Bayesian Analysis of Masked Series System Lifetime Data

The problem of analyzing series system lifetime data with masked or partial information on cause of failure is recent, compared to that of the standard competing risks model. A generic Gibbs sampling scheme is developed in this article towards a Bayesian analysis for a general parametric competing risks model with masked cause of failure data. The masking probabilities are not subjected to the symmetry assumption and independent Dirichlet priors are used to marginalize these nuisance parameters. The developed methodology is illustrated for the case where the components of a series system have independent log-Normal life distributions by employing independent Normal-Gamma priors for these component lifetime parameters. The Gibbs sampling scheme developed for the required analysis can also be used to provide a Bayesian analysis of data arising from the conventional competing risks model of independent log-Normals, which interestingly has so far remained by and large neglected in the literature. The developed methodology is deployed to analyze a masked lifetime data of PS/2 computer systems.     

ANALYSIS OF MASKED DATA IN A DEPENDENT COMPETING RISKS MODEL UNDER UNKNOWN ENVIRONMENT

ABSTRACT

System failure data is often analyzed to estimate component reliabilities. Due to cost and time constraints, the exact component causing the failure of the system cannot be identified in some cases. This phenomenon is called masking. Further, it is sometimes necessary for us to take account of the influence of the operating environment. Here we consider a series system, operating under unknown environment, of two components whose failure times follow the Marshall-Olkin bivariate exponential distribution. We present a maximum likelihood approach for obtaining estimators from the masked data for this system. From a simulation study, we found that the relative errors of the estimates are almost well behaved even for small or moderate expected number of systems whose cause of failure is identified.     

Constant Stress Accelerated Life Test on a Multiple-Component Series System under Weibull Lifetime Distributions

We will discuss the reliability analysis of the constant stress accelerated life test on a series system connected with multiple components under independent Weibull lifetime distributions whose scale parameters are log-linear in the level of the stress variable. The system lifetimes are collected under Type I censoring but the components that cause the systems to fail may or may not be observed. The data are so called masked for the latter case. Maximum likelihood approach and the Bayesian method are considered when the data are masked. Statistical inference on the estimation of the underlying model parameters as well as the mean time to failure and the reliability function will be addressed. Simulation study for a three-component case shows that Bayesian analysis outperforms the maximum likelihood approach especially when the data are highly masked.     

Accelerated life tests for log-normal series system with dependent masked data under Type-I progressive hybrid censoring

This article considers the constant stress accelerated life test for series system products, where independent log-normal distributed lifetimes are assumed for the components. Based on Type-I progressive hybrid censored and masked data, the expectation-maximization algorithm is applied to obtain the estimation for the unknown parameters, and the parametric bootstrap method is used for the standard deviation estimation. In addition, Bayesian approach combining latent variable with Gibbs sampling is developed. Further, the reliability functions of the system and components are estimated at use stress level. The proposed method is illustrated through a numerical example under different masking probabilities and censoring schemes.     

Bayesian Analysis of Masked Data in Step-stress Accelerated Life Testing

This article considers a k level step-stress accelerated life testing (ALT) on series system products, where independent Weibull-distributed lifetimes are assumed for the components. Due to cost considerations or environmental restrictions, causes of system failures are masked and type-I censored observations might occur in the collected data. Bayesian approach combined with auxiliary variables is developed for estimating the parameters of the model. Further, the reliability and hazard rate functions of the system and components are estimated at a specified time at use stress level. The proposed method is illustrated through a numerical example based on two priors and various masking probabilities.     

Robust Bayesian analysis for parallel system with masked data under inverse Weibull lifetime distribution

Abstract

This paper considers the statistical analysis of masked data in a parallel system with inverse Weibull distributed components under type II censoring. Based on Gamma conjugate prior, the Bayesian estimation as well as the hierarchical Bayesian estimation for the parameters and the reliability function of system are obtained by using the Bayesian theory and the hierarchical Bayesian method. Finally, Monte Carlo simulations are provided to compare the performances of the estimates under different masking probabilities and effective sample sizes.     

Bayesian analysis of competing risks with partially masked cause of failure

Summary. Bayesian analysis of system failure data from engineering applications under a competing risks framework is considered when the cause of failure may not have been exactly identified but has only been narrowed down to a subset of all potential risks. In statistical literature, such data are termed masked failure data. In addition to masking, failure times could be right censored owing to the removal of prototypes at a prespecified time or could be interval censored in the case of periodically acquired readings. In this setting, a general Bayesian formulation is investigated that includes most commonly used parametric lifetime distributions and that is sufficiently flexible to handle complex forms of censoring. The methodology is illustrated in two engineering applications with a special focus on model comparison issues.     

