中文摘要：

针对柔性作业车间调度问题（FJSP）,提出一种改进的双种群混合遗传算法,其中一个种群侧重全局搜索,另一个种群负责局部搜索。采用基于工序与基于机器相结合的编码机制,提出一种初始种群产生方法,提高初始种群的多样性;通过交换精英个体的方式实现两个种群间的协同优化,提高算法的精度和收敛速度。对比仿真结果验证了该算法求解FJSP问题的有效性。

英文摘要：

An improved genetic algorithm named DPHGA（double population hybrid genetic algorithm）was proposed to solve flexible job shop scheduling problem（FJSP）.DPHGA was comprised of two populations,in which one population focused on global search,and the other was responsible for the local search.An operation-based mechanism and a machine-based coding mechanism were used to generate the initial population.A method of generating initial population was then proposed to increase the diversity of population.Through exchanging the elite individuals of two populations,a collaborative optimization approach was proposed to improve the performance.Simulation results show that DPGA is an effective approach for solving FJSP.