alternatives. Before applying free association as a method, it is essential to create the right atmosphere, both environmentally and psychologically, for participants to talk about their feelings. Data to be submitted for analysis also needs to be classified under suitable design attribute and feeling categories. In Kansei engineering studies participants generally need to do ratings for comparison of several products with cumbersome number of words and this may affect the quality of the raw data in a negative way even before the analysis. The free association method used in this study may help reducing long data collection times in Kansei engineering studies and reduce the time for participants. In order to use RS effectively in decision-making, the data should be collected with care. The number of responses for a specific feeling is also likely to affect the extraction quality of the necessary design attributes. A limitation of this paper is that design solutions were proposed considering only one psychological feeling, in order not to make the paper too complex. Each health centre has a different type of waiting area design that may affect the expressed patient and staff Kansei. People in different demographic and cultural conditions may have different preferences for emotional and physical needs. Thus, particular design solutions need to be explored for waiting areas. This study contributes to the literature in several ways. First, with this study free association is proposed to provide a deeper understanding of meanings of design attributes for human feelings. To investigate attribute interrelations from the qualitative responses, the RS methodology is found suitable. The interactions between affective values and physical design attributes were shown by examining the relationship between calm feeling and design solutions. To create calm feeling in waiting areas, privacy, colours, location of play areas for children and plants interact in waiting areas. Good design of lighting, seating arrangements and minimal noise are also needed.

It is argued that by distinguishing important feelings from intangible and tangible quality characteristics that generate Kansei values, waiting experiences can be created that connect with people on a deeper level and transform the environment into spaces of greater significance. It is also suggested in this study that interactions between design attributes need to be considered to understand and reflect human feelings in product design.


405 Managerial implications From a managerial standpoint, applying Kansei engineering in servicescapes has a potential to contribute to important knowledge about affective design. Waiting areas can be described in terms of Affective Qualities, Technical Qualities and Interaction Qualities. The desired Affective Qualities from waiting areas are Relaxing, Pleasant and Arousing. Considering Pleasant Qualities; pleasantness, comfort and warm feelings are important. Calm is found to be the most important for Relaxing Quality. On the other hand, fresh feeling together with bracing appeared important to provide Arousing Quality. The following design recommendations have been generated from the present study and literature, in order to contribute to calm feeling. Privacy is related to waiting in small sitting groups. It is also related to reduced noise levels in the design of waiting areas (e.g. design of reception, child play area and examination areas of the health centres) and interaction of staff that pay attention to patients’ privacy. There are few studies related to the examination of privacy in healthcare settings (Barlas, 2001). There is a clear need for additional studies that examine privacy and confidentiality in, e.g. waiting areas and nurses’ stations (Ulrich and Zimring, 2004). Nilsson et al. (2004) confirm that it is important to feel safe, private, to be able to do things, find out information and being able to choose to wait in peace and quietness in a waiting room (from a study at a primary healthcare centre in Sweden). Patients prefer warm (e.g. red, orange) colours instead of bright colours to give a calming effect, which is also confirmed by Arneill and Devlin (2003). Cold colours such as blue and green have also been found to have a relaxing effect (Birren, 1978). Noise is an important factor, and most of the patients state that they do not like high sound levels (staff talking to each other, noisy children, mobile phones, TV, radio) in the waiting areas. Also, child play-areas built in waiting areas may also distract some patients waiting experience negatively due to a high sound level. Green plants, such as indoor plants (e.g. green and palm like plants) were mentioned by several participants as an opportunity for positive distraction. Indoor plants have also been shown to reduce perceived stress and reduce physical discomfort also in previous studies (Lohr and Pearson-Mims, 2000). Providing windows with scenes of nature was also found to have positive effects on patients (Ulrich, 1991). Lighting was often mentioned by the participants. The light directly influences an individual’s perception of the quality of space and awareness of physical, emotional, psychological and spiritual aspects of space (Kurtich and Eakin, 1993). According to Benya (1983) the areas to be improved in lighting of health centres are colour rendering, reduction of glare, more daylight, softer lighting and an emphasis on residential aspects of lighting.

