Research on the Two-Stage Location Problem of Logistics Distribution Centers with the Consideration of Energy Saving

随着新能源汽车的快速发展，电动物流车配送中心选址与路径规划问题(EVLRP)逐渐成为物流行业关注的焦点。然而现有研究主要聚焦于传统燃油车的选址与路径问题，忽视了电动车辆特性及建成环境等因素，导致选址与路径规划结果实际可行性不足。针对电动物流车能源补给特性和城市选址空间异质性，构建两阶段电动物流车配送中心选址与路径优化模型。第一阶段首先通过改进AP聚类算法生成合理的候选选址策略，将熵权法和TOPSIS方法引入相似度矩阵构建，充分考虑需求点周围建成环境因素，然后构建基于综合成本的配送中心选址优化模型；第二阶段以电动物流车能耗最小为目标函数，构建节能导向的电动物流车配送路径规划模型，并采用自适应模拟退火算法对最优路径进行求解。通过对苏州工业园区148个需求点和19个充电站的实际数据进行实例分析，对该方法的有效性进行验证。实例结果表明，改进后的模型能够显著降低物流配送过程中的能耗与总成本，较改进前的模型综合成本减少28.9%。研究构建包含选址策略生成、路径求解算法和成本优化模型的完整技术框架，为物流企业实现低碳运营和成本优化提供了理论依据与实施方案。With the rapid development of new energy vehicles, the problem of electric logistics vehicle distri-bution center location and route planning (EVLRP) has gradually become a focal point in the logis-tics industry. However, existing studies primarily focus on location and routing issues for tradition-al fuel-powered vehicles, neglecting the unique characteristics of electric vehicles and the built en-vironment. As a result, the feasibility of location and route planning outcomes is often insufficient in practical applications. To address the energy replenishment characteristics of electric logistics vehicles and the spatial heterogeneity of urban location selection, this study develops a two-stage optimization model for electric logistics vehicle distribution center location and route planning. In the first stage, an improved Affinity Propagation (AP) clustering algorithm is employed to generate reasonable candidate location strategies. The entropy weight method and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) are incorporated into the similarity matrix con-struction to fully account for the built environment factors surrounding demand points. A distribu-tion center location optimization model is then formulated based on comprehensive cost consider-ations. In the second stage, an energy-efficient route planning model for electric logistics vehicles is developed, aiming to minimize energy consumption. The adaptive simulated annealing algorithm is used to solve for the optimal delivery routes. The proposed method is validated through a case study using real-world data from 148 demand points and 19 charging stations in Suzhou Industrial Park. The results indicate that the improved model significantly reduces both energy consumption and total logistics costs, with an overall cost reduction of 28.9% compared to the baseline model. This study establishes a comprehensive technical framework encompassing location strategy gen-eration, route optimization algorithms, and cost optimization models. The findings provide theo-retical support and practical solutions for logistics enterprises to achieve low-carbon operations and cost optimization.