面向榆林中小型煤矿的煤炭销售数据分析及可视化方法  

Visualization of Small and Medium Coal Mine Sales Data

作　　者：冯治东[1] 刘格 何郁娜 郭红波[1] FENG Zhi-dong;LIU Ge;HE Yu-na;GUO Hong-bo(School of Information Engineering, Yulin University, Yulin 719000,China)

机构地区：[1]榆林学院信息工程学院

出　　处：《榆林学院学报》2020年第2期75-79,共5页Journal of Yulin University

基　　金：国家自然科学基金资助项目(51864046);陕西省教育厅专项计划项目(19Jk0998)

摘　　要：榆林中小型煤矿企业销售过程存在“粗放式管理”、“定价随意”、“数据不透明”等问题,且随着企业内部销售数据不断增加,以人力进行煤炭销售数据分析及可视化已经给企业造成了巨大的资源浪费。本研究通过实地调研榆林周边中小型煤矿企业销售模式,对煤种、客户和业务员三要素进行深入挖掘,针对购买吨数与结余数,以结余和吨数为特征属性,利用Python pandas对煤炭销售数据进行数据处理,采用K-Means聚类算法实现了客户的自动分类,借助matplotlib和tkinter对最终结果集进行可视化展示,在此基础上,开发了一款面向榆林中小型煤矿的煤炭销售数据分析及可视化组件。实例研究结果表明,该组件能够从不同角度对销售数据进行分析,降低煤矿企业的数据分析成本,为企业决策提供辅助支持。The sales process of Yulin's small and medium-sized coal mining enterprises has problems such as“extensive management”,“free pricing”,“data opacity”,etc.,and as the company's internal sales data continues to increase,manual analysis and visualization of coal sales data has caused enterprises a huge waste of resources.In this study,we investigated the sales model of small and medium-sized coal mines around Yulin,conducted in-depth mining of the three factors of coal type,customer,and salesperson.For the tonnage and balance of purchases,the balance and tonnage were the characteristic attributes.Sales data was processed for data,and K-Means clustering algorithm was used to realize automatic classification of customers.The final result set was visualized with matplotlib and tkinter,and a coal sales data analysis and visualization for Yulin small and medium coal mines was developed Components.The case study results show that this component can analyze sales data from different angles,reduce the data analysis cost of coal mining enterprises,and provide auxiliary support for enterprise decision-making.

关 键 词：榆林中小型煤矿 煤炭销售 数据分析 可视化 

分 类 号：C935[经济管理—管理学]