逐步增加的Ⅱ型截尾下系统可靠性的Bayes分析及寿命预测  

Bayes Analysis of System Reliability and Prediction Based on Progressively Type-Ⅱ Censored Tests Squares Estimator is BLUE

作　　者：李凌[1] 师义民[1] 鲁娟[2] 

机构地区：[1]西北工业大学应用数学系,西安710072 [2]西安建筑科技大学建筑设计院,西安710055

出　　处：《数学的实践与认识》2007年第18期57-64,共8页Mathematics in Practice and Theory

基　　金：国家自然科学基金(N0.70471057);陕西省教育厅自然科学基金资助项目(N0.03Jk065)

摘　　要：在逐步增加的型截尾模型下,研究部件寿命服从双参数指数分布的冷贮备串联系统可靠性指标的Bayes估计及单样本场合未来观测值的预测问题.在两个参数均未知的情形下,分别在平方损失(SE)、LINEX损失和熵(General Entropy,GE)损失函数下给出两个参数及可靠性指标的Bayes估计,对于超参数的确定,给出一种新的方法;并讨论了单样本场合未来观测值的预测问题,给出预测分布及预测区间;最后利用随机模拟方法进行比较,并对结果进行了讨论.Suppose that the life of unit is distributed as two parameters exponential distribution. The Bayes estimations of the reliability performances for cold standby series system are derived based on progressively type-Ⅱ censored samples. Bayes estimators are obtained using Squared Error （SE）, LINEX and General Entropy loss function. This was done with respect to the values of the parameter are unknown, and a new method is given to estimate hyper-parameters value. Bayesian predictive density function, which is necessary to obtain bounds for predictive interval of future record is derived and discussed. At last, some numerical results are discussed using Monte-Carlo simulation.

关 键 词：双参数指数分布 逐步增加的Ⅱ型截尾 冷贮备串联系统 BAYES估计 预测分布 损失函数 

分 类 号：O213.2[理学—概率论与数理统计]