Multiobjective Simulation Optimization Using Stochastic Kriging
Jianxia Zhang, Yizhong Ma, Lian-yan Zhu · Atlantis Press eBooks · 2016
Multiobjective simulation optimization aims at finding Pareto optimal solutions and a common approach is to rely on metamodels to alleviate computational costs of the optimization process. We present a stochastic Kriging based multiobjective optimization algorithm to estimate the Pareto fronts of multiobjective simulation optimization problems. After the objective functions were replaced by stochastic Kriging metamodels using limit simulation costs, techniques developed for deterministic multiobjective optimization can be applied to these metamodels. Numerical experiment of a multiobjective (s, S) inventory system illustrates the potential of stochastic Kriging in stochastic simulation optimization which is especially useful in Operations Research and Management Science.