A Brief Overview of Critical Chain Project Management

Critical Chain in the Single Project Environment

Goldratt (1997) introduced the concept of Critical Chain Project Management for Single Projects (CCPM-SP or Critical Chain) to begin to address the problems associated with the more traditional methods of PERT/CPM and Gantt charts. As presented later, CCPM-SP addresses many of the guidelines listed previously, but not all of them. Guideline I concerns recognition of project type. Guidelines II and VII deal with the development of the project network, while Guideline XII is concerned with multiple projects. The resulting project network is a feasible, but not necessarily optimal, project plan.

Figure 2-2 shows a typical activity-on-node PERT/CPM project network. Realistically, the completion of project activities requires the use of resources. Furthermore, resources are typically limited-there are only X number of programmers, or bulldozers, or whatever. Assume that the project shown in Fig. 2-2 is to be completed using three different resources. Figure 2-5 shows the same AON network as Fig. 2-2, but with the addition of resources. The shading on the diagram denotes resource use: A and B, C and D, and E and F share a common resource.

The reader will quickly note that the activity times have been reduced by 50 percent. This reduction, at least partially, addresses Guidelines IV, VIII, and X. In addition, there has been another arrow added to the diagram. This dashed arrow represented the priority of resource use within the project-addressing Guidelines III and VI. By using the PERT/CPM technique of forward- and backward-passes through the network considering this newly added dashed arrow, the ES/EF and LS/LF times can be determined. Those activities with zero slack are critical activities. However, the sequence of activities (A-D-C-F) does not correspond to a PERT/CPM path, so the term chain is used to denote the difference between a PERT/CPM path (which considers only technological precedence) and a CCPM-SP chain (which considers both technological and resource precedence). Since all of the activities on the chain A-D-C-F have zero slack, this chain is called the Critical Chain (CC).

Other additions to the diagram are the boxes labeled FB and PCB. These boxes denote feeding buffers and the project completion buffer, respectively. These buffers exist to address Guidelines V, VIII, and IX. Time was taken out of each of the activities in the project, resulting in a.5 probability that each activity will be completed on time. The buffers exist to increase the probability of on-time project completion. The PCB adds time to the end of the project. In this case, since the CC is 20.5 days the PCB would be 10.25 days-the project can then be promised to be delivered in 30.75 days. The feeding buffers exist to protect the CC from variation of non-critical activities. If activities B and E were started on the LS date and suffered any delay, then the CC would be jeopardized. The feeding buffers require that these activities be started sometime before their LS date. (Actual determination of buffer size is left to later chapters.) FIGURE 2-5 Typical activity-on-node project network with resource contention identified (shading shows same resource use).

In practice, CCPM-SP requires that all activities on the CC be monitored and started as soon as the previous activity ends in order to take advantage of early completions. This process addresses Guideline XI. Additionally, all resources on the CC are monitored to ensure that multitasking is eliminated or minimized to address Guideline V.

Brief Review of Critical Chain Literature

In the book Critical Chain, Goldratt (1997) first published the concept of CCPM. Like several of his prior texts, the book outlined the concept in a narrative fashion and does not seem to have been intended to be a "how-to" manual for CCPM. Rather, its purpose seems to have been to provide a basis for a stream of research that might be pursued by him and others. Pittman (1994) and Walker (1998) examined the single and multiple project environments (respectively) and sought to expose the assumptions and practice of scheduling and controlling projects by traditional methods. Their work provides the basis for the gedankens presented earlier in this chapter.

Hoel and Taylor (1999) sought to provide a method (via simulation) for determining the appropriate size for the buffers required by CCPM. Rand (2000) introduced CCPM to the project management literature framing CCPM as an extension of TOC. He concluded that CCPM not only dealt with the technical aspects of project management (like PERT/CPM) but also that CCPM dealt with how senior management manages human behavior in the construction of the project network as well as the execution of the network. Steyn (2000) followed this research with an investigation of the fundamentals of CCPM. He concluded that a major impediment to implementing CCPM is that it requires a fundamental change in the way project management is approached and that such a change is likely to meet with resistance.

However, Herroelen and Leus (2001) argued that while CCPM was as important to project management as TOC was to production scheduling, CCPM oversimplified the issue of scheduling and rescheduling. Herroelen, Leus, and Demeulemeester (2002) continued much of the same argument in a later paper. Likewise, Raz, Divr, and Barnes re-examined CCPM and concluded that project performance is often a function of the skills and capabilities of project leaders and that "some CCPM principles do make sense in certain situations" (2003, 31). McKay and Morton (1998) as well as Pinto (1999) were concerned that CCPM might be misapplied by managers who failed to understand the underpinnings of CCPM and who attempted to adopt it without fully changing their fundamental approach to the management of projects.

