Research on short-term passenger flow prediction of urban rail transit based on ARIMA algorithm

Abstract

The continuous growth of urban population has led to an increasing demand for urban public transport among citizens. In response, many cities have implemented urban rail transit systems to alleviate internal traffic congestion. Accurate short-term passenger flow prediction is crucial for the efficient operation of intelligent subway systems. Therefore, it is essential to establish or select a suitable model for predicting the short-term passenger flow of subways. This study utilizes ten days of travel data from the Beijing subway between May 1 and May 10, 2019. After conducting preliminary preprocessing on the 10-day subway appearance data, an ARIMA prediction model is established for initial forecasting. The accuracy of passenger volume forecast is evaluated by analyzing the variance between predicted values and actual values at specific time intervals, with results indicating that the ARIMA model demonstrates strong predictive capability.