Reliability evaluation for Weibull distribution under multiply type-І censoring

Abstract

The multiply type-І censoring represented that all units in life test were terminated at different times. For estimations of Weibull parameters, it was easy to compute the maximum likelihood estimation (MLE) and least-squares estimation (LSE) while it was hard to build confidence intervals (CI). The concept of generalized confidence interval (GCI) was introduced to build CIs of parameters under multiply type-I censoring. Further, GCI based on LSE and GCI based on MLE were proposed. It is mathematically proved that the former is exact and the latter is approximate. Besides, a Monte Carlo simulation study and an illustrative example also turn out that the GCI method based on LSE yields rather satisfactory results by comparison with the ones based on MLE. It should be clear that the GCI method is a sensible choice to evaluate reliability under multiply type-I censoring.

Keywords

multiply type-І censoring / generalized confidence interval / maximum likelihood estimation / least-squares estimation

References

[1] XuZ-j, ZhengJ-j, BianX-y, LiuY. A modified method to calculate reliability index using maximum entropy principle [J]. Journal of Central South University, 2013, 20(3): 1058-1063

[2] LiuJ, LiY. An improved adaptive response surface method for structural reliability analysis [J]. Journal of Central South University, 2012, 19(3): 1148-1154

[3] XieG-h, ZhangJ-s, LiuR-g. Application of matrix-based system reliability method in complex slopes [J]. Journal of Central South University, 2013, 20(2): 812-820

[4] ZhangC-y, BaiG-c. Extremum response surface method of reliability analysis on two-link flexible robot manipulator [J]. Journal of Central South University, 2012, 19(1): 101-107

[5] WangX-l, JiangP, GuoB, ChengZ-j. Real-time reliability evaluation based on damaged measurement degradation data [J]. Journal of Central South University, 2012, 19(1): 3162-3169

[6] WangF-s, ZhangJ-r, WangP-y, HuoS-h, YueZ-f. Reliability analysis of laminated composite under compression and shear loads [J]. Journal of Central South University, 2012, 19(7): 2712-2717

[7] CastetJ F, SalehJ H. Satellite and satellite subsystems reliability: Statistical data analysis and modeling [J]. Reliability Engineering & System Safety, 2009, 94(11): 1718-1728

[8] OlteanuD, FreemanL. The evaluation of median-rank regression and maximum likelihood estimation techniques for a two-parameter Weibull distribution [J]. Quality Engineering, 2010, 22(4): 256-272

[9] ThomanD R, BainL J, AntleC E. Maximum likelihood estimation, exact confidence intervals for reliability, and tolerance limits in the weibull distribution [J]. Technometrics, 1970, 12(2): 363-371

[10] BalakrishnanN, KateriM. On the maximum likelihood estimation of parameters of Weibull distribution based on complete and censored data [J]. Statistics & Probability Letters, 2008, 78(17): 2971-2975

[11] JoarderA, KrishnaH, KunduD. Inferences on Weibull parameters with conventional type-I censoring [J]. Computational Statistics & Data Analysis, 2011, 55(1): 1-11

[12] SkinnerK R, KeatsJ B, ZimmerW J. A comparison of three estimators of the Weibull parameters [J]. Quality and Reliability Engineering International, 2001, 17(4): 249-256

[13] FothergillJ. Estimating the cumulative probability of failure data points to be plotted on Weibull and other probability paper [J]. IEEE Transactions on Electrical Insulation, 1990, 25(3): 489-492

[14] ZhangL F, XieM, TangL C. A study of two estimation approaches for parameters of Weibull distribution based on WPP [J]. Reliability Engineering & System Safety, 2007, 92(3): 360-368

[15] GenschelU, MeekerW Q. A comparison of maximum likelihood and median-rank regression for Weibull estimation [J]. Quality Engineering, 2010, 22(4): 236-255

[16] YangZ, XieM, WongA C M. A unified confidence interval for reliability-related quantities of two-parameter Weibull distribution [J]. Journal of Statistical Computation and Simulation, 2007, 77(5): 365-378

[17] TanZ-b. A new approach to MLE of Weibull distribution with interval data [J]. Reliability Engineering and System Safety, 2009, 94: 394-403

[18] KrishnamoorthyK, LinY, XiaY. Confidence limits and prediction limits for a Weibull distribution based on the generalized variable approach [J]. Journal of Statistical Planning and Inference, 2009, 139(8): 2675-2684

[19] WeerahandiS. Generalized confidence intervals [J]. J Am Statist Assoc, 1993, 88: 899-905

[20] RoyA, MathewT. A generalized confidence limit for the reliability function of a two-parameter exponential distribution [J]. Journal of Statistical Planning and Inference, 2005, 128(2): 509-517

[21] MitraP K, SinhaB K. A generalized p-value approach to inference on common mean [J]. Journal of Statistical Planning and Inference, 2007, 137(11): 3634-3642

[22] KrishnamoorthyK, MathewT. Inferences on the means of lognormal distributions using generalized p-values and generalized confidence intervals [J]. Journal of Statistical Planning and Inference, 2003, 115(1): 103-121

[23] FernN A J. On calculating generalized confidence intervals for the two-parameter exponential reliability function [J]. Statistics, 2007, 41(2): 129-135

[24] WuW-H, HsiehH-Neng. Generalized confidence interval estimation for the mean of delta-lognormal distribution: An application to new zealand trawl survey data [J]. Journal of Applied Statistics, 2014, 41(7): 1471-1485

[25] ThomanD R, BainL J, AntleC E. Inferences on the parameters of the Weibull distribution [J]. Technometrics, 1969, 11(3): 445-460