Combining Machine Learning Models to Improve Estimated Time of Arrival Predictions
DOI:
https://doi.org/10.59490/ejtir.2025.25.1.7488Keywords:
machine learning, flight predictability, estimated time of arrivalAbstract
All aviation stakeholders require accurate estimated times of arrival in order to run flight operations as efficiently as possible. The time of arrival, however, is difficult to predict because it is affected by the uncertainties of the previous flight phases, with take-off time variability being the most significant contributor. At present, estimated time of arrival predictions are computed by the Enhanced Traffic Flow Management System, which collects data from a variety of sources to provide the best estimate throughout the entire duration of the flight. This paper introduces a novel approach that leverages existing machine learning models to enhance the accuracy of estimated time of arrival predictions, also during the pre-departure phase. More specifically, the first model (Knock-on) anticipates rotational reactionary delays arising from unrealistic available turn-around times; the second model (FADE) forecasts the evolution of air traffic flow management delays for regulated flights; and the third model, AirborneTime, was trained to identify systematic discrepancies between reported and actual airborne times. Using a dataset comprised of historical traffic and meteorological data collected during one year, this paper presents a comprehensive evaluation of this ensemble of models, referred to as PETA, against the current predictions across various time horizons, ranging from 6 hours before departure to the moment of take-off. The results indicate that the proposed solution surpasses the existing system in approximately two-thirds of the predictions. When the proposed solution performs better, the average and median improvements are 14 minutes and 7 minutes, respectively. However, when it underperforms, the average and median deteriorations are 7 minutes and 4 minutes, respectively. The optimal time frame appears to be between 2 and 6 hours before the departure time. This quantitative data is supported by feedback from European airlines, air navigation service providers and airports who used PETA in a live trial.
Downloads
Metrics
References
Strottmann Kern, C., Medeiros, I.P., Yoneyama, T. (2015). Data-driven aircraft estimated time of arrival prediction. 2015 Annual IEEE Systems Conference (SysCon) Proceedings, Vancouver, BC, pp. 727–733. https://doi.org/10.1109/SYSCON.2015.7116837
Ayhan, S., Costas, P., Samet, H. (2018). Predicting Estimated Time of Arrival for Commercial Flights. Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, New York, NY, pp. 33–42. https://doi.org/10.1145/3219819.3219874
Wang, Z., Liang, M., Delahaye, D. (2020a). Automated data-driven prediction on aircraft Estimated Time of Arrival. Journal of Air Transport Management, 88, 101840. https://doi.org/10.1016/j.jairtraman.2020.101840
Wang, G., Liu, K., Chen, H., Wang, Y., Zhao, Q. (2020b). A High-precision Method of Flight Arrival Time Estimation based on XGBoost. 2020 IEEE 2nd International Conference on Civil Aviation Safety and Information Technology (ICCASIT), Weihai, China, pp. 883–888. https://doi.org/10.1109/ICCASIT50869.2020.9368723
Christien, R., Favennec, B., Pasutto, P., Trzmiel, A., Weiss, J., Zeghal, K. (2021). Predicting arrival delays in the terminal area five hours in advance with machine learning. 14th USA Europe Air Traffic Management Research and Development Seminar, Virtual Event.
Silvestre, J., Santiago, M., Bregón, A., Martínez-Prieto, M.A., Álvarez-Esteban, P.C. (2021). On the Use of Deep Neural Networks to Improve Flights Estimated Time of Arrival Predictions. Engineering Proceedings, 13(1). https://doi.org/10.3390/engproc2021013003
Zhang, J., Peng, Z., Yang, C., Wang, B. (2022). Data-driven flight time prediction for arrival aircraft within the terminal area. IET Intelligent Transport Systems, 16(2), 263–275. https://doi.org/10.1049/itr2.12142
Ma, Y., Du, W., Chen, J., Zhang, Y., Lv, Y., Cao, X. (2023). A Spatiotemporal Neural Network Model for Estimated-Time-of-Arrival Prediction of Flights in a Terminal Maneuvering Area. IEEE Intelligent Transportation Systems Magazine, 15(1), 285–299. https://doi.org/10.1109/MITS.2021.3132766
Wang, L., Mao, J., Li, L., Li, X., Tu, Y. (2023). Prediction of estimated time of arrival for multi-airport systems via “Bubble” mechanism. Transportation Research Part C: Emerging Technologies, 149, 104065. https://doi.org/10.1016/j.trc.2023.104065
EUROCONTROL (2024). EATIN - EUROCONTROL Air Traffic Network Information. Accessed: 2024-08-04. https://www.eurocontrol.int/project/eatin
Dalmau, R. (2024). Probabilistic and explainable tree-based models for rotational reactionary flight delay prediction. CEAS Aeronautical Journal. https://doi.org/10.1007/s13272-024-00750-w
Dalmau, R., Genestier, B., Anoraud, C., Choroba, P., Smith, D. (2021). A Machine Learning Approach to Predict the Evolution of Air Traffic Flow Management Delay. 14th USA Europe Air Traffic Management Research and Development Seminar, Virtual Event.
Lundberg, S.M., Erion, G., Chen, H., et al. (2020). From local explanations to global understanding with explainable AI for trees. Nature Machine Intelligence, 2, 56–67. https://doi.org/10.1038/s42256-019-0138-9
Downloads
Published
How to Cite
License
Copyright (c) 2025 Ramon Dalmau, Aymeric Trzmiel, Stephen Kirby
This work is licensed under a Creative Commons Attribution 4.0 International License.
