基于q-Weibull分布的医疗设备可靠性分析

RELIABILITY ANALYSIS OF MEDICAL EQUIPMENT BASED ON q-WEIBULL DISTRIBUTION

实现了基于q-Weibull分布的医疗设备可靠性分析,为医疗机构修正运营维修管理策略提供依据。Weibull分布故障率函数是单调的,不能全面描述复杂的医疗设备全生命周期运行状态,因此引入q-Weibull分布预测医疗设备的剩余寿命,使用基于轮廓误差函数的方法简化q-Weibull分布参数估计过程,并以血滤机和无影灯灯座为例验证了该方法的有效性和可行性,通过均方误差(Mean Squared Error,MSE)、赤池信息准则(Akaike Information Criterion,AIC)和决定系数R^(2)比较了两种分布的优劣。对于血滤机和无影灯灯座两种分布都表现出相同的预测趋势,但从R^(2)和MSE对比来看,q-Weibull分布的拟合精度更高,尤其是基于q-Weibull分布的血滤机的MSE值(2.8181×10^(-3))远小于Weibull分布的MSE值(9.465);当血滤机和无影灯灯座运行50天后,其剩余寿命估计值分别为254.3909天和291.0111天。以上数据验证了q-Weibull分布的有效性和拟合精度的优良性,该方法值得在医疗设备可靠性研究中进行进一步的研究和推广。

A reliability analysis of medical equipment is implemented based on the q-Weibull distribution to provide a basis for healthcare organizations to revise their operational maintenance management strategies.The Weibull distribution failure rate function is monotonic and cannot fully describe the full life cycle operation of complex medical equipment.Therefore,this study introduces the q-Weibull distribution to predict the remaining life of medical equipment,uses a method based on the contour error function to simplify the q-Weibull distribution parameter estimation process,and verifies the validity and feasibility of the method using a hemofiltration apparatus and a lamp-holder of surgical shadowless lamp as examples,and compares the advantages and disadvantages of the two distributions by mean square error(MSE),the Akaike information criterion(AIC) and the coefficient of determination R^(2).Both distributions for the hemofiltration apparatus and lamp-holder of surgical shadowless lamp show the same predictive trend,but the R^(2) and MSE comparisons show that the q-Weibull distribution had a better fit accuracy,especially the MSE(2.818 1×10~(-3))for the hemofiltration apparatus based on the q-Weibull distribution is much smaller than the MSE(9.465) for the Weibull distribution.When the hemofiltration apparatus and lamp-holder of surgical shadowless lamp are operated for 50 days,their estimated remaining life is 254.390 9 days and 291.011 1 days respectively.The above data verifies the validity and fitting accuracy of the q-Weibull distribution,which is worthy of further research and promotion in the reliability research of medical equipment.