基于改进灰色-马尔可夫链方法的公路货运量预测

科学的货运量预测对公路发展战略的制定具有十分重要的意义．针对灰色模型的预测结果精确度受原始数据变化幅度的影响较大，且要求累加生成的数据列具有指数性质的缺点，采用带波动的多项式来替代GM（1，1）模型中的指数形曲线，并通过马尔可夫链对其预测结果进行修正，从而建立改进的灰色-马尔可夫链预测模型，同时利用该改进模型对我国公路货运量进行预测，经分析表明预测结果具有较高的精度，预测方法具有一定的可行性和有效性，预测结果可指导公路建设与管理．

Prediction of Highway Freight .Volumes Based on Improved Gray-Markov Chain

The freight volume prediction is very important to decision-making of highway development strategies. In the paper, aiming at shortcoming of tradition Gray prediction model, the fluctuation polynomial was utilized to displace the exponent curve in the model of GM（1,1）, and the prediction results were adjusted by Markov chain, then the prediction model based on improved Gray-Markov chain was established, and the highway freight volumes were predicted by the model. The analysis results show the prediction results are more accurate than the Gray-Markov chain model, and the method is feasible and effective. The prediction results can direct the construction and management of highway.