A Multi-Source Dynamic Temporal Point Process Model for Train Delay Prediction

Train delay prediction is a key technology for intelligent train scheduling and passenger services. The authors propose a train delay prediction model that takes into account the asynchrony of train events, the dynamics of train operations, and the diversity of influencing factors. Firstly, the authors consider train operations as discrete sequences of train events and propose a train arrival neural temporal point process (TANTPP) framework focused on predicting train delays that explicitly models the asynchrony of train events. Secondly, the authors introduce a multi-source dynamic spatiotemporal embedding method for the feature encoder in the TANTPP framework, which enhances the capability to capture the features of train operation networks. Thirdly, to better capture the distribution of train events in the TANTPP framework, the authors utilize a log-normal mixture hybrid method to learn the probability density distribution of train arrival events. Finally, the experimental result on real-world datasets demonstrates that the TANTPP model outperforms current state-of-the-art models, reducing the mean absolute error (MAE) by 10.85%, the RMSE by 9.8%, the RRSE by 3.78% and the mean absolute percentage error (MAPE) by 10.11% on average. To the best of the authors' knowledge, this is the first study to utilize neural temporal point processes to enhance train delay prediction.

• Record URL:
  • https://doi.org/10.1109/TITS.2024.3430031
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/oclc/41297384
• Supplemental Notes:
  • Copyright © 2024, IEEE.
• Authors:
  • Zhang, Dalin
  • Du, Chenyue
  • Peng, Yunjuan
  • Liu, Jiqiang
  • Mohammed, Sabah
  • Calvi, Alessandro
• Publication Date: 2024-11

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; Maps; References; Tables;
• Pagination: pp 17865-17877
• Serial:
  • IEEE Transactions on Intelligent Transportation Systems
  • Volume: 25
  • Issue Number: 11
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
  • ISSN: 1524-9050
  • Serial URL: http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=6979

Subject/Index Terms

• TRT Terms: Arrivals and departures; Neural networks; Predictive models; Railroad traffic; Stopped time delays; Train operations
• Subject Areas: Operations and Traffic Management; Planning and Forecasting; Railroads;

Filing Info

• Accession Number: 01948369
• Record Type: Publication
• Files: TRIS
• Created Date: Mar 11 2025 5:10PM