A Review of AI-Driven Railway Traffic Management Using Digital Twin Technology

Zunaira Fatima; Yathi Varun Somoju; Yash Biradar; Diana Moses

doi:10.55524/ijirem.2026.13.2.14

Abstract

The railway system is among the most extensive and complex parts of the world’s transport system. Traffic management is required to achieve punctuality, optimize capacity, and avoid delays from cascading to other trains. Classic approaches to rail operations have involved definition of dispatching via a manual approach and centralized rule-based control, resulting in poor scalability in regions with lots of mixed-traffic density. This technical report examines technologies that improve rail traffic management using modern optimization methods, machine learning, reinforcement learning, and digital twin methods. Conventional operations research techniques (e.g., mixed integer linear program (MILP), branch & bound, model predictive control) to improve efficiency of scheduling/travel time include AI-driven, deep reinforcement learning-based delay prediction; XGBoost-based prediction of delays, predictive maintenance frameworks. The results of the comparison show that optimization-based techniques continue to provide effective support in strategic planning and creating timetables, whereas they provide limited potential for dynamic, real-time control due to the lack of adaptability to changing operating conditions. Reinforcement learning techniques, on the other hand, will provide greater flexibility in dealing with uncertain, rapidly changing dispatching situations. Supervised models will provide good performance for delay prediction/disruption prediction (e.g., XGBoost), when applied to these rail systems. Lastly, this review suggests that the most promising future direction for intelligent railway traffic management systems will come from the development of hybrid Digital Twin architectures, which will integrate predictive analytics with adaptive control algorithms.

Keywords

Artificial Intelligence, Digital Twin, Railway Traffic Management, Reinforcement Learning

