The key factors affecting passenger choice behavior are analyzed based on machine learning methods, so as to effectively enhance the matching degree between the plan and demand while optimizing the train operation plan. Firstly, an analytical framework for travel choice behavior of high-speed railway passengers is proposed, which includes data collection, prediction algorithm design, and interpretability analysis, exploring the key features influencing high-speed railway passengers’ choice behavior. Secondly, considering constraints such as passenger flow demand, train capacity, variable coupling, departure and arrival stations, trains passing through sections, and number of stops, the optimization model of train operation plan for high-speed railway is constructed with the objectives of minimizing operation costs and the cost of total passenger stops. Then, in order to solve the mixed integer nonlinear programming model, an approximate linearization method for the nonlinear model is proposed, which is based on constrained deformation and the Big-M method. Finally, taking the Wuhan-Guangzhou south section of the Wuhan-Guangzhou High-Speed Railway as the research background, the effectiveness of the proposed model and solution method is verified. The results show that the number of stops is the most critical feature affecting travel choice behavior of passengers. After considering the minimum cost of total passenger stops in the optimization model goal, the cost reduction of total passenger stops in Wuhan-Guangzhou south direction of Wuhan-Guangzhou High-Speed Railway reaches about 91.11%. Moreover, the solution efficiency is significantly improved after introducing the Big-M method.
ZHANGChuntian, QIJianguo, YANGLixing, et al. Robust Train Operation Plan Based on Uncertain Passenger Demands for High-Speed Railway Corridors [J]. Journal of Transportation Systems Engineering and Information Technology, 2022, 22 (1): 115-123. in Chinese
NIUFeng, QIJianguo, QINJin. Optimization Model for Train Stopping Plan on High-Speed Railway Corridor with Uncertain Passenger Demands [J]. Journal of the China Railway Society, 2016, 38 (7): 1-7. in Chinese
PUSong, HongxiaLÜ, CHENDingjun, et al. High Speed Railway Passenger Train Line Planning Optimization Based on Improved Column Generation Algorithm [J]. Journal of the China Railway Society, 2015, 37 (9): 1-7. in Chinese
FUHuiling, NIELei, YANGHao, et al. Research on the Method for Optimization of Candiate-Train-Set Based Train Operation Plans for High-Speed Railways [J]. Journal of the China Railway Society, 2010, 32 (6): 1-8. in Chinese
QINJin, TANYuchao, ZHANGWei, et al. Train Planning Optimization for Intercity Railway Based on Space-Time Network [J]. Journal of the China Railway Society, 2020, 42 (2): 1-10. in Chinese
SHIFeng, DENGLianbo, HUOLiang. Bi-Level Programming Model and Algorithm of Passenger Train Operation Plan [J]. China Railway Science, 2007, 28 (3): 110-116. in Chinese
SHIFeng, LIYanlin, HUXinlei, et al. Service Level Oriented Optimization of Train Operation Plan for High Speed Railway [J]. China Railway Science, 2018, 39 (5): 127-136. in Chinese
SUNGuofeng, JINGYun, MAYawen. Optimization of Dynamic Train Line Planning Considering Multi-Dimensional Travel Demand of Passengers [J]. Journal of the China Railway Society, 2022, 44 (11): 10-18. in Chinese
YEYuling, WANGYishi. Research on Travel Mode Choice Behavior in Shanghai-Hangzhou Transport Corridor [J]. Journal of the China Railway Society, 2010, 32 (4): 13-17. in Chinese
ZHANGJia'nan, ZHAOPeng. Research on Passenger Choice Behavior of Trip Mode in Comprehensive Transportation Corridor [J]. China Railway Science, 2012, 33 (3): 123-131. in Chinese
[23]
JINGY, LIUY K, ZHANGZ H, et al. Passenger Travel Behaviour on Chinese High‐Speed Railways Using Machine Learning Based on Revealed-Preference Data [J]. Expert Systems, 2019, 36 (4): e12422.
WANGWenxian, NIShaoquan, HongxiaLÜ, et al. Boarding Choice for High-Speed Railway Passengers Based on Disaggregate Theory [J]. Journal of Transportation Systems Engineering and Information Technology, 2015, 15 (3): 13-18, 43. in Chinese
ZHAOPeng, ZHAIRuxue, SONGWenbo. Passenger Choice Behavior of High-Speed Railway Considering Individual Heterogeneity [J]. Journal of Beijing Jiaotong University, 2019, 43 (2): 117-123. in Chinese
[28]
LEED, DERRIBLES, PEREIRAF C. Comparison of Four Types of Artificial Neural Network and a Multinomial Logit Model for Travel Mode Choice Modeling [J]. Transportation Research Record: Journal of the Transportation Research Board, 2018, 2672 (49): 101-112.
[29]
ZHUZ J, ZENGJ, GONGX L, et al. Analyzing Influencing Factors of Transfer Passenger Flow of Urban Rail Transit: a New Approach Based on Nested Logit Model Considering Transfer Choices [J]. International Journal of Environmental Research and Public Health, 2021, 18 (16): 8462.
[30]
LIH Y, LIX, XUX Y, et al. Modeling Departure Time Choice of Metro Passengers with a Smart Corrected Mixed Logit Model - a Case Study in Beijing [J]. Transport Policy, 2018, 69: 106-121.
[31]
SALASP, DE LA FUENTER, ASTROZAS, et al. A Systematic Comparative Evaluation of Machine Learning Classifiers and Discrete Choice Models for Travel Mode Choice in the Presence of Response Heterogeneity [J]. Expert Systems with Applications, 2022, 193: 116253.
[32]
CHANGX M, WUJ J, LIUH, et al. Travel Mode Choice: a Data Fusion Model Using Machine Learning Methods and Evidence from Travel Diary Survey Data [J]. Transportmetrica A: Transport Science, 2019, 15 (2): 1587-1612.
[33]
HUS H, XIONGC F, CHENP, et al. Examining Nonlinearity in Population Inflow Estimation Using Big Data: an Empirical Comparison of Explainable Machine Learning Models [J]. Transportation Research Part A: Policy and Practice, 2023, 174: 103743.
SUNGuofeng, JINGYun, LIHebi, et al. An Interpretable Machine Learning Framework-Based Approach for Predicting Passenger Flow Distribution in Train Riding Sections [J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 24 (2): 249-262. in Chinese
[36]
SHIR, XUX Y, LIJ M, et al. Prediction and Analysis of Train Arrival Delay Based on XGBoost and Bayesian Optimization [J]. Applied Soft Computing, 2021, 109: 107538.