Article Preview
TopIntroduction
Software applications are developed by following a general project management paradigm specified by the Project Management Institute (PMI) and further refined by academics and practitioners within the MIS discipline (Lagerstrӧm, et al., 2012; Dennis, et al., 2012). This approach is used to improve the chances of success and to protect a firm’s investment in the development process (Marchewka, 2009). Despite this reliance on a highly structured methodology, it is often the case that IT projects fail, sometimes to the tune of hundreds of millions of dollars (cf. e.g., Nelson, 2007). While there are a number of reasons why IT projects fail, most failures are the result of people and process problems (Kappelman, et al., 2006; Nelson, 2007). At the top of the causal chain for these problems is project scheduling (Ropponen & Lyytinen, 2000). Scheduling involves assigning resources to sets of tasks that must be accomplished in a specific sequence. The number of resources that can be assigned is limited by the number available and this constraint complicates the scheduling process. This has come to be known as the Resource-Constrained Project Scheduling Problem (RCPSP) and has been the subject of a number of studies in the past 40 years (cf., e.g., Demeulemeester & Herroelen, 1997).
IT Projects are considered failures if they don’t work, don’t work correctly, are significantly over budget or are significantly late. In recent years despite extensive research and sophisticated project management software, projects continue to fail due to schedule and budget overruns (Herroelen & Leus, 2005). Data from the Standish Group’s Chaos report shows that the average percentage of projects that are considered as either a failure or challenged to be 71% from 2011 through 2015 (Standish Group, 2015). Such abysmal performance has prompted additional research into the reasons for both failure (cf. e.g., Kappleman, et al., 2006) and success (cf. Herroelen & Leus, 2005). Successful projects have good user involvement, support from managers, clear requirements and proper planning, and there is general agreement in the literature that technical problems are almost never the cause of IT project failure (Kappleman, et al., 2006). The clear majority of unsuccessful projects occurs because of either process or people mistakes (Nelson, 2007). In fact, Nelson (2007) identified poor estimation and/or scheduling as the primary reason for failure in 54% of the 99 projects in his study.
That poor scheduling ranks so highly among the causes of failure seems to be an odd result for two reasons. First, information technology development is all about creating leading edge software. One would therefore think that IT workers could apply this technology to their own discipline and use it to correct such problems with scheduling, yet there doesn’t seem to be such an ongoing effort. In defense of the IT discipline, however, project management is more than applying algorithms in a computer program, and so the fact that IT has not come up with effective solutions is not so surprising. Second, the amount of research that has been devoted to devising optimal scheduling methods has been vast and great strides have been made; however much of that work has not yet made its way into practice (Herroelen, 2005). The intractability of the problem seems to be what draws researchers, but on the other hand it is what continues to elude practitioners.