Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly lines
Abstract A multi‐agent iterative optimisation method based on deep reinforcement learning is proposed for the balancing and sequencing problem in mixed model assembly lines. Based on the Markov decision process model for balancing and sequencing, a balancing agent using a deep deterministic policy g...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-09-01
|
| Series: | IET Collaborative Intelligent Manufacturing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/cim2.12061 |
| _version_ | 1811340645757878272 |
|---|---|
| author | Youlong Lv Yuanliang Tan Ray Zhong Peng Zhang Junliang Wang Jie Zhang |
| author_facet | Youlong Lv Yuanliang Tan Ray Zhong Peng Zhang Junliang Wang Jie Zhang |
| author_sort | Youlong Lv |
| collection | DOAJ |
| description | Abstract A multi‐agent iterative optimisation method based on deep reinforcement learning is proposed for the balancing and sequencing problem in mixed model assembly lines. Based on the Markov decision process model for balancing and sequencing, a balancing agent using a deep deterministic policy gradient algorithm, a sequencing agent using an Actor–Critic algorithm, as well as an iterative interaction mechanism between these agents' output solutions are designed for realising the global optimisation of mixed model assembly lines. The exchange of solution information including assembly time and station workload in the iterative interaction realises the coordination of the worker assignment policy at the balancing stage and the production arrangement policy at the sequencing stage for the minimisation of work overload and idle time at stations. Through the comparative experiments with heuristic rules, genetic algorithms, and the original deep reinforcement learning algorithm, the effectiveness of the proposed method is demonstrated and discussed for small‐scale instances as well as large‐scale ones. |
| first_indexed | 2024-04-13T18:45:03Z |
| format | Article |
| id | doaj.art-1a35659d9edb401faa25137c96c592d2 |
| institution | Directory Open Access Journal |
| issn | 2516-8398 |
| language | English |
| last_indexed | 2024-04-13T18:45:03Z |
| publishDate | 2022-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Collaborative Intelligent Manufacturing |
| spelling | doaj.art-1a35659d9edb401faa25137c96c592d22022-12-22T02:34:36ZengWileyIET Collaborative Intelligent Manufacturing2516-83982022-09-014318119310.1049/cim2.12061Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly linesYoulong Lv0Yuanliang Tan1Ray Zhong2Peng Zhang3Junliang Wang4Jie Zhang5Institute of Artificial Intelligence Donghua University Shanghai ChinaCollege of Mechanical Engineering Donghua University Shanghai ChinaDepartment of Industrial and Manufacturing Systems Engineering The University of Hong Kong Hong Kong ChinaInstitute of Artificial Intelligence Donghua University Shanghai ChinaInstitute of Artificial Intelligence Donghua University Shanghai ChinaInstitute of Artificial Intelligence Donghua University Shanghai ChinaAbstract A multi‐agent iterative optimisation method based on deep reinforcement learning is proposed for the balancing and sequencing problem in mixed model assembly lines. Based on the Markov decision process model for balancing and sequencing, a balancing agent using a deep deterministic policy gradient algorithm, a sequencing agent using an Actor–Critic algorithm, as well as an iterative interaction mechanism between these agents' output solutions are designed for realising the global optimisation of mixed model assembly lines. The exchange of solution information including assembly time and station workload in the iterative interaction realises the coordination of the worker assignment policy at the balancing stage and the production arrangement policy at the sequencing stage for the minimisation of work overload and idle time at stations. Through the comparative experiments with heuristic rules, genetic algorithms, and the original deep reinforcement learning algorithm, the effectiveness of the proposed method is demonstrated and discussed for small‐scale instances as well as large‐scale ones.https://doi.org/10.1049/cim2.12061assemblingcontrol engineering computinggenetic algorithmsgradient methodsiterative methodslearning (artificial intelligence) |
| spellingShingle | Youlong Lv Yuanliang Tan Ray Zhong Peng Zhang Junliang Wang Jie Zhang Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly lines IET Collaborative Intelligent Manufacturing assembling control engineering computing genetic algorithms gradient methods iterative methods learning (artificial intelligence) |
| title | Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly lines |
| title_full | Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly lines |
| title_fullStr | Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly lines |
| title_full_unstemmed | Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly lines |
| title_short | Deep reinforcement learning‐based balancing and sequencing approach for mixed model assembly lines |
| title_sort | deep reinforcement learning based balancing and sequencing approach for mixed model assembly lines |
| topic | assembling control engineering computing genetic algorithms gradient methods iterative methods learning (artificial intelligence) |
| url | https://doi.org/10.1049/cim2.12061 |
| work_keys_str_mv | AT youlonglv deepreinforcementlearningbasedbalancingandsequencingapproachformixedmodelassemblylines AT yuanliangtan deepreinforcementlearningbasedbalancingandsequencingapproachformixedmodelassemblylines AT rayzhong deepreinforcementlearningbasedbalancingandsequencingapproachformixedmodelassemblylines AT pengzhang deepreinforcementlearningbasedbalancingandsequencingapproachformixedmodelassemblylines AT junliangwang deepreinforcementlearningbasedbalancingandsequencingapproachformixedmodelassemblylines AT jiezhang deepreinforcementlearningbasedbalancingandsequencingapproachformixedmodelassemblylines |