Variable taxi-out time prediction based on machine learning with interpretable attributes
This paper presents a machine learning-based approach for predicting the taxi-out time, with the departure process decomposed into two components: the time taken to travel from the gate to the departure queue, and the time spent in the departure queue. Gradient-Boosted Decision Tree (GBDT) models ar...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/178975 |
| _version_ | 1826110449517592576 |
|---|---|
| author | Lim, Yixiang Alam, Sameer Tan, Fengji Lilith, Nimrod |
| author2 | School of Mechanical and Aerospace Engineering |
| author_facet | School of Mechanical and Aerospace Engineering Lim, Yixiang Alam, Sameer Tan, Fengji Lilith, Nimrod |
| author_sort | Lim, Yixiang |
| collection | NTU |
| description | This paper presents a machine learning-based approach for predicting the taxi-out time, with the departure process decomposed into two components: the time taken to travel from the gate to the departure queue, and the time spent in the departure queue. Gradient-Boosted Decision Tree (GBDT) models are trained to predict the two components using different feature sets, and a comparison of both model shows that they can provide better prediction accuracy compared with conventional methods, with a Root Mean Squared Error (RMSE) of 1.79 minutes and 0.92 minutes when predicting the taxiing and queuing times respectively, and 78% and 96% of predictions falling within a «2 minute error margin. Predictions from the GBDT model are analysed and interpreted using SHAP (SHapley Additive exPlanations) values, a wellrecognised technique for providing interpretability to many different black-box models, and allowing feature importance to be evaluated at global (model) and local (individual prediction) levels. In particular, the most important feature groups for the taxiing and queuing models are respectively the route features and runway queuing features. The model explainability provides a pathway towards the certification of machine learning techniques in Air Traffic Controller (ATCO) decision support tools. |
| first_indexed | 2024-10-01T02:34:36Z |
| format | Journal Article |
| id | ntu-10356/178975 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T02:34:36Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1789752024-07-20T16:48:05Z Variable taxi-out time prediction based on machine learning with interpretable attributes Lim, Yixiang Alam, Sameer Tan, Fengji Lilith, Nimrod School of Mechanical and Aerospace Engineering Saab-NTU Joint Lab Engineering Variable Taxi Time Machine Learning This paper presents a machine learning-based approach for predicting the taxi-out time, with the departure process decomposed into two components: the time taken to travel from the gate to the departure queue, and the time spent in the departure queue. Gradient-Boosted Decision Tree (GBDT) models are trained to predict the two components using different feature sets, and a comparison of both model shows that they can provide better prediction accuracy compared with conventional methods, with a Root Mean Squared Error (RMSE) of 1.79 minutes and 0.92 minutes when predicting the taxiing and queuing times respectively, and 78% and 96% of predictions falling within a «2 minute error margin. Predictions from the GBDT model are analysed and interpreted using SHAP (SHapley Additive exPlanations) values, a wellrecognised technique for providing interpretability to many different black-box models, and allowing feature importance to be evaluated at global (model) and local (individual prediction) levels. In particular, the most important feature groups for the taxiing and queuing models are respectively the route features and runway queuing features. The model explainability provides a pathway towards the certification of machine learning techniques in Air Traffic Controller (ATCO) decision support tools. Published version This work was conducted under the Saab-NTU Joint Lab with support from Saab AB (publ). 2024-07-15T04:34:20Z 2024-07-15T04:34:20Z 2024 Journal Article Lim, Y., Alam, S., Tan, F. & Lilith, N. (2024). Variable taxi-out time prediction based on machine learning with interpretable attributes. Transactions of the Japan Society for Aeronautical and Space Sciences, 67(3), 136-144. https://dx.doi.org/10.2322/tjsass.67.136 0549-3811 https://hdl.handle.net/10356/178975 10.2322/tjsass.67.136 2-s2.0-85193781029 3 67 136 144 en Transactions of the Japan Society for Aeronautical and Space Sciences © 2024 The Authors. JSASS has the license to publish of this article. This is an open access article distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. application/pdf |
| spellingShingle | Engineering Variable Taxi Time Machine Learning Lim, Yixiang Alam, Sameer Tan, Fengji Lilith, Nimrod Variable taxi-out time prediction based on machine learning with interpretable attributes |
| title | Variable taxi-out time prediction based on machine learning with interpretable attributes |
| title_full | Variable taxi-out time prediction based on machine learning with interpretable attributes |
| title_fullStr | Variable taxi-out time prediction based on machine learning with interpretable attributes |
| title_full_unstemmed | Variable taxi-out time prediction based on machine learning with interpretable attributes |
| title_short | Variable taxi-out time prediction based on machine learning with interpretable attributes |
| title_sort | variable taxi out time prediction based on machine learning with interpretable attributes |
| topic | Engineering Variable Taxi Time Machine Learning |
| url | https://hdl.handle.net/10356/178975 |
| work_keys_str_mv | AT limyixiang variabletaxiouttimepredictionbasedonmachinelearningwithinterpretableattributes AT alamsameer variabletaxiouttimepredictionbasedonmachinelearningwithinterpretableattributes AT tanfengji variabletaxiouttimepredictionbasedonmachinelearningwithinterpretableattributes AT lilithnimrod variabletaxiouttimepredictionbasedonmachinelearningwithinterpretableattributes |