自适应遗传算法求解模糊作业车间调度问题  

Adaptive Genetic Algorithm for Fuzzy Job-shop Scheduling Problem

作　　者：杨建斌[1] 孙树栋[1] 牛刚刚[1] 王萌[1] 

机构地区：[1]西北工业大学系统集成与工程管理研究所,西安710072

出　　处：《机械科学与技术》2013年第1期16-21,共6页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金项目(51075337;50705076)资助

摘　　要：研究模糊作业车间调度问题(FJSSP),用三角模糊数表示模糊加工时间,用半梯形模糊数表示模糊交货期,以最大化最小客户满意度为调度目标,建立了模糊环境下Job-shop调度问题的模型。提出了一种自适应遗传算法,该算法采用基于优先列表的编码方式,提高了编码效率;在进化过程中对种群采用精英保留策略,确保最优个体不被破坏;并对自适应交叉变异算子进行了改进,使种群最优个体参与进化。仿真结果证明所提算法在寻优能力及收敛性能方面均有所改善。The fuzzy job-shop scheduling problem is studied in this paper. Fuzzy processing times were denoted by triangular fuzzy numbers and fuzzy due dates were presented by half trapezoid fuzzy numbers. Based on the agree- ment index to maximize the minimum customer satisfaction, the model of fuzzy job-shop scheduling problem was proposed. Furthermore, a new adaptive genetic algorithm was proposed, which promoted the coding efficiency by utilizing the coding method based on preferred list. Also, elitism preservation was adopted in the evolution process and adaptive crossover-mutation operator was improved so that the optimum individual can participate in the evolu- tion. At last, the improvement of the new algorithm in both search ability and convergence performance were veri- fied by the simulation results.

关 键 词：模糊作业车间调度 自适应遗传算法 精英保留 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]