Rosandich, Ryan GErquicia, Santiago2007-08-082007-08-082005-06-01CTS 05-06https://hdl.handle.net/11299/963Monte Carlo simulation is the currently accepted method for quantifying uncertainty in projects. It was the goal of the research presented in this report to develop a purely computational technique, based on traditional probability theory, for quantifying project uncertainty with accuracy equal to or greater than that of Monte Carlo simulation. Series and parallel operators were developed for combining independent task uncertainties in project networks. The operators were used to compute overall project uncertainty given individual task uncertainty, and to calculate slack and the degree of criticality for each task. Additional techniques were developed to deal with networks where the series and parallel operator were not enough, specifically those with path dependencies. Results equal or exceed the accuracy of Monte Carlo simulation, but computational times exceed those of Monte Carlo simulation for networks with many dependencies.42