The dual-resource-constrained re-entrant flexible flow shop a constraint programming approach and a hybrid genetic algorithm

The dual-resource-constrained re-entrant flexible flow shop scheduling problem represents a specialised variant of the flow shop scheduling problem, inspired by real-world scenarios in screen printing industries. Besides the well-known flow shop structure, stages consist of identical parallel machin...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International journal of production research Jg. 63; H. 5; S. 1803 - 1824
Hauptverfasser: Mlekusch, Johanna, Hartl, Richard F.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: London Taylor & Francis 04.03.2025
Taylor & Francis LLC
Schlagworte:
Computing time
constraint programming
Constraints
Decoding
dual-resource
flexible re-entrant flow shop scheduling
Genetic algorithms
Heuristic
hybrid genetic algorithm
Job shop scheduling
Production
Production scheduling
Screen printing
Workforce
ISSN:0020-7543, 1366-588X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The dual-resource-constrained re-entrant flexible flow shop scheduling problem represents a specialised variant of the flow shop scheduling problem, inspired by real-world scenarios in screen printing industries. Besides the well-known flow shop structure, stages consist of identical parallel machines and operations may re-enter the same stage multiple times before completion. Moreover, each machine must be operated by a skilled worker, making it a dual-resource-constrained problem according to the existing literature. The objective is to minimise the total length of the production schedule. To address this problem, our study employs two methods: a constraint programming model and a hybrid genetic algorithm with a single-level solution representation and an efficient decoding heuristic. To evaluate the performance of our methods, we conducted a computational study using different problem instances. Our findings demonstrate that the proposed hybrid genetic algorithm consistently delivers high-quality solutions, particularly for large instances, while also maintaining a short computational time. Additionally, our methods improve existing benchmark results for instances from the literature for a subclass of the problem. Furthermore, we provide managerial insights into how dual-resource constraints affect the solution quality and the efficiency associated with different workforce configurations in the described production setting.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0020-7543
1366-588X
DOI:10.1080/00207543.2024.2392198