Answering this criticism, Steyn (2002) sought to apply TOC to a variety of other areas of project management beyond the creation and execution of project schedules. He recognized the multidisciplinary nature of project management and how it affects cash flow, stakeholder needs, and risk management. Yeo and Ning (2002) began work on integrating supply chain management with project management. Sonawane (2004) incorporated systems dynamics with CCPM to create a "modern" project management system. Similarly, Lee and Miller (2004) applied systems thinking to multiple projects along with CCPM, and Trietsch (2005) argued that CCPM is, in fact, a more holistic approach to project management than traditional methods.

Herroelen and Leus concede that CCPM "seems practical and well thought-out...nevertheless, for single projects, the unconditional focus on a 'Critical Chain' seems useless..." (2004, 1616). Srinivasan, Best, and Chandrasekaren (2007) presented a case study that clearly appears to contradict this conclusion. The Warner Robins Air Logistics Center (WR-ALC) is charged with the repair and overhaul of C-5 transport aircraft. After an eight-month implementation period starting in 2005 and without the addition of any resources, WR-ALC returned five additional aircraft to the operational fleet by reducing the number of in-service planes from 12 to 7. The replacement value of these aircraft is $2.4 billion and does not consider nonmonetary benefits such as increased responsiveness and casualty avoidance during wartime.

Summary and Conclusions

The literature of project management relates to the practice of project management and to the theory of project management. The practice emphasizes the large number of project failures, and the theory focuses on fine-tuning algorithms in an attempt to minimize computer time or project duration. Certainly, the two themes should converge to provide a simple but effective approach to project management. One approach to refocusing the theoretical literature is to take a different perspective to project management. Our systems approach attempts to specifically identify several of the sources of project failure.

The purposes of this chapter were twofold. First, we examined the macro issues associated with project management. Second, the micro issues of project networks were examined. The overall objective of this research is not to propose solutions to each of the surface problems revealed in this chapter but to identify and logically link these surface problems to their underlying causes as well as to project failure. One must fully understand the core problems of project management and its environment before proposing a comprehensive solution to these core problems. Without this systems perspective, a proposed solution may create more problems than it solves.

This chapter provides evidence that, not unlike other business environments, the management of single and multiple projects has certain core problems that must be recognized prior to the development of new tools for planning and control. Recent research in the areas of both single and multiple project planning and control has recognized the shortcomings of the PERT/CPM method. Twelve guidelines have been proposed by which the effective planning and control of both single and multiple projects might be improved.

These guidelines reflect fundamental changes to the way single and multiple projects are currently planned and controlled. The objective of any improved planning and control technique should not be to find the optimal solution to each of the problems found in project management, but rather that a feasible or realistic solution be found for all of the problems in project management (such as that presented by Goldratt, 1997 and Newbold, 1999). Furthermore, solutions for any one problem should not be developed in isolation of the other problems. Additionally, complex solutions should not be developed for such solutions are difficult for practitioners to both understand and apply. A realistic planned project completion date that is met is better than an optimal solution that is never met.

The practitioners have voiced the strongest criticisms of current project management methods. This is because the current reward systems are based not upon the method used but upon the results received-on time, on budget, and to full specifications. Recognizing the inadequacies of PERT-based methods in achieving the desired results, practitioners attempt to modify these project management methods.

Practitioners recognize the effects of variability and finite capacity when their projects are completed late and over-budget, but do not understand the underlying reasons for the observed effects. They intuitively "know" that PERT/CPM assumptions are causing their project to fail, but have not recognized that their own behavior is also a cause. These behaviors, such as project managers seeking to delay spending or to avoid late penalties, or resource managers increasing planned activity duration to protect their resources, are driven by policies and measures. (A full discussion of policies and measures, how they influence behavior, and how to align individual behaviors with corporate goals is far beyond the scope of this chapter.) Important recommendations to practitioners may be made because of the research presented in this chapter. Project managers should understand that estimates of activity duration are prone to over-estimation and, counter-intuitively, often lead to poor project performance. Project managers should also understand that multiple projects are interdependent due to the shared use of common resources. As such, decisions made with respect to one project may have detrimental effects on other projects, even projects that have not yet started. Additionally, a reward system should be developed that recognizes the completion rather than the duration of both activities and projects. Resource managers need to understand the concept of a Critical Chain and must also take advantage of early activity completions. Finally, project managers should not base planned project completion dates on PERT-based plans, but rather on some method that recognizes the shared use of common resources and the existence of statistical fluctuation within and across projects.

