Multi-objective Flow Shop Scheduling UsingDifferential Evolution

摘要

This paper proposes an effective Differential Evolution (DE) basedrnhybrid algorithm for Multi-objective Permutation Flow Shop Scheduling Problemrn(MPFSSP), which is a typical NP-hard combinatorial optimization problem. In thernproposed Multi-objective Hybrid DE (MOHDE), both DE-based searchingrnoperators and some special local searching operators are designed to balance thernexploration and exploitation abilities. Firstly, to make DE suitable for solvingrnMPFSSP, a largest-order-value (LOV) rule based on random key representation isrndeveloped to convert the continuous values of individuals in DE to jobrnpermutations. Then, to enrich the searching behaviors and to avoid prematurernconvergence, a Variable Neighborhood Search (VNS) based local search withrnmultiple different neighborhoods is designed and incorporated into thernMOHDE. Simulation results and comparisons with the famous random-weightrngenetic algorithm (RWGA) demonstrate the effectiveness and robustness of ourrnproposed MOHDE.