A Simulation-Based Robust Optimization Model for Supply Chain Network Design

Nowadays, in a hotly competitive environment, production-distribution network design is a critical decision that has significant impacts on a supply chain's long-term performance. Generally speaking, stochastic optimization and robust optimization models are two types of optimization models involving uncertainty. In this paper, we present a simulation-based robust optimization method for supply chain in uncertain environment, in which the demands of customers are assumed to be random variable, and the operation costs are considered as fuzzy numbers. The method based on scenario analysis is chosen to describe the circs of uncertain parameter. We establish model and develop a hybrid intelligent algorithm based on genetic algorithm to solve the proposed model. Finally simulation is used to evaluate performance of supply chain configuration and illustrate the effectiveness of model and solution algorithm. The approach is proved to be robust and could handle the large scale of the real world problems.