融合搜索指数的卷烟需求预测研究  

Research on Cigarette Demand Prediction Based on Search Index

作　　者：黄飞杰[1] 张卫东[2] 侯石鹏 宋红文[3] 陶林 HUANG Feijie;ZHAGN Weidong;HOU Shipeng;SONG Hongwen;TAO Lin(Internet Research Center,Guangxi Zhongyan Industry Co.,Ltd.,Nanning,China,530000;School of Computer Science and Technology,Southwest University of Science and Technology,Mianyang,China,6210103;School of Manufacturing Science and Engineering,Southwest University of Science and Technology,Mianyang,China,621010)

机构地区：[1]广西中烟工业有限责任公司互联网研究中心,广西南宁530000 [2]西南科技大学计算机科学与技术学院,四川绵阳621010 [3]西南科技大学制造科学与工程学院,四川绵阳621010

出　　处：《福建电脑》2022年第10期17-20,共4页Journal of Fujian Computer

基　　金：四川省科技厅项目《物流配送中心成本可视化动态控制系统研究》(No.2017GZ0187)资助。

摘　　要：在总量控制、稍紧平衡的行业背景下,为了解决实际需求与实际投放误差造成的结构性缺货问题,本文研究搜索指数与卷烟历史销量的相关性,筛选与价位段销量相关性大的关键词搜索数据,构建了多元线性回归模型的数据输入。对比发现,加入搜索指数的预测模型,能够有效提升预测精度。通过对卷烟价位段需求展开精准预测,能指导厂商根据预测结果投放符合市场需求趋势的商品,减少结构性缺货的情况,并开展品牌培育营销活动,有效提升商品销量和营销效果。In the context of total control, slightly tight balance industry background, in order to solve the actual demand and the actual error of structural out of stock, this paper studies the search index and cigarette sales history correlation, screening and price segment sales correlation large keyword search data, build the multiple linear regression model of data input. The comparison shows that adding the prediction model of the search index can effectively improve the prediction accuracy. By making accurate prediction of the cigarette price segment demand, it can guide the manufacturers to put the commodities in line with the market demand trend according to the forecast results, reduce the structural shortages, and carry out brand cultivation and marketing activities, which can effectively improve the commodity sales and marketing effect.

关 键 词：预测模型 搜索指数 卷烟 

分 类 号：F272.1[经济管理—企业管理]