The 22nd COTA International Conference of Transportation Professionals (CICTP 2022)

In order to improve the level of public transport services, cover the blind spots of public transport stations and improve the market competitiveness of public bicycles, it is therefore necessary to analyse the public bicycle rental and return characteristics and carry out short-time traffic demand forecasting analysis. Based on the public bicycle IC card data, this paper introduces rental and return coefficients to cluster public bicycle stations, analyses the spatial and temporal characteristics of the rental and return demand of various stations, determines the factors affecting public bicycle travel, and establishes an improved ARIMA model that takes into account station location, temperature and other factors to forecast the short-time demand of various public bicycle stations. The model was validated by combining the demand data of public bicycle rental and return in Ningbo, and the ARIMA(9,2,2) model was found to be effective in fitting the demand data of the third category of stations.

Public bicycle;IC Card data;Short-term demand forecast;ARIMA model