The establishment of freight volume forecasting model is very important to the development of logistics industry. The traditional freight volume forecasting model can not deal with nonlinear problems effectively and can not accurately predict the freight volume. In view of the above problems, GDP, the lump sum of retail of consumer goods and employment are taken as input variables, and railway freight volume is taken as output variable to forecast the monthly national volume of railway freight. In this paper, GA-BP neural network is established to predict the national railway freight volume based on BP network, which is optimised by genetic algorithm. Based on genetic algorithm and BP network, this model can further optimize the weights and thresholds, which can realize more powerful nonlinear mapping ability and better deal with complex nonlinear problems. The experimental results show that GA-BP neural network model’s prediction results have better stability and accuracy, which testifies the model’s feasibility and provides a reference for freight volume prediction.
Research on prediction model of National Railway Freight Volume based on GA-BP network
01.07.2022
1061642 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Railway Freight Volume Prediction Based on Support Vector Regression (SVR)
Trans Tech Publications | 2014
|Prediction Models for Railway Freight Volume Based on Artificial Neural Networks
Trans Tech Publications | 2014
|Random Forest Regression Model Application for Prediction of China’s Railway Freight Volume
Springer Verlag | 2021
|