基于遗传算法的多目标柔性车间作业调度方法  

GENETIC ALGORITHM BASED MULTIPLE OBJECTIVE FLEXIBLE JOB SHOP SCHEDULING METHOD

作　　者：许秀林[1] 胡克瑾[2] 

机构地区：[1]南通职业大学电子工程系,江苏南通226007 [2]同济大学经济管理学院,上海200092

出　　处：《计算机应用与软件》2012年第7期266-270,共5页Computer Applications and Software

摘　　要：多目标柔性车间作业调度通常将多个目标进行无量纲处理,加权转换成单一目标函数用于解的优化筛选,但权重选择难免存在一定的随意和偏好,影响调度效果。针对这一问题,提出单目标决策下的多目标调度解决方案,从多个目标中选择一个重要的目标作为决策目标,其他目标作为算法搜索的导向目标。其中对遗传算法变异算子进行改造,将随机性变异转换成目标诱导性变异。实例仿真结果表明了算法的有效性和可行性。Multiple objective flexible job shop scheduling usually handles multiple objects as dimensionless, i. e. , weighted transforming them into one single object function for solutions optimal filtration. However the selection of weights is inevitably influenced by casualty and preference, so that the scheduling performance is also influenced. To solve this problem, a multiple objective scheduling solution under single objective decision is proposed. It selects one important object from multiple objects as decisive object while the remaining as inductive objects. In the genetic algorithm, the mutation operator is reformed from random variation to induced variation. Simulation results show that the algorithm is effective and feasible.

关 键 词：遗传算法 车间作业调度 企业资源规划 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]