基于遗传算法的多目标设施定位方法  

A Multi-Objectives Method of Facility Location Based on Genetic Algorithm

作　　者：李波[1] 马大奎[1] 崔欣[1] 

机构地区：[1]天津大学,天津30007

出　　处：《中国机械工程》2008年第7期828-833,共6页China Mechanical Engineering

基　　金：国家自然科学基金资助项目(70572045);教育部新世纪优秀人才支持计划资助项目(NCET-06-0236)

摘　　要：提出了一种基于遗传算法的多目标设施定位方法。设计了一种新的整数和浮点数结合的编码方式;通过列举法使得每代个体满足约束条件,避免了罚函数的使用;可调整的双参数变异算子使进化过程能够较好地跳出局部最优解;应用基于Pareto排序的适应值分配方式得到目标函数的非劣解集,为决策者提供了多种选择方式。仿真实验说明了该方法的有效性。This paper proposed a multi-objective model of facility location. Based on genetic algorithm（GA） ,a new hybrid coding method using the integer and floating number was presented. The enumerable method guarantees each individual in the population to satisfy the constrains, thus the punishment function was avoided. The two adjustable parameters made the evolution process jump out of the local best solution perfectly. The application of the fitness assignment based on Pareto ranking attained inferior solution set of the multi-objective and multiple options were provided to the decisionmaker. Finally, the simulation result shows that the algorithm is effective.

关 键 词：设施定位 多目标 遗传算法 非劣解集 

分 类 号：TP138[自动化与计算机技术—控制理论与控制工程]