system lifetime data
Recently Published Documents


TOTAL DOCUMENTS

17
(FIVE YEARS 1)

H-INDEX

6
(FIVE YEARS 0)

Test ◽  
2017 ◽  
Vol 27 (4) ◽  
pp. 787-810 ◽  
Author(s):  
M. Hermanns ◽  
E. Cramer

2015 ◽  
Vol 62 (7) ◽  
pp. 550-563 ◽  
Author(s):  
Jian Zhang ◽  
Hon Keung Tony Ng ◽  
Narayanaswamy Balakrishnan

2014 ◽  
Vol 687-691 ◽  
pp. 1198-1201
Author(s):  
Bin Liu ◽  
Yi Min Shi ◽  
Jing Cai ◽  
Mo Chen

The Type-II generalized progressively hybrid censored scheme with masked data is presented. Based on masked system lifetime data, using the expectation maximization algorithm and the Quasi-Newton method, we obtain the Maximum Likelihood Estimation (MLE) of the components distribution parameters in the Weibull case. Finally, Monte Carlo simulation is presented to illustrate the effect.


2014 ◽  
Vol 687-691 ◽  
pp. 1015-1018 ◽  
Author(s):  
Jing Cai ◽  
Yi Min Shi ◽  
Hong Bo Yue

This article considers a step-stress partially accelerated life tests for series system model where independent and non-identical Burr XII-distributed lifetimes are assumed for the components. Based on Type-I progressive hybrid censored and masked data, expectation maximum algorithm combined with auxiliary variables is developed for estimating the model parameters and the acceleration factor. In addition, the asymptotic confidence intervals are constructed by the parametric bootstrap method. Furthermore, the proposed method is illustrated through a simulation study under various masking levels.


2014 ◽  
Vol 551 ◽  
pp. 626-632
Author(s):  
Mo Chen ◽  
Yi Min Shi ◽  
Li Jin

We consider the series system with three independent and non-identical components, each of which has Burr XII distributed lifetime. Based on progressively type-II censored and masked system lifetime data, the maximum likelihood estimates (MLE) of the components parameters and reliability function are obtained. By introducing a latent variable, Bayesian estimators of the components parameters and the reliability function are also developed using Gibbs sampling method. Furthermore, in the numerical simulation study, the MLE and Bayesian estimates are compared under different removal probabilities and different times.


Sign in / Sign up

Export Citation Format

Share Document