Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs)
Rescheduling is essential in real-world production to adjust schedules when significant disturbances render existing ones non-optimal. Manufacturers are often required to reschedule production tasks as quickly as possible. This paper proposes a rapid production rescheduling framework for flow shop u...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10179223/ |
| _version_ | 1797771686253166592 |
|---|---|
| author | Pakkaporn Saophan Warut Pannakkong Raveekiat Singhaphandu Van-Nam Huynh |
| author_facet | Pakkaporn Saophan Warut Pannakkong Raveekiat Singhaphandu Van-Nam Huynh |
| author_sort | Pakkaporn Saophan |
| collection | DOAJ |
| description | Rescheduling is essential in real-world production to adjust schedules when significant disturbances render existing ones non-optimal. Manufacturers are often required to reschedule production tasks as quickly as possible. This paper proposes a rapid production rescheduling framework for flow shop under machine failure disturbance, called PPGA-ANNs, with the goal of minimizing makespan while ensuring sufficient computational efficiency. The framework begins with a scheduling knowledge creation phase conducted before starting production. It applies the proposed Perturbation Population Genetic Algorithm (PPGA) to solve generated scenarios of flow shop production with machine failure problems. The performance of the PPGA is compared to other research algorithms and to the standard genetic algorithm (GA). The same data set from a widely used scheduling benchmark is used for all algorithms to confirm the effectiveness of the PPGA. Artificial neural networks (ANNs) are then applied to store the scheduling knowledge obtained from the PPGA. In the knowledge implementation phase, when a machine failure problem occurs during production, the rescheduling solution is provided by the ANNs if the machine failure problem is identical to a generated scenario. Otherwise, the rescheduling solution is provided by the PPGA, using the initial solution obtained from the ANNs. Based on the experimental results, the PPGA-ANNs framework demonstrates better performance in makespans than benchmark algorithms. Additionally, it provides faster solutions, particularly for new machine failure problems. In conclusion, the proposed framework is capable of minimizing the makespan with a short computational time for real-world production, addressing the limitations of existing state-of-the-art meta-heuristic algorithms. |
| first_indexed | 2024-03-12T21:41:10Z |
| format | Article |
| id | doaj.art-21d201f95af6407d948fc71f6d5c799f |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-12T21:41:10Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-21d201f95af6407d948fc71f6d5c799f2023-07-26T23:00:13ZengIEEEIEEE Access2169-35362023-01-0111757947581710.1109/ACCESS.2023.329457310179223Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs)Pakkaporn Saophan0https://orcid.org/0000-0001-6128-6071Warut Pannakkong1Raveekiat Singhaphandu2Van-Nam Huynh3https://orcid.org/0000-0002-3860-7815School of Knowledge Science, Japan Advanced Institute of Science and Technology, Nomi, JapanSchool of Manufacturing Systems and Mechanical Engineering, Sirindhorn International Institute of Technology, Thammasat University, Khlong Nueng, ThailandSchool of Knowledge Science, Japan Advanced Institute of Science and Technology, Nomi, JapanSchool of Knowledge Science, Japan Advanced Institute of Science and Technology, Nomi, JapanRescheduling is essential in real-world production to adjust schedules when significant disturbances render existing ones non-optimal. Manufacturers are often required to reschedule production tasks as quickly as possible. This paper proposes a rapid production rescheduling framework for flow shop under machine failure disturbance, called PPGA-ANNs, with the goal of minimizing makespan while ensuring sufficient computational efficiency. The framework begins with a scheduling knowledge creation phase conducted before starting production. It applies the proposed Perturbation Population Genetic Algorithm (PPGA) to solve generated scenarios of flow shop production with machine failure problems. The performance of the PPGA is compared to other research algorithms and to the standard genetic algorithm (GA). The same data set from a widely used scheduling benchmark is used for all algorithms to confirm the effectiveness of the PPGA. Artificial neural networks (ANNs) are then applied to store the scheduling knowledge obtained from the PPGA. In the knowledge implementation phase, when a machine failure problem occurs during production, the rescheduling solution is provided by the ANNs if the machine failure problem is identical to a generated scenario. Otherwise, the rescheduling solution is provided by the PPGA, using the initial solution obtained from the ANNs. Based on the experimental results, the PPGA-ANNs framework demonstrates better performance in makespans than benchmark algorithms. Additionally, it provides faster solutions, particularly for new machine failure problems. In conclusion, the proposed framework is capable of minimizing the makespan with a short computational time for real-world production, addressing the limitations of existing state-of-the-art meta-heuristic algorithms.https://ieeexplore.ieee.org/document/10179223/Artificial neural networkflow shop productiongenetic algorithmmachine failureproduction rescheduling |
| spellingShingle | Pakkaporn Saophan Warut Pannakkong Raveekiat Singhaphandu Van-Nam Huynh Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs) IEEE Access Artificial neural network flow shop production genetic algorithm machine failure production rescheduling |
| title | Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs) |
| title_full | Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs) |
| title_fullStr | Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs) |
| title_full_unstemmed | Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs) |
| title_short | Rapid Production Rescheduling for Flow Shop Under Machine Failure Disturbance Using Hybrid Perturbation Population Genetic Algorithm-Artificial Neural Networks (PPGA-ANNs) |
| title_sort | rapid production rescheduling for flow shop under machine failure disturbance using hybrid perturbation population genetic algorithm artificial neural networks ppga anns |
| topic | Artificial neural network flow shop production genetic algorithm machine failure production rescheduling |
| url | https://ieeexplore.ieee.org/document/10179223/ |
| work_keys_str_mv | AT pakkapornsaophan rapidproductionreschedulingforflowshopundermachinefailuredisturbanceusinghybridperturbationpopulationgeneticalgorithmartificialneuralnetworksppgaanns AT warutpannakkong rapidproductionreschedulingforflowshopundermachinefailuredisturbanceusinghybridperturbationpopulationgeneticalgorithmartificialneuralnetworksppgaanns AT raveekiatsinghaphandu rapidproductionreschedulingforflowshopundermachinefailuredisturbanceusinghybridperturbationpopulationgeneticalgorithmartificialneuralnetworksppgaanns AT vannamhuynh rapidproductionreschedulingforflowshopundermachinefailuredisturbanceusinghybridperturbationpopulationgeneticalgorithmartificialneuralnetworksppgaanns |