Systems Design Issues in Planning and Implementation: Lessons Learned and Strategies for Management Mahesh S. Raisinghan...
19 downloads
443 Views
78KB 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
Systems Design Issues in Planning and Implementation: Lessons Learned and Strategies for Management Mahesh S. Raisinghani
Idea Group Publishing
Systems Design Issues in Planning and Implementation
IDEA GROUP PUBLISHING
1331 E. Chocolate Avenue, Hershey PA 17033-1117, USA Tel: 717/533-8845; Fax 717/533-8661; URL-http://www.idea-group.com
1
16#$!
g n i h s i l b u P p u o r G a e SystemsigDesign Issues in Planning d I t h r Copy and Implementation: Lessons Learned and Strategies for hing s Management Publi
up o r G a e d I t h g i r y EXECUTIVE SUMMARY Cop Mahesh S. Raisinghani University of Dallas, USA
Telecommunications Company (TC) [company identity is concealed] produced a sales management application through internal and contract resources. This application, Schedule Graph (SG) System, was designed to automate the sales schedule process that had previously been a paper and pencil process. The system was designed and implemented in a matter of months to reduce cost and deliver an application that was long overdue. The project had been proposed for years, but funding issues had routinely delayed initiation. The sales development organization worked on the design and development for this application for approximately six months. The application was released with numerous software, hardware and network problems. The effects on the customer community, the information systems department and other stakeholders were sharp and far reaching. This case study investigates the lessons learned with this application and the implications for theory and practice. It can be instrumental to information systems managers, academicians and students to learn from the success and pitfalls of other organizations related to information systems development and management.
g n i h s i l b u P p u o r G a e d I t h g i r y p o C BACKGROUND g n i h lis b u P p u o r G a e d I t h g i r y p Co TC is a Fortune 100 sales company in the telecommunications industry with 50 regional sales offices across the country. TC employs thousands of people with both domestic and international operations, however, the user base for the SG System is approximately 150 employees nationwide. TC wanted to automate a sales scheduling process. Previously, directory sales representatives had been scheduled in markets or canvasses by a paper and pencil process. This process was very time consuming and led to tremendous frustration among sales managers. The burdensome manual process took valuable time away from sales coaching and selling activities that produced revenue dollars. The sales calendar stems 12 months and is typically updated on a weekly basis as personnel and markets can change rapidly in their business.
Copyright © Idea Group Publishing. Copying without written permission of Idea Group Publishing is prohibited.
2 Raisinghani
SETTING THE STAGE The sales managers had been requesting an automated solution for years in an effort to end what had rapidly become an administrative job rather than a sales job. The sales organization conducted business and processed sales online, a ‘Paper-less Automated Sales System.’ While the system itself was paper-less, the output as a hard copy report was paper-intensive in spite of on-line capabilities. While the automation was highly desirable and the efficiencies could not be argued, there were financial considerations and constraints that continued to push the project to the back burner for several years. The sales organization, in a time of declining sales, realized it was time to redirect its focus back into the development of its employees in an effort to strengthen its sales position. This meant that the manual processes needed to be removed by investing in process automation. An application needed to be developed to automate the sales schedule process. Time and money were significant factors in the development of the SG System.
