Prediction of high-speed train delay propagation based on causal text information

Abstract

The delay-causing text data contain valuable information such as the specific reasons for the delay, location and time of the disturbance, which can provide an efficient support for the prediction of train delays and improve the guidance of train control efficiency. Based on the train operation data and delay-causing data of the Wuhan–Guangzhou high-speed railway, the relevant algorithms in the natural language processing field are used to process the delay-causing text data. It also integrates the  train operating-environment information and delay-causing text information so as to develop a cause-based train delay propagation prediction model. The Word2vec model is first used to vectorize the delay-causing text description after word segmentation. The mean model or the term frequency-inverse document frequency-weighted model is then used to generate the delay-causing sentence vector based on the original word vector. Afterward, the  train operating-environment features and delay-causing sentence vector are input into the extreme gradient boosting (XGBoost) regression algorithm to develop a delay propagation prediction model. In this work, 4 text feature processing methods and 8 regression algorithms are considered. The results demonstrate that the XGBoost regression algorithm has the highest prediction accuracy using the test features processed by the continuous bag of words and the mean models. Compared with the prediction model that only considers the train-operating-environment features, the results show that the prediction accuracy of the model is significantly improved with multiple regression algorithms after integrating the delay-causing feature.

Keywords

High-speed rail / Delay propagation / Cause of delay / Word2vec / Natural language processing

References

[1.] Wen C Li Z Huang P . Cause-specific investigation of primary delays of wuhan-guangzhou HSR. Trans Lett, 2020 12 7 1-14

[2.] Wen C Li Z Lessan J . Statistical investigation on train primary delay based on real records: evidence from wuhan-guangzhou HSR. Int J Rail Transp, 2017 5 3 170-189

[3.] Ye Y, Zhu B, Huang P et al (2022) OORNet: a deep learning model for on-board condition monitoring and fault diagnosis of out-of-round wheels of high-speed trains. Measurement 199. https://doi.org/10.1016/j.measurement.2022.111268

[4.] Kecman P Goverde RMP. Online data-driven adaptive prediction of train event times. IEEE Trans Intell Transp Syst, 2015 16 1 465-474

[5.] Kecman P, Corman F, Meng L (2015) Train delay evolution as a stochastic process. In: the 6th International Conference on Railway Operations Modelling and Analysis, Tokyo, pp 007–1–19

[6.] Milinković S Marković M Vesković S . A fuzzy petri net model to estimate train delays. Simul Model Pract Theory, 2013 33 144-157

[7.] Carey M Carville S. Testing schedule performance and reliability for train stations. J Op Res Soc, 2000 51 6 666-682

[8.] Huang P, Spanninger T, Corman F (2022). Enhancing the understanding of train delays with delay evolution pattern discovery: a clustering and bayesian network approach. IEEE Transactions on Intelligent Transportation Systems. https://doi.org/10.1109/TITS.2022.3140386

[9.] Nm A Sm B Kst C . Analyzing passenger train arrival delays with support vector regression. Trans Res Part C: Emerging Technol, 2015 56 251-262

[10.] Chao W, Lessan J, Fu L et al (2017) Data-driven models for predicting delay recovery in high-speed rail. In: the 4th International Conference on Transportation Information and Safety (ICTIS), Edmonton

[11.] Huang P Chao W Fu L . A deep learning approach for multi-attribute data: a study of train delay prediction in railway systems. Inf Sci, 2019 516 234-253

[12.] Huang P Wen C Fu L . A hybrid model to improve the train running time prediction ability during high-speed railway disruptions. Saf Sci, 2020 122 104510

[13.] Shi R Xu X Li J . Prediction and analysis of train arrival delay based on XGBoost and Bayesian optimization. Appl Soft Comput, 2021 109 107538

[14.] Huang P Li Z Wen C . Modeling train timetables as images: a cost-sensitive deep learning framework for delay propagation pattern recognition. Expert Syst Appl, 2021 177 114996

[15.] Wang Y Wen C Huang P. Predicting the effectiveness of supplement time on delay recoveries: a support vector regression approach. Int J Rail Transp, 2021 10 3 375-392

[16.] Li Z Huang P Wen C . Prediction of train arrival delays considering route conflicts at multi-line stations. Trans Res Part C: Emerging Technol, 2022 138 103606

[17.] Olsson NOE Haugland H. Influencing factors on train punctuality—results from some norwegian studies. Transp Policy, 2004 11 4 387-397

[18.] Xu P Corman F Peng Q. Analyzing railway disruptions and their impact on delayed traffic in Chinese high-speed railway. IFAC-PapersOnLine, 2016 49 3 84-89

[19.] Li H Parikh D He Q . Improving rail network velocity: a machine learning approach to predictive maintenance. Transp Res Part C, 2014 45 17-26

[20.] Lee WH Yen LH Chou CM. A delay root cause discovery and timetable adjustment model for enhancing the punctuality of railway services. Trans Res Part C Emerging Technol, 2016 73 49-64

[21.] Hassan A Mahmood A. Convolutional recurrent deep learning model for sentence classification. IEEE Access, 2018 6 13949-13957

[22.] Kim Y (2014) Convolutional neural networks for sentence classification. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, pp 1746–1751

[23.] Hou Y Wen C Huang P . Delay recovery model for high-speed trains with compressed train dwell time and running time. Railw Eng Sci, 2020 28 4 424-434

Funding

National Natural Science Foundation of China(71871188); Research and development project of China National Railway Group Co., Ltd(P2020X016); China Railway Chengdu Group Co. Ltd(CX 2120)