在串聯系統中，通常假設系統物件是彼此獨立的。本文考慮在型一設限下，以二元 Marshall-Olkin 韋伯分布去配適兩物件的壽命。而在串聯系統中，時常無法觀測到造成系統失效的物件，此時該系統為隱蔽資料。當系統壽命被設限或隱蔽時，資料中具遺失值，故此，我們使用期望值-最大化演算法去求得模型中未知參數的最大概似估計。同時，我們也利用遺失資訊法則去計算最大概似估計的費雪訊息矩陣，以近似最大概似估計的標準誤差。更進一步地我們將其分別推廣到定應力和階段應力加速壽命試驗，並比較各模型中所得之模型參數估計及系統與個別物件的平均壽命、分位數和可靠度函數之統計推論。由模擬結果顯示，我們所提出的方法頗為精確，另外我們也成功地分析了一組真實資料。

Problems in series systems usually assume the lifetime distributions of components are independent. In this thesis, we consider to model the lifetime of a twocomponent series system under Type-1 censoring scheme by the Marshall-Olkin bivariate Weibull distribution. It is often to include masked data in which the component that causes the system to fail is not observed. When the data are masked or censored, there exist missing variables in the model. We apply the EM-algorithm to find the MLEs of the unknown parameters. In addition, we calculate the Fisher information via missing information principle to approximate the standard errors of the MLEs. Furthermore, we extend these results to the constant and step-stress accelerated life tests. Statistical inference on the lifetime distribution as the mean lifetimes, reliability functions and the quantiles of system and components are derived. Simulation study shows that the proposed methods perform accurately. A real data set is analyzed successfully.

[1] Balakrishnan, N. and Mitra, D. Left truncated and right censored Weibull data and
likelihood inference with an illustration. Computational Statistics and Data Analysis,
56(12):4011–4025, 2012.
[2] Berger, J. O. and Sun, D. Bayesian analysis for the poly-Weibull distribution. Journal of
the American Statistical Association, (88):1412–1418, 1993.
[3] Escobar, L. A. and Meeker, W. Q. A review of accelerated test models. Statistical Science,
(21):552577, 2006.
[4] Fan, T. H. and Hsu T. M. Statistical inference of a series system with masked data
with correlated log-normal lifetime distributions under Type-I censoring. International
Conference on Advances in Probability and Statistics-Theory and Applications, December
30 2011.
[5] Fan, T. H. and Hsu, T. M. Accelerated life tests of a series system with masked interval data
under exponential lifetime distributions. IEEE Transactions on Reliability, 61(3):798–807,
2012.
[6] Fan, T. H. and Wang, W. L. Accelerated life tests for Weibull series systems with masked
data. IEEE Transactions on Reliability, 60(3):557–568, 2011.
[7] Fan, T. H., Hsu, T. M. and Peng, K. J. Bayesian inference of a series system on Weibull
step-stress accelerated life tests with dependent masking. To appear on Journal of Quality Technology and Quantitative Management, 2013.
[8] Khamis, I. H. Comparison between constant and step-stress tests for Weibull models.
International Journal of Quality and Reliability Management, (14):74–81, 1997.
[9] Khamis, I. H. and Higgins, J. J. Optimum 3-step step-stress tests. IEEE Transactions on
Reliability, (45):341–345, 1996.
[10] Kim, C. M. and Bai, D. S. Analyses of accelerated life test data under two failure modes.
International Journal of Reliability, Quality and Safety Engineering, 9(2):111–125, 2002.
[11] Klein, J. P. and Basu, A. P. Weibull accelerated life test when there are competing causes
of failure. Communications in Statistics-Theory and Methods, (10):2073–2100, 1981.
[12] Klein, J. P. and Basu, A. P. Accelerated life tests under competing Weibull causes of
failure. Communications in Statistics-Theory and Methods, (11):2073–2100, 1982.
[13] Kundu, D. and Dey, A. K. Estimating the parameters of the Marshall-Olkin bivariate
Weibull distribution by EM algorithm. Computational Statistics and Data Analysis,
53(4):956–965, 2009.
[14] Kundu, D. and Guptab, A. K. Bayes estimation for the Marshall–Olkin bivariate Weibull
distribution. Computational Statistics and Data Analysis, 57(1):271–281, 2013.
[15] Kuo, L. and Yang, T. Bayesian reliability modeling for masked system lifetime data.
Statistics and Probability Letters, (47):229–241, 2000.
[16] Lin, D. K. J. and F. M. Guess. Bayes estimation of component-reliability from masked system-life data. IEEE Transactions on Reliability, (45):233–237, 1996.
[17] Lin, H. H. Competing risks model of Marshall-Olkin bivariate exponential distribution
under hybrid censoring. Master Thesis, Graduate Institute of Statistics, National Central
University, Jungli, Taiwan, 2012.
[18] Louis, T. A. Finding the observed information matrix using the EM algorithm. Journal
of the Royal Statistical Society, (44):226–233, 1982.
[19] Marshall, A. W. and Olkin, I. A multivariate exponential distribution. Journal of the
American Statistical Association, (62):30–44, 1967.
[20] McLachlan, G. J. and Krishnan, T. The EM Algorithm and Extensions. Wiley, New York.,
second edition, 2008.
[21] Meeker, W. Q. and Escobar, L. A. Statistical Method for Reliability Data. NewYork, 1998.
[22] Meeker, W. Q. and Nelson, W. Optimum accelerated life-tests for Weibull and extreme
value distributions. IEEE Transactions on Reliability, R-24:321–332, 1975.
[23] Miyakawa, M. Analysis of incomplete data in competing risk model. IEEE Transactions
on Reliability, (33):293–296, 1984.
[24] Moeschberger, M. L. Life tests under dependent competing causes of failure. Technometrics,
16(1):39–47, 1974.
[25] Mukhopadhyay, C. Maximum likelihood analysis of masked series system lifetime data.
Journal of Statistical Planning and Inference, (136):803–838, 2006.
[26] Mukhopadhyay, C. and Basu, A. P. Bayesian analysis of competing risks: k independent
exponentials. Technical Report No.516, 1993.
[27] Nandi, S. and Dewan, I. An EM algorithm for estimating the parameters of bivariate
Weibull distribution under random censoring. Computational Statistics and Data Analysis,
54(6):1559–1569, 2010.
[28] Nelson, W. Hazard plotting methods for analysis of life data with di↵erent failure modes.
Journal of Quality Technology, 2:126–49, 1970.
[29] Nelson, W. Accelerated life testing-step-stress models and data analysis. IEEE Transactions
on Reliability, (29):103–108, 1980.
[30] Oakes, D. Direct calculation of the information matrix via the EM algorithm. Journal of
the Royal Statistical Society, (61):479–482, 1999.
[31] Mazzuchi, T. A., Soyer, R. and Vopatek A. Linear Bayesian inference for accelerated
Weibull model. Lifetime Data Analysis, (3):63–75, 1997.
[32] Usher, J. S. Weibull component reliability-prediction in the presence of masked data.
IEEE Transactions on Reliability, (45):229–232, 1996.
[33] Usher, J. S. and Hodgson, T. J. Maximum likelihood analysis of component reliability
using masked system life-test data. IEEE Transactions on Reliability, (37):550– 555, 1988.
[34] Wang, W. and Kececioglu, D. B. Fitting the Weibull log-linear model to accelerated
life-test data. IEEE Transactions on Reliability, (49):217–223, 2000.