Port Container Throughput Forecast Based On Ant Colony Optimization Algorithm

Since the beginning of the 21 st century,economic globalization has deeply affected all aspects of current economic development.Regional economic integration has developed with the trend,and the port as an important economic transportation hub has its central strategic position and day.The economic development of any region and country is based on the port as an important support.Under the initiative of “One Belt,One Road”,the optimization and construction of the port has become an important cornerstone for the creation of the “Maritime Silk Road”.Therefore,in order to enable port construction to accurately serve the needs of the current market,accurately predicting port throughput is of great practical significance for improving port freight efficiency and economic efficiency,and is irreplaceable for planning the development of China's maritime industry significance.The main contents of this paper are as follows:Firstly,the origin and basic ideas of the ant colony algorithm are introduced.The TSP problem is used to analyze the ant colony algorithm in detail.Then the advantages and application status of the ant colony algorithm are briefly described.Thirdly,because SVR has excellent regression ability for small samples,this paper selects SVR as the prediction model.In order to make the prediction accuracy higher,this paper constructs an SVR model based on ant colony optimization algorithm,which is the gamma parameter of SVR using ant colony algorithm.Perform optimization and introduce the other parameters of SVR in detail.Finally,select the container throughput monthly data of Ningbo Zhoushan Port(March 2003 to July 2018)for empirical analysis,and use the data from March 2003 to January 2018 as a training set to predict the next 6 months.Port container throughput.The data is preprocessed first,then the model is modeled in Python language.At the same time,other predictive models are validated.The model is combined with SVR model,ARIMA model,GM(1,1)model,Bayesian.The optimized SVR model is used to compare and analyze the prediction results.The results show that the ant colony optimization based SVR model has better prediction results and provides a theoretical basis for port construction.