基于多目标算法与动态仿真的带调整时间的拆卸线平衡优化方法  

Optimization of Disassembly Line Balancing Problems with Setup Times Based on Multi-objective Algorithm and Dynamic Simulation

作　　者：李六柯 张则强[1] 胡扬[1] 邹宾森 

机构地区：[1]西南交通大学机械工程学院,成都610031

出　　处：《中国机械工程》2017年第17期2115-2124,共10页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(51205328;51405403);教育部人文社会科学研究青年基金资助项目(12YJCZH296);四川省应用基础研究计划资助项目(2014JY0232)

摘　　要：针对拆卸方向改变对作业时间的影响,以最小化拆卸成本、无效作业率和不平滑率为优化目标,构建了考虑调整时间的多目标拆卸线平衡问题模型,并提出了一种Pareto免疫遗传算法与仿真技术相结合的求解方法。算法融合了遗传算子和免疫算子,能有效避免解出现退化和陷入局部最优。通过求解25项拆卸任务算例,并与现有5种算法进行对比,验证了所提算法的有效性。进而将所提模型和算法应用于某打印机拆卸线实例,进一步考虑拆卸时间不确定、存在故障等实际情况,运用仿真技术分析与优化,结果表明所提方法能有效降低等待和堵塞造成的无效作业时间,提高拆卸效率。In view of the influences of disassembly direction changes on operation times,a multiobjective model of disassembly line balancing problem considering the setup times was established involving three optimization objectives,i.e.minimizing the disassembly costs,minimizing the invalid work rates and minimizing the non-smooth rates.And a Pareto immune genetic algorithm integrated with the simulation technology was proposed.The proposed algorithm introduced the genetic operator and the immune operator which might avoid the local optimum and ensure the global convergence.The proposed algorithm was applied to a 25-task disassembly case,the results of which indicate the effectiveness of the proposed algorithm by comparing with the other 5algorithms.Further,the proposed model and the algorithm were applied to design a printer disassembly line.Finally,considering the uncertainty of the disassembly times and failures in the actual disassembly productions,simulation technology was used to analyze and optimize the disassembly line.And the simulation results show that the proposed method may effectively reduce the invalid times caused by waiting and blocking and improve the disassembly efficiency.

关 键 词：拆卸线平衡问题 多目标优化 调整时间 免疫遗传算法 仿真 

分 类 号：TH165[机械工程—机械制造及自动化]