Study on forecasting method of highway port cargo volume

According to the problem of predicting highway port freight volume, the BP neural network combination forecasting model is proposed based on the study of commonly used forecasting methods including time series forecast, gray system forecast and combination forecast method. Combined with the freight traffic condition of Suifenhe highway port, the combination forecasting model is verified. The experimental results indicate that this method is very effective to forecast the highway freight.