论文部分内容阅读
针对柔性作业车间调度问题(Flexible Job-shop Scheduling Problem,FJSP)中的不同性能指标优化,提出一种改进的元胞遗传算法。结合柔性作业车间调度的特点,设计一种基于工序编码和设备分配的双层编码,在交叉变异时分别对两层编码进行操作,同时在变异时引入贪婪式变异以加快收敛速度。为了克服传统遗传算法早熟和收敛慢的特点,设计了根据邻居个体自适应的选择算子。将该改进的元胞遗传算法求解柔性作业车间调度问题并同其他遗传算法的测试结果进行比较,表明所提出的改进元胞遗传算法在求解柔性作业车间调度问题上的有效性。
Aiming at the optimization of different performance indexes in Flexible Job-shop Scheduling Problem (FJSP), an improved CGA is proposed. Combining with the characteristics of flexible job shop scheduling, a two-layer coding based on process coding and device allocation is designed. Two layers of coding are respectively operated during crossover variation, and greedy mutation is introduced to speed up the convergence speed. In order to overcome the characteristics of traditional genetic algorithm, such as premature convergence and slow convergence, a selection operator adaptive to individual neighbors is designed. The improved CGA is used to solve the problem of flexible job shop scheduling and compared with the test results of other genetic algorithms. It shows that the improved CGA is effective in solving the problem of flexible job shop scheduling.