A reinforcement learning-driven brain storm optimisation algorithm for multi-objective energy-efficient distributed assembly no-wait flow shop scheduling problem

A reinforcement learning-driven brain storm optimisation idea (RLBSO) is proposed in this paper to solve multi-objective energy-efficient distributed assembly no-wait flow shop scheduling problem. The objectives of the problem include minimising the maximum assembly completion time ( ), minimising t...

Full description

Saved in:
Bibliographic Details
Published in:International journal of production research Vol. 61; no. 9; pp. 2854 - 2872
Main Authors: Zhao, Fuqing, Hu, Xiaotong, Wang, Ling, Xu, Tianpeng, Zhu, Ningning, Jonrinaldi
Format: Journal Article
Language:English
Published: London Taylor & Francis 03.05.2023
Taylor & Francis LLC
Subjects:
Algorithms
Assembly
brain storm optimisation
Clustering
clustering mechanism
Completion time
Critical path
Energy consumption
Energy-efficient
Factories
Industrial plants
Job shop scheduling
Multiple objective analysis
no-wait flow shop
Optimization
product assignment rule
Q-learning mechanism
Resource allocation
ISSN:0020-7543, 1366-588X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A reinforcement learning-driven brain storm optimisation idea (RLBSO) is proposed in this paper to solve multi-objective energy-efficient distributed assembly no-wait flow shop scheduling problem. The objectives of the problem include minimising the maximum assembly completion time ( ), minimising the total energy consumption (TEC) and achieving resource allocation balanced . Four operations, which are critical factory insert, critical factory swap, critical factory insert to other factories, critical factory swap with other factories, are designed to optimise the objective of maximum assembly completion time. Q-learning mechanism is utilised to guide the selection of operations to avoid blind search in the iteration process. The learning mechanism based on clustering mechanism in brain storm optimisation algorithm is utilised to assign products to factories in the objective space according to the processing time of products to balance the resources allocation. The speed of operations on non-critical path is slowed down to reduce TEC regarded with the characteristics of no-wait flow shop scheduling problem. The experimental results under 810 large-scale instances by RLBSO show that the RLBSO outperforms the comparison algorithm for addressing the problem.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0020-7543
1366-588X
DOI:10.1080/00207543.2022.2070786