An Improved Nondominated Sorting Genetic Algorithm-II for Multi-objective Flexible Job-shop Scheduling Problem Considering Worker Assignments

Abstract

This study proposed a multi-objective flexible job-shop scheduling problem (MOFJSP) considering worker assignments which minimizing three objectives, namely maximal completion time, total workload of machines and labor cost. An improved nondominated sorting genetic algorithm-II (INSGA-II) is developed to solve the proposed MOFJSP. A left-shift decoding method is applied to effectively utilize the existing idle time intervals of machines. Specifically, a probability-based uniform crossover operator called “PEPPX” is adopted and the updating rule for the parameter of choosing probability is well tuned to balance the exploration and exploitation in different stages of the iterations. Besides, variable neighborhood search (VNS) with three critical-path based local search operators is integrated to search better solutions from the current solution of the elitist population. Numerical experiments on several problem instances shows the performance of the proposed INSGA-II.