Efficient sampling for simulation-based optimization under uncertainty

We address the efficiency issue for simulation-based optimization under uncertainty. In such a case, there are several design alternatives to simulate and each simulation has its own uncertainty to manage or reduce. We present a very efficient sampling approach to manage the overall uncertainty so that the total simulation time can be minimized. We also compare other allocation procedures, including a popular two-stage procedure in simulation literature. Numerical testing shows that our approach is much more efficient than all compared methods. Comparisons with other procedures show that our approach can achieve a speedup factor of 3~4 for a 10-design example. The speedup factor is even higher with the problems having a larger number of designs