Performance Evaluation of Complex Decentralized and Centralized Multi-Echelon Distribution Supply Chain

Abstract

In this paper, we consider a multi-echelon multi-products distribution system, made up of a single supplier, and multiple distributors, wholesalers and retailers. Each participant places its orders to the upstream (previous) node according to a continuous review policy. To be in the real complexity of this system, we have presented demand, lead time and capacity as stochastic variables. In a multi-echelon distribution system, several decision centers interact to keep up the performance of this system. In this context, there are two control system architectures whereas each one differs from another to afford to the supply chain a better ability to meet the customers’ demand, namely, the centralized and the decentralized architecture. We first develop a generic discrete event simulator, which can be used as a support for simulation of both centralized and decentralized control system. Subsequently, the results of our model provide a performance analysis based on comparing local and global decision making.