基于神经网络的青岛港口物流需求量预测研究  

Research on Demand Forecasting of Port Logistics in Qingdao Based on Neural Network

作　　者：王佳颖 顼玉卿 李媛 WANG Jia-ying;XU Yu-qing;LI Yuan(School of Urban Geology and Engineering,Hebei University of Geosciences,Shijiazhuang,Hebei 050031,China)

机构地区：[1]河北地质大学城市地质与工程学院,石家庄050031

出　　处：《青岛职业技术学院学报》2021年第6期70-76,共7页Journal of Qingdao Technical College

基　　金：河北省社会科学基金项目(HB19GL065)。

摘　　要：港口物流需求是制定港口资源分配规划与进出口贸易的重要依据。为了确保港口物流需求的准确性,构建港口物流需求预测模型。对于指标的选取则从社会经济、人文和港口物流发展三个层面进行选择,采用主成分分析方法进行主成分得分预测并科学的选取神经网络的输入层内容,在此基础上,采用三层BP(Back Propagation)神经网络模型对青岛港口物流需求预测进行实证分析。通过对其未来三年预测,结果表明:BP神经网络有较强的的泛化能力,对青岛港口物流需求的预测有较好表现。Port logistics demand is an important basis for the development of port resource allocation planning and import and export trade.In order to ensure the accuracy of port logistics demand,the port logistics demand forecasting model is built.The indicators are selected from three aspects of social economy,culture and port logistics development.The principal component analysis method is used to predict the principal component score and scientifically select the input layer content of the neural network.The three-layer BP(Back Proporation)neural network model is used to make an empirical analysis on the logistics demand prediction of Qingdao port.The results of the prediction in the next three years show that the BP neural network has a strong generalization ability and has a good performance in the logistics demand prediction of Qingdao port.

关 键 词：主成分分析 BP神经网络 港口物流 需求预测 

分 类 号：F552[经济管理—产业经济]