Researchers have identified many of these surface problems in their studies; however, no comprehensive examination of the causes of these surface problems has been undertaken. We feel this approach is not enough. To provide a practical framework for reducing project failures, a systems approach must be taken to identify both macro and micro surface problems, core drivers (environmental factors), and core problems with the PERT/CPM methodology. We do not propose a comprehensive solution to addressing project management; however, we do provide some guidelines to start a dialog with other researchers in developing a more effective and practitioner-friendly approach to project management. Researchers should use these guidelines as a starting point to develop algorithms that are more robust. Goldratt's Critical Chain method offers promise in addressing many of these problems. It has been used effectively in a limited but growing number of different environments. That method and others need to be developed and refined to provide a systems perspective encompassing the needs of project managers, resource managers, and organization managers.

Policies, procedures, measures, planning, and control methods need to be re-examined as indicated by the current reality trees of single and multiple project organizations. Underlying conflicts among the goals and measures of managers create many of the surface problems seen in a project management environment. These conflicts must be resolved by providing supporting policies, procedures, and measures. Given that these can be devised and successfully implemented, a systems perspective must be utilized to identify all of the core drivers in a given environment and the planning and control system so structured to accommodate these core drivers. The project environment has several common core drivers that must be incorporated into any planning and control methodology. We tried to identify a number of these and provide guidelines for managers to consider in planning and controlling projects. These guidelines should also provide the foundation for further research into developing and testing effective methodologies for planning and controlling projects.

Academia needs to shift emphasis from defining a good algorithm from one that minimizes computer time or finds the shortest completion time to determining ways to construct networks to guarantee completion of the project on the plan and methods of immunizing projects against statistical fluctuation. The recognition that in the presence of statistical fluctuation and dependent events, lateness accumulates are essential. Methods of eliminating or minimizing the effect of the accumulated lateness on project completion are needed. Strategic buffering of resources, paths, and networks in single and multiple projects must also be studied.

References

Abdel-Hamid, T. K. 1993. "A multiproject perspective of single-project dynamics," Journal of Systems and Software 22(3):151165.

Allam, S. I. G. 1988. "Multi-project scheduling: a new categorization for heuristic scheduling rules in construction scheduling problems," Construction Management and Economics 6(2):93115.

Avots, I. 1970. "Why does project management fail?" Management Review 59(10):3641.

Badiru, A. B. 1992. "Critical resource diagram: A new tool for resource management," Industrial Engineering 24(10):5859, 65.

Badiru, A. B. 1993. "Activity-resource assignments using critical resource diagramming." Project Management Journal 24(3):1521.

Bildson, R. A. and Gillespie, J. R. 1962. "Critical Path Planning-PERT Integration", Operations Research 10(6):909912.

Black, K. 1996. "Causes of project failure: A survey of professional engineers," PM Network 2124.

Bock, D. B. and Patterson, J. H. 1990. "A comparison of due date setting, resource assignment, and job preemption heuristics for the multiproject scheduling problem," Decision Sciences 21(2):387402.

Brooks, F. P. 1995. The Mythical Man-Month. Anniversary Edition. Boston: Addison-Wesley.

Brown, D. 2001. "Lack of skills to blame for project failures," Canadian HR Reporter 14(17): 112.

Clark, C. E. 1961. Comments on the Proceeding Paper (The PERT Model for the Distribution of an Activity Time). Operations Research 10(3):348.

Coulter III, C. 1990. "Multiproject management and control," Cost Engineering 32(10):1924.

Davis, E. W. 1966. "Resource allocation in Project Network Models-A survey," Journal of Industrial Engineering 17(4):177188.

Davis, E. W. 1973. "Project scheduling under resource constraints: Historical review and categorization of procedures," AIIE Transactions 5(4):297313.

Davis, K. R., Stam, A., and Grzybowski, R. A. 1992. "Resource constrained project scheduling with multiple objectives: A decision support approach," Computers & Operational Research 19(7):657669.

Dean, B. V., Denzler, D. R., and Watkins, J. J. 1992. "Multiproject staff scheduling with variable resource constraints," IEEE Transactions on Engineering Management 39(1):5972.

Deckro, R. F., Winkofsky, E. P., Hebert, J. E., and Gagon, R. 1991. "A decomposition approach to multi-project scheduling," European Journal of Operational Research 51(1):110118.

Dumond, J. 1992. "In a multiresource environment: How much is enough?" International Journal of Production Research 30(2):395410.

Dumond, E. J. and Dumond, J. 1993. "An examination of resourcing policies for the multi-resource problem," International Journal of Operations Management 13(5):5476.

Feldman, J. I. 2001. "The seven deadly sins of project estimating," Information Strategy 18(1):3036.

Goldratt, E. M. 1997. Critical Chain. Great Barrington, MA: North River Press.

Granger, C. H. 1964. "The hierarchy of objectives," Harvard Business Review 42(3):6374.

Gutierrez, G. J. and Kouvelis, P. 1991. "Parkinson's law and its implications for project management," Management Science 17(8):9901001.

Healy, T. L. 1961. "Activity subdivision and PERT probability statements," Operations Research 341348.

