基于机器学习算法的卷烟营销智能客户拜访策略研究  

Research on intelligent customer visit strategy for cigarette marketing based on machine learning algorithms

作　　者：翁金香 王浩名 周洋 胡红春[2] WENG Jinxiang;WANG Haoming;ZHOU Yang;HU Hongchun(Xinyang Branch of Henan Tobacco Company,Xinyang 464000,China;Staff Development Institute of China NationTobacco Corporation,Zhengzhou 450008,China)

机构地区：[1]河南省烟草公司信阳市公司,河南信阳464000 [2]中国烟草总公司职工进修学院,河南郑州450008

出　　处：《现代电子技术》2024年第4期153-158,共6页Modern Electronics Technique

摘　　要：为提高卷烟行业客户经理的工作效率和服务质量,提出一种基于机器学习算法的卷烟营销智能客户拜访策略。在构建卷烟客户价值分类模型的基础上,利用K均值聚类机器学习算法对零售客户进行分类,合理设置LKH超启发算法和Dijkstra最短路径算法的参数,对客户经理拜访路径进行最优规划和智能导航。仿真结果表明,基于机器学习算法的卷烟营销智能客户拜访策略显著提高了客户经理人均拜访户数、商户满意度,大大缩减了户均在途时间和商户拜访服务覆盖周期。文中提出的策略有助于推动卷烟营销工作质量变革、效率变革和动力变革。In order to improve the work efficiency and service quality of customer managers in the cigarette industry,an intelligent customer visit strategy for cigarette marketing based on machine learning algorithms is proposed.Based on the construction of a cigarette customer value classification model,the K-means clustering machine learning algorithm is used to classify retail customers.The optimal planning and intelligent navigation of the customer manager's visit path are realized by means of the reasonable setting of the parameters of LKH hyper-heuristic algorithm and the Dijkstra shortest path algorithm.The simulation results show that the cigarette marketing intelligent customer visit strategy based on machine learning algorithms can improve the per capita number of visits by customer managers and merchant satisfaction,greatly reducing the per capita in transit time and merchant visit service coverage cycle.The proposed strategy is helpful in promoting changes in the quality,efficiency,and motivation of cigarette marketing work.

关 键 词：卷烟营销 机器学习 智能客户拜访策略 K均值聚类 超启发算法 最短路径算法 商户满意度 在途时间 

分 类 号：TN919-34[电子电信—通信与信息系统]