Travel demand forecast of a new high-speed railway station during COVID 19

ABSTRACT

Passenger flow at a new high-speed railway station presents significant uncertainty during COVID-19, which brings a huge challenge to the daily management and operation of the station. To detect the future development trend of demand and reduce the impact of its fluctuation on the daily operation of the station, three classical forecast methods are applied to predict the passenger flow in and out of the station during workdays in this paper. Furthermore, the performance of these methods is compared by conducting a case study of Huairou South Station. The results show that the ARIMA model (autoregressive integrated moving average model) shows better performance than the neural network model and Bass model (Bass diffusion model). Finally, a revised ARIMA model is introduced to predict the passenger flow of the National Day. © 2021 IEEE.