Large-scale information system becomes increasingly important in the informatization of the communities and enterprises. Limited researches had focused on the methodologies and tools which are crucial to the development of large-scale information system (IS). This report explores these issues and analyzes the large-scale information system development (ISD) situation in China. In addition, a China Telecom EAP system case is studied.
Nowadays, the rapid development of information technology (IT) has brought an enormous amount data pressure on the society. It becomes a significant problem to process such huge data, because their features are complex (Zykov, 2006). Large scale information systems are a mainstay of the informatization of a country. How successful it is developed and how efficient it deals with various data have a great significance both on technology and society. With this dependency, more efficient and in time development of large complex information system is in demand, and this IS should be easily to adapt with the varying requirements (Hevner & Linger, 1996).
To develop a successful large-scale IS, careful planning and reliable methodological foundations are needed (Kitano et al., 1993). Despite lots researches and report had focus on Information system development, there is limited ones had looked into large-scale information system development in China. The evaluation and selection of methodologies and tools for large ISD in practice are little. For these reasons, it is essential to study the methodologies and tools based on large-scale information systems, especially in China. This report will start with a brief overview of Chinese ISD situation and give some basic information of large-scale IS. Then, it will focus on the methodologies and tools for large-scale IS. The features and comparison of these methodologies will be shown. Besides, the MIS (management information system) of China Telecom will be discussed to show how the large-scale IS brings breakthrough to large Chinese corporations.
Information system development situation in China
With the development of knowledge economy, especially after China entered WTO, Chinese government has been persisting in using information technology to propel and accelerate industrialization. The development of "Digital City", the spread of ERP (Enterprise Resource Planning) and the appearance of the "Virtual Enterprise" concept are all the results of informatization.
IS, as the information carrier, plays a more and more important role in digitizing of the communities, enterprises and organizations. The development of information system has become one of the hottest technologies. Currently, some large enterprises and public institutions have made a significant investment in the development of information system. For instance, banks, supermarket, IT business, book shop and telecommunication enterprise all have built their internal large IS.
However, as a populous nation with numerous enterprises and public institutions, China is still in the primary stage of information industry when consider the overall level. To improve this level, the demand of ISD is growing rapidly, especially the large-scale IS.
Large-scale information system
The size of an information system is a relative concept. Generally speaking, large-scale IS can be defined by some features. Firstly, the goals and functions is multiple and complex. Secondly, the data collected and processed is massive and various. Besides, the structure and external environment is more complex. In addition, for some nation-wide and world-wide information systems, the scopes of them are extensive and apply to a wide range of areas.
Large-scale systems can be divided into division-wide systems, corporate-wide systems and nation-wide systems (Kitano et al., 1993). In China, common large-scale information system on the basis of application system includes government information system, management information system (MIS) of large enterprise and organization, local MIS and social information service system. For example, the ERP management system of China telecom is one kinds of large scale MIS for enterprise.
A real world case: ERP system of China Telecom
China Telecom (China Telecommunications Corporation), an outsized nationalized telecom operator in China, has been one of the Top 500 Global Corporations for many years as well as the world's largest fixed line operator (China Telecommunications Corporation, 2010). The chairman of China telecom stated in their Annual Report 2005 that China Telecom had contributed a lot to the informatization of community and accelerated the process to an information-based national economy. This arises in two ways. On one hand, China Telecom mainly provides the integrated information services which include the fixed-line telephone, mobile service, applications services and Internet connection. On the other hand, China Telecom had developed different management information systems to support these services. Figure 3-3 shows some "login" interfaces of management information systems of China Telecom.
As the figure shows, China Telecom has lots information system with different purpose and functions. However, China Telecom is a so large corporation that they need a management system to manage all the process for business. Therefore, they have the challenges to build a large MIS which complies with international, reporting rules, integrate all major business functions and achieve real-time management (China Telecommunications Corporation, 2010).
Their solutions were building an ERP (Enterprise Resource Planning) management system. ERP system has shared data stores, which can help China Telecom manage all the information and functions of their business and company. This system can provide support in accounting and controlling, production and material management, quality management and plant maintenance, sales and marketing, project management, and human resources (Bobek et al., 2002). ERP based E-business improves business performance. The key benefits of ERP system for China Telecom included initial focus on essential business needs, comprehensive business insight and analysis, enhanced internal collaboration and increased employee productivity.
ANALYSIS AND FINDINGS
From the situation discussed above, it can be found that it is essential to develop large-scale IS to drive the informatization of China. To make the large-scale ISD successful, methodologies and tools selected appropriately to use is very important.
