摘要(英) |
In traditional reliability life test, usually the data are the failure (censored) times of the test units. However, most of the analyses do not consider the variabilities in the real life, such as the use-rate, load, temperature, humidity, etc. With new technology, we can install sensors or smart chips in many products so that the variabilities of
the above characteristics can be collected. In this thesis, we consider the life test of multi-components series systems, when the cycles-to-failure and use-rate information are available for those systems with censors (chips). On the other hand, only the times-to-failure can be collected from the systems without censors (chips). EM-algorithm is employed to obtain the MLEs along with their estimated standard errors computed based on the observed information matrix via the missing information principle. It turns out that the proposed method not only provides accurate results but also saves much computing time than the existing method. |
論文目次 |
1 Introduction.............................................1
2 Maximum Likelihood Estimates for Censored Data...........5
2.1 The Likelihood Function of the Connected Group.........6
2.2 The Likelihood Function of the Not-Connected Group.....8
2.3 The Full Likelihood Function..........................10
2.4 The EM-Algorithm......................................11
2.5 The Information Matrix................................16
2.6 Reliability Inference.................................20
2.7 The Experimental Results..............................22
2.7.1 Simulation of censored data under theta_T=(5,2,4,3,0.8)’............................................23
2.7.2 Simulation of censored data under theta_T=(5,2,4,3,1)’..............................................29
2.7.3 Simulation of censored data under theta_T=(5,2,4,3,1.2)’............................................34
3 Maximum Likelihood Inference for Masked Data............39
3.1 Likelihood for Imputed Components in the Not-Connected Group.....................................................40
3.2 EM-algorithm in Masked Data...........................41
3.3 The Information Matrix in Masked Data.................43
3.4 The Experimental Results with Masked Data.............45
3.4.1 Simulation of masked data under theta_T=(5,2,4,3,0.8)’............................................45
3.4.2 Simulation of masked data under theta_T=(5,2,4,3,1)’..............................................50
3.4.3 Simulation of masked data under theta_T=(5,2,4,3,1.2)’............................................54
4 Identification of the Masking Component.................59
4.1 Support Vector Classification.........................59
4.2 Identification for the Masked Data Using SVM..........65
4.3 Using Maximum Probability as a Classifier.............66
4.4 Comparison............................................70
5 Conclusions and Future Work.............................72
Bibliography..............................................74
Appendices................................................78
A. The Iterative Formulas in the EM-Algorithm for Unmasked Data......................................................78
B. Derivation of the Complete Information.................80
C. Derivation the Missing Information.....................81
D. Derivation the Missing Information in Masked Data......85 |
參考文獻 |
[1] Balakrishnan N. and Kateri M. (2008) "On the maximum likelihood estimation of parameters of Weibull distribution based on complete and censored data". Statistics and Probability Letters, Vol. 78, pp. 2971-2975.
[2] Balakrishnan N. and Mitra D. (2011) "Likelihood inference for lognormal data with left truncation and right censoring with an illustration". Journal of Statistical Planning and Inference, Vol. 141, pp. 3536-3553.
[3] Boser B. E., Guyon I. M. and Vapnik V. (1992) "A training algorithm for optimal margin classifiers". In Proceedings of the Fifth Annual Workshop of Computational Learning Theory, Pittsburgh.
[4] Burges C. J. C. (1998) "A tutorial on support vector machines for pattern recognition". Data Mining and Knowledge Discovery, Vol 2, pp. 121-167.
[5] Chang C. C. and Lin C. J. (2011) "LIBSVM: a library for support vector machines". ACM Transactions on Intelligent Systems and Technology, Vol. 2, pp. 27:1-27:27.
[6] Cortes C. and Vapnik V. (1995) "Support-vector Networks". Machine Learning, Vol. 20, pp. 273-297.
[7] Cristianini N. and Taylor J. S. (2000) An introduction to support vector machines. Cambridge University Press, Cambridge, UK.
[8] Efron B. (1979) "Bootstrap method: another look at the jackknife". Annals of Statistics, Vol. 17, pp. 1-26.
[9] Fan T. H. and Hsu T. M. (2012) "Accelerated life tests of a series system with masked interval data under exponential lifetime distributions". IEEE Transactions on Reliability, Vol. 61, pp. 798-808.
