Probabilistic analysis of project scheduling, how to incorporate uncertainty into project management

Abstract

Project evaluation must consider both investment costs and completion times, both of which are subject to uncertainty. This article presents an approach to explicitly incorporate the uncertainty in the duration of activities necessary to complete a project. The PERT and CPM methods, widely adopted in industry and government since the 1950s, help manage the interdependence of project activities. Although originally differentiated by their handling of time and cost, these methods share the limitation of not adequately considering uncertainty. They use three-time estimates (optimistic, most likely, and pessimistic) for a "probabilistic analysis" of the critical path, but this approach does not address multiple scenarios nor adjust the critical path based on actual time variations. To overcome these limitations, an Excel model was developed that allows for a more appropriate probabilistic analysis of the project's completion time. The article details the objectives of the new procedure and the steps necessary to incorporate uncertainty, using a simple example of constructing an industrial plant. Conclusions and recommendations are presented at the end of the article.