MONTE CARLO SIMULATION WITH EXACT ANALYSIS FOR STOCHASTIC PERT NETWORKS

摘要

This paper deals with one of methods for estimating the probability distribution of the project completion time in stochastic PERT network. This method combines Monte Carlo simulation with exact analysis. The exact probability distribution for a subnet is used to improve an estimator of probability distribution for larger networks. In the paper computational results for different distributions of task duration are presented. More complicated then unimodal distributions are considered also. Both symbolic and simulation calculation were made with MatLab and own procedures.