An improved iterated greedy algorithm for the distributed hybrid flowshop scheduling problem
Chao Lu, Jun Zheng, Lvjiang Yin, Renyi Wang
Abstract
This study attempts to solve the distributed hybrid flowshop scheduling problem (DHFSP) with the makespan criterion. First, a mixed-integer linear programming model for the DHFSP is formulated. Then, an improved iterated greedy (IIG) algorithm is developed to handle this DHFSP. In IIG, a new initialization strategy is designed to improve the quality of the initial solution. A hybrid operator, which combines the perturbation operator and destruction/construction operator, is proposed to enhance the global search ability. According to the characteristics of the DHFSP, a new local search method, which integrates four neighbourhood structures, is designed to strengthen the exploitation capability. The best parameter configuration of IIG is investigated through design of experiments, and the validity of each improved part of IIG is verified by performing extensive experiments. Finally, IIG is compared with other optimization algorithms on 100 large-scale instances. The experimental results show that IIG is effective in addressing this DHFSP.