[10] Fan T. H. and Wang W. L. (2011) "Accelerated life tests for Weibull series systems with masked data". IEEE Transactions on Reliability, Vol. 60, pp. 557-569.
[11] Friedman J. H. (1996) Another approach to polychotomous classification. Technical report, Stanford Department of Statistics. http://www-stat.stanford.edu/jhf/ftp/poly.ps.Z.
[12] Guess F. M., Usher J. S. and Hodgson T. J. (1991) "Estimating system and component
reliabilities under partial information on cause of failure". Journal of Statistical Planning and Inference, Vol. 29, pp. 75-85.
[13] Hong Y. and Meeker W. Q. (2010) "Field-failure and warranty prediction based on auxiliary use-rate information". Technometrics, Vol. 52, pp. 148-159.
[14] Huang C. M., Lee Y. J., Lin D. K. J. and Huang S. Y. (2007) "Model selection for support vector machine via uniform design". Computational Statistics and Data Analysis, Vol. 52, pp. 335-346.
[15] Kreßel U. (1999) "Pairwise classification and support vector machines". Advances in Kernel Methods - Support Vector Learning, Schölkopf B., Burges C. J. C. and Smola A. J., editors. pp. 255-268, Cambridge, MA, MIT Press.
[16] Knerr S., Personnaz L. and Dreyfus G. (1990) "Single-layer learning revisited: A stepwise procedure for building and training a neural network". Neurocomputing: Algorithms, Architectures and Applications, Fogelman-Soulie and Herault, editors. NATO ASI Series, Springer.
[17] Louis T. A. (1982) "Finding the observed information matrix when using the EM algorithm". Journal of the Royal Statistical Society, Series B, Vol. 44, pp. 226-233.
[18] Meeker W. Q. and Escobar L. A. (1998) Statistical methods for reliability data. New York: Wiley.
[19] Meeker W. Q., Escobar L. A. and Hong Y. (2009) "Using accelerated life test results to predict product field reliability". Technometrics, Vol. 51, pp. 146-161.
[20] Miyakawa M. (1984) "Analysis of incomplete data in competing risk model". IEEE Transactions on Reliability, Vol. R-33, pp. 293-296.
[21] Mukhopadhyay C. (2006) "Maximum likelihood analysis of masked series system lifetime data". Journal of Statistical Planning and Inference, Vol. 136, pp. 803-838.
[22] Newton M. A. and Raftery A. E. (1994) "Approximate bayesian inference with the weighted likelihood bootstrap". Journal of the Royal Statistical Society, Series B, Vol. 56, pp. 3-48.
[23] Ng H. K. T. and Wang Z. (2009) "Statistical estimation for the parameters of Weibull distribution based on progressively type-I interval censored sample". Journal of Statistical Computation and Simulation, Vol. 79, pp. 145-159.
[24] Ng H. K. T., Chan P. S. and Balakrishnan N. (2002) "Estimation of parameters from progressively censored data using EM algorithm". Computational Statistics and Data Analysis, Vol. 39, pp. 371-386.
[25] Pal M. (2005) "Multiclass approaches for support vector machine based land cover classification". Proceedings of the 8th Annual International Conference, Map India.
[26] Sarhan A. M. (2001) "Reliability estimations of components from masked system life data". Reliability Engineering and System Safety, Vol. 74, pp.107-113.
[27] Sarhan A. M. (2003) "Estimation of system components reliabilities using masked data". Applied Mathematics and Computation, Vol. 136, pp. 79-92.
[28] Stacy E. W. (1962) "A generalization of the gamma distribution". The Annals of Mathematical Statistics, Vol. 33, pp. 1187-1192.
[29] Usher J. S. (1996) "Weibull component reliability-prediction in the presence of masked data". IEEE Transactions on Reliability, Vol. 45, pp. 229-232.
[30] Usher J. S. and Hodgson T. J. (1988) "Maximum likelihood analysis of component reliability using masked system life-test data". IEEE Transactions on Reliability, Vol. 37, pp. 550-555.
[31] Vapnik V. (1998) Statistical learning theory. New York: Wiley.
[32] Yang G. (2007) Life cycle reliability engineering. New Jersey: Wiley. |