Abstract: To improve the collaborative operation performance of closed-loop supply chain network,a three-stage closed-loop supply chain system under the uncertain environment of market demands and returned products available quantity was considered,which consisted of a joint recycling center being responsible for disassembling and cleaning,multiple factories being responsible for manufacturing and remanufacturing,multiple distribution centers being responsible for selling and recycling,and a recycling-production-distribution triple-level collaborative planning model was developed.To coordinate the operation of the system as a whole,the model considered the interrelationship of materials flowing among members at different levels,as well as their own operation objectives.According to the characteristics of decision interactions,the triple-level iterative solution approach was designed based on Double Chromosomes Adaptive Genetic Algorithm (DCAGA).A numerical example abstracted from a real supply chain of automobile engine was suggested to demonstrate the effectiveness and applicability of the proposed model and solution approach.

Key words: closed-loop supply chain, collaborative planning, uncertain environment, multi-level program, double chromosomes adaptive genetic algorithm

摘要： 为了提升闭环供应链网络的协同运作绩效,在市场需求和回收产品供应数量不确定环境下,针对由一个进行拆卸、检验的联合回收中心,多个进行制造和再制造的生产工厂,以及多个进行销售和下游回收的分销中心组成的三阶段闭环供应链系统,建立了回收—生产—分销三层协同计划模型。该模型充分考虑了各级成员企业之间物料流动的相互关联和各自不同的运作目标,能够确保整个系统协同运作。基于决策交互的特点,设计了分层迭代整体求解策略,并在单层求解过程中设计了双倍体自适应遗传算法。最后,通过一个汽车发动机生产供应网络的仿真算例,验证了模型和方法的适用性和有效性。

关键词: 闭环供应链, 协同计划, 不确定环境, 多层规划, 双倍体自适应遗传算法