TQM 20,4

406

Security-safety is related with cleanliness, order of furniture and environment, hygienically designed material, reception’s placement, alarm buttons, how staff notice and give attention to the patients, size of waiting area, warm colours, soft shapes, stable design of furniture and giving a professional impression with the environment and the staff. Functionality is related with lighting (especially reading lamps, low lighting and spotlights), low sound level, reduced noise (no mobile phones, separate play area for children), layout with open, airy and spacious design (e.g. wardrobes), easy access to emergency exit, big windows for daylight as well service design attributes such as getting queue numbers. Security-safety, functionality and privacy appear as the main design issues for Technical Qualities. Welcoming environments including staff that cares and gives attention to patients are the important characteristics to improve Interaction Quality with the staff working at primary health centres. Waiting area C is perceived with positive qualities such as Relaxing Quality and Exciting Quality while waiting areas D and F are perceived with Pleasant Quality. If the environment is unpleasant, increasing arousal level will move consumers into the distressing region (Lovelock and Wirtz, 2004). Waiting areas A, B and E are perceived with Unpleasant and Gloomy (interaction of unpleasant and sleepy) Qualities that may affect satisfaction of patients from the service given at the health centre. Further, developments to reduce the resources that are needed to study the relationship between design attributes and affective values have high priority for future research. Regarding design of waiting areas in healthcare, different age groups also need to be considered as well as type of health care unit. References Abdi, H. and Valentin, D. (2007), “Multiple correspondence analysis”, in Salkind, N. (Ed.), Encyclopedia of Measurement and Statistics, Sage, Thousand Oaks, CA. Arneill, A.B. and Devlin, S.A. (2003), “Healthcare environments and patient outcomes: a review of the literature”, Environment and Behaviour, Vol. 35 No. 5, pp. 665-94. Barlas, D. (2001), “Comparison of the auditory and visual privacy of emergency department treatment areas with curtains versus those with solid walls”, Annals of Emergency Medicine, Vol. 38 No. 2, pp. 135-9. Benya, J.R. (1983), “Light loads”, Progressive Architecture, April, pp. 127-32. Birren, F. (1978), Color Human Response, Litton Educational Publishing, New York, NY. Bitner, M.J. (1992), “Servicescape: the impact of physical surroundings on customers and employees”, Journal of Marketing, Vol. 56, pp. 57-71. Brady, M. and Cronin, J. (2001), “Some new thoughts on conceptualising perceived service quality: a hierarchical approach”, Journal of Marketing, Vol. 65 No. 3, pp. 34-49. Childs, T.H.C., Dalgarno, K.W. and Mckay, A. (2006), “Delivering mass produced bespoke and appealing products”, Special Issue on Advanced Manufacturing Technology, JSME International Journal Series C, Vol. 49 No. 1, pp. 2-10. Cozby, P.C. (1989), Methods in Behavioral Research, Mayfield Publishing Company, Mountain View, CA.

