Delay recovery model for high-speed trains with compressed train dwell time and running time

Abstract

Modeling the application of train operation adjustment actions to recover from delays is of great importance to supporting the decision-making of dispatchers. In this study, the effects of two train operation adjustment actions on train delay recovery were explored using train operation records from scheduled and actual train timetables. First, the modeling data were sorted to extract the possible influencing factors under two typical train operation adjustment actions, namely the compression of the train dwell time at stations and the compression of the train running time in sections. Stepwise regression methods were then employed to determine the importance of the influencing factors corresponding to the train delay recovery time, namely the delay time, the scheduled supplement time, the running interval, the occurrence time, and the place where the delay occurred, under the two train operation adjustment actions. Finally, the gradient-boosted regression tree (GBRT) algorithm was applied to construct a delay recovery model to predict the delay recovery effects of the train operation adjustment actions. A comparison of the prediction results of the GBRT model with those of a random forest model confirmed the better performance of the GBRT prediction model.

References

[1.] Mattsson L-G. Railway capacity and train delay relationships 2007 Berlin Springer

[2.] Keiji K, Naohiko H, Shigeru M. Simulation analysis of train operation to recover knock-on delay under high-frequency intervals. Case Stud Transp Policy 2015, 3 1 92-98

[3.] Kariyazaki K, Hibino N, Morichi S. Simulation model for estimating train operation to recover knock-on delay earlier. Asian Transp Stud 2013, 2 284-294

[4.] Krasemann JT. Design of an effective algorithm for fast response to the re-scheduling of railway traffic during disturbances. Transp Res Part C: Emerg Technol 2012, 20 1 62-78

[5.] Yuan J, Hansen IA. Optimizing capacity utilization of stations by estimating knock-on train delays. Transp Res Part B Methodol 2007, 41 2 202-217

[6.] Cadarso L, Marín Á, Maróti G. Recovery of disruptions in rapid transit networks. Transp Res Part E: Logist Transp Rev 2013, 53 15-33

[7.] D'Ariano A, Pranzo M, Hansen IA. Conflict resolution and train speed coordination for solving real-time timetable perturbations. IEEE Trans Intell Transp Syst 2007, 8 2 208-222

[8.] Dollevoet T, Huisman D, Kroon L, Schmidt M, Schöbel A. Delay management including capacities of stations. Transp Sci 2014, 49 2 185-203

[9.] Cheng Y. Hybrid simulation for resolving resource conflicts in train traffic rescheduling. Comput Ind 1998, 35 3 233-246

[10.] Roberts C, Easton JM, Kumar AVS, Kohli S. Innovative applications of big data in the railway industry 2017 Pennsylvania IGI Global

[11.] De Fabris S, Longo G, Medeossi G (2010) Automated analysis of train event recorder data to improve micro-simulation models. In: Timetable planning and information quality, pp 125–134 10.2495/978-1-84564-500-7/1.

[12.] Bendfeldt J, Mohr U, Muller L. RailSys: a system to plan future railway needs. WIT Trans Built Environ 2000, 50 249-255

[13.] Yamamura A, Koresawa M, Adachi S, Tomii N. Taking effective delay reduction measures and using delay elements as indices for Tokyo’s metropolitan railways. WIT Trans Built Environ 2014, 135 3-15

[14.] Naohiko H, Osamu N, Shigeru M, Hitoshi I, Norio T. Recovery measure of disruption in train operation in Tokyo Metropolitan Area. Transp Res Procedia 2017, 25 4374-4384

[15.] Liebchen C, Lübbecke M, Möhring R, Stiller S (2009) The concept of recoverable robustness, linear programming recovery, and railway applications. In: Robust and online large-scale optimization. Springer, Berlin, pp 1–27

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

[17.] Kecman P, Goverde RM. Predictive modelling of running and dwell times in railway traffic. Public Transp 2015, 7 3 295-319

[18.] Khadilkar H. Data-enabled stochastic modelling for evaluating schedule robustness of railway networks. Transp Sci 2016, 51 4 1161-1176

[19.] Guo J, Meng L, Kecman P, Corman F (2015) Modeling delay relations based on mining historical train monitoring data: a Chinese railway case. In: Proceedings of the 6th international seminar on railway operations modeling and analysis (RailTokyo 2015), March 23–26, Chiba Institute of Technology, Tokyo, Japan

[20.] Huang P, Wen C, Yang Y, Jiang C, Chen Y, Li J (2017) Delay propagation mechanism of high-speed railway. In: 96th TRB annual meeting. Transportation Research Board, Washington DC

[21.] Wen C, Lessan J, Fu L, Huang P, Jiang C (2017) Data-driven models for predicting delay recovery in high-speed rail. In: 4th international conference on transportation information and safety (ICTIS), IEEE, pp 144–151

[22.] Jiang C, Huang P, Lessan J, Fu L, Wen C. Forecasting primary delay recovery of high-speed railway using multiple linear regression, supporting vector machine, artificial neural network, and random forest regression. Can J Civ Eng 2019, 46 5 353-363

[23.] Yuan J, Hansen IA (2008) Closed form expressions of optimal buffer times between scheduled trains at railway bottlenecks. In: 11th international IEEE conference on intelligent transportation systems, IEEE, pp 675–680

[24.] Safapour E, Kermanshachi S, Alfasi B, Akhavian R. Identification of schedule performance indicators and delay recovery strategies for low-cost housing projects. Sustainability 2019, 11 21 6005

[25.] Alkhonaini M, El-Sayed H (2018) Minimizing delay recovery in migrating data between physical server and cloud computing using reed-solomon code. In: 20th international conference on high performance computing and communications (HPCC), Exeter, United Kingdom, pp 718–724

[26.] Yasufumi O, Yoshiki M, Norio T. Analysis of delay recovery operation of railway drivers using decision trees. IEEJ Trans Ind Appl 2018, 138 11 877-883 in Japanese)

[27.] Palmqvist C-W, Olsson NOE, Winslott-Hiselius L (2017) An empirical study of timetable strategies and their effects on punctuality. In: 7th international conference on railway operations modeling and analysis, Lille, France, 5–7 April, 2017

[28.] Goverde R, Hansen I. TNV-prepare: analysis of Dutch railway operations based on train detection data. Comput Railw 2000, 7 779-788

[29.] Pope P, Webster J. The use of an F-statistic in stepwise regression procedures. Technometrics 1972, 14 2 327-340

[30.] Elith J, Leathwick JR, Hastie T. A working guide to boosted regression trees. J Anim Ecol 2008, 77 4 802-813

[31.] Zhang H, Yang Q, Shao J, Wang G. Dynamic streamflow simulation via online gradient-boosted regression tree. J Hydrol Eng 2019, 24 10 04019041

[32.] Ivatt PD, Evans MJ. Improving the prediction of an atmospheric chemistry transport model using gradient-boosted regression trees. Atmos Chem Phys 2020, 20 13 8063-8082

[33.] Fushiki T. Estimation of prediction error by using K-fold cross-validation. Stat Comput 2011, 21 2 137-146

Funding

Department of Science and Technology of Sichuan Province (CN)(2018JY0567); National Natural Science Foundation of China(71871188)