Herroelen, W. and Leus, R. 2001. "On the merits and pitfalls of critical chain scheduling," Journal of Operations Management 19(5):559577.

Herroelen, W. and Leus, R. 2004. "Robust and reactive project scheduling: A review and classification of procedures," International Journal of Production Research 42(8):15991620.

Herroelen, W., Leus, R., and Demeulemeester, E. 2002. "Critical chain project scheduling: Do not oversimplify," Project Management Journal 33(4):4960.

Hoel, K. and Taylor, S. G. 1999. "Quantifying buffers for project schedules," Production and Inventory Management Journal (40)2:4347.

Hughes, M. W. 1986. "Why projects fail: The efforts of ignoring the obvious," Industrial Engineering 1418.

James, G. 2000. "Beware of consultants peddling snake oil," Computerworld 34(39):40.

Kelley, J. E. 1962. "Critical-path planning and scheduling mathematical basis," Operations Research 296320.

Kerzner, H. 1994. Project Management: A Systems Approach to Planning, Scheduling, and Controlling. 5th ed. New York: Van Nostrand Reinhold.

Kim, S. and Leachman, R. C. 1993. "Multi-project scheduling with explicit lateness costs," IIE Transactions 25(2):3444.

Klingel Jr., A. R. 1966. "Bias in PERT project completion time calculations for a real network," Management Science 13(4):194201.

Krakow, I. H. 1985. Project Management with the IBM PC Using Microsoft Project, Harvard Project Manager, Visischedule, Project Scheduler. Bowie, MD: Prady Communications Co.

Krakowski, M. 1974. "PERT and Parkinson's law," Interfaces 5(1):3540.

Kurtulus, I. 1985. "Multiproject scheduling: Analysis of scheduling strategies under unequal delay penalties," Journal of Operations Management 5(3):291307.

Kurtulus, I. and Davis, E. W. 1982. "Multi-project scheduling: Categorization of heuristic rules performance," Management Science 28(2):161172.

Kurtulus, I. and Narula, S. C. 1985. "Multi-project scheduling: Analysis of project performance," IIE Transactions 17(1):5866.

Lawrence, S. R. and Morton, T. E. 1993. "Resource-constrained multi-project scheduling with tardy cost: Comparing myopic, bottleneck, and resource pricing heuristics," European Journal of Operational Research 64(2):168187.

Lee, B. and Miller, J. 2004. "Multi-project software engineering analysis using systems thinking," Software Process Improvement and Practice 9(3):173214.

Lee, S. M., Park, O. E., and Economides, S. C. 1978. "Resource planning for multiple projects," Decision Sciences 9(1):4967.

Levy, F. K., Thompson, G. L., and Wiest, J. D. 1962. "The ABCs of Critical Path Method," Harvard Business Review 98108.

Lowery, G. and Stover, T. 2001. Managing Projects with Microsoft Project 2000 for Windows. New York: John Wiley & Sons.

Malcolm, D. G., Roseboom, J. H., and Clark, C. E. 1959. "Application of a technique for research and development program evaluation," Operations Research 646669.

Marks, N. E. and Taylor, H. L. 1966. "CPM/PERT: A diagrammatic scheduling procedure," Studies in Personnel and Management (18). Austin: Bureau of Business Research, Graduate School of Business, University of Texas.

Matta, N. F. and Ashkenas, R. N. 2003. "Why good projects fail anyway," Harvard Business Review 109114.

McKay, K. N. and Morton, T. E. 1998. "Critical chain," IIE Transactions 30(8):759762.

Meredith, J. R. and Mantel, S. J. 2003. Project Management: A Managerial Approach. 5th ed. New York: John Wiley & Sons.

Middleton, C. J. 1967. "How to set up a project organization," Harvard Business Review 7382.

Miller, R. W. 1962. "How to plan and control with PERT," Harvard Business Review 93104.

Millstein, H. S. 1961. "Comments on the proceeding paper (Healy)," Operations Research 349350.

Mohanty, R. P. and Siddiq, M. K. 1989. "Multiple projects-Multiple resources constrained scheduling: A multiobjective approach," Engineering Costs & Production Economics 18(1): 8392.

Neimat, T. 2005. "Why IT projects fail," The Project Perfect White Paper Collection. http://www.projectperfect.com.au.

Newbold, R. 1999. Project Management in the Fast Lane: Applying the Theory of Constraints. Boca Raton, FL: St. Lucie Press.

Paige, H. W. 1963. "How PERT-cost helps the general manager," Harvard Business Review 8795.

Parkinson C. N. 1957. Parkinson's Law and Other Studies in Administration. New York: Random House.

Payne, J. H. 1995. "Management of multiple simultaneous projects: A state-of-the-art review," International Journal of Project Management 13(3):163168.