Optimization Design of Automotive Parts Logistics Network Based on Facility Location-Routing

To address inefficiencies in Company X’s automotive parts logistics network—caused by the rapid expansion and geographic dispersion of its suppliers (371 in East China, with uneven spatial distribution: concentrated in the Yangtze River Delta, scattered in northern Jiangsu/southern Zhejiang, and significant supply volume gaps) and the inherent limitations of the existing two-stage (supplier-central distribution center) structure—this study proposes a three-stage optimization framework (supplier-distribution center- central distribution center) by adding distribution centers to reduce costs and improve efficiency. Focusing on the East China network, the methodology is systematic: Firstly, K-means clustering was applied to analyze 371 suppliers, with 10 iterations identifying 6 candidate distribution centers (1 excluded for proximity to the central distribution center); Secondly, a linear programming model integrating fixed (distribution center), transportation, and transshipment costs is built to minimize total logistics expenses; Thirdly, a genetic algorithm optimizes milk-run routes to cut trip frequency. Empirical results show significant improvements: the optimized network selects Wuxi Huishan K1 (Fengwu Intelligent Manufacturing Park) and Shanghai Anting J1 (central distribution center) as collaborative nodes, restricts delivery radius to ≤300 km (down from over 500 km), saves 25,481 yuan in annual transportation costs, eases the original hub’s load, and enhances vehicle scheduling efficiency. This study contributes threefold: (1) a “clustering-selection + model optimization” framework for supplier dispersion; (2) a cost-driven location model validated via sensitivity analysis of fixed/transshipment/transportation costs; (3) practical verification via Company X’s case, providing references for similar automotive parts enterprises.