Time Prediction in Ship Block Manufacturing Based on Transfer Learning
Accurate time prediction is critical to the success of ship block manufacturing. However, the emergence of new ship types with limited historical data poses challenges to existing prediction methods. In response, this paper proposes a novel framework for ship block manufacturing time prediction, int...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-11-01
|
| Series: | Journal of Marine Science and Engineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-1312/12/11/1977 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850267217381294080 |
|---|---|
| author | Jinghua Li Pengfei Lin Dening Song Zhe Yan Boxin Yang Lei Zhou |
| author_facet | Jinghua Li Pengfei Lin Dening Song Zhe Yan Boxin Yang Lei Zhou |
| author_sort | Jinghua Li |
| collection | DOAJ |
| description | Accurate time prediction is critical to the success of ship block manufacturing. However, the emergence of new ship types with limited historical data poses challenges to existing prediction methods. In response, this paper proposes a novel framework for ship block manufacturing time prediction, integrating clustering and the transfer learning algorithm. Firstly, the concept of distributed centroids was innovatively adopted to achieve the clustering of categorical attribute features. Secondly, abundant historical data from other types of blocks (source domain) were incorporated into the neural network model to explore the effects of block features on manufacturing time, and the model was further transferred to blocks with limited data (target domain). Leveraging the similarities and differences between source and target domain blocks, actions involving freezing and fine-tuning parameters were adopted for the predictive model development. Despite a small sample size of only 80, our proposed block time prediction method achieves an impressive mean absolute percentage error (MAPE) of 8.62%. In contrast, the MAPE for the predictive model without a transfer learning algorithm is notably higher at 14.97%. Experimental validation demonstrates the superior performance of our approach compared to alternative methods in scenarios with small sample datasets. This research addresses a critical gap in ship block manufacturing time prediction. |
| format | Article |
| id | doaj-art-bf1f25f1b29b431ea9850737fd9c77ea |
| institution | OA Journals |
| issn | 2077-1312 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Marine Science and Engineering |
| spelling | doaj-art-bf1f25f1b29b431ea9850737fd9c77ea2025-08-20T01:53:53ZengMDPI AGJournal of Marine Science and Engineering2077-13122024-11-011211197710.3390/jmse12111977Time Prediction in Ship Block Manufacturing Based on Transfer LearningJinghua Li0Pengfei Lin1Dening Song2Zhe Yan3Boxin Yang4Lei Zhou5College of Mechanical and Electrical Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Shipbuilding Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Mechanical and Electrical Engineering, Harbin Engineering University, Harbin 150001, ChinaShanghai Waigaoqiao Shipbuilding Co., Ltd., Shanghai 200137, ChinaCollege of Mechanical and Electrical Engineering, Harbin Engineering University, Harbin 150001, ChinaCollege of Mechanical and Electrical Engineering, Harbin Engineering University, Harbin 150001, ChinaAccurate time prediction is critical to the success of ship block manufacturing. However, the emergence of new ship types with limited historical data poses challenges to existing prediction methods. In response, this paper proposes a novel framework for ship block manufacturing time prediction, integrating clustering and the transfer learning algorithm. Firstly, the concept of distributed centroids was innovatively adopted to achieve the clustering of categorical attribute features. Secondly, abundant historical data from other types of blocks (source domain) were incorporated into the neural network model to explore the effects of block features on manufacturing time, and the model was further transferred to blocks with limited data (target domain). Leveraging the similarities and differences between source and target domain blocks, actions involving freezing and fine-tuning parameters were adopted for the predictive model development. Despite a small sample size of only 80, our proposed block time prediction method achieves an impressive mean absolute percentage error (MAPE) of 8.62%. In contrast, the MAPE for the predictive model without a transfer learning algorithm is notably higher at 14.97%. Experimental validation demonstrates the superior performance of our approach compared to alternative methods in scenarios with small sample datasets. This research addresses a critical gap in ship block manufacturing time prediction.https://www.mdpi.com/2077-1312/12/11/1977ship block manufacturingtime predictionclusteringtransfer learning algorithmneural network |
| spellingShingle | Jinghua Li Pengfei Lin Dening Song Zhe Yan Boxin Yang Lei Zhou Time Prediction in Ship Block Manufacturing Based on Transfer Learning Journal of Marine Science and Engineering ship block manufacturing time prediction clustering transfer learning algorithm neural network |
| title | Time Prediction in Ship Block Manufacturing Based on Transfer Learning |
| title_full | Time Prediction in Ship Block Manufacturing Based on Transfer Learning |
| title_fullStr | Time Prediction in Ship Block Manufacturing Based on Transfer Learning |
| title_full_unstemmed | Time Prediction in Ship Block Manufacturing Based on Transfer Learning |
| title_short | Time Prediction in Ship Block Manufacturing Based on Transfer Learning |
| title_sort | time prediction in ship block manufacturing based on transfer learning |
| topic | ship block manufacturing time prediction clustering transfer learning algorithm neural network |
| url | https://www.mdpi.com/2077-1312/12/11/1977 |
| work_keys_str_mv | AT jinghuali timepredictioninshipblockmanufacturingbasedontransferlearning AT pengfeilin timepredictioninshipblockmanufacturingbasedontransferlearning AT deningsong timepredictioninshipblockmanufacturingbasedontransferlearning AT zheyan timepredictioninshipblockmanufacturingbasedontransferlearning AT boxinyang timepredictioninshipblockmanufacturingbasedontransferlearning AT leizhou timepredictioninshipblockmanufacturingbasedontransferlearning |