Dijkstra, K., Pieterse, M. and Pruyn, A. (2006), “Physical environmental stimuli that turn healthcare facilities into healing environments through psychologically mediated effects: systematic review”, Journal of Advanced Nursing, Vol. 56 No. 2, pp. 166-81. Freud, S. (1913), On Beginning the Treatment? Standard Edition of the Complete Psychological Works of Sigmund Freud XII, Hogarth Press, London, p. 135. Greenacre, M.J. (1984), Theory and Application of Correspondence Analysis, Academic Press, London. Hackley, C. (2003), Doing Research Projects in Marketing Management and Consumer Research, Taylor & Francis, London. Holbrook, M.B. (1998), Consumer Value: A Framework for Analysis and Research, Routledge, London. Jiao, J., Zhang, Y. and Helander, M. (2006), “A Kansei mining system for affective design”, Expert Systems with Applications, Vol. 30 No. 4, pp. 658-73. Katz, K., Larson, B. and Larson, R. (1991), “Prescription for the waiting in line blues: entertain, enlighten, and engage”, Sloan Management Review, Vol. 32 No. 2, pp. 44-53. ¨ Kuller, R. (1980), “Architecture and emotions”, in Mikellides, B. (Ed.), Architecture for People, Studio Vista, London. Kurtich, J. and Eakin, G. (1993), Interior Architecture, Van Nostrand Reinhold, New York, NY. Leather, P., Beale, D., Santos, A., Watts, J. and Lee, L. (2003), “Outcomes of environmental appraisal of different hospital waiting areas”, Environment and Behavior, Vol. 35 No. 6, pp. 842-69. Lohr, V.I. and Pearson-Mims, C.H. (2000), “Physical discomfort may be reduced in the presence of interior plants”, HortTechnology, Vol. 10, pp. 53-8. Lovelock, C. and Wirtz, J. (2004), Services Marketing: People, Technology, Strategy, Prentice-Hall, Upper Saddle River, NJ. Nagamachi, M. (1995), “Kansei engineering: a new ergonomic consumer-oriented technology for product development”, International Journal of Industrial Ergonomics, Vol. 15, pp. 3-11. Nagamachi, M., Matsubara, Y. and Wilson, J.R. (1999), “Comparative study of kansei engineering analysis between Japan and UK”, Japanese Journal of Ergonomics, Vol. 35, pp. 432-3. Nagasawa, S. (2006), Customer Experience Management Influencing on Human Kansei to MOT, Springer-Verlag, Berlin, Lecture Notes in Computer Science. Nilsson, B., Pahle´n, M. and Brunnstro¨m, L. (2004), Design med omtanke: en bok om design fo¨r ha˚llbar utveckling, Svensk byggtja¨nst, Stockholm. Nishino, T., Nagamachi, M. and Tanaka, H. (2006), “Variable precision Bayesian rough set model and its application to kansei engineering”, in Peters, J.F. and Skowron, A. (Eds), Transactions on Rough Sets V: LNCS, Springer-Verlag, Berlin. Osgood, C.E., Suci, G.J. and Tannenbaum, P.H. (1957), The Measurement of Meaning, University of Illinois Press, Champagne, IL. Parasuraman, A., Zeithaml, V.A. and Berry, L.L. (1988), “SERVQUAL: a multiple item scale for measuring consumer perceptions of service quality”, Journal of Retailing, Vol. 64 No. 1, pp. 12-40. Parker, I. (2004), Qualitative Psychology, McGraw-Hill Education, Berkshire. Pawlak, Z. (1992), Rough Sets: Theoretical Aspects of Reasoning about Data, Kluwer Academic Publishers, Dordrecht. Predki, B., Slowinski, R., Stefanowski, J., Susmaga, R. and Wilk, Sz (1998), “ROSE – software implementation of the rough set theory”, in Polkowski, L. and Skowron, A. (Eds), Rough


407

TQM 20,4

408

Sets and Current Trends in Computing: Lecture Notes in Artificial Intelligence, Vol. 1424, Springer-Verlag, Berlin. Russell, J.A., Ward, L.M. and Pratt, G.A. (1981), “Affective quality attributed to environments – a factor analysis study”, Environment and Behaviour, Vol. 13, pp. 259-88. Schu¨tte, S. (2005), “Engineering emotional values in product design: kansei engineering in development”, doctoral thesis, Institute of Technology, Linko¨ping University, Linko¨ping. Schu¨tte, S., Eklund, J., Ishihara, S. and Nagamachi, M. (2007), “Affective meaning: the kansei engineering approach”, in Schifferstein, H.N.J. and Hekkert, P. (Eds), Product Experience, Elsevier Science Publishers, Barking. SPSS (2006), SPSS Base 15.0 for Windows User’s Guide, SPSS, Chicago, IL. Ulrich, R.S. (1991), “Effects of interior design on wellness: theory and recent scientific research”, Journal of Health Care Interior Design, Vol. 3, pp. 97-109. Ulrich, R. and Zimring, C. (2004), The Role of the Physical Environment in the Hospital of the 21st Century: A Once-in-a-Lifetime Opportunity, Report to The Center for Health Design for the Designing the 21st Century Hospital Project, The Center for Health Design, Concord, CA. Watson, D., Hubbard, B. and Wiese, D. (2000), “General traits of personality and affectivity as predictors of satisfaction in intimate relationships: evidence from self- and partner-ratings”, Journal of Personality, Vol. 68 No. 3, pp. 413-49. Zeithaml, V.A. and Bitner, M.J. (2003), Services Marketing, McGraw-Hill, Singapore. Corresponding author Ebru Ayas can be contacted at: [email protected]


The TQM Journal, Volume 20 Issue 4: The International Review Of Organizational Improvement Kansei affective Engineering Design

Kansei Affective Engineering

International Journal of Powder Metallurgy Volume 45 Issue 4

International Journal of Powder Metallurgy Volume 44 Issue 4

International Journal of Powder Metallurgy Volume 43 Issue 4

International Journal of Powder Metallurgy Volume 46 Issue 4

International Review of Neurobiology, Volume 20

Journal of Organizational Change Management - Volume 20 Issue 3 (2007) - Special Issue: Organizational Change: Over-Psychologized and Under-Socialized?

