Requirements for a Software Process Maintenance Framework for Executive Information Systems in the Telecommunications Industry

This paper presents the interview findings of the current situation in software development for Executive Information Systems (EIS) in the Thai telecommunications industry and identifies critical success factors (CSFs) that can serve as requisite requirements for a software process maintenance framework in this context. The results show that software development teams do not perceive existing formal routines as an efficient way to manage software development processes and eventually to deliver quality results. One of the factors that play a central role in quality software development is an efficient software process. Efficiency requires project management activities to enable the proper task execution. Unfortunately, existing agile methods, e.g., Scrum and XP offer only inadequate support for project management, e.g., limited support for subcontracting and developing software that demands high level of quality control. For continuously efficient software processes, this study thus proposes a framework aimed at the establishment and improvement of software development processes covering adequate software engineering and project management perspectives. The overall goal of the resulting framework therefore is to contribute to the improvement of software process development. KEY-WORDS: Software Development, Project Management, Software Process Improvement, Software Process Maintenance 1. MOTIVATION AND BACKGROUND The best known traditional software development approach still is the waterfall approach, which in fact is the oldest original approach. It is a systematic and sequential pattern reaching from an initial feasibility study to the maintenance of the developed IS. However, there are several limitations, e.g., well-defined requirements, timeconsuming, too much documentation and high cost (Jirachiefpattana, 1996). Agile methods, e.g., eXtreme programming (XP), Scrum, Dynamic Systems Development Method (DSDM) and Feature Driven Development (FDD), were thus developed to overcome these limitations. They are gaining recognition in the software development community due to their response to market expectation, i.e., innovative and high quality software (Highsmith & Cockburn, 2001). Moreover, CSFs for successful agile software development are identified by a multitude of studies (Attarzadeh & Ow, 2008; Ceschi et al., 2005; Chow & Cao, 2008; Gemuenden & Lechler, 1997). However, software development methods should be efficient (Kumar & Welke 1992). Efficiency requires project management activities to enable the proper execution of software development tasks. Project management thus provides the backbone for efficient software development (Abrahamsson et al., 2003; Gilb, 1988). From this view, some agile methods (e.g., Scrum, DSDM and FDD) are supplemented with guidelines on project management that allow for rapid delivery of quality products. Nevertheless, there generally is no comprehensive project management support (Abrahamsson et al., 2003). Scrum which is definitely the most popular (Shalloway & Trott, 2009) offers limited project management support, e.g., for subcontracting, developing with large teams, developing Journal of Global Management Research 8 Copyright© Institut Fidal Inc (2010) software that demands high quality control, and distributed development environments (Highsmith & Cockburn, 2001; Ionel, 2008; Shalloway et al., 2009; Turk et al., 2002). Albeit, researchers such as Turk et al. (2002) suggest that traditional project management practices are an applicable way, so far no integrated method offering adequate project management support to overcome these Scrum‘s limitations has been identified. PMBOK is the broadest and most widely used standard reference of industry best project management practices (Thomas & Tilke, 2009) and definitely compatible with agile ways (PMI, 2008). Hence, there is not a great need but also a great opportunity to develop an integrated PMBOK and Scrum approach. Besides, the quality of the software development process results in the quality of software (Humphrey, 1989). A software development process generally deals with how it can be implemented, but not so much with what processes should be implemented. Thus, only ―how‖ cannot guarantee that software quality will be delivered. Software process improvement (SPI) can produce the quality of the software development process (Huang et al., 2005) that results in software quality (Lehman, 1991). CMMI is a well-accepted model for improving the performance of software development processes and software quality, and referring to what processes should be implemented to achieve successful software development (SEI, 2002). Hence, CMMI is considered as an effective way to maintain software process for continuous efficiency. Motivated by the situation described above, this paper thus proposes a software process maintenance framework, which in this context means efficient software process establishment and improvement. This paper also focuses on EIS development in the Thai telecommunications industry as a case study. Recently, the use of EIS has significantly increased since the success of EIS in developed countries stimulates a number of executives to adapt this Information System (IS) into their organizations in order to compete in an increasingly competitive environment. EIS are different from Transaction Processing Systems (TPS), Management Information Systems (MIS), and Decision Support Systems (DSS) in terms of problems addressed, users and data used. TPS serve operational management by performing and recording the daily routine transactions necessary to conduct the business and solve structured problems which have standard solutions. Both MIS and DSS provide middle management. However, there are different characteristics for the way in which MIS deal with summarized and compressed data from the TPS and sometimes does an analysis of that summarized data to solve structured problems, while DSS use data from TPS, MIS, and external sources to solve semi-structured problems which only part of the problem has a structured quality. EIS provide information for top management to solve unstructured problems which have no standard solutions for resolving the situation, so that they can identify problems and opportunities by combining internal and external information that is relevant to decision making (Jirachiefpattana, 1996; Laudon & Laudon, 2009). EIS can directly aid and support communications, coordination, planning and control functions of managers and executives in an organization. Supporting this, Nord & Nord (1995) argue that utilizing EIS software can provide valuable benefits, i.e., better communication, increased confidence in decision making, and eventually increased profits. Moreover, Telecommunications still is one of the most rapidly evolving competitive markets and one of the fastestgrowing areas of technology in the world. Thailand‘s telecommunications sector is worth mentioning that it has continued to grow, in the last five years. It generated 8.1 billion dollars in service revenue in 2008 (Pyramid Research, 2009). The main engine of growth is a broadband service that its growth rate was more than 700% from 2003 to 2004 (The World Bank, 2008). Lately reported, Thailand was one of five Asian countries ranked among the world‘s top ten fastest-growing consumer broadband markets in 2008 (Point Topic, 2009). EIS require fundamental revision and software development methods that must be able to deal with rapid evolution (Jirachiefpattana et al., 1996). Unfortunately, EIS is likely to be more difficult in Thailand due to difficult software development environments, e.g., organizational culture, lack of user participation and inappropriate methods (Jirachiefpattana et al., 1996). ―Do the problems identified in prior research still exist in EIS development in the Thai context? Do the problems involve both software engineering and project management aspects? What are factors affecting successful EIS development?‖ are three fundamental questions of this study. The main methods used during this research are qualitative analysis via interviews and literature review methods. For the interviews, two organizations were identified and are thought to representative of EIS in the Thai telecommunications industry. Field data collection was performed during March and April 2009. Questions about the software development process were developed for project managers, developers and coordinators. These questions were answered in structured interviews that ranged in length from one to two hours. Journal of Global Management Research 9 Copyright© Institut Fidal Inc (2010) 2. A LOOK AT THE CURRENT SITUATION IN EIS DEVELOPMENT IN THE THAI TELECOMMUNICTIONS INDUSTRY For getting an idea of the current situation in the EIS development in the Thai telecommunications (by focusing on Internet services), we use findings of interviews with software development teams working for two companies: True Corporation Public Company Limited and TOT Public Company Limited. These companies are the two biggest broadband Internet Service Providers (ISP) in the Bangkok region and have their own optical fiber cable networks in Bangkok and in the vicinity. Even though there are many ISP in Thailand, most of them still lease bandwidth from one of these two companies. With 85% True Corporate also has the largest market share (Thailand Guru, 2009; The World Bank, 2008). However, the size of companies does not affect the model of EIS development. In the organizational context of EIS implementation, the results reveal that the executives could sometimes not provide adequate participation in the projects. Subordinates did not have full authorities when it came to making decisions. Communication processes during EIS in organizations also were quite complicated, e.g., executives or users do neither have good cooperation nor do they participate well. These limitations resulted in