Overview of an Economic Analysis of Simulation Selection Problems

This paper summarizes a new approach that we recently proposed for ranking and selection problems, one that maximizes the expected NPV of decisions made when using stochastic or discrete-event simulation. Our formulation assumes that facilities exist to simulate a fixed number of alternative projects, and we pose the problem as a “stoppable” version of a Bayesian bandit problem. Under relatively general conditions, a Gittins index can be used to indicate which system to simulate or implement. We provided an asymptotic approximation for the index that is appropriate when simulation outputs are normally distributed with known but potentially different variances for the different systems. Costs included variable costs of simulation and discounting due to simulation analysis time.