a e d I t h g i r y CASE DESCRIPTION p Co
g n i h s i l b u P p u o r G
g n i h lis b u P p u o r G a e d I t h g i r y p Co g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co
In-house application developers as well as contract resources were involved in the design and development of the scheduling system. Rapid Application Development (RAD) was used to produce the SG system. The systems development life cycle was significantly reduced to save time and money for a project that had been long awaited in the customer community. Many applications that are produced from the RAD framework are developed in isolation, since that contributes to its speed to market. The deliverables and outcomes of RAD are the same as for the traditional Structured Development Life Cycle (SDLC)–a systems development plan, which includes the application being developed, a description of the user and business requirements for the application, logical and physical designs for the application, and the application’s construction and implementation, with a plan for its continued maintenance and support. However, the traditional SDLC is indifferent as to the specific tools and techniques used to create and modify each of the deliverables listed above; RAD puts a heavy emphasis on the use of computer-based tools to support as much of the development process as possible in order to enhance the speed of development. In this case study, the focus is on the use of RAD instead of SDLC. However the SG System suffered from not involving other organizational business units. This is noted as a significant drawback with RAD because traditional development stages are able to have greater overall business understanding, as speed is not the primary concern (Hoffer et al., 1998). In fact, David E.Y. Sarna (Eskow, 1997), who is well known for his work with RAD, argues that network planning and monitoring are important issues that can often be overlooked when development takes place in isolation. Specifically for optimal application and system performance, server(s) must have adequate memory, processing capacity, and Redundant Array of Individual Disks (RAID) are recommended for back-up and data security. After the SG System was released to production, significant time was spent rewriting the application, deploying upgraded hardware, integrating network technology and developing a support staff to maintain the application. Additionally, the project and development costs were significantly increased with the activities required to stabilize the system. Approximately $350,000 was spent to purchase additional hardware and maintenance, and almost $550,000 was invested in application coding changes for the system to run more efficiently. Although $900,000 was not a significant percentage of TC’s overall budget, proactive systems planning could have minimized or eliminated this expense and caused less frustration for the approximately 150 SG System users nationwide. It is also important to note that the additional maintenance cost is not uncommon in applications developed using the RAD methodology due to lack of attention to internal controls (e.g., inconsistent internal
Systems Design Issues in Planning and Implementation
3
designs within and across systems), lack of scalability designed into the system and lack of attention to later systems administration built into the system (e.g., not integrated into overall enterprise data model and missing system recovery features). In contrast, Office Depot jeopardized short-term profitability in 1996 when the company decided to delay most of its new application development for almost a year. The decision was made in an effort to stabilize an application portfolio before it was released. While the company suffered in the absence of short-term gains, the additional time spent on development has had a long-term positive effect on their bottom line due to better quality results from the applications produced (Hoffman, 1998). This is a sharp contrast to TC’s results with their release of SG. As noted in the SG discussion above, speed was the primary motivator in the development process. In the long run, neither time nor money was saved as a result of the rewrites and additional hardware acquisitions, that were expended in an effort to rescue a system that was the product of poor planning and design. The current system architecture of the SG system is illustrated in Figure 1 and described in the following section.
g n i h s i l b u P p u o r G a e d I t h g i r y p o C g n i h s i l Current System Overview b u P up o r G a e d I t h g i r y Cop g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co
Purpose: Software: Tools: Client: Database:
On-line interactive tool for sales force market scheduling In-house vendor product Oracle Developer 2000: Oracle, Procedure Builder, SQL Plus & SQL Net Sales Managers Oracle version 8i
Figure 1: Current SG System Architecture
Application Server with ORACLE DBMS
Print Server connected to the shared laser printer
UserWorkstations
UserWorkstations
4 Raisinghani
System Architecture DEC Server: Oracle database resides in a DEC cluster - VMS version 6.3. Oracle Database: Users are connected to the Oracle database via TCP/IP and SQLNET. User PC Configuration: • Processor: Pentium • Memory: 64 MB • Network: Token Ring or Ethernet • User Interface: Windows 95/98/2000 • Communications: OnNet 1.1 FTP TCP/IP & SQL Net 2.1.4C Application Server: Application is distributed on local Banyan Servers to the desktop. Network Protocol: Application generates two significant types of network traffic. Banyan traffic is generated from the PC to the local Banyan Server. Additionally, SQL Net traffic is generated across the Wide Area Network (WAN) for database connectivity.
