Research On Fluctuation Of Holiday Railway Passenger Flow And Application On Forecast Of Passenger Flow

ABSTRACT:Analysis of fluctuation characteristics of holiday passenger flow and railway passengers flow forecasting is an important content of holiday passenger traffic organizing, the main basis of scientifically and reasonably make out passenger transport plan and passenger products and a effective way for passenger traffic enterprise to enhance the passenger market competitiveness.This paper mainly researched on fluctuation characteristics and short-time forecast of holiday passenger flow. The objective is to find the fluctuation characteristics of holiday passenger flow through analysis of passenger flow and then be used in passenger flow forecasting in order to guide the transport organization program especially the train plan of extra passenger train.In this paper, firstly determined research object based on reclassifying the holiday passenger flow, and then studied the fluctuation characteristics of holiday passenger flow between origination-city and destination-city, and fond the morphological change features and the influence cycle of three days and seven days'holiday passenger flow which provided the foundation for holiday fluctuation characteristics researching.The paper proposed the definitions and calculation methods of the fluctuating coefficient of holiday passenger flow based on fluctuation characteristics of everyday passenger flow. According to trend characteristics of holiday passenger flow given a clustering of temporal sequences method based on variation trend. After applying the method to selected samples, fond the main influence factors which can affect the fluctuation characteristics of holiday passenger flow.In order to apply fluctuations of passenger flow to holiday passenger flow forecasting, converted the passenger flow forecast to two questions, the one is making sure the fluctuating coefficient, the other one is daily passenger flow forecasting. In the study of the former question, the factors of holiday passenger flow are applied to evaluate origination-city and destination-city. In connection with the later, LS-SVM is applied to forest, and use genetic algorithm to select parameters of LS-SVM in order to improve the accuracy of forecasting.Finally, the proposed method is applied to ticket data of Guangzhou to Zhuzhou in order to make the empirical analysis. The results validated the proposed methods are effective and feasible.