基于自适应遗传算法的大型关重件车间布局优化  

Layout Optimization of Key-Important Parts Workshop Based on Improved Adaptive Genetic Algorithm

作　　者：张青雷[1,2] 党文君 段建国 ZHANG Qing-lei;DANG Wen-jun;DUAN Jian-guo(School of Mechanical Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;China(Shanghai)Free Trade Zone Supply Chain Institute,Shanghai Maritime University,Shanghai 201306,China)

机构地区：[1]上海理工大学机械工程学院,上海200093 [2]上海海事大学,中国(上海)自贸区供应链研究院,上海201306

出　　处：《机械设计与制造》2021年第1期236-239,共4页Machinery Design & Manufacture

基　　金：上海市(临港)产业转型升级发展专项资金项目(T-220217001);G80大功率低速船用曲轴研制;地方院校能力建设专项计划(18040501600)。

摘　　要：针对大型关重件生产车间的多行布局问题,建立了同时考虑物流费用与物流时间的多目标优化数学模型,运用了一种改进的自适应遗传算法,其交叉概率和变异概率可随群体的适应度自动改变,使算法避免陷入局部最优;加入精英保留策略,使每代中的最优个体都能够得到保留,避免交叉和变异操作遗失全局最优解。最后结合某船用曲轴车间实例,分别运用该算法和标准遗传算法对车间布局模型进行计算,通过数据和性能分析表明该算法有效的解决了遗传算法易陷入早熟及收敛速度慢的问题;优化后的车间物流费用及物流时间减少约35%,证明该方法在车间布局中的可行性和有效性。In order to solve the problem of multi-row layout of heavy duty parts production workshop,a multi-objective optimization mathematical model considering logistics cost and logistics time is established.An improved adaptive genetic algorithm was used.The crossover probability and mutation probability of the multi-row layout can be automatically changed according to the population’s fitness.The algorithm avoids falling into a local optimum;an elite retention strategy is added so that the best individual in each generation can be preserved,and crossover and mutation operations are lost to avoid the global optimal solution.Finally,a marine crankshaft shop example was used to calculate the shop floor layout model by using the algorithm and the standard genetic algorithm respectively.The results of data and performance analysis show that this algorithm can solve the problems of genetic algorithm falling into early maturity and slow convergence.The optimized workshop logistics cost and logistics time are reduced by about 35%,which proves the feasibility and effectiveness of this method in workshop layout.

关 键 词：关重件 车间布局 多目标优化 自适应遗传算法 

分 类 号：TH16[机械工程—机械制造及自动化] TB491[一般工业技术]