Simulation-based optimization and its application in multi-echelon network stochastic inventory system

For a three-echelon network stochastic inventory system, it is difficult for the analytical method to get the optimal inventory control policy. The simulation model, in which the customer arrival time follows a Poisson process, all of the customer demand, the customer purchasing behavior and the lead time is stochastic and the manufacture production capacity is limited, is built on the basis of the discrete event system simulation theory. Then the simulation-based optimization method is used to solve the problem by combining the simulation model and the particle swarm optimization algorithm together. The simulation sample not only demonstrates the feasibility and the effectiveness of the simulation-based optimization method by comparisons and analyses but also shows its applicability in the supply chain management.