多目标粒子群算法在混装线再平衡中的应用  

Multi-objective particle swarm algorithm in mixed-model assembly line rebalancing

作　　者：戴隆州 吴永明[1] 李少波[1] 罗利飞[1] Dai Longzhou;Wu Yongming;Li Shaobo;Luo Lifei(Key Laboratory of Advanced Manufacturing Technology of Ministry of Education,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学现代制造技术教育部重点实验室,贵阳550025

出　　处：《计算机应用研究》2018年第1期145-149,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(51505094);贵州省科学技术基金计划项目(黔科合基础(2016)1037);贵州省应用基础研究计划重大项目[黔科合JZ字(2014)2001];贵州大学引进人才科研项目[贵大人基合字(2014)60号];贵州大学研究生创新基金资助项目

摘　　要：企业对产品进行创新改进,带来装配线上装配任务的变化,从而造成已平衡装配线的失衡。针对上述变化给企业混流装配线带来的影响进行了研究,以最小化生产节拍、工作站间的负荷和工人完成新装配任务的调整成本为优化目标来建立混装线再平衡的数学模型;设计了一种新的多目标粒子群算法求解模型,算法中引入各粒子动态密集距离去筛选外部文档的非劣解和指导全局最优值的更新,在控制解的容量的同时保持Pareto解集分布均匀;此外,引入变异机制,提高了种群的全局搜索能力。结合具体实例的验证表明,该改进多目标粒子群算法能有效地解决混装线再平衡问题。In the enterprise,with the innovation of products,assembly tasks on the assembly line have changed,resulting in the balance of the assembly line lost the balance.Aiming at the impact of these changes brought to the assembly line,this paper studied this problem and established a mathematical model for mixed-model assembly line rebalancing,in which the optimization objectives were minimizing the cycle time and station load,and the adjustment cost of completing the new assembly tasks by operators.Moreover,this paper proposed a new and improved multi-objective particle swarm optimization(MOPSO)algorithm to solve the model.The algorithm used dynamic crowding distance to update the Pareto solution set and guide the selection of global optimal solution,made solution more evenly distribute while maintaining the size of the solution set.Besides,it introduced the particle mutation strategy into the algorithm to improve the global searching ability of the population.Finally,with specific examples of the verification results show that the improved MOPSO algorithm can effectively solve the problem of mixed-model assembly lines rebalancing.

关 键 词：混装线 再平衡 多目标优化 粒子群算法 

分 类 号：TP278[自动化与计算机技术—检测技术与自动化装置]