Analysis and forecast of passenger flow based on public transportation IC card and GPS data

Bus passenger flow data are the basis of traffic planning department and public transportation operation department for management, operation and dispatching. It has an important significance for traffic flow accurately and predicting traffic flow. But the traditional methods obtaining passenger flow only can get little information by field investigation which are not comprehensive. This paper utilizes IC card data and GPS data that passengers get on and get off to estimate the station. Based on the estimation, we have proposed a method to forecast passenger flow. The results show that the method of estimating the passenger flow based on the data of the public transport IC card and the GPS data has more than eighty percent accuracy. The forecasting method of autoregression has the highest accuracy rate compared with Holt-Winters and decision tree method.