Adaptive simulation budget allocation for determining the best design

We consider the problem of allocating a given simulation budget among a set of design alternatives in order to maximize the probability of correct selection. Prior work has focused on deriving static rules that predetermine the number of simulation replications to be allocated to each design. In contrast, we formulate the problem as a Markov decision process (MDP) and propose a dynamic myopic scheme to adaptively allocate simulation samples based on current estimates of the means and variances of the design alternatives. We provide numerical examples to illustrate the performance of the proposed dynamic allocation rule.