Maximizing Section Throughput Using AI‐Powered Precise Train Traffic Control
DOI:
https://doi.org/10.18178/JAAI.2026.4.2.143-151Keywords:
train traffic control, section throughput, Multi-Agent Reinforcement Learning (MARL), railway automation, predictive scheduling, Long Short-Term Memory (LSTM), movement authority, real-time optimization, Application Programming Interface (API), Coordinated Universal Time (UTC)Abstract
Railway networks transport millions of passengers and vast quantities of freight daily, yet a substantial share of operational delays arises from manual or semi-automated traffic management systems that cannot adapt swiftly enough to evolving track conditions. This paper proposes an Artificial Intelligencedriven framework for maximizing section throughput—the number of trains traversing a defined track segment per unit time—while enforcing safe headways and minimizing schedule deviation. The system pairs real-time sensor telemetry with a Multi-Agent Reinforcement Learning (MARL) controller that issues movement authorities and adjusts speed profiles at block boundaries, supported by an Long Short-Term Memory (LSTM)-based predictive layer that forecasts congestion bottlenecks up to 40 minutes ahead. Simulation experiments on a 187 km single-track corridor with 14 intermediate stations show the AI controller raises section capacity by up to 23% under normal conditions and 39% under severe disturbance compared to a rule-based baseline, while reducing mean delay propagation by 41%. The architecture deploys above legacy interlocking systems via a standardized Application Programming Interface (API) bridge, preserving existing safety certifications, and recorded zero safety violations across all evaluation episodes.
References
[1] McKinsey & Company. (2022). The rail sector's journey to net zero: A strategic perspective on European railway performance. Retrieved from https://www.mckinsey.com/
[2] Ye, H., Sun, Z., Fang, J., & Jiang, Z. (2019). Predictive train control using integrated deep learning. IEEE Access, 7, 24753–24763. https://doi.org/10.1109/ACCESS.2019.2898973
[3] Dollevoet, T., Huisman, D., Schmidt, M., & Schöbel, A. (2012). Delay management with rerouting of passengers. Transportation Science, 46(1), 74–89. https://doi.org/10.1287/trsc.1110.0375
[4] Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. (2017). Proximal policy optimization algorithms. arXiv preprint, arXiv:1707.06347.
[5] Abril, M., Barber, F., Ingolotti, L., Salido, M. A., Tormos, P., & Lova, A. (2008). An assessment of railway capacity. Transportation Research Part E: Logistics and Transportation Review, 44(5), 774–806. https://doi.org/10.1016/j.tre.2007.04.001
[6] Bešinović, N., Goverde, R. M. P., Quaglietta, E., & Roberti, R. (2022). Neural network-based prediction for train rescheduling. Transportation Research Part C: Emerging Technologies, 138, 103605. https://doi.org/10.1016/j.trc.2022.103605
[7] Carey, M., & Lockwood, D. (1995). A model, algorithms and strategy for train pathing. Journal of the Operational Research Society, 46(8), 988–1005. https://doi.org/10.1057/jors.1995.135
[8] Chang, C. S., & Sim, S. S. (1997). Optimising train movements through coast control using genetic algorithms. Proceedings of IEE—Electric Power Applications, vol. 144, no. 1. (pp. 65–73). https://doi.org/10.1049/ip-epa:19970818
[9] Corman, F., D’Ariano, A., Pacciarelli, D., & Pranzo, M. (2010). A tabu search algorithm for rerouting trains during rail operations. Transportation Research Part B: Methodological, 44(1), 175–192. https://doi.org/10.1016/j.trb.2009.05.004
[10] European Union Agency for Railways (ERA). (2019). ERTMS/ETCS System Requirements Specification (Subset-026 v3.6.0). Retrieved from https://www.era.europa.eu/
[11] Goverde, R. M. P. (2005). Punctuality of railway operations and timetable stability analysis [Doctoral dissertation, Delft University of Technology]. TU Delft Repository. Retrieved from https://repository.tudelft.nl/
[12] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
[13] Wang, Y., Tang, T., Ning, L., & van den Boom, T. (2018). Passenger demand oriented train scheduling and rolling stock circulation planning for an urban rail transit line. Transportation Research Part B: Methodological, 118, 193–212. https://doi.org/10.1016/j.trb.2018.10.006
[14] Lowe, R., Wu, Y., Tamar, A., Harb, J., Abbeel, P., & Mordatch, I. (2017). Multi-agent actor-critic for mixed cooperative-competitive environments. Advances in Neural Information Processing Systems (NeurIPS), 30, 6379–6390.
[15] Yin, J., Yang, L., Tang, T., Gao, Z., & Ran, B. (2017). Energy-efficient metro train rescheduling. Transportation Research Part B: Methodological, 91, 178–210. https://doi.org/10.1016/j.trb.2016.07.005
[16] International Union of Railways (UIC). (2004). UIC Code 406: Capacity. Retrieved from https://uic.org/
[17] Zhang, Q., Han, B., & Li, D. (2008). Modeling and simulation of passenger alighting and boarding movement in Beijing metro stations. Transportation Research Part C: Emerging Technologies, 16(5), 635–649. https://doi.org/10.1016/j.trc.2008.01.002
[18] Lüthi, M., Medeossi, G., & Nash, A. (2009). Increasing railway capacity and reliability through integrated real-time rescheduling. Transportation Research Record, 2117, 56–65. https://doi.org/10.3141/2117-08
[19] Pachl, J. (2002). Railway Operation and Control. VTD Rail Publishing.
[20] Zhong, Q., Zhao, Y., Zhu, X., & Liu, H. (2021). Attention-based reinforcement learning for train regulation in metro systems. IEEE Transactions on Intelligent Transportation Systems, 23(7), 8279–8292. https://doi.org/10.1109/TITS.2021.3055525
[21] Quaglietta, E., Pellegrini, P., Goverde, R. M. P., Albrecht, T., Jaekel, B., & Hansen, I. A. (2016). Impact of a stochastic and dynamic setting on railway dispatching stability. Journal of Rail Transport Planning & Management, 6(1), 1–22. https://doi.org/10.1016/j.jrtpm.2015.11.002
Published
Issue
Section
License
Copyright (c) 2026 by the authors.
This work is licensed under a Creative Commons Attribution 4.0 International License.
