基于改进粒子群优化算法的混流装配线演进平衡  

Mixed assembly line evolution balancing based on improved particle swarm algorithm

作　　者：吴永明[1] 戴隆州 李少波[1] 罗利飞[1] WU Yongming DAI Longzhou LI Shaobo LUO Li f ei(Key Laboratory of Modern Manufacturing Technology, Guizhou University, Guiyang 550025, Chin)

机构地区：[1]贵州大学现代制造技术教育部重点实验室,贵州贵阳550025

出　　处：《计算机集成制造系统》2017年第4期781-790,共10页Computer Integrated Manufacturing Systems

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资助项目(黔科合基础(2016)1037);贵州省应用基础研究计划重大资助项目(黔科合JZ字(2014)2001);贵州大学引进人才科研资助项目[贵大人基合字(2014)60号]~~

摘　　要：随着产品需求的多样化、装配工艺及技术进步、设备更新等动态变化,装配线平衡方案需不断调整,甚至重新规划与演进平衡。为了探究上述因素对混流装配线演进平衡的影响,提出了实现装配线演进平衡的方法,建立了以最小化装配线的生产节拍、站间平滑指数、演进平衡调整成本为优化目标的混流装配线演进平衡数学模型,并通过改进粒子群优化算法进行优化。在该算法中,为增加粒子的多样性和搜索能力,克服传统粒子群优化算法快速收敛等问题,以粒子进化的成功率来更新算法中的惯性因子,将群体中非最优粒子中的有利信息迁移到群体中的最优粒子上,从而加快算法的搜索速度。结合某企业的生产实例验证了该方法的有效性和可行性。With the dynamic changes such as different customer needs, equipment updates and improvements in as- sembly process and technology, the assembly line needed to be constantly adjusted, even balanced again. Evolution balance factors were researched for Mixed-Model Assembly Line （MMAL） in many ways, and a method was pro- posed for MMAL evolution balancing. A balance mathematical model, in which the minimum cycle time, smooth- ness index between workstation stations and adjustment costs were as the optimization target was established for MMAL evolution and optimized through an Improved Particle Swarm Optimization （IPSO） algorithm simultaneous- ly. To increase particle diversity and improve search speed, the inertia factor was updated by using the success rate of particle evolution in IPSO, in which the favorable information in non-optimal particles was added to the optimal particle in the group. The effectiveness and feasibility of the method were validated by optimizing assembly line bal- ancing of a manufacturing enterprise.

关 键 词：混流装配线 演进平衡 粒子群优化算法 优化 

分 类 号：F245[经济管理—劳动经济]