基于BP神经网络的港口物流需求预测研究  

The Research of Port Logistics Demand Prediction Based on BP Neural Network

作　　者：陈雄寅 CHEN Xiong-yin(Liming Vocational University,Quanzhou 362000;Zhejiang Normal University,Jinhua 321004,China)

机构地区：[1]黎明职业大学,福建泉州362000 [2]浙江师范大学,浙江金华321004

出　　处：《物流工程与管理》2022年第12期11-14,20,共5页Logistics Engineering and Management

基　　金：2022年福建省职业教育研究课题项目(GB2022028);2022年全国高校、职业院校物流教改教研课题项目(JZW2022270)。

摘　　要：为构建港口物流需求预测指标体系,将经济发展规模、产业结构、进出口贸易情况、社会消费状况等腹地经济指标作为输入指标,将港口货物吞吐量和集装箱吞吐量等港口物流需求指标作为输出指标,应用Pearson相关分析法对输入和输出指标的相关性进行分析,得出所有指标的相关性均大于0.83,P值均小于0.01,表明上述指标之间存在强相关性和高显著性。在此基础上,建立由上述指标组成的物流需求BP神经网络预测模型。最后以泉州港为例,采用BP神经网络模型预测泉州港的港口物流需求,并与多元线性回归预测模型预测的结果进行比较,结果表明BP神经网络预测模型的精准性和稳定性更高。In order to build the prediction index system for port logistics demand,the hinterland economic indicators such as economic development scale,industrial structure,import and export trade and social consumption are taken as input indicators,and the port logistics demand indicators such as port cargo throughput and container throughput are taken as output indicators.The correlation between input and output indicators is analyzed by Pearson correlation analysis method,and it is concluded that the correlation of all indicators is greater than 0.83,and P values are less than 0.01,indicating that there is a strong correlation and high significance between the above indicators.On this basis,BP neural network prediction model of logistics demand composed of the above indicators is established.Finally,taking Quanzhou port as an example,BP neural network model is used to predict the port logistics demand of Quanzhou port,which is compared with the results predicted by multiple linear regression prediction model,and the results show that BP neural network prediction model are more accurate and stable.

关 键 词：港口物流 物流需求预测 预测指标体系 泉州港 

分 类 号：F550[经济管理—产业经济]