Probabilistic analysis of schedule risks for building construction

grantor: University of TorontoResearch was undertaken to develop a method to assist in the determination of the lower and upper activity duration values for schedule risk analysis by Monte Carlo simulation. Belief network was the tool used for this purpose. The development of belief network model consisted of four steps. First, construction schedule risks were identified through a literature review, an expert review, and a group review by a team of experts. Second, cause-effect relationships among these risks were identified through an expert survey. This led to the development of the structure of belief network model. Third, probabilities for various combinations of parents for each risk variable were obtained through an expert interview survey and incorporated into the model. And fourth, sensitivity analysis was performed besides testing with historical data to verify the model. At the end, a user interface was developed using MS Visual Basic for making the model user friendly.M.A.Sc