Prediction of Freight Volume Based on Grey Correlation and Improved Grey Neural Network
Abstract
Freight volume prediction not only plays an important role in the rational allocation of logistics resources, but also has an important impact on the formulation of related policies. This paper first determines the key factors that affect the freight volume through gray correlation analysis, takes the key factors as the input of the grey neural network model, and improves the weights and thresholds of the gray neural network through genetic algorithms to avoid the model from falling into a local optimum. The prediction results of three different models show that the gray neural network based on genetic algorithm optimization has higher prediction accuracy, which proves that the model is reasonable and reliable and can provide a reference for freight volume prediction. The model can also be applied to prediction in other fields, and it also proves the advantages of the combined model.