企业异常用水预警管理数据

通过对企业用水量情况在线智能监测，从而得到企业异常用水预警管理数据。该数据可以为金融机构评估企业信贷能力时作为参考依据，在信贷反欺诈、贷前审批、额度管理和贷后监测等业务环节，金融机构可以从用水视角评估企业经营健康状况，消减金融机构与企业间的信息不对称，进一步完善金融机构对企业信用评价的指标体系。1.数据采集：人工采集和自动采集。2.数据处理：对采集到的原始数据进行清洗、分析、整理等方式，获取所需要的数据：时间、用户代码、统一社会信用代码、本期水量、水量月均值、水量标准差、预警指标、预警等级等。对用水企业名进行了代码匿名化处理。3.算法1）水量月均值：一年的企业用水量数据，除以12个月，得到月均用水量；2）水量标准差：每月用水量减去其月均值的平方和，所得结果除以12，再把所得值开根号；3）企业月度用水量：当期月企业用水量；4）预警判断：企业当期水量与月均水量间的差值高于3倍标准差时，预警指标3+，提示一级预警；当企业当期水量与月均水量间的差值高于2倍标准差时，预警指标2+，提示二级预警；企业当期水量与月均水量间的差值高于1倍标准差，预警指标1+，提示三级预警。4.数据应用：可以为金融机构信贷业务提供信息支持，通过对企业用水量在线智能监测，实现企业异常用水波动预警提醒信息。在信贷反欺诈、贷前审批、额度管理和贷后监测等业务环节，金融机构可以从用水视角评估企业经营健康状况，消减金融机构与企业间的信息不对称，进一步完善金融机构对企业信用评价的指标体系。

This dataset is generated via online intelligent monitoring of enterprise water consumption, providing abnormal water usage early warning management data. This data can serve as a reference for financial institutions when evaluating enterprises' creditworthiness. In business links including credit anti-fraud, pre-loan approval, quota management and post-loan monitoring, financial institutions can assess enterprises' operational health from the perspective of water consumption, reduce information asymmetry between financial institutions and enterprises, and further improve the credit evaluation index system for enterprises adopted by financial institutions.

1. Data Collection: Manual and automatic data collection.
2. Data Processing: Clean, analyze and organize the collected raw data to acquire required datasets including time, user code, unified social credit code, current period water consumption, monthly average water consumption, water consumption standard deviation, early warning indicators and early warning levels. The names of water-using enterprises are anonymized using code assignment.
3. Algorithms:
1) Monthly average water consumption: Calculate the annual water consumption data of the enterprise and divide it by 12 months to obtain the monthly average water consumption;
2) Water consumption standard deviation: Subtract the monthly average water consumption from each month's water consumption, sum the squared differences, divide the resultant value by 12, then take the square root of the obtained value;
3) Monthly enterprise water consumption: The water consumption of the enterprise in the current month;
4) Early warning judgment: When the difference between the current period's water consumption and the monthly average water consumption of the enterprise exceeds 3 times the standard deviation, the early warning indicator is marked as 3+, prompting a first-level early warning; when the difference exceeds 2 times the standard deviation, the early warning indicator is marked as 2+, prompting a second-level early warning; when the difference exceeds 1 time the standard deviation, the early warning indicator is marked as 1+, prompting a third-level early warning.
4. Data Application: The dataset can provide information support for the credit business of financial institutions. Through online intelligent monitoring of enterprise water consumption, early warning reminders for abnormal water usage fluctuations of enterprises can be realized. In business links such as credit anti-fraud, pre-loan approval, quota management and post-loan monitoring, financial institutions can evaluate the operational health of enterprises from the perspective of water consumption, reduce information asymmetry between financial institutions and enterprises, and further improve the credit evaluation index system for enterprises used by financial institutions.

搜集汇总

数据集介绍

特点

该数据集主要用于企业异常用水预警管理，通过监测企业用水量，为金融机构提供信贷评估参考。数据包含10731条记录，每月更新，涵盖多个关键字段如预警提醒、水量月均值等，适用于信贷反欺诈、贷前审批等金融业务场景。