a e d I t h g i r y p Co
g n i h s i l b u P p u o r G
There are a number of considerations that must be reviewed when new software is being developed. Additionally, there are a number of technical and customer groups that must be involved in this process. • Network: The network will likely play a part in any new application or system that is developed today. In the case of SG, the network implications were not investigated before the system was released to production. The system was not written to efficiently run across the network. When the application was released, it was determined that the network response was too slow because the application creates too much traffic across the wide and local area networks. The application required extensive bandwidth for customers to access as well as update the schedule graphs. Additionally, the client hardware was not powerful enough to mask the network response issues. To correct the current situation, the application was rewritten and eventually loaded and launched from the client’s PC. The network services group created a special distribution utility that would load any software updates from the server to the desktop. The client checks the server at log-in and if there is a new version of software, it downloads the changes to the desktop. This process should have been in place before the application was released rather than after as damage control. • Hardware Considerations: Applications for the most part are more sophisticated today with GUI interface and Web-enabled technology that dictates greater desktop processing power, memory, hard drive space, etc. With this in mind, it is important to understand what equipment your client community will be using to access the application you are designing. If an organization’s customers have not yet upgraded to Pentium processors, you will need to address this in the design or possibility in the project budget as recommendation for new equipment may be appropriate. In the case of SG, the customer community had been accustomed to working on “dumb” terminals (DEC: VT420 & VXT2000) and only recently been given recycled personal computers that ranged in models from Intel 80386 to low-level Pentium personal computers. When SG was released, customers with 486 machines could use the application, but with very poor response, and those with 386 machines could not access the application because they did not have the minimum configuration. Needless to say, this caused significant customer issues and dissatisfaction from the client community. To correct this situation, the computer equipment group replaced the customer’s equipment with Pentium II and III model computers. Again, a situation that could have been avoided with proper planning and requirements investigation. • Customer and Supplier Involvement: Customer involvement is very important in the design of any new system. This is especially true with rapid application development as the development time is shorter. The need for customer input and involvement is significantly higher because there is no time for misunderstandings or multiple reworks of the requirements or the code design.
g n i h lis b u P p u o r G a e d I t h g i r y p Co g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co
Systems Design Issues in Planning and Implementation
5
Additionally, a strong partnership with customers will likely yield important feedback and help the information technology (IT) organization to produce quality software by better understanding customer needs. Increasingly, organizations collaborate to complement their core competencies. New product development for TC, for example, is often a collaborative process, with customers and suppliers contributing complementary knowledge and skills. Information technology facilitates interorganizational learning since organizations collaborate closely through virtual integration. The role of information technology in lower and higher levels of interorganizational learning, cognitive and affective trust, and virtual and humanistic interorganizational collaboration should be leveraged (Scott, 2000). • Planning – Application Testing & Design: The notion of ‘write once, run anywhere’ may or may not be realistic yet in the software industry when many software vendors, of which Java is a good example, do not offer the same templates, training and features (Kobielus, 1998). However, planning is an important means to get there. Without proper planning before, during and after the designing and testing phases, there is little chance for implementation success. Quality in the form of customer needs is what must be planned in the design of any system software. The software must be functional and easy to use based on their frame of reference. Additional concerns include the speed at which the software performs based on customer equipment and interfaces. Anticipating future user needs and allowing flexibility in the design models are further examples of situations that should be addressed in the planning process to help ensure quality results. Thorough testing both from a technical perspective and a customer perspective are also vital events that require planning to achieve a successful implementation. Thorough testing of the technology including data integrity and product integration are key quality metrics. Additionally, customers must be involved in the testing process to verify functionality and utility of the system. • Maintenance & Support: Before a system is released to production, there should be a clearly defined support organization to sustain the application and resolve any technical or training issues. In the case of SG, a support organization was not established or trained nor was any formal documentation in place with information about the system design. This was a difficult obstacle to overcome. It was almost six months before a technical review document was distributed and system overview sessions were held for key support groups including: operations, production control, network operations and help desk personnel. These activities should be part of a project plan for the release of any new software or system. These activities should be performed long before the system is released, as it will