基于组合模型的浙江省物流需求预测  

Logistics Demand Forecast in Zhejiang Province Based on Combined Model

作　　者：姬栋媛 陆芬 JI Dongyuan;LU Fen(School of Management,Wuhan University of Science and Technology,Wuhan 430081,China;Center for Service Science and Engineering,Wuhan University of Science and Technology,Wuhan 430081,China)

机构地区：[1]武汉科技大学管理学院,湖北武汉430081 [2]武汉科技大学服务科学与工程研究中心,湖北武汉430081

出　　处：《物流科技》2024年第23期22-25,共4页Logistics Sci Tech

基　　金：湖北省教育厅科学技术研究项目“考虑随机学习效应的双渠道供应链战略库存决策研究”(B2020005)。

摘　　要：在物流业的发展中,物流需求的预测和分析是进行物流资源分配和规划的重要依据,做好物流需求的预测才能更好地使物流服务的需求和供给达到一个相对平衡的状态,从而进一步加快物流业的发展。组合预测相对于单一预测模型会更精准,因此文章分别使用灰色GM1,1模型、线性回归模型以及ARIMA模型进行预测,并使用方差倒数给三种预测模型赋予权重,最终建立组合预测模型。最后,文章还依据浙江省过去的物流需求趋势给出了一些建议,旨在进一步加强浙江省的物流发展。In the development of the logistics industry,the prediction and analysis of logistics demand is an important basis for the allocation and planning of logistics resources,and the prediction of logistics demand can better make the demand and supply of logistics services reach a relatively balanced state,thereby further accelerating the development of the logistics industry.Combinatorial prediction is more accurate than a single prediction model,so this paper uses the grey GM1,1 model,linear regression model and ARIMA model to make predictions,and uses the reciprocal variance to assign weights to the three prediction models,and finally establishes the combined prediction model.Finally,this paper also gives some suggestions based on the past logistics demand trends in Zhejiang Province,aiming to further strengthen the logistics development of Zhejiang Province.

关 键 词：物流需求预测 灰色预测 线性回归 时间序列 组合预测 

分 类 号：F272[经济管理—企业管理]