Abstract: To solve the multi-objective enterprise remanufacturing logistics network optimization problem under the carbon tax policy,a mixed integer programming model with minimum total cost and carbon emission was established by considering facilities location and transportation modes.The constraint handling mechanism was designed to modify the multi-objective particle swarm optimization algorithm,and the effectiveness of model and algorithm were verified with the operation data of cement enterprise.Through comparing the modified multi-objective particle swarm algorithm with constrained non-dominated sorting genetic algorithm-Ⅱ,the superiority of modified multi-objective particle swarm optimization algorithm in solving sophisticated problems was verified.

Key words: remanufacturing, logistics network optimization, mixed integer programming, modified multi-objective particle swarm optimization algorithm

摘要： 为解决碳税政策下的多目标企业再制造物流网络优化问题,综合考虑设施选址与运输方式选择,建立了总成本和碳排放最小化的混合整数规划模型。设计了约束处理机制改进多目标粒子群优化算法,并结合水泥企业运营管理数据验证了模型和算法的有效性。将改进多目标粒子群算法与第二代约束非支配排序遗传算法比较,凸显了改进多目标粒子群算法在求解中、大规模问题时的优越性。

关键词: 再制造, 物流网络优化, 混合整数规划, 改进多目标粒子群优化算法