Abstract: To improve the resource utilization and entire efficiency of remanufacturing disassembly,according to the concept of remanufacturing service,a model of remanufacture disassembly service was put forward.The layout and characteristics of three types of disassembly lines namely single product disassembly,multi product mixed disassembly and robot based flexible disassembly were analyzed in this model.Aiming at the line balancing problem of remanufacture disassembly service,the single product-dual objectives model for disassembly line balancing (DLBP) model was formulated with the optimization goals of maximizing load balancing rate of disassembly line and giving priority to disassembling high-value parts.To simplify the parameter setting and improve the computational efficiency,an Improved Teaching-Learning-Based Optimization (ITLBO) algorithm was developed.On the basis of the traditional TLBO and random key initialization,the algorithm was endowed with capability of “self-learning” to improve its partial searching ability.Through the comparison analysis for two group benchmark cases of ITLBO and genetic algorithm,the effectiveness and feasibility were verified.An application example of reducer disassembly was given to verify the practicability of the model and algorithm.

Key words: remanufacturing, disassembly service, disassembly line balancing, multi-objective optimization, teaching-learning-based optimization

摘要： 为了提高再制造拆卸的资源利用率和整体效率,根据再制造服务的理念提出一种再制造拆卸服务模式,分析了该模式下单产品拆卸、多产品混流拆卸和基于机器人的柔性拆卸3种类型生产线的布局和特点。针对再制造拆卸服务生产线平衡问题,以最大化拆卸线负载均衡率和优先拆卸高价值的零部件为优化目标,建立了单产品双目标的拆卸线平衡模型,为简化参数设置并提高运算效率,提出一种改进型教与学的优化算法,在随机键法初始化以及教与学的基础上,赋予自学习能力,以增强算法的局部搜索效力。通过对比分析该算法与遗传算法对两组经典案例的求解结果,验证了该算法的可行性与有效性,以减速器拆卸为工程实例,验证了模型与算法的实用性。

关键词: 再制造, 拆卸服务, 生产线平衡, 多目标优化, 教与学优化算法