Rapid Algorithm for Generating and Selecting Optimal Metro Train Speed Curves Based on Alpha Zero and Expert Experience


  • Dewang CHEN School of Transportation, Fujian Institute of Technology
  • Zhongjie WU School of Transportation, Fujian Institute of Technology
  • Yuqi LU School of Transportation, Fujian Institute of Technology
  • Wendi ZHAO School of Transportation, Fujian Institute of Technology
  • Zhiming LIN School of Transportation, Fujian Institute of Technology




rail transit, speed control, data generation, curve optimisation, urban rail transit, train speed


According to the current research status of urban rail transit’s fully automatic operation (FAO), the train driving speed curves are usually obtained through simulation and calculation. The train driving speed curves obtained by this method not only have low efficiency but also are not suitable for complex road conditions. Inspired by AlphaZero, a reinforcement learning algorithm that utilised vast amounts of artificial data to defeat AlphaGo, an AI Go program, this paper investigates and analyses methods for rapidly generating a large number of speed curves and selecting those with superior performance for train operation. Firstly, we use the powerful third-party library in Python as the basis, combined with the idea of AlphaZero, to produce artificial speed curves for metro train driving. Secondly, we set relevant parameters with reference to expert experience to quickly produce massive reasonable artificial speed curves. Thirdly, we analysed relevant indicators such as energy consumption, running time error and passenger comfort to select some speed curves with better comprehensive performance. Finally, through the many observations with different running distances and different speed limits, we found that the speed curves produced and selected by our algorithm are more productive, diverse and conducive to the research of train driving operation than the actual data from traditional manual driving and ATO (automatic train operation) system.


Wei Y, et al. Research on the basic theoretical issues of smart metro construction. Science China Technological Sciences. 2019;62:2038–2051. DOI: 10.1007/s11431-019-9551-4.

He N, et al. Rail-induced traffic in China. Promet – Traffic&Transportation. 2017;29(5):511–520. DOI: 10.7307/ptt.v29i5.2235.

Zhou Q, et al. Analysis of the impacts of passenger demand on the profitability of different types of urban rail transit. Promet – Traffic&Transportation. 2023;35(1):71–86. DOI: 10.7307/ptt.v35i1.4185.

Yang X, et al. A novel prediction model for the inbound passenger flow of urban rail transit. Information Sciences. 2021;566:347–363. DOI: 10.1016/j.ins.2021.02.036.

Zhang Q, et al. A two-layer modelling framework for predicting passenger flow on trains: A case study of London underground trains. Transportation Research Part A: Policy and Practice. 2021;151:119–139. DOI: 10.1016/j.tra.2021.07.001.

Jong JC, Chang S. Algorithms for generating train speed profiles. Journal of the eastern ASIA society for transportation studies. 2005;6:356–371. DOI: 10.11175/easts.6.356.

Miyatake M, Ko H. Optimization of train speed profile for minimum energy consumption. IEEJ Transactions on Electrical and Electronic Engineering. 2010;5(3):263–269. DOI: 10.1002/tee.20528.

Jonaitis J. Determination of extreme train running parameters along a railway line segment. Transport. 2006;21(2):123–130. DOI: 10.1080/16484142.2006.9638053.

Albrecht T, Binder A, Gassel C. Applications of real‐time speed control in rail‐bound public transportation systems. IET Intelligent Transport Systems. 2013;7(3):305–314. DOI: 10.1049/iet-its.2011.0187.

Zhao J, Deng W. Fuzzy multiobjective decision support model for urban rail transit projects in China. Transport. 2013;28(3):224–235. DOI: 3846/16484142.2013.829119.

Chen D, et al. Online learning algorithms for train automatic stop control using precise location data of balises. IEEE Transactions on Intelligent Transportation Systems. 2013;14(3):1526–1535. DOI: 10.1109/TITS.2013.2265171.

Ahmadi S, Dastfan A, Assili M. Improving energy‐efficient train operation in urban railways: Employing the variation of regenerative energy recovery rate. IET Intelligent Transport Systems. 2017;11(6):349–357. DOI: 10.1049/iet-its.2016.0256.

Gailienė I. Investigation into the calculation of superelevation defects on conventional rail lines. Transport. 2012;27(3):229–236. DOI: 10.3846/16484142.2012.719198.

