Abstract
Predicting the delay time of trains is an important task in intelligent transport systems, as an accurate prediction can provide a reliable reference for passengers and dispatchers of the railway system. However, due to the complexity of the railway system, interactions of various spatio-temporal variables make it difficult to find the rules of delay propagation. We introduce a Sequential Precoding Spatial-Temporal Network (SPSTN) model to predict the delay of trains. SPSTN consists of a Transformer encoder that captures long-term dependencies in time series, and spatio-temporal graph convolution blocks that model delay propagation at both temporal and spatial levels. Experiments on a subset of the British railway network show that SPSTN performs favorably against the state-of-the-art, which verifies that the combination of sequential precoding and spatio-temporal convolution can effectively model delay propagation on railway networks.
J. Fu and L. Zhong—These authors contributed equally to this work and should be considered co-first authors.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Barbour, W., Samal, C., Kuppa, S., Dubey, A., Work, D.B.: On the data-driven prediction of arrival times for freight trains on U.S. railroads. In: 21st International Conference on Intelligent Transportation Systems, ITSC 2018, pp. 2289–2296 (2018)
Bruna, J., Zaremba, W., Szlam, A., LeCun, Y.: Spectral networks and locally connected networks on graphs. In: 2nd International Conference on Learning Representations, ICLR 2014 (2014)
Heglund, J.S.W., Taleongpong, P., Hu, S., Tran, H.T.: Railway delay prediction with spatial-temporal graph convolutional networks. In: 23rd IEEE International Conference on Intelligent Transportation Systems, ITSC 2020, pp. 1–6 (2020)
Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. In: 5th International Conference on Learning Representations, ICLR 2017 (2017)
Li, J., Xu, X., Shi, R., Ding, X.: Train arrival delay prediction based on spatial-temporal graph convolutional network to sequence model. In: 24th IEEE International Intelligent Transportation Systems Conference, ITSC 2021, pp. 2399–2404 (2021)
Li, S., et al.: Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting. In: Advances in Neural Information Processing Systems: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, vol. 32, pp. 5244–5254 (2019)
Murali, P., Dessouky, M., Ordóñez, F., Palmer, K.: A delay estimation technique for single and double-track railroads. Transp. Res. Part E: Logist. Transp. Rev. 46(4), 483–495 (2010)
Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems: Annual Conference on Neural Information Processing Systems 2017, vol. 30, pp. 5998–6008 (2017)
Yu, B., Yin, H., Zhu, Z.: Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting. In: Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, IJCAI 2018, pp. 3634–3640 (2018)
Zhou, H., et al.: Informer: beyond efficient transformer for long sequence time-series forecasting. In: Thirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021, pp. 11106–11115 (2021)
Acknowledgements
This work is supported by The Center of National Railway Intelligent Transportation System Engineering and Technology (Contract No. RITS2021KF08), China Academy of Railway Sciences (Contract No. 2021YJ195).
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Fu, J., Zhong, L., Li, C., Li, H., Kong, C., Shao, J. (2023). SPSTN: Sequential Precoding Spatial-Temporal Networks for Railway Delay Prediction. In: Li, B., Yue, L., Tao, C., Han, X., Calvanese, D., Amagasa, T. (eds) Web and Big Data. APWeb-WAIM 2022. Lecture Notes in Computer Science, vol 13421. Springer, Cham. https://doi.org/10.1007/978-3-031-25158-0_37
Download citation
DOI: https://doi.org/10.1007/978-3-031-25158-0_37
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-25157-3
Online ISBN: 978-3-031-25158-0
eBook Packages: Computer ScienceComputer Science (R0)