Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches
Rising shipping emissions greatly affect greenhouse gas (GHG) levels, so precise fuel consumption forecasting is essential to reduce environmental effects. Precision forecasts using machine learning (ML) could offer sophisticated solutions that increase the fuel efficiency and lower emissions. Indee...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Sciendo
2025-03-01
|
| Series: | Polish Maritime Research |
| Subjects: | |
| Online Access: | https://doi.org/10.2478/pomr-2025-0008 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850072912168484864 |
|---|---|
| author | Pham Nguyen Dang Khoa Dinh Gia Huy Nguyen Canh Lam Dang Hai Quoc Pham Hoang Thai Nguyen Quyen Tat Tran Minh Cong |
| author_facet | Pham Nguyen Dang Khoa Dinh Gia Huy Nguyen Canh Lam Dang Hai Quoc Pham Hoang Thai Nguyen Quyen Tat Tran Minh Cong |
| author_sort | Pham Nguyen Dang Khoa |
| collection | DOAJ |
| description | Rising shipping emissions greatly affect greenhouse gas (GHG) levels, so precise fuel consumption forecasting is essential to reduce environmental effects. Precision forecasts using machine learning (ML) could offer sophisticated solutions that increase the fuel efficiency and lower emissions. Indeed, five ML techniques, linear regression (LR), decision tree (DT), random forest (RF), XGBoost, and AdaBoost, were used to develop ship fuel consumption models in this study. It was found that, with an R² of 1, zero mean squared error (MSE), and a negligible mean absolute percentage error (MAPE), the DT model suited the training set perfectly, while R² was 0.8657, the MSE was 56.80, and the MAPE was 16.37% for the DT model testing. More importantly, this study provided Taylor diagrams and violin plots that helped in the identification of the best-performing models. Generally, the employed ML approaches efficiently predicted the data; however, they are black-box methods. Hence, explainable machine learning methods like Shapley additive explanations, the DT structure, and local interpretable model-agnostic explanations (LIME) were employed to comprehend the models and perform feature analysis. LIME offered insights, demonstrating that the major variables impacting predictions were distance (≤450.88 nm) and time (40.70 < hr ≤ 58.05). By stressing the most important aspects, LIME can help one to comprehend the models with ease. |
| format | Article |
| id | doaj-art-9f119c2df4f94c6088ebd0916898f82c |
| institution | DOAJ |
| issn | 2083-7429 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Sciendo |
| record_format | Article |
| series | Polish Maritime Research |
| spelling | doaj-art-9f119c2df4f94c6088ebd0916898f82c2025-08-20T02:46:59ZengSciendoPolish Maritime Research2083-74292025-03-01321819410.2478/pomr-2025-0008Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning ApproachesPham Nguyen Dang Khoa0Dinh Gia Huy1Nguyen Canh Lam2Dang Hai Quoc3Pham Hoang Thai4Nguyen Quyen Tat5Tran Minh Cong6Ho Chi Minh City University of Transport, Ho Chi Minh city, Viet NamHo Chi Minh City University of Transport, Ho Chi Minh city, Viet NamThe Business School, Business Innovation Department, RMIT University, Ho Chi Minh city, Viet NamHo Chi Minh City University of Transport, Ho Chi Minh city, Viet NamHo Chi Minh City University of Transport, Ho Chi Minh city, Viet NamHo Chi Minh City University of Transport, Ho Chi Minh city, Viet NamNha Trang University, Nha Trang, Viet NamRising shipping emissions greatly affect greenhouse gas (GHG) levels, so precise fuel consumption forecasting is essential to reduce environmental effects. Precision forecasts using machine learning (ML) could offer sophisticated solutions that increase the fuel efficiency and lower emissions. Indeed, five ML techniques, linear regression (LR), decision tree (DT), random forest (RF), XGBoost, and AdaBoost, were used to develop ship fuel consumption models in this study. It was found that, with an R² of 1, zero mean squared error (MSE), and a negligible mean absolute percentage error (MAPE), the DT model suited the training set perfectly, while R² was 0.8657, the MSE was 56.80, and the MAPE was 16.37% for the DT model testing. More importantly, this study provided Taylor diagrams and violin plots that helped in the identification of the best-performing models. Generally, the employed ML approaches efficiently predicted the data; however, they are black-box methods. Hence, explainable machine learning methods like Shapley additive explanations, the DT structure, and local interpretable model-agnostic explanations (LIME) were employed to comprehend the models and perform feature analysis. LIME offered insights, demonstrating that the major variables impacting predictions were distance (≤450.88 nm) and time (40.70 < hr ≤ 58.05). By stressing the most important aspects, LIME can help one to comprehend the models with ease.https://doi.org/10.2478/pomr-2025-0008ship fuel consumptionfuel consumption forecastingmodel predictionmachine learning techniqueoperation efficiency |
| spellingShingle | Pham Nguyen Dang Khoa Dinh Gia Huy Nguyen Canh Lam Dang Hai Quoc Pham Hoang Thai Nguyen Quyen Tat Tran Minh Cong Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches Polish Maritime Research ship fuel consumption fuel consumption forecasting model prediction machine learning technique operation efficiency |
| title | Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches |
| title_full | Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches |
| title_fullStr | Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches |
| title_full_unstemmed | Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches |
| title_short | Forecasting and Feature Analysis of Ship Fuel Consumption by Explainable Machine Learning Approaches |
| title_sort | forecasting and feature analysis of ship fuel consumption by explainable machine learning approaches |
| topic | ship fuel consumption fuel consumption forecasting model prediction machine learning technique operation efficiency |
| url | https://doi.org/10.2478/pomr-2025-0008 |
| work_keys_str_mv | AT phamnguyendangkhoa forecastingandfeatureanalysisofshipfuelconsumptionbyexplainablemachinelearningapproaches AT dinhgiahuy forecastingandfeatureanalysisofshipfuelconsumptionbyexplainablemachinelearningapproaches AT nguyencanhlam forecastingandfeatureanalysisofshipfuelconsumptionbyexplainablemachinelearningapproaches AT danghaiquoc forecastingandfeatureanalysisofshipfuelconsumptionbyexplainablemachinelearningapproaches AT phamhoangthai forecastingandfeatureanalysisofshipfuelconsumptionbyexplainablemachinelearningapproaches AT nguyenquyentat forecastingandfeatureanalysisofshipfuelconsumptionbyexplainablemachinelearningapproaches AT tranminhcong forecastingandfeatureanalysisofshipfuelconsumptionbyexplainablemachinelearningapproaches |