Industrial Engineering, University of Kashan
Traditional scheduling problems with the batch processing machines (BPM) assume that machines are continuously available, and no time is needed for their preventive maintenance (PM). In this paper, we study a realistic variant of flowshop scheduling which integrates flow shop batch processing machines (FBPM) and preventive maintenance for minimizing the makespan. In order to tackle the given problem, we develop an electromagnetism-like (EM) algorithm, as a recent evolutionary technique, and propose a enhanced EM algorithm, in which the EM is hybridized with a diversification mechanism, and an effective local search to enhance the efficiency of the algorithm. The proposed algorithms are evaluated by comparison against two existing well-known EMs in the literature. For this purpose, we not only study the behavior of different operators and parameters of these algorithms by Taguchi experimental design method, but also investigate the impacts of the rise in problem sizes on the performance of the developed algorithm. The superiority of our hybrid EM is inferred from all the computational results obtained in various circumstances.