Entropy as a Measure of Uncertainty for PERT Network Completion Time Distributions and Critical Path Probabilities

This study proposes that the entropy function provides a simple and useful tool for project managers to better cope with project uncertainty, and therefore better manage projects. The entropy function has a long and established history as a measure of uncertainty in information theory. If selected activities on PERT networks are modified by reducing their activity times by one or more time units, simulations on the original and modified networks can generate output on the entropies of completion time distributions and critical path probabilities. Modified networks can be ranked on a scale of decreasing entropy (or decreasing uncertainty) to determine which activities on the original network have the greatest impact on an overall reduction in project uncertainty. In this manner, these activities would be identified as worthy of additional resources in the real world to implement reductions in their activity times.