Modeling the Duration of the Impact of Unplanned Disruptions on Passenger Trips Using Smartcard Data in Urban Rail Systems

Abstract

Many urban rail systems operate near capacity given the rapid increase in passenger demand, and unplanned disruptions are unavoidable. From a passenger perspective, the duration of trip delays is a major concern, and passenger trip delays may be longer than the train delays. Several studies have focused on predicting train delays, but the research on the duration of the disruption impacts on passenger trips is limited given that the duration is not observed directly. This paper proposes a probabilistic method to estimate the disruption impact duration using smartcard data, explores statistical and machine learning models to predict the duration of impacts on passengers, and identifies influencing factors including incident characteristics, operating conditions, infrastructure, external factors, and demand. The results highlight that prediction accuracies are acceptable for multiple linear regression, accelerated failure time, and random forest models. Disruptions caused by power failures have longer impact durations than other causes, followed by platform screen doors. The fixed block signaling system leads to a larger disruption duration than the moving block system. The study provides, for the first time, a data-driven approach to understanding the duration of the impact of disruptions on passenger trips using smartcard data which can facilitate timely and informed decision-making under unplanned disruptions.

Keywords

Unplanned disruptions / Impact duration on passenger trips / Smartcard data / Urban rail systems

References

[1] Currie G, Muir C. Understanding passenger perceptions and behaviors during unplanned rail disruptions. Transp Res Procedia, 2017, 25: 4392-4402

[2] Liu TY, Ma ZL, Koutsopoulos HN. Unplanned disruption analysis in urban railway systems using smart card data. Urban Rail Transit. Urban Rail Transit, 2021, 7: 177-190

[3] Valenti G, Lelli M, Cucina D. A comparative study of models for the incident duration prediction. Eur Transp Res Rev, 2010, 2: 103-111

[4] Ozbay K, Noyan N. Estimation of incident clearance times using Bayesian Networks approach. Accid Anal Prev, 2006, 38(3): 542-555

[5] Nam D, Mannering F. An exploratory hazard-based analysis of highway incident duration. Transp Res A Policy Pract, 2000, 34(2): 85-102

[6] Wei C-H, Lee Y. Sequential forecast of incident duration using artificial neural network models. Accident Analysis Prevention, 2007, 39(5): 944-954

[7] Weng JX, Zheng Y, Yan XD, Meng Q. Development of a subway operation incident delay model using accelerated failure approaches. Accident Anal Prevent, 2014, 73: 12-19

[8] Weng JX, Zheng Y, Qu XB, Yan XD. Development of a maximum likelihood regression tree-based model for predicting subway incident delay. Transp Res C, 2015, 57: 30-41

[9] Wang HY, Li LY, Pan PJ, Wang YK, Jin YH. Online detection of abnormal passenger out-flow in urban metro system. Neurocomputing, 2019, 359: 327-340

[10] Lapamonpinyo S, Derrible S, Corman F. Real-time passenger train delay prediction using machine learning: a case study with Amtrak passenger train routes. IEEE Open J Intell Transp Syst, 2022, 3: 539-550

[11] Shi ZB, Zhang N, Zhang YL. Hazard-based model for estimation of congestion duration in urban rail transit considering loss minimization. Transp Res Record J Transp Res Board, 2016, 2595: 78-87

[12] Gu JJ, Jiang ZB, Fan W, Wu JM, Chen JJ (2020) Real-time passenger flow anomaly detection considering typical time series clustered characteristics at metro stations. J Transp Eng A Syst 146(4)

[13] Chen XX, Yang CH, Xu XD, Gong YB (2016) Anomaly detection in metro passenger flow based on random matrix theory. In: IEEE intelligent transportation systems conference (ITSC), Auckland, NZ, pp 525–630

[14] Wang XH, Zhang Y, Liu H, Wang Y, Wang L-CH, Yin BC. An improved robust principal component analysis model for anomalies detection of subway passenger flow. Journal of Adv Transp, 2018, 2018: 12

[15] Malandri C, Fonzone A, Cats O. Recovery time and propagation effects of passenger transport disruptions. Physica A, 2018, 505: 7-17

[16] Yap M, Cats O. Predicting disruptions and their passenger delay impacts for public transport stops. Transportation, 2021, 48: 1703-1731

[17] Webb A, Kumar P, Khani A. Estimation of passenger waiting time using automatically collected transit data. Public Transport, 2020, 12: 299-311

[18] Camponogara E, Nazari LF (2015) Models and algorithms for optimal piecewise-linear function approximation. Math Probl Eng 2015

[19] Zhou XS, Cheng QX, Wu X, Li PH, Belezamo B, Lu JW A meso-to-macro cross-resolution performance approach for connecting polynomial arrival queue model to volume-delay function with inflow demand-to-capacity ratio. Multimodal Transp, 2022

[20] Chow WK, Qu L, Pang EC. Incidents on fire and ventilation provision in subway systems in Hong Kong. Int J Eng Perform Based Fire Codes, 2004, 10(3): 41-47.

[21] Hamad K, Ai-Ruzouq R, Zeiada Dabous SA, Khalil MA. Predicting incident duration using random forests. Transportmetrica A, 2020, 16(3): 1269-1293

[22] “Historical Weather”. Forecasts|World Weather. https://www.worldweatheronline.com/hong-kong-weather.aspx, accessed on 18 Jun 2021.

[23] Monjo R. Measure of rainfall time structure using the dimensionless n-index. Climate Res, 2016, 67(1): 71-86

[24] Stehman SV. Selecting and interpreting measures of thematic classification accuracy. Remote Sens Environ, 1997, 62(1): 77-89

[25] Wang P, Zhang QP. Train delay analysis and prediction based on bid data fusion. Transp Saf Environ, 2019, 1(1): 79-88