A deep reinforcement learning based approach for dynamic distributed blocking flowshop scheduling with job insertions

The distributed blocking flowshop scheduling problem (DBFSP) with new job insertions is studied. Rescheduling all remaining jobs after a dynamic event like a new job insertion is unreasonable to an actual distributed blocking flowshop production process. A deep reinforcement learning (DRL) algorithm...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IET collaborative intelligent manufacturing Ročník 4; číslo 3; s. 166 - 180
Hlavní autoři: Sun, Xueyan, Vogel‐Heuser, Birgit, Bi, Fandi, Shen, Weiming
Médium: Journal Article
Jazyk:angličtina
Vydáno: Wuhan John Wiley & Sons, Inc 01.09.2022
Wiley
Témata:
Algorithms
Completion time
Decision making
Deep learning
deep reinforcement learning
distributed blocking flowshop scheduling problem
dynamic scheduling
Factories
Genetic algorithms
Heuristic
job insertions
Job shops
Machine learning
Manufacturing
multi‐agent deep deterministic policy gradient
Neural networks
Preventive maintenance
Production scheduling
Scheduling
ISSN:2516-8398, 2516-8398
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:The distributed blocking flowshop scheduling problem (DBFSP) with new job insertions is studied. Rescheduling all remaining jobs after a dynamic event like a new job insertion is unreasonable to an actual distributed blocking flowshop production process. A deep reinforcement learning (DRL) algorithm is proposed to optimise the job selection model, and local modifications are made on the basis of the original scheduling plan when new jobs arrive. The objective is to minimise the total completion time deviation of all products so that all jobs can be finished on time to reduce the cost of storage. First, according to the definitions of the dynamic DBFSP problem, a DRL framework based on multi‐agent deep deterministic policy gradient (MADDPG) is proposed. In this framework, a full schedule is generated by the variable neighbourhood descent algorithm before a dynamic event occurs. Meanwhile, all newly added jobs are reordered before the agents make decisions to select the one that needs to be scheduled most urgently. This study defines the observations, actions and reward calculation methods and applies centralised training and distributed execution in MADDPG. Finally, a comprehensive computational experiment is carried out to compare the proposed method with the closely related and well‐performing methods. The results indicate that the proposed method can solve the dynamic DBFSP effectively and efficiently.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2516-8398
2516-8398
DOI:10.1049/cim2.12060