Sains Malaysiana 52(2)(2023): 641-653

http://doi.org/10.17576/jsm-2023-5202-24

 

Bayesian Estimation of Time to Failure Distributions Based on Skew Normal Degradation Model: An Application to GaAs Laser Degradation Data

(Anggaran Bayesian Masa untuk Taburan Kegagalan Berdasarkan Model Degradasi Normal Pencong: Aplikasi untuk Data Degradasi Laser GaAs)

 

LAILA NAJI BA DAKHN, MOHD AFTAR ABU BAKAR* & KAMARULZAMAN IBRAHIM

 

Jabatan Sains Matematik, Fakulti Sains dan Teknologi, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor Darul Ehsan, Malaysia

 

Diserahkan: 30 Jun 2022/Diterima: 16 Disember 2022

 

Abstract

In this paper, the Bayesian method which involves informative and weakly informative priors are considered to estimate the parameters and percentiles of the time to failure distribution. The parameters of the time to failure distribution and its percentiles are determined based on linear degradation model where the degradation parameter is assumed to follow the skew normal distribution. For the prior distributions, location and scale parameters of the skew normal distribution is assumed to follow the uniform distribution while the shape parameter is assumed to follow gamma distribution. Two gamma priors are considered, either informative or weakly informative prior, depending on the assumed values of the hyper parameters. The performance of the method under the different prior assumptions is compared using a simulation study based on Markov Chain Monte Carlo method as well as a real data application. It is found that the parameter estimation based on informative prior is more precise as opposed to the weakly informative prior, especially in the case of small sample size. In addition, the skew normal degradation model is compared to the log-logistic degradation model through a simulation study and a real application of GaAs laser data. When modeling the percentiles of the time to failure distribution, results found based on the skew normal distribution is generally found to be more precise, particularly for the higher percentile values.

 

Keywords: Bayesian method; linear degradation model; log-logistic distribution; skew normal distribution; time to failure distribution

 

Abstrak

Dalam kertas ini, kaedah Bayesan yang melibatkan prior bermaklumat dan kurang bermaklumat dipertimbangkan untuk menganggar parameter dan persentil untuk taburan masa kegagalan. Parameter dan persentil bagi taburan masa kegagalan ditentukan berdasarkan model degradasi linear yang mana parameter degradasi diandaikan mengikuti taburan normal pencong. Untuk taburan prior, parameter skala dan lokasi bagi taburan normal pencong diandaikan mengikuti taburan seragam manakala parameter bentuk diandaikan mengikuti taburan gama. Dua prior gama yang dipertimbangkan, iaitu sama ada bermaklumat atau kurang bermaklumat, bergantung kepada nilai parameter hiper yang diandaikan. Prestasi kaedah berkenaan di bawah andaian yang berbeza dibandingkan menerusi kajian simulasi berdasarkan kaedah Rantai Markov Monte Carlo dan juga aplikasi data sebenar. Didapati bahawa penganggaran parameter berdasarkan prior bermaklumat adalah lebih persis berbanding prior kurang bermaklumat, khususnya apabila saiz sampel kecil. Seterusnya, model degradasi normal pencong dibandingkan dengan model degradasi log-logistik menerusi kajian simulasi dan aplikasi data laser GaAs. Bila memodelkan persentil bagi taburan masa kegagalan, secara amnya, hasil menunjukkan bahawa keputusan berdasarkan taburan normal pencong adalah lebih persis, khususnya untuk persentil yang bernilai tinggi.

 

Kata kunci: Kaedah Bayesian; model degradasi linear; taburan log-logistik; taburan masa kegagalan; taburan normal pencong

 

RUJUKAN

Al-haj Ebrahem, M., Al-momani, N. & Eidous, O. 2021. Variable scale kernel density estimation for simple linear degradation model. Electronic Journal of Applied Statistical Analysis 14(2): 359-372. https://doi.org/10.1285/i20705948v14n2p359

Albert, J. 2008. Bayesian Computation with R. 3rd ed. New York: Springer.

Alhamidie, A.A., Kamarulzaman Ibrahim, Alodat, M.T. & Wan Zawiah Wan Zin. 2019. Bayesian inference for linear regression under alpha-skew-normal prior. Sains Malaysiana 48(1): 227-235.

Azzalini, A. 1985. A class of distributions which includes the normal ones. Scandinavian Journal of Statistics 12(2): 171-178.

Dakhn, L.N.B., Ebrahem M.A-H. & Eidous, O. 2017. Semi-parametric method to estimate the time-to- failure distribution and its percentiles for simple linear degradation model. Journal of Modern Applied Statistical Methods Article 16(2): 322-346. https://doi.org/10.22237/jmasm/1509495420

Bayes, C.L. & Branco, M.D. 2007. Bayesian inference for the skewness parameter of the scalar skew-normal distribution. Brazilian Journal of Probability and Statistics 21: 141-163.

