基于灰色模型与神经网络的铁路货运量预测

Prediction of Railway Freight Volume Based on Grey Model and Neural Network

准确把握铁路货运量的变化规律,对于优化货运组织提高货运效率具有重要意义。对此,提出了一种基于灰色模型与神经网络的铁路货运量预测算法,先用传统灰色预测模型对铁路货运量数据进行初步预测,然后用BP神经网络对初步预测值进行修正。结果表明,该预测算法与实际铁路货运量的相对平均误差可控制在5%以内,预测精度高于单一算法,可应用于货运决策中广泛存在的铁路货运量趋势分析问题。

Accurate grasp of the change law of railway freight volume is of great significance for optimizing freight organization and improving freight transport efficiency.In this regard,a railway freight volume forecasting algorithm based on grey model and neural network is proposed.Firstly,the traditional grey prediction model is used to predict the railway freight volume data,and then the BP neural network is used to revise the preliminary prediction value.The results show that the relative average error between the prediction algorithm and the actual railway freight volume can becontrolled within5%,and the prediction accuracy is higher than the single algorithm.It can be applied to the trend analysis of railway freight volume which is widely existed in freight transportation decision-making.