Railway passenger flow forecasting based on time series analysis with big data

Abstract

With the rapid development of high-speed railway in China, the research of railway passenger flow forecasting has become a key research direction. The forecast of railway passenger flow can help formulate a reasonable price, improve the organization of passenger terminals, optimize the allocation of railway vehicles resources and improve the service capability of passenger transport equipment, which are of great significance to improve the efficiency of railway passenger transport. In this study, a comprehensive forecasting model based on time series analysis is proposed for railway passenger flow forecasting. To solve problems which cannot be handled with traditional programming model in the context of big data, time series analysis is introduced into the solution. Railway Passenger Flow Forecasting model based on Time Series Analysis is established with the combination of the long-term trend factor, the seasonal factor and the weather factor. Railway passenger flow data obtained from the Railway Bureau are used for the case study. The change rule of passenger flow was researched under different conditions, the railway passenger flow in the next two weeks was forecast, and the corresponding optimization of vehicle configuration and station docking scheme are proposed. Sensitivity analysis shows good stability and robustness of the model.