A simulation-based optimization method for production-distribution network design

Production-distribution network design is a critical decision that has significant impacts on a supply chain's long-term performance. Supply chain dynamics, such as demand fluctuation, and transportation instability, are omitted in most mathematical models due to tractability. We present a simulation-based optimization method in this paper for multi-criteria production-distribution network design. The method consists of a multiobjective optimizer and a simulation module. The optimizer, based on a multiobjective genetic algorithm, is used to direct the search for compromised solutions regarding to various criteria. Candidate solutions are evaluated by the discrete-event simulation module, developed in a flexible manner that enables automatic simulation of various supply chain structures without human intervention. The method is applied to a case study from automotive industry. A set of Pareto-optimal solutions are obtained, including decisions on the open/close decisions on facilities, order splitting ratios and inventory control policies.

[1]  Carlos A. Coello Coello,et al.  An updated survey of GA-based multiobjective optimization techniques , 2000, CSUR.

[2]  Michael C. Fu,et al.  Feature Article: Optimization for simulation: Theory vs. Practice , 2002, INFORMS J. Comput..

[3]  Zuo-Jun Max Shen,et al.  An Inventory-Location Model: Formulation, Solution Algorithm and Computational Results , 2002, Ann. Oper. Res..

[4]  Michael C. Fu,et al.  Optimization for Simulation: Theory vs. Practice , 2002 .

[5]  Terry P. Harrison,et al.  Global Supply Chain Management at Digital Equipment Corporation , 1995 .

[6]  Mehrdad Tamiz,et al.  Multi-objective meta-heuristics: An overview of the current state-of-the-art , 2002, Eur. J. Oper. Res..

[7]  Hongwei Ding,et al.  A multiobjective optimization method for strategic sourcing and inventory replenishment , 2004, IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004.

[8]  Marc Goetschalckx,et al.  Strategic production-distribution models: A critical review with emphasis on global supply chain models , 1997 .

[9]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[10]  Ray J. Paul,et al.  Simulation optimisation using a genetic algorithm , 1998, Simul. Pract. Theory.

[11]  Russell D. Meller,et al.  The interaction of location and inventory in designing distribution systems , 2000 .

[12]  Arthur M. Geoffrion,et al.  Twenty Years of Strategic Distribution System Design: An Evolutionary Perspective , 1995 .