Short-Term Passenger Flow Prediction of High-Speed Railway Based on Copula Theory
Yuedi Yang, Beijing Jiaotong UniversityShow Abstract
Jun Liu, Beijing Jiaotong University
Minshu Ma, Beijing Jiaotong University
Xudong Jia, California State Polytechnic University, Pomona
Xuchao Chen, Beijing Jiaotong University
ABSTRACT: Short-term passenger flow prediction is an important task in the operation of high speed railway. In this paper, two improved models are developed based on the Copula theory: the first one (the passenger flow prediction model) uses the multivariate Copula function. It considers the impacts of multiple factors in passenger flow prediction. The second one is the passenger flow combined prediction model which improves the multivariate Copula function by considering the correlation between high-speed railway passenger flow and airfare. The paper then presents a case study to compare the original Copula model and the two improved models in forecasting high-speed rail passenger flow. It is found that the model supported by the multivariate Clayton Copula function and the combined prediction model supported by the multivariate Copula function have the best predication accuracy. The average relative error rates for these two models are about 6%.
A Dynamic Pricing Model for the Solution of the Railway Revenue Management Problem: A Case Study of the Wuhan-Guangzhou HSR
Xiushan Jiang, Beijing Jiaotong UniversityShow Abstract
Wenyang Tan, Central South University
High-speed railway (HSR) is currently regarded as one of the most significant mode for inter-city passenger transport. But its operating revenue in China is not satisfied because of the single, inflexible pricing mechanism. This paper proposes a dynamic pricing model of HSR based on revenue management. The logit model is used to describe the customer’s discrete choice, and an optimal pricing policy is given so as to maximize the total revenue. The proposed approach is applied to Wuhan-Guangzhou HSR in China for demonstration. Several numerical results prove that pricing policy increase the total revenue at 25.74%.
Modeling the Influence of Disturbances in a High-Speed Rail System
Ping Huang, Southwest Jiaotong UniversityShow Abstract
Ping Huang, Southwest Jiaotong University
Qiyuan Peng, Southwest Jiaotong University
Liping Fu, University of Waterloo
Chaozhe Jiang, Southwest Jiaotong University
Yuxiang Yang, Southwest Jiaotong University
Accurately predicting the influence of disturbances in High-Speed Railways (HSR) has great significance for improving real-time train dispatching and operation management. In this paper, we established distribution models to estimate the inferences of high-speed train disturbances (HSTD), including the number of affected trains (NAT) and total delayed time (TDT). We extracted data about the disturbances and their affected train groups from historical train operation records of Wuhan-Guangzhou (W-G) HSR in China and used a K-Means clustering algorithm to classify them into four categories. Five common distribution models were applied to fit the distributions of NAT and TDT of each clustered category and the models with best goodness-of-fit were determined according to the Kolmogorov-Smirnov (K-S) test. The validation results show that the models accurately revealed the characteristics of HSTD and that these models can be used in real-time dispatch to predict the NAT and TDT once the type of delays and its basic characters are known.
Multi-Stage Pricing and Seat Allocation Model for High-Speed Rail with Multi-Train Services
Xinlei Hu, Central South UniversityShow Abstract
Feng Shi, Central South University
Guangming Xu, Central South University
Runfa Wu, Central South University
Tangjian Wei, Central South University
This paper deals with the multi-stage determination of price and seat allocation within the booking horizon for the high-speed rail (HSR) revenue management (RM) problem with multi-train services. We use an elastic demand functions to describe the relationship between passenger demand and weighted average price for each O-D in each decision period, and then we propose a multi-stage pricing and seat allocation model with a nonlinear mathematical programming formulation. Since the objective function is nonlinear and non-differentiable, we convert it into the differentiable formulation first. Then, we transform the model into an unconstrained programming by adding nonlinear constraints into the objective with a penalty function. Based on the Davidon-Fletcher-Powell (DFP) algorithm framework, we design a solution approach. Compared with Matlab optimization toolbox for nonlinear programming, our solution approach can obtain similar solutions with far less CPU time cost under different scenarios. Numerical experiments show that the proposed model and algorithm can be used to solve the RM problem for all trains within a high-speed rail line in the real world.