基于时间序列分析的港口集装箱吞吐量预测分析

Prediction of the Container Throughput in a Port Based on Time Sequence Analysis

集装箱运输是现今国际运输的主要方式之一,长期监测港口集装箱吞吐量并对未来趋势进行相关的预测研究,对于港口基础设施建设、区域经济发展以及国家战略发展方面都有着非常深远的意义。本文针对影响港口集装箱吞吐量复杂性因素的非线性特点,以中国某港口2001年1月至2014年12月共14 a的集装箱吞吐量数据为例,利用统计分析软件R对该港口集装箱吞吐量数据进行时间序列分析,提出了一个合理的ARIMA模型来描述集装箱吞吐量的变化,并对模型进行检验优化,应用该模型预测十个月后的集装箱吞吐量数据。结果表明,利用统计分析软件R对港口集装箱吞吐量数据进行时间序列分析后得到的预测模型可以很好的描述港口集装箱吞吐量变化情况,据此可以为港口在不同时期制定最优措施提供参考。

Container transportation is one of the main modes of international transportation. Long - term monitoring port container throughput and predicting future trends are of importance to the port infrastructure construction, regional economic development and national strategy development. In this paper, aiming at the nonlinear characteristics of the factors affecting the complexity of port con-tainer throughput, the container throughput data in a port in China between January 2001 and December 2014 were investigated and statistical analysis software R was used to conduct time series analysis on the data. A reasonable model was put forward to describe the change of container throughput data, which was tested and optimized. The container throughput data after ten months was forecasted by using the model. Results showed that the prediction model after time series analysis can well describe the change of the port con-tainer throughput, therefore can provide reference for the optimal measures in different periods.