Bayesian Networks and Critical Path Method for Project Scheduling

Introduction

Although all projects are unique in some fashion, they all share the common management constraint of uncertainty. This paper provides a summary and critique of Khodakarami, Fenton, and Neil's (2007) study, "Project Scheduling: Improved Approach to Incorporate Uncertainty Using Bayesian Networks," followed by an analysis of how the article relates to the concepts of project integration management, project scope management, and project time management. A summary of the research and its important findings are provided in the conclusion.

Summary and Critique of the Study

A fundamental characteristic of successful risk management is the ability to effectively manage uncertainty. Uncertainty resulting from a lack of comparable past experience (a major source of uncertainty), the need to trade off between performance measures, and a dearth of accurate information can all cause problems in project scheduling (Khodakarami et al., 2007). The Khodakarami et al. study examines how the widely used scheduling technique — the critical path method — can be incorporated into Bayesian networks to provide project managers with improved uncertainty management capabilities by offering quantifiable measures of the different aspects of uncertainty. This integration is noteworthy because the critical path method has been shown to provide improved decision information for project planning as well as generating information valuable for project management purposes more broadly (Lauren, 2008).

In their study, Khodakarami et al. report that risk management is a key area of project management as described by the Guide to the Project Management Body of Knowledge, 2000 edition (PMBOK Guide), and that a central issue in risk management is the ability to handle uncertainty. According to Khodakarami et al., "Current project risk management processes induce a restricted focus on managing project uncertainty because the term 'risk' has become associated with 'events' rather than the more general sources of significant uncertainty" (p. 39). Although the critical path method is among the most widely used project management tools, it does not take uncertainty into account — a gap that is filled by integrating Bayesian networks. According to these researchers, "The most well-established approach to handling uncertainty in these circumstances is the Bayesian approach. Where complex causal relationships are involved, the Bayesian approach is extended by using Bayesian networks. The challenge is to incorporate the critical path method into Bayesian networks" (Khodakarami et al., p. 40).

To their credit, the authors preface their discussion of Bayesian networks with a concise description of the critical path methodology and its notation, which helps readers understand the need for incorporating an uncertainty component. In this regard, Khodakarami et al. report that, "Critical path method is very simple and provides very useful and fundamental information about a project and its activities schedule. However, because of its single-point estimate assumption, it is too simplistic to be used in complex projects" (p. 41). The authors' extensive use of acronyms, though, makes it difficult to follow some of their more esoteric descriptions of the integration process, notwithstanding their use of illustrative graphics in support of their assertions. Despite this limitation, the authors describe several aspects of their innovative approach that can contribute to project management practice in meaningful ways, including those discussed in the sections below.

Project integration management involves the processes required for overall project coordination (Kendrick, 2003). The methodology developed by Khodakarami et al. allows for the development of various "what-if" scenarios that can model the effects of different processes at different points in time (Khodakarami et al., 2007).

Project Integration Management

The methodology developed by Khodakarami et al. may be less useful for project scope management applications. Project scope management describes the project processes required to ensure that the project "defines the deliverable(s) and that it includes all the work required, and only the work required, for successful project completion" (Kendrick, 2003, p. 10). The work breakdown structure of the critical path method is regarded as part of project scope management as defined in the PMBOK Guide (Kendrick, 2007).

Conclusion

The research showed that the critical path method is among the most widely used project management approaches today. One of the major constraints of the critical path method, however, is its inability to integrate uncertainty — a constraint that is directly addressed by the integration of Bayesian networks with the critical path model. The model developed by Khodakarami et al. remains untested in real-world settings, and the authors concede that more research will be needed before these techniques can be scaled up to satisfy the requirements of project managers in practice.