An efficient collaborative multi-swap iterated greedy algorithm for the distributed permutation flowshop scheduling problem with preventive maintenance

Qiu-Yang Han,Hong-Yan Sang,Quan-Ke Pan,Biao Zhang,Heng-Wei Guo
DOI: https://doi.org/10.1016/j.swevo.2024.101537
IF: 10.267
2024-04-01
Swarm and Evolutionary Computation
Abstract:Under the context of globalization, distributed multi-factory production model is becoming the mainstream of the manufacturing industry because it provides enterprises with flexible and efficient production solutions. In real-world environments, the wear and tear of machines is unavoidable, so it is necessary to maintain equipment for sustainable processing and production. This paper explores the distributed permutation flowshop scheduling problem with preventive maintenance operations (DPFSP/PM) and introduces a collaborative multi-swap iterated greedy (CMSIG) algorithm to minimize the makespan. In the initialization phase, an improved heuristic algorithm is proposed, which re-swaps inserted jobs based on NEH2 to generate initial solutions. In the destruction phase, an adaptive destroy method is proposed. The factory is selected based on a tournament strategy and a critical factory strategy, and then the jobs are greedily reinserted in the construction phase. In the local search phase, four swap operators operate together to enhance their ability for exploring the solution space. Finally, extensive experiments show that the proposed CMSIG algorithm has significant advantages in solving DPFSP/PM compared with other algorithms.
computer science, artificial intelligence, theory & methods
What problem does this paper attempt to address?