Optimizing Safety Stock Placement in Large Real-World Automotive Supply Networks Using the Guaranteed-Service Model

Abstract

This paper presents an optimization model for the placement of safety stocks in multi-echelon supply networks using the Guaranteed-service Model. Our model handles complex network topologies and multiple products while examining the impact of service level and service time on total costs, formulated with mixed-integer linear programming. We utilize a unique network dataset acquired through data mining of financial databases to generate scenarios that reflect the complexities of real-world supply networks of five major automotive corporations. Experimental results validate the effectiveness of a dynamic-programming based solver in obtaining optimal solutions within large general network topologies. Furthermore, a sensitivity analysis reveals a negative correlation between safety stock costs and the maximum allowed service and a positive correlation between safety stock costs and the service level.