Date of Award


Document Type


Degree Name

Master of Science


Department of Operational Sciences

First Advisor

Dennis C. Dietz, PhD


The use of control variates is a well-known variance reduction technique for discrete event simulation experiments. Currently, internal control variates are used almost exclusively by practitioners and researchers when using control variates. The primary objective of this study is to explore the variance reduction achieved by the replicative use of an external, analytical model to generate control variates. Performance for the analytical control variates is compared to the performance of typical internal and external control variates for both an open and a closed queueing network. Performance measures used are confidence interval width reduction, realized coverage, and estimated Mean Square Error. Results of this study indicate analytical control variates achieve comparable confidence interval width reduction with internal and external control variates. However, the analytical control variates exhibit greater levels of estimated bias. Possible causes and remedies for the observed bias are discussed and areas for future research and use of analytical control variates conclude the study.

AFIT Designator


DTIC Accession Number