Research on Cargo Throughput Forecasting at Haikou Port Based on a Combined Model

In order to predict the cargo throughput of Haikou harbor more accurately, a combination model is used to predict the cargo throughput. Firstly, the influencing factors of port cargo throughput are analyzed qualitatively, and the influencing factors of cargo throughput are analyzed by the grey correlation analysis method to get the influencing indexes with high correlation strength, and then the GM(1,1) grey-Markov prediction model is established to predict the influencing indexes in the coming year, and the BP neural network prediction model is established to predict the future cargo throughput of port in the coming year. The port cargo throughput in future years is predicted by establishing a BP neural network prediction model. In order to verify the effectiveness of the model, Haikou Port was selected as a case study for analysis. The results show that the relative error of the predicted value of the combined model in 2024 is 1.02%, while the relative error of the predicted value of the single gray GM(1,1)-Markov model is 4.37%, which is a significant improvement in accuracy, and the model predicts better, and the cargo throughput of Haikou port shows an upward trend in the next five years, which will further promote the development of the economy of Hainan Province.