基于灰色关联分析与遗传神经网络的襄阳市物流货运量预测

Prediction of Logistics Freight Transportation Volume in Xiangyang Based on Gray Relative Analysis and Genetic Neural Network

首先运用灰色关联分析法对襄阳市货运量及其影响因素进行分析,得出影响襄阳市货运量变化的主要因素有地区生产总值(GDP)、第二产业产值、第三产业产值、全社会交通运输行业固定资产投资、进出口总值、社会消费品零售总额和人均GDP,然后将这些因素作为遗传神经网络的输入神经元,建立物流货运量预测的遗传神经网络模型,并对其进行训练;最后用训练好的模型预测2017-2019年襄阳市物流货运量。结果显示,基于灰色关联分析的遗传神经网络模型预测的平均相对误差为0.2153%,预测精度较高,在物流货运量预测方面具有一定的参考价值。

Firstly,the paper used the gray relational analysis method to analyze freight transportation volume in Xiangyang and its influencing factors.According to the analysis result,we isolated the main influencing factors as regional GDP,production value of secondary industries and tertiary industries,fixed investments of the whole society in transportation industry,gross value of imports and exports,total retail sales of consumer goods and GDP per capita.Then,taking these factors as the input neurons of the genetic neural network,the paper established and trained the genetic neural network model for logistics freight transportation volume prediction.Finally,the trained model was used to predict the logistics freight transportation volume of Xiangyang city from 2017 to 2019,with result showing that the genetic neural network model based on gray relative analysis has high prediction accuracy with the average relative error of 0.2153%,which has certain reference value in logistics freight transportation volume prediction.