摘要
研究了具有模糊加工时间和模糊交货期的多目标作业车间调度问题,首先给出了基于模糊优先规则的编码新方式,染色体的每一位表示在GT算法迭代过程中,对应机器上发生的某次冲突,根据该基因位对应的优先规则消除。然后设计了基于个体密集距离的多目标进化算法,该算法利用密集距离进行外部档案维护和适应度赋值。最后将多目标进化算法应用于模糊作业车间调度问题,以最大化最小一致指标和最小化模糊最大完成时间,并和其他算法比较。计算结果验证了多目标进化算法在模糊调度方面良好的搜索性能。
Multi-objective job shop scheduling with fuzzy processing time and fuzzy due date was studied. A new fuzzy priority rules-based representation method was firstly presented. Each gene in the chromosome represented that in the procedure of GT algorithm, the conflict occurred in the corresponding machine was resolved by the corresponding priority rule. Secondly, Multi-Objective Evolutionary Algorithm （CMOEA） based on individual crowding measurement was designed, and external archives maintenance and fitness assignment were conducted through this crowding measurement. Finally, CMOEA was applied to six fuzzy job shop scheduling to maximize the minimum agreement index and minimize the maximum fuzzy complete time and was compared with other algorithms. The computational results demonstrated the good performance （in fuzzy scheduling） of CMOEA.
出处
《计算机集成制造系统》
EI
CSCD
北大核心
2006年第2期174-179,共6页
Computer Integrated Manufacturing Systems
基金
国家自然科学基金资助项目(70071017)。~~
关键词
模糊作业车间调度
密集距离
优先规则
多目标进化算法
fuzzy job shop scheduling
crowding measurement
priority rule
multi-objective evolutionary algorithm
