应用粒子群优化(PSO)进行了考虑机器调整时间、工件运输时间以及提前/拖期惩罚的作业车间调度问题的研究,分析了各时间约束对调度的影响,在此基础上设计了一种解决多时间约束调度问题的混合离散粒子群(HDPSO)算法.该算法在初始阶段采用反向学习机制初始化以提高初始解质量,引入记忆池的概念,在每次迭代中利用记忆池中精英解对当代种群搜索加以指导,以增加粒子与优秀群体间的交流并提高收敛速度及跳出局部最优的能力,最后采用一种针对问题的变邻域搜索策略提高了算法收敛精度.实例仿真验证了该算法的有效性.
The job-shop scheduling considering the ness/tardiness punishment was studied by the time constraints of processing time, setup time, transit time and earli- application of particle swarm optimization ( PSO), and the influences of the constraints on the scheduling were analyzed. Then, a hybrid discrete (HD) PSO (HDPSO) algorithm tbr solving the job-shop scheduling with multi-time constraints was designed. The algorithm uses the opposition-based learning to initialize population to improve the quality of the initial population. It also implants the memory mecha- nism into the discrete PSO to speed up convergencel At last the algorithm adopts a modified variable neighborhood search (VNS) to strengthen the local search ability. The effectiveness of the proposed algorithm was demonstrated by the experiments on different simulation examnles.