Job-shop scheduling considering rescheduling in uncertain dynamic environment

This paper formalizes the job-shop scheduling problem in uncertain dynamic environment with a mathematical model which takes jobs under processing and waiting for processing as scheduling objects. The model adopts 3-stage rescheduling based on the rolling window rescheduling strategy. We proposed a layered hybrid ant-colony and genetic algorithm (HACGA) to perform the scheduling optimization which considers the minimum completion time, minimum cost, max utilization rate, and minimum deviation degree as objectives. The outer layer of the algorithm uses ant-colony algorithms with bidirectional convergence to determine the machine assignment and the inner layer of the algorithm uses genetic algorithm with neighborhood search to optimize the order of manufacturing jobs. We conduct a numerical simulation to show the feasibility and effectiveness of the algorithm.