Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Department of Operational Sciences

First Advisor

Kenneth W. Bauer, Jr. PhD


This research makes significant contributions towards improving the efficiency of simulation studies using an external analytical model. The foundation for this research is the analytical control variate (ACV) method. The ACV method can produce significant variance reduction, but the resulting point estimate may exhibit bias. A Monte Carlo sampling method for resolving the bias problem is developed and demonstrated through a queueing network example. The method requires knowledge of the parameters and approximate distributions of the random variables used to produce the ACV. Often, some of these parameters or distributions are not known. Both parametric and non-parametric alternatives to the Monte Carlo method are explored for these cases. Significant variance reduction using an ACV indicates that the outputs of both models are highly correlated. This relationship is exploited and a new methodology is developed for conducting searches of a simulation design space using an analytical model vice a simulation model. The justification for the new surrogate search method is based on validating the analytical model to the simulation model. The effectiveness of the method is demonstrated on two simulation models including the HQ AMC Mobility Analysis Support System (MASS) model.

AFIT Designator


DTIC Accession Number



The author's Vita page is removed.