Abstract: In order to overcome the short comings of the study of reverse logistics vehicle routing optimization in a reasonable combination of product recovery price adjustment and vehicle routing optimization scheduling, taking the intelligent recycling bin as the research object, and considering the multi-frequency recovery and vehicle sharing scheduling strategy, this paper proposed a reverse logistics vehicle routing optimization scheme based on product recovery pricing. First, this paper established a linear function between the collection quantity and the recycling price. Next, it established a reverse logistics operating cost model including shared vehicle transportation cost, the vehicle maintenance cost, and the penalty cost of the time window violation and environmental externality benefit, and proposed the maximum product profit model of the recycling center. Then, it designed a K-means clustering algorithm according to the characteristics of the model, to consider the space location, recycling frequency, and time window constraints of the intelligent recycling bin, and therefore proposed an improved genetic algorithm-particle swarm optimization (GA-PSO) hybrid algorithm which combined the strong global search ability of GA and the fast conergence speed of PSO. After that, it adopted the elite retention strategy to enhance the efficiency of the hybrid algorithm. A compaison of the hybrid genetic algorithm (HGA), genetic algorithm-tabu search (GA-TS) and hybrid ant colony optimization (HACO) verified the validity of the proposed model and algorithm. Fnally, it studied the proposed method based on a real-world case study of the intelligent reverse logistics network in Chongqing, China, and analyzed and discussed the recycling frequency and vehicle sharing at different product recovery pricing. The results show that the model and algorithm proposed n this paper can be used for effetively selection of the optimal pricing strategy. resource sharing of recycling vehicles, and reasonable vehicle routing optimization scheduling, and can effectively reduce the transportation cost of reverse logistics while maximizing the revenue of the recovery center, which can provide decision reference and method support for reverse logistics enterprises in the product recovery pricing strategy and vehicle routing optimization scheduling.

Key words: product recovery pricing, reverse logistics, multi-frequency recovery, vehicle routing optimization, GA-PSO（genetic algorithm-particle swarm optimization）hybrid algorithm

摘要： 针对逆向物流车辆路径优化问题研究在产品回收定价调整和车辆路径优化调度结合方面存在的不足，以智能回收箱为研究对象，考虑多频次回收和车辆共享调度策略，提出基于产品回收定价的逆向物流车辆路径优化方案。首先，构建了智能回收箱回收量与回收定价的线性函数；然后，构建了包含共享车辆运输成本、维护成本、违反时间窗惩罚成本和环境外部性收益之和最小化的逆向物流回收运营成本模型，并建立了回收中心产品的最大化收益模型；其次，根据模型特点设计了考虑智能回收箱地理位置、回收频次和回收时间窗的时空聚类算法，进而提出一种改进的混合算法，该混合算法结合了遗传算法全局搜索能力强与粒子群算法收敛速度快的特点进行了算法间的优势互补，同时采用了精英保留策略，增强了混合算法的搜索性能，并通过与HGA算法、GA-TS算法和HACO算法进行比较分析，验证了模型和算法的有效性；最后，结合重庆市某智能回收物流网络的实际数据进行优化研究，分析了不同产品定价下的回收频次和车辆共享调度情况。结果表明，本文所提的模型和算法能够进行产品回收定价策略的有效选择、产品回收车辆的资源共享以及合理的车辆路径优化调度，并可在回收中心获得最大化收益的同时有效降低逆向物流的运输成本，进而为逆向物流企业进行产品回收定价和车辆回收路径优化调度提供方法支持和决策参考。

关键词: 产品回收定价, 逆向物流, 多频次回收, 车辆路径优化, GA-POS混合算法