References

Prasad, C. and S. S. Jamuar, “Optimising Indian Railways infrastructure by AI,” Journal of Mobile Multimedia, vol. 17, no. 1–3, pp. 157–174, 2021. Available from: https://doi.org/10.13052/jmm1550-4646.17138
Chowdhury, K., P. Koley and A. Chakraborty, “RSTGCN: Railway-centric spatio-temporal graph convolution network,” arXiv preprint, 2025. Available from: https://doi.org/10.48550/arXiv.2510.01262
Ministry of Railways, FOIS Implementation Report. Government of India, 2010.
Comptroller and Auditor General (CAG) of India, FOIS Real-Time Tracking Limitations Audit Report, 2010. Available from: https://doi.org/10.1016/j.trpro.2020.08.021
Sharma, R. C., I. Hossain and A. Kumar, “Improving on-time performance: Predicting train delays with machine learning techniques,” in Sustainable Railway Ecosystem, Modeling, and Control, pp. 155–174, 2024. Available from: https://link.springer.com/chapter/10.1007/978-981-97-0437-8_9
CAMTECH, Centralized Traffic Control Handbook, 2022.
CTC Tundla, Implementation Case Study. Indian Railways Report, 2022.
Narouwa, M. et al., “Enabling network technologies for flexible railway connectivity,” IEEE Access, vol. 12, pp. 151532–151553, 2024. Available from: https://doi.org/10.1109/ACCESS.2024.3479879
Ramachandran, A., “Optimising locomotive scheduling in railway operations,” Indian Engineering Journal, 2025. Available from: https://iejournal.in/assets/pdf/5)%20IEJ%20May%202025-11-18.pdf
Garrisi and C. Cervelló-Pastor, “Train-scheduling optimization model for railway networks with multiplatform stations,” Sustainability, vol. 12, no. 1, p. 257, 2020. Available from: https://doi.org/10.3390/su12010257
M. Mazzarello and E. Ottaviani, “A traffic management system for real-time traffic optimisation in railways,” Transportation Research Part B: Methodological, vol. 41, pp. 246–274, 2007. Available from: https://doi.org/10.1016/j.trb.2006.02.005
A.-Q. Tian, J.-S. Pan and H.-X. Lv, “Optimizing train scheduling in heavy-haul railways using diversified cooperative deep reinforcement learning,” Transportation Research Record, vol. 2680, no. 2, pp. 286–312, 2026. Available from: https://doi.org/10.1177/03611981251364832
M. Aronsson, M. Bohlin and P. Kreuger, “MILP formulations of cumulative constraints for railway scheduling: A comparative study,” in ATMOS 2009, 2009. Available from: https://doi.org/10.4230/OASIcs.ATMOS.2009.2141
Sartor, C. Mannino, T. Nygreen and L. Bach, “A MILP model for quasi-periodic strategic train timetabling,” Omega, vol. 116, p. 102798, 2023. Available from: https://doi.org/10.1016/j.omega.2022.102798
D’Ariano, D. Pacciarelli and M. Pranzo, “A branch and bound algorithm for scheduling trains in a railway network,” European Journal of Operational Research, vol. 183, no. 2, pp. 643–657, 2007. Available from: https://doi.org/10.1016/j.ejor.2006.10.034
Mannino and A. Mascis, “Optimal real-time traffic control in metro stations,” Operations Research, vol. 57, no. 4, pp. 1026–1039, 2009. Available from: https://doi.org/10.1287/opre.1080.0642
S. Trivedi, “Timetabling of rail commuter services using constraint logic programming,” International Journal of Current Engineering and Technology, vol. 6, no. 1, pp. 81–83, 2016.
F. Corman, A. D’Ariano, D. Pacciarelli and M. Pranzo, “A tabu search algorithm for rerouting trains during rail operations,” Transportation Research Part B: Methodological, vol. 44, no. 1, pp. 175–192, 2010. Available from: https://doi.org/10.1016/j.trb.2009.05.004
J.-Y. Huang, “Using ant colony optimization to solve train timetabling problem of mass rapid transit,” 2006. Available from: https://doi.org/10.2991/jcis.2006.38
I. Afolayan, A. Ghosh, J. F. Calderín and A. D. Masegosa, “Emerging trends in machine learning assisted optimization techniques across intelligent transportation systems,” IEEE Access, vol. 12, pp. 173981–174005, 2024. Available from: https://doi.org/10.1109/ACCESS.2024.3501775
M. Khalilzadeh, D. Pamucar and A. Heidari, “Reducing train delays with machine learning-based predictive maintenance for railways,” Decision Making: Applications in Management and Engineering, vol. 8, no. 2, pp. 265–284, 2025. Available from: https://doi.org/10.31181/dmame8220251514
S. Hiraguri, “Evaluation of train control method using prediction control,” Quarterly Report of RTRI, vol. 49, pp. 163–167, 2008. Available from: https://doi.org/10.2219/rtriqr.49.163
L. Pullagura and J. Katiravan, “Train delay prediction using machine learning,” International Journal of Engineering and Advanced Technology, vol. 9, pp. 1312–1315, 2019. Available from: https://doi.org/10.35940/ijeat.A2088.129219
B. Kersbergen, T. van den Boom and B. De Schutter, “Distributed model predictive control for railway traffic management,” Transportation Research Part C: Emerging Technologies, vol. 68, pp. 462–489, 2016. Available from: https://doi.org/10.1016/j.trc.2016.05.006
Högdahl and M. Bohlin, “A combined simulation-optimization approach for robust timetabling on main railway lines,” Transportation Science, vol. 57, no. 1, pp. 52–81, 2022. Available from: https://doi.org/10.1287/trsc.2022.1158
L. Nielsen, L. Kroon and G. Maróti, “A rolling horizon approach for disruption management of railway rolling stock,” European Journal of Operational Research, vol. 220, pp. 496–509, 2012. Available from: https://doi.org/10.1016/j.ejor.2012.01.037
S. Kaewunruen, J. Sresakoolchai and Y. Lin, “Digital twins for managing railway maintenance and resilience,” Open Research Europe, vol. 1, p. 91, 2021. Available from: https://doi.org/10.12688/openreseurope.13806.2
Fuller, Z. Fan, C. Day and C. Barlow, “Digital Twin: Enabling technologies, challenges and open research,” IEEE Access, vol. 8, pp. 108952–108971, 2020. Available from: https://doi.org/10.1109/ACCESS.2020.2998358
Deo et al., “Implementation of a deep learning-based ICT model to solve the problem of enroute confirmation of waitlisted tickets in Indian Railways,” 2025. Available from: https://doi.org/10.1007/978-981-96-3352-4_23

Cites this article as

Z. Fatima, Y. V. Somoju, Y. Biradar, D. Moses, "A Review of AI-Driven Railway Traffic Management Using Digital Twin Technology", International Journal of Innovative Research in Engineering and Management (IJIREM), Vol-13, Issue-2, Page No-103-107, 2026. Available from: https://doi.org/10.55524/ijirem.2026.13.2.14