It is crucial to understand the concept of ISD methodology first. Generally speaking, methodology can be defined from technical aspect as the framework which contains the structure, planning, and process controlling (Centers for Medicare & Medicaid Services, 2005). However, there is a debate for decades that methodology is not just a study of technical method. Maddison et al. (1984) augmented that methodology should also include documentation, management and training for developers of information systems. There are large amounts of researchers support for this wider scope that methodology should reflect social changing and different perspectives and experience. The following discussion will refer to both the academic and practice issues.
There are a variety of methodologies over these years, each of them has its own advantages and disadvantages. Loosely speaking, they can be categorized as structure life-cycle, prototyping, object-oriented and CASE methodologies. Prototyping is usually used for small-medium ISD and CASE is a new one which is immaturity. They are all not suit for the large-scale ISD. Among these methodologies, the structured life-cycle and object-oriented methodologies are the most outstanding and widely used for large-scale information system development.
Since mid-seventies, the structured life-cycle methodology has played a dominant role in the methodology for ISD. In this methodology, the process of information system is divided into different phases based on ISLC (information system life cycle). Every phase has its tasks in isolation, and the tasks in different stage are finished systematically. This methodology diversifies a complex process, makes it into separated and easier section which is convenient for the distribution of responsibility and can reduce the difficulty of the entire development. However, structured life-cycle methodology will take a long time to complete the ISD, because a new phase can start only until the previous phase is finished. This weakness may lead to ISD crisis in such a day with a rapid development of IT.
To solve the timelines problem, object-oriented (OO) methodology was raised and became the trend in nineties. This methodology based on the object, specific tools are used to realize the exchange between object description and system structure. There are lots advantages of OO methodology. Firstly, it is simple. OO model makes the real world object clear and reduces the complexity of structure. Secondly, it is easy to modify and extend the system. New features can be added and the changing operating environments can be responded by inventing a few new objects and modifying some existing ones. Furthermore, OO allows objects be maintained separately and reduce the costs, which is suit for the complex ISD. In addition, the object in one IS can be reused in other systems. This advantage increases the development efficiency. At the same time, this methodology cannot be applied without specific tools to support. Besides, OO is a bottom- up methodology. In large ISD, it may leads to a pool structure of IS by using bottom-up methodology only rather than dividing the system by top-down methodology (structure life-cycle is a top-down methodology).
With these advantages and disadvantages of structure life-cycle methodology and object-oriented methodology, how to select and use them for large-scale ISD is a problem. Perhaps for the historical factor, methodology on basis of linear ISD structured life-cycle still seems to be the most widely used (Gasson, 1995). However, for large-scale information system, the development life-cycle is long and has a wide range. There are great amount of maintenance work during the development, let alone after the system has been put into practice. It is obvious that the object-oriented approach has strengths on the maintenance of IS. Consequently, the object-oriented methodology will become the tendency of large-scale ISD. However, just like mentioned above, a top-down method is essential to structure the system. A combination of these two methodologies can be used. In practice, this is regularly the fact.
It can be seen that during different life cycle of SID, different methodologies are used.
Gasson (1995) pointed in his report that the concept of "methodology" got from academic research was limited to the design stage of the ISD life-cycle, and some methodologies do not apply to the reality. In practice, a combination of methods used to ISD is common. In addition, many large enterprises and organizations will build unique methods which is adapt to their own situations. According the research down by Russo et al. (1996), 37% of the methods used in organizations are a mix of the common methods and internal methods designed or modified by their own organizations or projects.
There is another fact in practice, information system development is not just a technical activity, but also brings a social change. It can be seen as the product of a new social system which is based on the information technology. Therefore, to support the definition with a wider scope, alternative development methods on management aspects have been developed (Gasson, 1995). These methods concentrate on the development strategy while the traditional ones are focus on the development process. Gasson had given a classification which includes, for example, project management, risk management, and business process re-designs. Management approaches can help reduce ISD crisis and lead to an efficiency work. The management of the methodology should be paid attention to as well as methodology itself. It refers to different perspectives, experiences and how methodologies are used (Gasson, 1995).
Generally, tools should be a part of methodology. This report states tools separately is to address the CASE tool. Tools for ISD should support both the information systems and the methods used. It is important to select appropriate and optimal tools. Tolvanen (1998) reported that the development of technology had caused almost all the tasks of ISD down with lots of tools. Nevertheless, the CASE (Computer-Aided System Engineering) had developed to cover all kinds of computer tools which include business modeling, requirements capture and implementation tools. Furthermore, CASE is combined with 4GLs (the fourth-generation programming language) which are widely used to build medium- to large-scale systems. For this reason, there is a trend that CASE tool will be more and more popular and there has been much reliance on it in practice. In large organization, CASE tool is paid attention to by Information system managers, since it can help them integrate the system throughout the organization (Gasson, 1995). From the above, CASE will be the best tool of large-scale ISD, for either the technical facts or the social issues.