dramatically increase the success of the project and the satisfaction of the customer community. The field of information technology moves very quickly, and it can be difficult to determine what new technology will be embraced by corporate America based on stability and business needs. Ravichandran and Rai (2000) identify top management leadership, a sophisticated management infrastructure, process management efficacy and stakeholder participation as important elements of a quality-oriented organizational system for software development. Businesses typically adopt a standard, as it is easier to implement, control and maintain. Over the next five years, “adaptive architectures” will be a primary design point for enterprise-wide technical architectures (Meta Group, Inc., 1998). The adaptive nature will allow more rapid change to be made to business applications and processes with reduced impact. It is predicted that by the end of next year, half of global 2000 businesses will cross-train technologists and business strategists (Meta Group, Inc., 1998). The merger should produce better information technology products and services which is a strong profit source. Global markets have increased significantly with Internet commerce and opened many more business opportunities. The Internet is and will be a driving force for years to come (Stevenson, 1995). Microsoft went through a reengineering process two years ago as part of a business strategy to
g n i h s i l b u P p u o r G a e d I t h g i r y p o C g n i h s i l b u P up o r G a e d I t h g i r y Cop g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co
6 Raisinghani
dominate the Web as it has the desktop (Porter, 1997). Finally, as a result of business process reengineering and corporate transformation due to the digital economy, improving process has become an important area. From a systems development perspective, it is important to keep in mind that the issue is not process; it is what programmers are asking process to do, and where to apply it. Armour (2001) points out three laws of software process that are relevant in the context of this case study: 1) Process only allows us to do things we already know how to do. 2) People can only define software processes at two levels: too vague and too confining. 3) The last type of knowledge to consider as a candidate for implementation into an executable software form will be the knowledge of how to implement knowledge into an executable form.
g n i h s i l b u P p u o r G Facing the Organization Current Challenges/Problems a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co
In the last decade, there has been a great deal of attention and discussion on Business Process Redesign that is largely tied to the Total Quality Movement (Lee, 1995). Work processes in many large corporations have been under executive scrutiny to improve quality and customer satisfaction. Continuous improvement in all business processes is the corporate goal. Business processes have two important characteristics: 1) internal and external customers, and 2) cross functional boundaries (Davenport, 1993). Developing new software is a good example of a high impact business process. If we relate this to the design of SG, there should have been internal and external customers and crossfunctional involvement in the development process. Controlling business processes is important to their success and also a key to instituting incremental quality improvements. Information technology organizations have started to take on a customer advocate role as a means of controlling business and customer processes. “IT Capabilities should support business processes, and business processes should be in terms of the capabilities IT can provide” (Davenport and Short, 1994). Adherence to business processes maximizes efficiency and positive results. Competition in the information technology field is fierce, specifically with respect to software development. This has had a strong impact on information technology organizations. Many system vendors are offering additional services to corporate clients including multi-platform support, application development and integration services as a means of increasing revenue beyond hardware sales. This is positive for users because it will likely drive prices down and create more product choices for customers. The supplier, however, will only gain competitive advantage if the product is innovative, quality rich, service friendly and priced reasonably. Hardware suppliers are recognizing the importance of service and emphasizing this with customers (Vijayan, 1997). There are several focal points that will determine the survival of specialized vendors or inter-company resources and they can be narrowed down to the following: 1) quality (features, ease of use, tools, etc.); 2) service (performance, ability to enhance, etc.); 3) price (Anderson et al., 1997). Competition can be tempered with current strategy and careful planning. Technology is a key ingredient in any corporate growth strategy (Way, 1998). The business units can be aligned properly to supply quality software on demand with aggressive planning including tools, requirements, design, testing, etc. Among the key networking O.S. companies, Microsoft has a long-term plan to fully integrate its development environments. Microsoft believes the result will reduce software development time and reduce product prices as a result of the efficiencies gained (i.e., training time and money, etc.) (Gaudin, 1997). There is an inherent difficulty that surrounds software development. The nature of software is complex and unstable. Software, in spite of successful implementations, requires change over time whether it stems from customer request or vendor recommendation. “I believe the hard part of building software to be the specification, design and testing of this conceptual construct, not the labor of representing it and testing the fidelity of the representation” (Brooks, 1985). Realizing that complexity is an inherent and unavoidable obstacle, it becomes more important to focus on planning and designing the appropriate technology to avoid unnecessary updates. A well-designed system will lead to greater stability and reduce maintenance costs.
