Reliability Analysis of Multiple Causes of Failure in Presence of Independent Competing Risks

The failures of complex systems always arise from different causes in reliability test. However, it is difficult to evaluate the failure effect caused by a specific cause in presence of other causes. Therefore, a generalize reliability analysis model, which takes into account of the multiple competing causes, is highly needed. This paper develops a statistical reliability analysis procedure to investigate the reliability characteristics of multiple failure causes under independent competing risks. We mainly consider the case when the lifetime data follow log-location-scale distributions and may also be right-censored. Maximum likelihood (ML) estimators of unknown parameters are derived by applying the Newton–Raphson method. With the large-sample assumption, the normal approximation of the ML estimators is used to construct the asymptotic confidence intervals in which the standard error of the variance-covariance matrix is calculated by using the delta method. In particular, the Akaike information criterion is utilized to determine the appropriate fitted distribution for each cause of failure. An illustrative numerical experiment about the fuel cell engine (FCE) is presented to demonstrate the feasibility and effectiveness of the proposed model. The results can facilitate continued advancement in reliability prediction and reliability allocation for FCE, and also provide theoretical basis for the application of reliability concepts to many other complex systems. Copyright © 2014 John Wiley & Sons, Ltd.

[1]  V T Farewell,et al.  The analysis of failure times in the presence of competing risks. , 1978, Biometrics.

[2]  Erhard Cramer,et al.  Progressively Type-II censored competing risks data from Lomax distributions , 2011, Comput. Stat. Data Anal..

[3]  E. Kaplan,et al.  Nonparametric Estimation from Incomplete Observations , 1958 .

[4]  Wei Huang,et al.  Reliability analysis of electronic devices with multiple competing failure modes involving performance aging degradation , 2003 .

[5]  Debasis Kundu,et al.  Analysis of incomplete data in presence of competing risks among several groups , 2006, IEEE Transactions on Reliability.

[6]  Sumit Kumar,et al.  On progressively censored competing risks data for Weibull distributions , 2009, Comput. Stat. Data Anal..

[7]  Uditha Balasooriya,et al.  Competing causes of failure and reliability tests for Weibull lifetimes under type I progressive censoring , 2004, IEEE Transactions on Reliability.

[8]  Keming Yu,et al.  New Inference for Constant-Stress Accelerated Life Tests With Weibull Distribution and Progressively Type-II Censoring , 2014, IEEE Transactions on Reliability.

[9]  Gareth E. Haslam,et al.  Assessing fuel cell vehicle innovation and the role of policy in Japan, Korea, and China , 2012 .

[10]  Debanjan Mitra,et al.  Likelihood Inference Based on Left Truncated and Right Censored Data From a Gamma Distribution , 2013, IEEE Transactions on Reliability.

[11]  H. Akaike.  A new look at the statistical model identification , 1974 .

[12]  Robert K. Dixon,et al.  Development and demonstration of fuel cell vehicles and supporting infrastructure in China , 2011 .

[13]  Chanseok Park,et al.  Parametric inference of incomplete data with competing risks among several groups , 2004, IEEE Transactions on Reliability.

[14]  R. A. Leibler,et al.  On Information and Sufficiency , 1951 .

[15]  Tie Chen,et al.  STATISTICAL RELIABILITY ANALYSIS FOR FUEL CELL VEHICLE BASED ON FIELD TEST DATA , 2014 .

[16]  R. J. Herman,et al.  Maximum Likelihood Estimation For Multi-Risk Model , 1971 .

[17]  Wei Xie,et al.  Maximizing system availability through joint decision on component redundancy and spares inventory , 2014, Eur. J. Oper. Res..

[18]  N. Sugiura.  Further analysts of the data by akaike' s information criterion and the finite corrections , 1978 .

[19]  T. Ishioka,et al.  Maximum likelihood estimation of Weibull parameters for two independent competing risk , 1991 .