大数据挖掘用于电力企业评价体系构建
|
摘要
随着大数据时代的来临,电力企业需要探索新的管理模式.从数据出发,对大量历史监测指标数据进行统计分析,从中筛选出影响企业发展的关键指标,提取关键指标的历史特征.采用层次聚类和Kmeans聚类两种聚类方法对关键指标进行聚类分析,并且结合管理经验,与专家讨论对指标类簇进行合理的命名,最终将关键指标分为了服务电网、运营指数、风险指数三个能够反映公司管理程度的一级指数,并且在一级指数下又分为了若干个子指数,建立了适合新源控股公司管理发展的评价体系.与传统的公司评价体系构建的方法相比,基于大数据挖掘方法构建的评价体系更能够体现出各个指标之间的内在联系,从而能够更好的为决策者提供管理建议.
With the advent of the era of big data,electric power company need to explore new management models. Based on the data,this paper analyzes a large number of historical monitoring data, selects key indexes that affect the development of the enterprise, and extracts the historical characteristics of key indexes.In order to cluster analysis of key indexes, hierarchical clustering and Kmeans clustering are used, and we discuss with experts the reasonable naming of indexes clusters combining with management experience.Finally, the key indexes are divided into three indicators: service index, operation index and risk index, which can reflect the management degree of the company, and under the first-level index, it is divided into several sub-indices. Compared with the traditional method of corporate evaluation system construction, the evaluation system based on big data mining method can better reflect the internal relationship between various indexes, this will provide more management advice to decision makers.
With the advent of the era of big data,electric power company need to explore new management models. Based on the data,this paper analyzes a large number of historical monitoring data, selects key indexes that affect the development of the enterprise, and extracts the historical characteristics of key indexes.In order to cluster analysis of key indexes, hierarchical clustering and Kmeans clustering are used, and we discuss with experts the reasonable naming of indexes clusters combining with management experience.Finally, the key indexes are divided into three indicators: service index, operation index and risk index, which can reflect the management degree of the company, and under the first-level index, it is divided into several sub-indices. Compared with the traditional method of corporate evaluation system construction, the evaluation system based on big data mining method can better reflect the internal relationship between various indexes, this will provide more management advice to decision makers.
引文
[1]王昊.浅谈我国电力企业管理的现状及创新措施[J].科学技术创新,2012(5):219-219.
[2]陈明.电力企业配电网工程项目风险管理研究[D].华北电力大学;华北电力大学(北京),2017.
[3]王伟,王高红,周翔,等.抽水蓄能电站服务电网评价指标体系及方法[J].电网与清洁能源,2010, 26(8):88-92.
[4]赵凯卉.”互联网+”时代企业管理创新的若干研究[J].商场现代化,2017(9):149-150.
[5]陈敏娜,梁海华.K-means聚类算法在企业环境效率评价的应用及R语言实现[J].数学的实践与认识,2018(2):307-315.
[6]张颖.配电网建设项目风险管理研究[J].企业改革与管理,2017(5):21-21.
[2]陈明.电力企业配电网工程项目风险管理研究[D].华北电力大学;华北电力大学(北京),2017.
[3]王伟,王高红,周翔,等.抽水蓄能电站服务电网评价指标体系及方法[J].电网与清洁能源,2010, 26(8):88-92.
[4]赵凯卉.”互联网+”时代企业管理创新的若干研究[J].商场现代化,2017(9):149-150.
[5]陈敏娜,梁海华.K-means聚类算法在企业环境效率评价的应用及R语言实现[J].数学的实践与认识,2018(2):307-315.
[6]张颖.配电网建设项目风险管理研究[J].企业改革与管理,2017(5):21-21.