Estimating project completion times - simulation and analytic approach

Abstract

According to the PERT method, a project completion time is normally distributed with a mean equal to the critical path's length. The expected length of the critical path is the sum of the expected activities' durations of that path. During the course of our research, we realized that in many cases the conditions of the Central Limit Theorem are not satisfied; hence, the project length is often not normally distributed. In other words, the expected project length is greater than the sum of the activities' duration along the critical path After several iterations of reducing the project length using the CPM method, we achieved a project network with several critical paths. The project length distributes according to the Maxima among all paths. In particular, the project length is longer than any critical path length. We analyzed project completion time using both the analytic approach and simulation. Our results show that in many cases it is biased to use the critical path length when attempting to estimate project completion time. The results of the current study provide users with better estimates and upper bound to the project completion time. These results can be used to improve risk management programs.