基于并行多目标遗传算法的制造伙伴优化研究  

Manufacturing Partner Optimization with Parallel Multi-objective Genetic Algorithm

作　　者：董朝阳[1] 孙树栋[1] 张波[2] 

机构地区：[1]西北工业大学机电学院,西安710072 [2]西安建筑科技大学机电工程学院,西安710055

出　　处：《机械科学与技术》2007年第5期542-547,共6页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家863计划(2003AA411110);教育部博士点基金项目(20040699025)资助

摘　　要：建立了制造伙伴优化的数学模型,并指出该模型属于复杂的多目标、多选择、多约束背包问题,提出了并行多目标妥协遗传算法进行求解:算法采用基于排列的编码方式,由多个子种群独立进化并定期交换最佳个体,适应度计算采用自适应权重方法及基于距离度量的妥协方法,并通过适应度共享保持种族多样性,最终求得决策者可接受的妥协解。We establish the mathematical model for manufacturing partner optimization, which belongs to, in our opinion, a complicated knapsack problem of multiple objectives, multiple choices and multiple constraints. Then the method of encoding based on permutation, evolves independently from numerous subspecies and periodically exchanges their individuals. Their fitness is calculated with the adaptive weight approach and the compromise approach based on distance measurement, and is shared to keep subspecies diversity, thus obtaining the compromise solution acceptable to a decision-maker.

关 键 词：制造伙伴优化 背包问题 并行多目标遗传算法 

分 类 号：TH166[机械工程—机械制造及自动化]