Since the railway freight volume has a great impact on the development of national economy, forecasting railway freight volume has become an important step in the overall railway construction planning. To establish the combination model of grey correlation analysis and BP neural network. At the same time, through the gray correlation degree, to take the total population, per capita disposable income of urban residents, railway mileage, increase of primary industry and secondary industry as evaluation indicators. Based on the above indicators, establish a prediction model of railway freight volume based on BP neural network, and then test the model. The result shows that the average relative error is 2.06%. The model,has high prediction speed and accuracy, is an effective forecasting method of railway freight volume and can assist the overall planning of railway.

Full text: IJEIR_2436_FINAL.pdf