多目标柔性车间调度的二级精选算法

摘要

Multi-object flexible job shop scheduling problem is more practical which is a multi-objective combinatorial optimization problem. Using traditional methods to find the optimal solution is very difficult because it will produce a large number of solution space. Second-order optimization method is a combinatorial optimization algorithm which is based on genetic algorithm (GA) of primary single-objective optimization and high selected method of multi-objective decision-making system optimization. In the initial optimization stage, finding one Pareto solution set which is most concerned using GA. In the high selected stage, considering multi-objective requirement and using analytic hierarchy process (AHP), technique for order preference by similarity to ideal solution (TOPSIS) of multiple attribute decision making to select optimal solution in Pareto solution set. The machining example showed the proposed method is feasible and it can help decision makers to choose a satisfactory solution.%多目标柔性车间调度问题与实际更加符合,是典型的多目标组合优化问题,运用传统算法求解会产生大量的解空间,找到最优解是非常棘手的问题.基于此,提出了二阶优化方法,即基于遗传算法的初级单目标优化和基于多目标决策体系的高级精选优化的组合优化算法.初级优化阶段,采用改进的遗传算法,选用企业最关心的单目标选出一组Pareto解集；高级精选优化阶段,考虑多目标要求,采用多属性决策体系的层次分析法和TOPSIS算法从Pareto中选出最优解.某加工实例表明该方法是可行的,能很好地帮助决策者选择出一个满意的解.