基于改进FFA算法的柔性制造车间调度方法

摘要

To solve the discrete manufacturing of flexible manufacturing workshop in a small precision mold processing enterprise, a modified heuristic algorithm(five factors scheduling algorithm,FFA) was proposed. This algorithm took into account five factors that may affect the sort of selection, i. e. , processing start time of current task procedures, processing working time of current procedure, remaining processing time of current device, the total remaining processing time of the task after current procedure, and the remaining processing time after the task reaching the next device upon the completion of current procedure. When the constructed evaluation index was the smallest through maximizing the first three factors and minimizing the last two factors, the task was prioritized. In addition, the evaluation index formulae was improved to select the appropriate coefficient for five factors, hence reducing the calculation complexity. Finally, compared with the genetic algorithm, it is proved that the modified algorithm has more advantages in shortening maximum completion time and improving calculation speed.%针对某小型精密模具加工企业柔性制造车间离散型生产方式下调度困难的情况,笔者提出了一种改进的启发式算法——改进五因素调度算法(five factors scheduling algorithm,FFA).算法考虑了可能影响选择排序的5个因素,即当前任务待排工序可能开始加工时间、待排工序工时、可用设备剩余加工时间、本工序完成后本任务剩余加工时间以及本工序之后本任务将要到达的紧后设备上的剩余加工时间,最大化前3个因素,最小化后2个因素,所构造评价指数最小时,该任务优先排序.并对构造的评价指数公式进行了改进,为5个因素选取合适系数,降低了计算复杂度.最后通过试验与遗传算法进行比较,证明了该算法在缩短最大完工时间和提高计算速度方面更有优势.