多车型冷链物流同时送取货车辆路径优化

Research on the Vehicle Routing Optimization with Heterogeneous Fleet for Simultaneous Delivery and Pickup in Cold Chain Logistics

摘要: 在“双碳”目标驱动下，冷链物流亟需兼顾低碳减排与运营成本双重目标。针对现有研究多维度协同不足、碳排放测算简化等问题，文章提出时空耦合的多车型协同机制，构建车型配置–中心协作–正逆向路径的三维决策模型，集成制冷成本、时间窗柔性处罚等冷链特性约束，建立多目标优化函数。针对模型特点，设计动态自适应遗传算法，通过种群演化状态感知和样本集中度反馈，动态调整交叉率和变异率。仿真实验得出：相比单一车型配送，多车型策略降低总成本25.15%，减少碳排放12%；多中心模式相较于单中心配送距离缩减38.79%，碳排放降低31.03%；同时送取货模式较独立模式节约成本36.95%，验证了模型在资源协同配置与低碳转型中的有效性，为冷链物流企业提供了“车型–网络–流向”多要素联动的决策范式。

Abstract: Driven by the “dual carbon” goal, cold chain logistics urgently needs to balance the dual goals of low-carbon emissions reduction and operating costs. Aiming at the problems of insufficient multidimensional collaboration and simplified carbon emission measurement in existing research, a spatiotemporal coupled multi-vehicle collaboration mechanism is proposed, and a three-dimensional decision model of vehicle configuration, center collaboration, and forward-reverse reverse paths is constructed. The cold chain characteristic constraints, such as refrigeration cost and flexible time window penalty, are integrated, and a multi-objective optimization function is established. Design a dynamic adaptive genetic algorithm based on the characteristics of the model, which dynamically adjusts the crossover and mutation rates through population evolution state perception and sample concentration feedback. Through simulation experiments, it was found that compared to single-vehicle delivery, the multi-vehicle strategy reduces total costs by 25.15% and carbon emissions by 12%. Compared to single-center distribution, the multi-center mode reduces the delivery distance by 38.79% and carbon emissions by 31.03%. The simultaneous delivery and pickup mode saves 36.95% of costs compared to the independent mode, verifying the effectiveness of the model in resource collaborative allocation and low-carbon transformation and providing a decision-making paradigm of “vehicle model-network-flow” multi-factor linkage for cold chain logistics enterprises.