Minimizing opportunity cost in selecting the best feasible design

Constrained ranking and selection (R&S) refers to the problem of selecting the best feasible design where both main objective and constraint measures need to be estimated via stochastic simulation. Despite the growing interests in constrained R&S, none has considered other selection qualities than a statistical measure called the probability of correct selection (PCS). In contrast, several new developments in other R&S literatures have considered financial significance as the selection quality. This paper aims to lay the foundation of using other selection qualities by attempting to minimize the opportunity cost in allocating the limited simulation budget. The opportunity cost is defined and two allocation rules which minimize its upper bound are presented together with a fully-sequential heuristic algorithm for implementation.