基于灰色马尔科夫模型的深圳港集装箱吞吐量预测

Container Throughput Prediction of Shenzhen Port Based on Grey-Markov Model

为了更准确的预测港口集装箱吞吐量,以深圳港2003-2017年集装箱吞吐量数据为基础,通过数学方法将灰色预测模型和马尔科夫预测模型两种数学模型结合起来,建立灰色马尔科夫模型,然后利用灰色预测模型与灰色马尔科夫模型分别进行深圳港集装箱吞吐量的预测计算。结果表明:灰色马尔科夫预测模型能大幅降低波动性较大的时间序列的预测误差,尤其适用于中长期的预测;同时,传统的灰色预测GM(1,1)模型的预测精度也得到明显提高。根据建立的预测模型,计算出了深圳港2018-2020年的集装箱吞吐量预测值。

In order to predict port container throughput more accurately, On the basis of container throughput of Shenzhen port from 2003 to 2017, combined with the Markov model and the Grey model to established Gray-Markov model. The Grey model and Gray-Marov model are used respectively to forecast the container throughput of Shenzhen port. The results show that the Grey-Markov prediction model can significantly reduce the prediction error of time series with high volatility, especially for medium and long-term prediction. At the same time, the prediction accuracy of the traditional GM(1,1) model for gray prediction has also been significantly improved. And then, based on the established prediction model, Shenzhen port throughput forecast value is calculated for 2018 and 2020.