Bryson, M.C. 1974. Heavy-tailed distributions: Properties and tests. Technometrics 16(1): 61-68. https://doi.org/10.1080/00401706.1974.10489150

Chen, X., Sun, X., Ding, X. & Tang, J. 2019. The inverse gaussian process with a skew-normal distribution as a degradation model. Communications in Statistics - Simulation and Computation. https://doi.org/10.1080/03610918.2018.1527351

Coro, G. 2017. Gibbs Sampling with JAGS : Behind the Scenes. https://www.researchgate.net/publication/313905185

Ebrahem, M.A-h., Alodat, M.T. & Amani Arman. 2009. Estimating the time-to-failure distribution of a linear degradation model using a Bayesian approach. Applied Mathematical Sciences 3(1): 27-42.

Ebrahem, M.A-h., Eidous, O. & Kmail, G. 2009. Estimating percentiles of time-to-failure distribution obtained from a linear degradation model using kernel density method. Communications in Statistics - Simulation and Computation 38(9): 1811-1822. https://doi.org/10.1080/03610910903145130

Eidous, O., Ebrahem, M.A-h. & Dakhn, L.N.B. 2017. Estimating the time-to-failure distribution and its percentiles for simple linear degradation model using double kernel method. Journal of Probability and Statistical Science 15(1): 121-134.

Giraldo Gómez, N. 2005. An example of a heavy tailed distribution. Matematicas 13(1): 43-50.

Guure, C.B. & Noor Akma Ibrahim. 2014. Approximate Bayesian estimates of Weibull parameters with Lindley’s method. Sains Malaysiana 43(9): 1433-1437.

Hamada, M. 2005. Using degradation data to assess reliability. Quality Engineering 17(4): 615-620. https://doi.org/10.1080/08982110500225489

Jin, Z. 2016. Semiparametric accelerated failure time model for the analysis of right censored data. Communications for Statistical Applications and Methods 23(6): 467-478. https://doi.org/10.5351/csam.2016.23.6.467

Meeker, W.Q., Escobar, L.A. & Lu, C.J. 1999. Accelerated degradation tests: Modeling and analysis. Statistics Preprints 2: 1-24.

Meeker, W.Q. & Escobar, L.A. 1998. Statistical Method for Reliability Data. New York: John Wiley and Sons, Inc.

Oliveira, R.P.B., Loschi, R.H. & Freitas, M.A. 2018. Skew-heavy-tailed degradation models: An application to train wheel degradation. IEEE Transactions on Reliability 67(1): 129-141. https://doi.org/10.1109/TR.2017.2765485

Pan, D., Liu, J-B. & Yang, W. 2018. A new result on lifetime estimation based on Skew-Wiener degradation model. Statistics and Probability Letters 138: 157-164. https://doi.org/10.1016/j.spl.2018.03.009

Plummer, M. 2003. JAGS : A program for analysis of Bayesian graphical models using Gibbs sampling. Distributed Statistical Computing. https://www.r-project.org/nosvn/conferences/DSC-2003/Drafts/Plummer.pdf

Puggard, W., Niwitpong, S-a. & Niwitpong, S. 2022. Comparison analysis on the coefficients of variation of two independent Birnbaum-Saunders distributions by constructing confidence intervals for the ratio of coefficients of variation. Sains Malaysiana 51(7): 2265-2281.

Rawashdeh, A., Ebrahem, M.A-h. & Momani, A. 2018. A Bayesian approach to estimate the failure time distribution of a log-logistic degradation model. METRON 76: 155-176. https://doi.org/10.1007/s40300-018-0141-7

Robins, J. & Tsiatis, A.A. 1992. Semiparametric estimation of an accelerated failuare time model with time-dependent covariates. Biometrika 79(2): 311-319.

Shafiq, M., Alamgir & Atif, M. 2016. On the estimation of three parameters lognormal distribution based on fuzzy life time data. Sains Malaysiana 45(11): 1773-1777.

Spiegelhalter, D.J., Best, N.G., Carlin, B.P. & Van Der Linde, A. 2002. Bayesian measures of model complexity and fit. Biostatistics 64(4): 583-616.

Thangjai, W., Niwitpong, S-A. & Niwitpong, S. 2021. A Bayesian approach for estimation of coefficients of variation of normal distributions. Sains Malaysiana 50(1): 261-278. https://doi.org/10.17576/jsm-2021-5001-25

Tsai, C-c. & Lin, C-t. 2015. Lifetime inference for highly reliable products based on skew-normal accelerated destructive degradation test model. IEEE Transactions on Reliability 64(4): 1340-1355.

 

*Pengarang untuk surat-menyurat; email: aftar@ukm.edu.my

   

sebelumnya