改进遗传蜂群算法求解分布式柔性作业车间调度问题

针对分布式柔性作业车间调度问题,提出一种改进遗传蜂群算法求解方案。算法采用基于机器编码的编码方案,根据编码特点和分布式柔性作业车间的特点,设计了一种基于编码相似度的交叉操作,可以避免在交叉过程中产生非法解,提高算法的运行效率,并通过在不同的交叉操作后,以不同概率进行两种变异操作的方式改进了雇佣蜂时期的搜索操作,改善了算法的迭代速度;采用排序选择策略替代原来跟随蜂时期的选择策略;改进侦查蜂的蜜源抛弃机制,通过对比已获得的全局最优解,对达到搜索上限的蜜源进行部分抛弃,防止破坏优质解再次陷入随机搜索。最后,通过对比不同算法对实例求解,验证本文算法的有效性。

AN IMPROVED GENETIC BEE COLONY ALGORITHM FOR DISTRIBUTED FLEXIBLE JOB SHOP SCHEDULING PROBLEM

An improved genetic bee colony algorithm was proposed to solve the distributed flexible job-shop scheduling problem. The algorithm adopts A coding scheme based on machine coding was adopt, and according to the characteristics of the encoding characteristics and distributed flexible job shop, crossover operation based on the similarity was designed, which could avoid illegal solutions in the process of cross and improve the efficiency of the algorithm, and through different crossover operation, the search operation in the employ bees was improved, the iteration speed of the algorithm was also improved. The sequencing selection strategy was adopted to replace the original strategy of following bees. The nectar source abandonment mechanism of scout bees was improved. By comparing the obtained global optimal solution, the nectar sources that reached the search upper limit were partially abandoned to prevent the destruction of high-quality solution from being trapped in random search again. Finally, the effectiveness of the proposed algorithm was verified by comparing different algorithms to solve examples.