A Combination Forecasting Model of Urban Ring Road Traffic Flow

A combination forecasting model of urban ring road traffic flow based on neural network, Kalman filter and ARIMA model is proposed in this paper, and the traffic flow data of Beijing third-ring-road (BTRR) are explored to test the validity of the model. The experimental results show that combination forecasting cannot improve the forecasting precision in contrast to which ARIMA model and neural network models are more accurate in the traffic congestion periods. However, combination forecasting model can improve forecasting precision comparing with single forecasting models in traffic non-congestion periods