A Monte Carlo Simulation Approach to Cost-Uncertainty Analysis

An important aspect of cost research is the measurement of the uncertainty inherent in the projection of system cost. Approaches to this problem have in the past centered on the decision makers intuition or in sensitivity analysis. Only recently have approaches utilizing such tools as statistical decision theory and probability theory been formulated. This study focuses on the Monte Carlo simulation approach to uncertainty in cost analysis. This approach requires a Expression of input estimates as probability distributions reflecting uncertainty. b Cost equations pertinent to a particular model. The Monte Carlo simulation approach then generates a The frequency distribution for system cost. b Statistical measures that illustrate the nature and magnitude of system cost uncertainty. Two models are developed, the Beta model and the Weibull model, each of which reflects a particular distribution form for the inputs. The relative costs and advantages of each model are compared. A users guide to the program and complete program listings are presented in an appendix.