基于聚类的港口吞吐量预测方法及其适用性分析

在统计分析历史数据的基础上,选取港口吞吐量、GDP值等指标,采用SPSS统计分析软件中的层次聚类分析法,将我国具有代表性的港口按照吞吐量增长规律分成平稳增长型、加速增长型和波动增长型3类。然后选择时间序列法、回归分析法、灰色模型理论和神经网络模型法,对不同类型的港口吞吐量预测的适用性进行了理论分析。最后以上海港和镇江港为实例进行计算,并对不同预测方法的适用性进行了验证。

Study on port throughput forecasting method and its application based on cluster analysis

Based on the statistical analysis of a large quantity of historical data,choosing port throughput,GDP and other factors as indexes,several representative ports were divided into three types：the normally increasing port,the accelerative increasing port and the fluctuating increasing port by increment law of port throughput according to cluster analysis in SPSS.Then the applicability of the time series models,regression model,grey models,and neural network model were analyzed.Finally,taking Shanghai port and Zhenjiang port as examples,the applicability of the forecasting models for different types of ports were verified in this paper.