Multi-objective Optimization Method of Flexible Job-Shop Lot-Splitting Scheduling

WANG Yun,FENG Yi-xiong,TAN Jian-rong,GAO Yi-cong

DOI: https://doi.org/10.3785/j.issn.1008-973x.2011.04.022

2011-01-01

Abstract:The multi-objective flexible job-shop scheduling optimization model was conducted in order to solve multi-objective optimization problem of flexible job-shop small batch lot-splitting scheduling.The model was concerned with makespan,tardiness penalty,manufacturing cost,count of batch and total workload.The optimal solutions were obtained by using improved strength Pareto evolutionary algorithm(SPEA).The algorithm was improved by using the fuzzy c-means clustering algorithm to accelerate the clustering procedure within the external population.A self-adaptive mutation operator was also introduced to enhance the diversity of solutions.A Pareto optimal set can be achieved in a single run with the constraint Pareto domination concept and the flexible representation schema.Then the preference sequence of Pareto solutions was achieved and a solution was extracted as the best compromise one based on set theory.The jobs were split into flexible size batch,and the batch routing and sequencing were simultaneously optimized by the method.The performance of the method was evaluated through simulation.The feasibility and validity of the method were proved in a workshop scheduling.