A multiobjective memetic algorithm with particle swarm optimization and Q-learning-based local search for energy-efficient distributed heterogeneous hybrid flow-shop scheduling problem

Most existing distributed hybrid flow-shop scheduling problems (DHFSPs) assume identical shops and lack consideration of heterogeneous shops. This study focuses on energy-efficient heterogeneous DHFSP. A multiobjective memetic algorithm with particle swarm optimization and Q-learning-based local sea...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Expert systems with applications Jg. 237; S. 121570
Hauptverfasser: Zhang, Wenqiang, Li, Chen, Gen, Mitsuo, Yang, Weidong, Zhang, Guohui
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 01.03.2024
Schlagworte:
Energy-efficient
Heterogeneous distributed scheduling
Hybrid flow-shop scheduling
Memetic algorithm
Particle swarm optimization
Q-learning
Memetic algorithm
Q-learning
Energy-efficient
Particle swarm optimization
Hybrid flow-shop scheduling
Heterogeneous distributed scheduling
ISSN:0957-4174, 1873-6793
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Most existing distributed hybrid flow-shop scheduling problems (DHFSPs) assume identical shops and lack consideration of heterogeneous shops. This study focuses on energy-efficient heterogeneous DHFSP. A multiobjective memetic algorithm with particle swarm optimization and Q-learning-based local search is proposed in order to optimize both makespan and total energy consumption. Particle swarm optimization with multi-group is specifically designed as a global search strategy to improve the fast convergence performance of solutions in multi-direction of Pareto front. To improve the problem-specific knowledge search, two local search strategies are designed to further improve the quality and diversity of solutions. In addition, Q-learning is utilized to guide variable neighborhood search to better balance the exploration and exploitation of algorithms. This study investigates the effect of parameter setting and conducts extensive numerical tests. The comparative results and statistical analysis demonstrate the superior convergence and distribution performance of the proposed algorithm. •Multi-group PSO as global search enhances multi-direction convergence of PF.•Two local search strategies cooperate with particle swarm optimization.•Inter-factory local search with insert and swap between critical factories.•Intra-factory local search with Q-learning and VNS within factories.•Two initialization methods increase the diversity of population.
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2023.121570