Systems Design Issues in Planning and Implementation
7
The release of SG was a challenging experience for the technical team as well as the customer community. How can future software releases be improved based on this experience: • Focus on Quality – Quality is an overriding business strategy for most global companies. Customer satisfaction is a key success factor for application groups developing software inhouse. If customers are not satisfied with the quality of the software produced, they can very easily find a contract vendor to develop any type of needed software. As previously mentioned, competition is fierce in the IT environment and customers have many choice which makes quality a distinguishing factor in success. • Importance of Planning – While there is no easy cookbook for software development, planning is an essential ingredient to a successful software implementation. Proper planning will help minimize rework and contribute to quality and customer satisfaction. Planning will also provide more opportunity for reuse of models, coding, and testing. • Understand your Customer – Extensive customer involvement is very important in the design and development of any new system, but particularly so with those developed under RAD. In the case of SG, customers should have been more involved in the requirements, design and implementation phases. • Initiate Cross-Functional Involvement – Successful software development requires participation from multiple information technology groups as well as customer groups. A business process should be in place to coordinate the involvement of all appropriate groups. It is suggested that within the next few years, “Architectural Webification” will replace ‘legacy-to-client/server migration’ as the dominant design structure. In the next few years, it is predicted that enterprise-wide technical architectures will have quality measures for ‘time-toimplementation’ and total-cost-of-ownership to benchmark: 1) logical horizon (linking any user to any node); 2) object heterogeneity (ability to share any information, service or process with objects from multiple platforms); and 3) systemic utility (scalability, portability, interoperability, availability, manageability, design transparency, and extensibility). It is expected that many Global 2000 companies will use a ‘software factory’ model to implement new application systems (META Group, Inc., 1998). This approach will require software developers to focus on assembly and reuse as opposed to a craftsman approach. With the strong influence of the Internet and electronic commerce on information technology, and more specifically its impact or potential impact on software development, systems designers and developers need to critically evaluate whether this medium would eliminate “middleware” issues and reduce development costs. Due to a constantly changing environment, a poor understanding of the user’s needs and preferences, as well as a lack of willingness to modify existing organizational structures and decision models, the full economic potential of Web Information Systems (WIS) has not been realized. Reference models in the case of WIS, have to integrate a conceptual data and navigational model and–by choosing a system-specific optimal level of abstraction–should equally be applicable for structured as well as unstructured information. For that reason Martin (1996) introduces the term “intelligent evolution”. With special emphasis on corporate business behavior, he compares three types of evolution with the classic Darwinian evolution based on the survival of the fittest: a. Internal (r)evolution during the pre-deployment phase: First order evolution, modifying a product or service (WIS) within a predesigned process and corporate structure. Second order evolution, modifying the process, methodology, or fundamental design of work (WIS methodology). b. External Evolution: Third order evolution, considering external factors outside the corporation (e.g., relationships with customers, other companies, governmental institutions, standardization committees, etc.).
g n i h s i l b u P p u o r G a e d I t h g i r y p o C g n i h s i l b u P up o r G a e d I t h g i r y Cop g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co
8 Raisinghani
One of the problems regarding modeling and developing hypermedia applications is the strong interdependency between presentation (user interface) and representation (explicit structuring) of published information. Many meta models and design methods for traditional client-server architectures lack the necessary object types for modeling this interdependency and are better suited for highly structured segments. Therefore, it is pertinent to design WIS with a close eye on the architecture of currently deployed systems and to compare them with each other.
