A hybrid optimization algorithm based on fitness landscape analysis for generalized job-shop scheduling problems
•JSP problem domains are formed due to different constraints or optimization objectives.•It is very valuable to design optimization algorithms with generalization for JSP domains.•Fitness landscape analysis can provide a basis for the design of optimization algorithms. Due to the diverse constraints...
Gespeichert in:
| Veröffentlicht in: | Computers & industrial engineering Jg. 208; S. 111390 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Journal Article |
| Sprache: | Englisch |
| Veröffentlicht: |
Elsevier Ltd
01.10.2025
|
| Schlagworte: | |
| ISSN: | 0360-8352 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Zusammenfassung: | •JSP problem domains are formed due to different constraints or optimization objectives.•It is very valuable to design optimization algorithms with generalization for JSP domains.•Fitness landscape analysis can provide a basis for the design of optimization algorithms.
Due to the diverse constraints and objectives inherent to job-shop scheduling problems (JSPs), this problem domain has emerged as a significant challenge. In existing research, the optimization algorithm must be tailored to specific production scenarios, which prolongs the development cycle and increases the cost. This paper proposes a hybrid optimization algorithm for addressing diverse JSPs within the problem domain (generalized JSPs) based on fitness landscape analysis. Firstly, a mathematical model of the generalized JSPs is constructed, and then the common features in different problems are obtained based on fitness landscape analysis. On this basis, this paper proposes a hybrid optimization algorithm and verifies it in three different JSPs. The superiority of the proposed algorithm is then verified in comparison with the existing best meta-heuristic algorithms for solving JSPs. |
|---|---|
| ISSN: | 0360-8352 |
| DOI: | 10.1016/j.cie.2025.111390 |