Feminist Review: Journal, Issue 44

Feminist Review: Journal,Issue 49

Israel Journal of Mathematics, Volume 4, Issue 4 (1966)

Feminist Review: Journal, Issue 43

Feminist Review: Journal, Issue 47 (Feminist Review)

Feminist Review: Journal, Issue 46 (Feminist Review)

Feminist Review: Journal, Issue 38 (Feminist Review)

Feminist Review: Issue 35 (Feminist Review Journal)

Feminist Review: Journal, Issue 40 (Feminist Review)

Journal of Organizational Change Management - Volume 18 Issue 1 (2005)

Feminist Review: Issue 37 (Feminist Review Journal)

International Review of Neurobiology Volume 4

Feminist Review: Journal, Issue 42 (''Feminist Review'' Journal)

Feminist Review: Journal, Issue 41 (Feminist Review Journal)

International Journal of Manpower - Volume 23 Issue 8 (2002) - Special Issue: Work values and organizational behaviour: towards the new millennium

Israel Journal of Mathematics, Volume 4, Issue 3 (1966)

Journal of Semantics, Volume 2, Issue 3-4, 1983

Israel Journal of Mathematics, Volume 4, Issue 1 (1966)

International Review of Industrial and Organizational Psychology, 2006 (International Review of Industrial and Organizational Psychology)

Israel Journal of Mathematics, Volume 2, Issue 4 (1964)

International Review of Industrial and Organizational Psychology

International Review of Industrial and Organizational Psychology, 2003 (Volume 18)

Israel Journal of Mathematics, Volume 4, Issue 2 (1966)

The TQM Journal, Volume 20 Issue 4: The International Review Of Organizational Improvement Kansei affective Engineering Design

Kansei Affective Engineering

International Journal of Powder Metallurgy Volume 45 Issue 4

International Journal of Powder Metallurgy Volume 44 Issue 4

International Journal of Powder Metallurgy Volume 43 Issue 4

International Journal of Powder Metallurgy Volume 46 Issue 4

International Review of Neurobiology, Volume 20

Journal of Organizational Change Management - Volume 20 Issue 3 (2007) - Special Issue: Organizational Change: Over-Psychologized and Under-Socialized?

Feminist Review: Journal, Issue 44

Feminist Review: Journal,Issue 49

Israel Journal of Mathematics, Volume 4, Issue 4 (1966)

Feminist Review: Journal, Issue 43

Feminist Review: Journal, Issue 47 (Feminist Review)

Feminist Review: Journal, Issue 46 (Feminist Review)

Feminist Review: Journal, Issue 38 (Feminist Review)

Feminist Review: Issue 35 (Feminist Review Journal)

Feminist Review: Journal, Issue 40 (Feminist Review)

Journal of Organizational Change Management - Volume 18 Issue 1 (2005)

Feminist Review: Issue 37 (Feminist Review Journal)

International Review of Neurobiology Volume 4

Feminist Review: Journal, Issue 42 (''Feminist Review'' Journal)

Feminist Review: Journal, Issue 41 (Feminist Review Journal)

International Journal of Manpower - Volume 23 Issue 8 (2002) - Special Issue: Work values and organizational behaviour: towards the new millennium

Israel Journal of Mathematics, Volume 4, Issue 3 (1966)

Journal of Semantics, Volume 2, Issue 3-4, 1983

Israel Journal of Mathematics, Volume 4, Issue 1 (1966)

International Review of Industrial and Organizational Psychology, 2006 (International Review of Industrial and Organizational Psychology)

Israel Journal of Mathematics, Volume 2, Issue 4 (1964)

International Review of Industrial and Organizational Psychology

International Review of Industrial and Organizational Psychology, 2003 (Volume 18)

Israel Journal of Mathematics, Volume 4, Issue 2 (1966)

Recommend Documents