基于历史运维大数据挖掘的电力企业绩效工资分配感知模型

Perceived Model of Performance Wage Allocation in Power Enterprises Based on Historical Operation and Maintenance Big Data Mining

针对传统电力企业绩效工资分配机制存在评价对象相对岗位下的多维片面性、岗绩分离模式动态博弈性较差、因素加权主变量均衡修正机制缺失、业务导向与目标导向量化考核及全景感知具有滞后性等固有弊端,提出了一种基于历史运维大数据挖掘的电力企业绩效工资分配感知模型。基于国家电网历史运维大数据,构建绩效工资分配与效益营收的物理映射,给出全息运维数据与绩效工资分配的定量耦合关系,引入机器学习算法,构建较长周期内的绩效工资分配感知稳定预测机制,形成具备潜在规律挖掘、定量映射生成、改革走向精准预测、效益营收预警等功能的绩效工资分配全局协同精准化预测架构。选取国网某供电公司为效能评价载体,开发了电力企业绩效工资分配感知的原型系统,并对其进行了多方面的实证分析以证明它的综合效能,一线运维分析结果表明,该原型系统具备绩效工资分配与效益营收动态建模、定量耦合关系实时修正、较长周期内的改革走向稳定预测等全方位绩效工资分配感知体系效能,在指标权重全局最优度量、因素加权主变量均衡修正、绩效工资分配感知稳定预测等层面具有明显优势。

In view of the inherent disadvantages of traditional performance-based salary distribution mechanism in electric power enterprises,such as multi-dimensional one sidedness of the evaluation object relative to the post,poor dynamic game of post performance separation mode,lack of factor weighted main variable balance correction mechanism,lag of quantitative assessment and panoramic perception of business orientation and target orientation,this paper puts forward a performance mining method of electric power enterprises based on historical operation and maintenance big data mining,and establishes a salary distribution perception model.Based on the historical operation and maintenance big data of the State Grid,this paper constructs the physical mapping between the performance wage distribution and the benefit revenue,gives the quantitative coupling relationship between the holographic operation and maintenance data and the performance wage distribution,introduces the machine learning algorithm,constructs the perceptual and stable prediction mechanism of the performance wage distribution in a long period,and forms the potential regular mining,quantitative mapping generation,accurate prediction and efficiency of the reform trend.The overall collaborative and accurate prediction framework of performance salary distribution is given with the functions of benefit,revenue and early warning.This paper selects a power supply company of State Grid as the carrier for efficiency evaluation,develops a prototype system of performance salary distribution perception of power enterprises and conducts a comprehensive efficiency empirical analysis.The results of the first-line operation and maintenance verification show that the prototype system has a full range of performance salary distribution,such as dynamic modeling of performance salary distribution and benefit revenue,real-time correction of quantitative coupling relationship,long-term stable prediction of reform trend,etc.The perceived system performance has obvious advantages in the aspects of the global optimal measurement of index weight,the correction of factor weighted main variable equilibrium,and the perceptual stability prediction of performance wage distribution.