The Project Manager's Challenge
Even so, some problems remain. I have addressed these previously, but because they are still not addressed in the latest Guide, I feel compelled to restate the following concepts again.
One problem facing the project manager is that on the one hand he or she wants to manage both schedule and cost to realistic expectations (using either most likely or weighted average estimates), and on the other hand wants a high degree of certainty that he or she will complete the project within both schedule and cost constraints.
This is a challenge because a realistic estimate, at best, represents a 50:50 probability that one can deliver the project for the agreed upon cost and schedule, or do better, as shown in Figure 3. In reality, because of the nature of the statistical chances of being correct when making estimates, our chances of completing the project on the favorable side of the estimate when using most likely estimates are actually less than 50:50. As shown in the graph, the "most likely" value has the greatest chance of being correct, but is still overly optimistic.
Because of the skew of the estimating probability curve shown in Figure 3, the weighted average drives the expected cost higher and duration longer. If that is the case, what project manager wants to stake their reputation on delivering less than half their projects on or below budget and/or on time or ahead of schedule? Yet we still want to manage to realistic expectations.
Figure 3 - Estimating Probability Curve
Contingency reserves can be used to increase the probability of completing the project within agreed-to targets to a higher confidence level. For example, perhaps you would like to commit to a schedule and budget that you are 80%, 90%, or even 95%, sure you will complete within targets (see Figure 4). This assumes that the contingency reserve is incorporated into the performance management baseline, which is the benchmark against which performance measurements are taken and reported.
But on the other hand, what project manager wants his or her project team to think that they have all this extra time and budget to work with, or management to think he or she has padded the estimates? Does anyone doubt that if we make the reserve amounts of time and budget visible that the team is likely to deliver against the longer schedule and higher budget figure, even if no risks materialize that require the allocation of these reserves? It is all too easy for project team members to lose their sense of urgency to keep the project moving forward and closely manage costs and time.
Figure 4 - Impact of changing the odds
In Part 2 of this paper, I will recommend
a solution to these issues.
21. Parish, Roger. "PMBOK® Guide Fourth Edition - Unraveling Project Reserves". March 2010.