基于数据挖掘关联规则分析法的呼吸机预防性维护机制与数据分析研究

Research and achievement analysis on preventive maintenance mechanism of ventilator based on association rules analysis method of data mining

目的:探讨基于数据挖掘关联规则分析法的呼吸机预防性维护机制在呼吸机设备管理中的应用价值。方法:选取医院临床在用的91台呼吸机,采用随机数表法将其分为对照组(45台)和实验组(46台),对照组采用常规定期预防性维护,实验组采用基于数据挖掘关联规则分析法的呼吸机预防性维护机制,制定支持度的最小支持阈值和置信度的最小置信阈值,按需进行预防性维护。对比两组呼吸机维护次数、故障数和故障类型的差异性。结果:实验组预防性维护年均次数低于对照组,差异有统计学意义(t=4.17,P<0.05);实验组年故障数和年平均故障数均低于对照组,差异有统计学意义(t=4.44,t=3.76;P<0.05);实验组呼吸机管路故障和清洁消毒故障低于对照组,意外碰撞故障高于对照组,差异有统计学意义(x^(2)=4.45,x^(2)=4.29,x^(2)=5.74;P<0.05)。结论:基于数据挖掘关联规则分析法的呼吸机预防性维护机制,可在一定程度上降低责任工程师工作量,减少人为因素引起的设备故障,提高呼吸机临床使用效率。

Objective:To explore the application value of ventilator preventive maintenance mechanism based on association rules analysis method of data mining in ventilator equipment management.Methods:A total of 91 ventilators in clinical use at the hospital were selected and randomly divided into control group(45)and experimental group(46).The control group adopted routine preventive maintenance,the experimental group adopted the preventive maintenance mechanism of ventilator based on association rules analysis method of data mining by establishing the minimum support thresholds and minimum confidence thresholds and performing preventive maintenance as needed.The differences of maintenance times,number of failures and failure types between the two groups of ventilators were compared.Results:The number of preventive maintenance units per year and the average number per year of the experimental group were lower than those of the control group,the difference was statistically significant(t=4.17;P<0.05);both the number of annual failures and the average annual number of failures in the experimental group are lower than those in the control group,the difference was statistically significant(t=4.44,t=3.76;P<0.05);ventilator pipe failure and cleaning/disinfection failure in the experimental group were lower than those in the control group and other failures such as accidental collision were higher than those in the control group(x^(2)=4.45,x^(2)=4.29,x^(2)=5.74;P<0.05).Conclusion:The preventive maintenance mechanism of ventilator based on data mining association rule analysis can reduce the workload of responsible engineers,reduce equipment failure caused by human factors,and improve the clinical efficiency of ventilators.