There are different kinds of CASE tools. The common ones are IBM Rational Rose, Model Maker, Power Designer, Enterprise Architect, MicrosoftVisual Visio and Visual Paradigm. Figure 1 to Figure 4 show 4 types UML model build by CASE tool.
Use Case Diagram describes the functionality and behavior of the system during requirements elicitation and analysis (Bruegge & Dutoit, 2004).
Class Diagram describes the structure of the system, includes objects, classes, attributes, operations and their associations (Bruegge & Dutoit, 2004).
Sequence Diagram presents the dynamic behavior of the system and show the communication among objects (Bruegge & Dutoit, 2004).
Deployment Diagram describes the relationship among run-time components and hardware nodes (Bruegge & Dutoit, 2004).
It can be seen from the figures that CASE tool makes the structure and design of IS more visual, easily to understand and modify. These features can speed up development process, reduce costs, and increase productivity of large-scale information system.
A successful information system should be well designed architecture, contain more performance and flexible, have less programming work and lower costs. The large-scale ISD is always a process with high cost and lots human resources. Consequently, before the system is designed, good methodologies should request an outline of the constraint condition such as costs, progress, schedule, equipments, and so on. This method avoids duplication and reduces costs. Furthermore, it can help to develop a system with high quality, which will decrease correction and modification.
In the procedure of information system development and maintenance, the sustainability of data sources is always a problem. Project team members should record the system development information during the process. This can help reduce the loss on data sources when a member leaves this project team.
To sum up, large-scale ISD in China had brought lots benefits to the enterprises and communities. China Telecom is a good example. However, it is necessary to improve the development of these kinds of system to drive the overall informatization level of China. Methodologies and tools appropriate for a large-scale IS should be selected by considered the technical, organizational, project and team factors, further than academic and technical aspects. In addition, practical methodologies not only contain the technical methods but also have managerial (social) approaches.
- Bobek. S, Potocan. V, Sternad. S, Spicka. H, (2002) Information Systems in Virtual Corporations: Issues for ERP Based E-business Systems. Available at: http://informingscience.org/proceedings/IS2002Proceedings/papers/bobek209Infor.pdf, viewed 10 December 2009.
- Centers for Medicare & Medicaid Services, (2005) Selecting a Development Approach. Available at: http://www.cms.hhs.gov/SystemLifeCycleFramework/downloads/SelectingDevelopmentApproach.pdf, viewed 25 December 2009.
- China Telecom Corporate, Company overview. Available at: http://en.chinatelecom.com.cn/corp/index.html, viewed 10 December 2009.
- China Telecom Corporation Limited, Annual Report 2005: Chairman's statement. Available at: http://www.chinatelecom-h.com/eng/ir/reports/annual2005/annual2005_05.pdf, viewed 25 December 2009.
- Gasson, S. (1995) The Role of Methodologies in It-related Organizational Change. Available at: http://188.8.131.52/bitstream/1860/1968/1/2006175275.pdf, viewed 15 December 2009.
- Hevner, A. R. Linger, R. C. (1996) Methods and Tools for Information Systems Development: Introduction to Minitrack.
- Kitano, H. Shimazu, H, Shibata, A. (1993) Case-Method: A Methodology for Building Large-Scale Case-Base Systems. Available at: http://www.aaai.org/Papers/AAAI/1993/AAAI93-046.pdf, viewed 10 December 2009.
- Maddison, R. Baker, G. Bhabuta, L. Fitzgerald, G. Hindle, K. Song, J. Stokes, N. & Wood, J. (1984) Feature Analysis of Five Information Systems Methodologies, in T. Bemelmans (Ed.) Beyond Productivity: Information Systems Development For Organizational Effectiveness, Elsevier Science Publishers, North Holland.
- Russo, N. Hightower, R. Pearson, J. (1996) The failure of methodologies to meet the needs of current development environments. In: Lessons learned from the use of methodologies (Proceedings of the 4th Conference on Information System Methodologies), (eds. N. Jayaratna, B. Fitzgerald), British Computer Society, pp. 387-394.
- Tolvanen, J. P. (1998) Incremental Method Engineering with Modeling Tools: Theoretical Principles and Empirical Evidence. Available at: http://users.jyu.fi/~jpt/doc/thesis/ime.html, viewed 20 December 2009.
- Weitz, L. (1990) Future of 4GLs lies in CASE integration - fourth-generation languages integrated with computer-aided software engineering tools Available at: http://findarticles.com/p/articles/mi_m0SMG/is_n2_v10/ai_8167986/, viewed 20 December 2009.
- Zykov, S. V. (2006) Web-based Enterprise Information System Development: The Integrated Methodology. Available at: http://arxiv.org/ftp/cs/papers/0607/0607119.pdf, viewed 10 December 2009.