On the comparison of solutions in stochastic simulation–optimization problems with several performance measures

In a stochastic simulation context, iterative methods of optimization which perform at each step of their optimization procedure a comparison between two different values of the objective function, need the use of statistical tests in order to properly evaluate and compare the simulation results. However, when the objective function to optimize is a multi-criteria function involving several performance measures, conflicting situations may arise. In particular, these situations are observed for models of manufacturing system, where the performance criteria to evaluate may be contradictory. In this case, the statistical test may lead to the rejection of acceptable solutions. To avoid this, we propose a simple and effective procedure to evaluate a multi-criteria function when it is computed as a weighted sum of different performance measures. This procedure allows retention of all solutions which present a significant improvement, while other solutions are rejected.