Cheng R, et al. Intelligent driving methods based on expert knowledge and online optimization for high-speed trains. Expert Systems with Applications. 2017;87:228–239. DOI: 10.1016/j.eswa.2017.06.006.

Liang Y, et al. A modified genetic algorithm for multi-objective optimization on running curve of automatic train operation system using penalty function method. International Journal of Intelligent Transportation Systems Research. 2019;17:74–87. DOI: 10.1007/s13177-018-0158-6.

Liu K, Wang XC, Qu Z. Research on multi-objective optimization and control algorithms for automatic train operation. Energies. 2019;12(20):3842. DOI: 10.3390/en12203842.

Dimitrova Stoilova S, Valentinov Stoev V. Methodology of transport scheme selection for metro trains using a combined simulation-optimization model. Promet – Traffic&Transportation. 2017;29(1):23–33. DOI: 10.7307/ptt.v29i1.2139.

Ran XC, et al. Energy‐efficient approach combining train speed profile and timetable optimisations for metro operations. IET Intelligent Transport Systems. 2020;14(14):1967–1977. DOI: 10.1049/iet-its.2020.0346.

Li S, et al. Integrated train dwell time regulation and train speed profile generation for automatic train operations on high-density metro lines: A distributed optimal control method. Transportation Research Part B: Methodological. 2021;148:82–105. DOI: 10.1016/j.trb.2021.04.009.

Huang Z, et al. Distributed cooperative cruise control for high‐speed trains with multi‐objective optimization. IET Control Theory & Applications. 2022;16(16):1645–1656. DOI: 10.1049/cth2.12331.

Scheiermann J, Konen W. AlphaZero-inspired game learning: Faster training by using MCTS only at test time. IEEE Transactions on Games. 2022. DOI: 10.1109/TG.2022.3206733.

Ouellet V, et al. Improve performance and robustness of knowledge-based FUZZY LOGIC habitat models. Environmental Modelling & Software. 2021;144:105138. DOI: 10.1016/j.envsoft.2021.105138.

Yin J, et al. Research and development of automatic train operation for railway transportation systems: A survey. Transportation Research Part C: Emerging Technologies. 2017;85:548–572. DOI: 10.1016/j.trc.2017.09.009.

Wu J, Huang Y. Simulation research on speed curve of high speed train based on object-oriented technology. IEEE Asia-Pacific Conference on Image Processing, Electronics and Computers (IPEC), Apr-2021. IEEE; 2021. p . 328–332. DOI: 10.1109/IPEC51340.2021.9421249.

Muniandi G. Blockchain‐enabled virtual coupling of automatic train operation fitted mainline trains for railway traffic conflict control. IET Intelligent Transport Systems. 2020;14(6):611–619. DOI: 10.1049/iet-its.2019.0694.

Pan D, Chen Z, Mei M. Energy efficiency emergence of high-speed train operation and systematic solutions for energy efficiency improvement. SN Applied Sciences. 2020;2:1–13. DOI: 10.1007/s42452-020-2692-5.

Zhao N, et al. Field test of train trajectory optimisation on a metro line. IET Intelligent Transport Systems. 2017;11(5):273–281. DOI: 10.1049/iet-its.2016.0214.

Nassiri P, et al. Train passengers comfort with regard to whole-body vibration. Journal of Low Frequency Noise, Vibration and Active Control. 2011;30(2):125–136. DOI: 10.1260/0263-0923.30.2.125.

Zheng X, et al. Method on generating massive virtual driving curves for high-speed trains of the Cross-Taiwan Strait Railway and its statistical analysis. The Journal of Supercomputing. 2023;1–24. DOI: 10.1007/s11227-023-05621-5.

Lu Y, Dewang C, Zhaolin Z. Algorithm for automatically generating a large number of speed curves of subway trains based on alphazero. Chinese Journal of Intelligent Science and Technology. 2021;3(2):179–184. DOI: 10.11959/j.issn.2096-6652.202118.




How to Cite

CHEN, D., WU, Z., LU, Y., ZHAO, W., & LIN, Z. (2024). Rapid Algorithm for Generating and Selecting Optimal Metro Train Speed Curves Based on Alpha Zero and Expert Experience. Promet - Traffic&Transportation, 36(3), 560–579. https://doi.org/10.7307/ptt.v36i3.304