A Bayesian Framework for the Incorporations of Priors and Sample Data in Simulation Experiments

In this article, we propose a theoretical framework to estimate performance measures in simulation experi- ments, incorporating both sample data from a random component and priors on input parameters of the simulation model. Our approach takes into account both the inherent uncertainty of the model as well as parameter uncertainty. We discuss the estimation of a conditional expectation under a Bayesian framework and point and variability estimators are proposed when direct sampling from the posterior distribution is not allowed. The application and properties of the proposed meth- odology are illustrated through an inventory model and simulation experiments using a Markovian model.