Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study
Travel time information is used as input or auxiliary data for tasks such as dynamic navigation, infrastructure planning, congestion control, and accident detection. Various data-driven Travel Time Prediction (TTP) methods have been proposed in recent years. One of the most challenging tasks in TTP...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/1/106 |
| _version_ | 1797499212987891712 |
|---|---|
| author | Irfan Ahmed Indika Kumara Vahideh Reshadat A. S. M. Kayes Willem-Jan van den Heuvel Damian A. Tamburri |
| author_facet | Irfan Ahmed Indika Kumara Vahideh Reshadat A. S. M. Kayes Willem-Jan van den Heuvel Damian A. Tamburri |
| author_sort | Irfan Ahmed |
| collection | DOAJ |
| description | Travel time information is used as input or auxiliary data for tasks such as dynamic navigation, infrastructure planning, congestion control, and accident detection. Various data-driven Travel Time Prediction (TTP) methods have been proposed in recent years. One of the most challenging tasks in TTP is developing and selecting the most appropriate prediction algorithm. The existing studies that empirically compare different TTP models only use a few models with specific features. Moreover, there is a lack of research on explaining TTPs made by black-box models. Such explanations can help to tune and apply TTP methods successfully. To fill these gaps in the current TTP literature, using three data sets, we compare three types of TTP methods (ensemble tree-based learning, deep neural networks, and hybrid models) and ten different prediction algorithms overall. Furthermore, we apply XAI (Explainable Artificial Intelligence) methods (SHAP and LIME) to understand and interpret models’ predictions. The prediction accuracy and reliability for all models are evaluated and compared. We observed that the ensemble learning methods, i.e., XGBoost and LightGBM, are the best performing models over the three data sets, and XAI methods can adequately explain how various spatial and temporal features influence travel time. |
| first_indexed | 2024-03-10T03:44:15Z |
| format | Article |
| id | doaj.art-3921a051c27e4700846c97ef24d0e97e |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T03:44:15Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-3921a051c27e4700846c97ef24d0e97e2023-11-23T11:22:59ZengMDPI AGElectronics2079-92922021-12-0111110610.3390/electronics11010106Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative StudyIrfan Ahmed0Indika Kumara1Vahideh Reshadat2A. S. M. Kayes3Willem-Jan van den Heuvel4Damian A. Tamburri5Jheronimus Academy of Data Science, Sint Janssingel 92, 5211 DA Hertogenbosch, The NetherlandsJheronimus Academy of Data Science, Sint Janssingel 92, 5211 DA Hertogenbosch, The NetherlandsDepartment of Industrial Engineering and Innovation Sciences, Eindhoven University of Technology, 5612 AZ Eindhoven, The NetherlandsDepartment of Computer Science and Information Technology, La Trobe University, Plenty Road, Melbourne, VIC 3086, AustraliaJheronimus Academy of Data Science, Sint Janssingel 92, 5211 DA Hertogenbosch, The NetherlandsJheronimus Academy of Data Science, Sint Janssingel 92, 5211 DA Hertogenbosch, The NetherlandsTravel time information is used as input or auxiliary data for tasks such as dynamic navigation, infrastructure planning, congestion control, and accident detection. Various data-driven Travel Time Prediction (TTP) methods have been proposed in recent years. One of the most challenging tasks in TTP is developing and selecting the most appropriate prediction algorithm. The existing studies that empirically compare different TTP models only use a few models with specific features. Moreover, there is a lack of research on explaining TTPs made by black-box models. Such explanations can help to tune and apply TTP methods successfully. To fill these gaps in the current TTP literature, using three data sets, we compare three types of TTP methods (ensemble tree-based learning, deep neural networks, and hybrid models) and ten different prediction algorithms overall. Furthermore, we apply XAI (Explainable Artificial Intelligence) methods (SHAP and LIME) to understand and interpret models’ predictions. The prediction accuracy and reliability for all models are evaluated and compared. We observed that the ensemble learning methods, i.e., XGBoost and LightGBM, are the best performing models over the three data sets, and XAI methods can adequately explain how various spatial and temporal features influence travel time.https://www.mdpi.com/2079-9292/11/1/106travel time predictionspatio-temporalXGBoostLightGBMLSTMhybrid models |
| spellingShingle | Irfan Ahmed Indika Kumara Vahideh Reshadat A. S. M. Kayes Willem-Jan van den Heuvel Damian A. Tamburri Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study Electronics travel time prediction spatio-temporal XGBoost LightGBM LSTM hybrid models |
| title | Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study |
| title_full | Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study |
| title_fullStr | Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study |
| title_full_unstemmed | Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study |
| title_short | Travel Time Prediction and Explanation with Spatio-Temporal Features: A Comparative Study |
| title_sort | travel time prediction and explanation with spatio temporal features a comparative study |
| topic | travel time prediction spatio-temporal XGBoost LightGBM LSTM hybrid models |
| url | https://www.mdpi.com/2079-9292/11/1/106 |
| work_keys_str_mv | AT irfanahmed traveltimepredictionandexplanationwithspatiotemporalfeaturesacomparativestudy AT indikakumara traveltimepredictionandexplanationwithspatiotemporalfeaturesacomparativestudy AT vahidehreshadat traveltimepredictionandexplanationwithspatiotemporalfeaturesacomparativestudy AT asmkayes traveltimepredictionandexplanationwithspatiotemporalfeaturesacomparativestudy AT willemjanvandenheuvel traveltimepredictionandexplanationwithspatiotemporalfeaturesacomparativestudy AT damianatamburri traveltimepredictionandexplanationwithspatiotemporalfeaturesacomparativestudy |