Search space reduction in ordinal optimization for performance evaluation of DEDS

Simulation plays a vital role in analyzing DEDS. However, using simulation to analyze complex systems can be time-consuming and expensive. Particularly, in the case of precise performance evaluation, computing budget, time constraint, and pseudo-random number generator limitations can become prohibitive. Ordinal optimization is an effective approach for improving the efficiency of simulation and optimization of DEDS. However, the ordinal optimization approach does not pay much attention to the problem of large search space. In the paper, for the reduction of search space, we propose a combined approach, ordinal optimization with orthogonal arrays. With this approach, the problem of large search space in stochastic optimization can become more manageable. Consequently, the proposed method can be more efficient for the reduction of simulation burden compared to the conventional ordinal optimization method