FURTHER READING
g n i h s i l b u P p u o r G
B. DeCori (1997). Bridging the Gap. AS/400 Systems Management, 25 (7), 47-50. M.K. McGee (1997). E-Commerce Applications in Three Months. Information Week, 643, 121-122. J. Hibbard (1998). Time Crunch. Information Week, 691, 42-52. K. Bull (1998). Making Magic. InfoWorld, 20 (25), 33. C. Wilde (1997). Do-it-yourself Service. Information Week, 659, 87-90.
a e d I t h g i r y REFERENCES p Co
g n i h lis b u P p u o r G a e d I t h g i r y p Co g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u ro BIOGRAPHICAL SKETCH a G e d I t h g i r y p o C
Anderson, J., Gallagher, S., Levitt, J., and Jurvis, J. (1997). “Technologies worth watching-groupware, suites, security, data marts, and the ever-popular Web will be hot this year,” Information, No. 612, January, , 29-37. Armour, Phillip G. (2001). “The laws of software process” Communications of the ACM, Association for Computing Machinery. Brooks, F. Jr. (1987). “No Silver Bullet: Essence and accidents of software engineering,” IEEE Computer, 20(4), 10-19. Eskow, D. (1997). “Dealing with the aftershock of a new SAP implementation,” Datamation, 43(4), 105108. Gaudin, S. (1997). “Visual Studio takes big step,” Computerworld,31(10), 2. Hoffer, J., George J. and Valacich, J. (1998). Modern Systems Analysis & Design, 486-497, 767, 780781, 806-807, 814-815. Hoffman, T., “Office Depot endures app dev delays to ensure tech future,” Computerworld, 32(11), 1, 97. Kobielus, J. (1998). “Write once, run anywhere: An impractical ideal,” Network World, 15(9), 45. Lee, T. (1995). “Workflow tackles the productivity paradox,” Datamation. META Group, Inc. White Paper, “Enterprise Architecture Strategies,” 1998. (http:// www.metagroup.com/newwhos.nsf/InterNotes/Link+Pages/eas+-+trends) Martin, J. (1996). Cybercorp: The New Business Revolution. New York: Amacom. Porter, P. (1997).“Microsoft gets with the Web,” Software Magazine, 17(7), 101. Rai, Arun, and Ravichandran, T. (2000). “Quality Management in Systems Development: An Organizational System Perspective”, MISQ, 24(3). Scott, Judy E. (2000). “Facilitating Interorganizational Learning with Information Technology” Journal of Management Information Systems, 17(2), 81 – 114. Stevenson, D. (1995). “Positioning enterprise architecture,” ISWorldNet, June 1995. Vijayan, J. (1997). “Hardware vendors profit from integration push,” Computerworld, 31(30), 28-29. Way, P. (1998). “The direct route to America,” Insurance & Technology, 23(1), 35-36.
Mahesh S. Raisinghani is a faculty member at the Graduate School of Management, University of Dallas, where he teaches MBA courses in Information Systems and E-Commerce, and serves as the Director of Research for the Center for Applied Information Technology. He is also the President and CEO of Raisinghani and Associates International, a diversified global firm with interests in software
Systems Design Issues in Planning and Implementation
9
consulting and technology options trading. As a global thought leader on E-Business and Global Information Systems, he serves as the local chair of the World Conference on Global Information Technology Management and the track chair for E-Commerce Technologies Management as well as a world representative for the Information Resources Management Association. He has published in numerous leading scholarly and practitioner journals, presented at leading world-level scholarly conferences and has recently published his book E-Commerce: Opportunities and Challenges. He is the invited editor of the special issue of the Journal of Electronic Commerce Research on Intelligent Agents in E-Commerce. Dr. Raisinghani was also selected by the National Science Foundation after a nationwide search to serve as one the panelists on the Information Technology Research Panel and Electronic Commerce Research Panel. He serves on the editorial review board for leading information systems publications and is included in the millennium edition of Who’s Who in the World, Who’s Who Among America’s Teachers and Who’s Who in Information Technology.
g n i h s i l b u P p u o r G a e d I t h g i r y p o C g n i h s i l b u P up o r G a e d I t h g i r y Cop g n i h s i l b u P p u o r G a e d I t h g i r y Cop g n i h lis b u P p u o r G a e d I t h g i r y p Co