基于韦布尔寿命模型的血液透析机关键性部件寿命分布的研究

Study on Life Distribution of Key Parts of Hemodialysis Machine Based on Weibull Lifetime Model

目的针对血液透析机临床应用特殊性,建立以三参数韦布尔分布为基础的血液透析机关键部件寿命分布模型,为血液透析机关键部件寿命分布及预防性更换提供理论研究依据。方法介绍并建立了三参数韦布尔寿命模型,收集并整合费森尤斯公司及第三方维修公司提供的2013至2019年血液透析机血泵、肝素泵、超滤泵、比例泵四类工单数据,结合最大似然估计法对韦布尔寿命模型参数进行计算估值,并使用我院血液透析机故障数据对韦布尔寿命模型进行拟合优度检验。结果我院血液透析机血泵、肝素泵、超滤泵、比例泵使用寿命与寿命模型分布相一致,与模型呈现较高的相关性,相关系数分别为0.92、0.83、0.95、0.80,对应各关键部件检验P值分别为0.89、0.21、0.91、0.31。结论使用韦布尔寿命模型对血液透析机关键性部件寿命进行预测切实可行,对部件预防性更换策略的制定具有一定的理论研究依据。

Objective To establish a life distribution model of the key parts of the hemodialysis machine based on the three-parameter Weibull distribution according to the particularity of clinical application of hemodialysis machine,which can provide a theoretical basis for the life distribution and preventive replacement of key components of the hemodialysis machine.Methods The three-parameter Weibull lifetime model was introduced and established.The repair order data of blood pump,heparin pump,ultrafiltration pump and proportioning pump used for the hemodialysis machine from 2013 to 2019 provided by Fresenius and the third maintenance company was collected and integrated.In this paper,the maximum likelihood estimation method was used to evaluate the parameters of the Weibull lifetime model,and the goodness of fit test was carried out on the Weibull lifetime model using the fault data of the hemodialysis machine in our hospital.Results The service life of blood pump,heparin pump,ultrafiltration pump and proportioning pump used for the hemodialysis machine in our hospital was consistent with the distribution of lifetime model,showing a high correlation with the model.The correlation coefficients were 0.92,0.83,0.95 and 0.80 respectively.The p-value of corresponding key components were 0.89,0.21,0.91 and 0.31 respectively.Conclusion It is feasible to predict the lifetime of the key parts of hemodialysis machine by using Weibull life model,which has a certain theoretical basis for the formulation of preventive replacement strategy.