Research On Rescheduling Strategy Of Job Shop Under Uncertainty

In the workshop production process of modern manufacturing mode, there are various uncertain production disturbances, which could reduce the effectiveness of the original schedule, and result in the increase of WIP inventory and the production delay. Rescheduling technology, as an effective means of coping with production disturbance, can correct the original schedule or regenerate a new scheduling scheme to ensure the stability of production system. Therefore, rescheduling research has theoretical and practical significance for optimal control of workshop production process. And aiming at the triggering and optimizing problems of manufacturing shop rescheduling, the rescheduling strategy of Job Shop under uncertain environment is proposed in this paper.First, the uncertain production disturbances of Job Shop are described and classified according to the influence, and the concept of latest completion time is put forward to reflect the of production completion time after disturbance. And a hybrid-driven rescheduling mechanism based on latest completion time is proposed to avoid production system shaking because of frequent rescheduling.Secondly, the rescheduling optimization method which combines right-shift rescheduling with complete rescheduling is determined by analyzing and comparing among right-shift, local and complete rescheduling. In the complete rescheduling, scheduling resource set is reset based on the kind of production disturbance, and then the rescheduling scheme is generated through improved Non-dominated Sorting Genetic Algorithm-Ⅱ (NSGA-Ⅱ). In order to enhance solution quality, the improved measure, namely the initial pop method is proposed, which combines random generation with optimal selection.Finally, aiming at the schedule selection problem of rescheduling, a rescheduling decision method based on improved TOPSIS is proposed. The set of optional schedules is taken as the research subject, and the decision index is established in order to depict performance of the optional schedules, quantitatively. And then the partial order of optional schedules is ascertained based on Kullback-Leibler divergence and grey correlation degree, so as to choose the optimal reschedule.