A New Technique for Estimating the Distribution of a Stochastic Project Makespan

A critical success dimension in projects is the ability to complete a project within an estimated duration. In that regard, effective project scheduling techniques in an uncertain environment is of interest in many organizations. In this paper, the authors use an analytic approach to analyze the behavior of time duration distributions of projects in stochastic activity networks, and propose a simple computation scheme for approximating their distribution. The findings offer an understanding of the large gap between PERT and simulation results, and the deviation of projects from their intended schedules. In addition to providing theoretical framework, the proposed approach also recommends a simple practical pragmatic technique that computes the time distribution of project duration. This is a simple and handy tool for the project manager that may replace simulation. As a byproduct, the earliest start time distribution for each activity is also estimated.