The Posterior Distribution of the Parameters of Component Lifetimes Based on Autopsy Data in a Shock Model

In this paper we consider a binary, monotone system whose component states are dependent through the possible occurrence of independent common shocks, i.e. shocks that destroy several components at once. The individual failure of a component is also thought of as a shock. Such systems can be used to model common cause failures in reliability analysis. The system may be a technological one, or a human being. It is observed until it fails or dies. At this instant, the set of failed components and the failure time of the system are noted. The failure times of the components are not known. These are the so-called autopsy data of the system. For the case of independent components, i.e. no common shocks, Meilijson (1981), Nowik (1990), Antoine et al . (1993) and GTsemyr (1998) discuss the corresponding identifiability problem, i.e. whether the component life distributions can be determined from the distribution of the observed data. Assuming a model where autopsy data is known to be enough for identifia bility, Meilijson (1994) goes beyond the identifiability question and into maximum likelihood estimation of the parameters of the component lifetime distributions based on empirical autopsy data from a sample of several systems. He also considers life-monitoring of some components and conditional life-monitoring of some other. Here a corresponding Bayesian approach is presented for the shock model. Due to prior information one advantage of this approach is that the identifiability problem represents no obstacle. The motivation for introducing the shock model is that the autopsy model is of special importance when components can not be tested separately because it is difficult to reproduce the conditions prevailing in the functioning system. In Gåsemyr & Natvig (1997) we treat the Bayesian approach to life-monitoring and conditional life- monitoring of components     

Identifiability of masking probabilities in competing risks models with emphasis on Weibull models

Abstract

Lifetime data with masked failure causes arise in both reliability engineering and epidemiology. The phenomenon of masking occurs when a subject is exposed to multiple risks. A competing risks model with masking probabilities is widely used for the masked failure data. However, in many cases, the model suffers from an identification problem. We show that the identifiability of masking probabilities depends on both the structure of data and the cause-specific hazard functions. Motivated by this result, two existing solutions are reviewed and further improved.     

The random partition masking model for interval-censored and masked competing risks data

We study the nonparametric maximum likelihood estimate (NPMLE) of the cdf or sub-distribution functions of the failure time for the failure causes in a series system. The study is motivated by a cancer research data (from the Memorial Sloan-Kettering Cancer Center) with interval-censored time and masked failure cause. The NPMLE based on this data set suggests that the existing masking models are not appropriate. We propose a new model called the random partition masking model, which does not rely on the commonly used symmetry assumption (namely, given the failure cause, the probability of observing the masked failure causes is independent of the failure time; see Flehinger et al. Inference about defects in the presence of masking, Technometrics 38 (1996), pp. 247–255). The RPM model is easier to implement in simulation studies than the existing models. We discuss the algorithms for computing the NPMLE and study its asymptotic properties. Our simulation and data analysis indicate that the NPMLE is feasible for a moderate sample size.     

Maximum likelihood analysis of masked series system lifetime data

Component lifetime parameters of a series system are estimated from system lifetimes and masked cause of failure observations. The time and cause of system failures are assumed to follow a competing risks model. The masking probabilities of the minimum random subsets are not subjected to the symmetry assumption. Sufficient regularity conditions are provided, justifying the maximum likelihood analysis. Maximum likelihood estimates of both the lifetime parameters and masking probabilities are generically computed via an EM algorithm. An appropriate set of asymptotically pivotal quantities are also derived. Such maximum likelihood based estimates are further refined by bootstrap. The developed techniques are illustrated by numerical examples of independent Weibull component lifetimes with distinct scale and shape parameters.     

Inference for a series system with dependent masked data under progressive interval censoring

This paper considers the statistical analysis of masked data in a series system with Burr-XII distributed components. Based on progressively Type-I interval censored sample, the maximum likelihood estimators for the parameters are obtained by using the expectation maximization algorithm, and the associated approximate confidence intervals are also derived. In addition, Gibbs sampling procedure using important sampling is applied for obtaining the Bayesian estimates of the parameters, and Monte Carlo method is employed to construct the credible intervals. Finally, a simulation study is proposed to illustrate the efficiency of the methods under different removal schemes and masking probabilities.     

Consistency of the Generalized MLE with Interval-Censored and Masked Competing Risks Data

We consider nonparametric estimation based on interval-censored competing risks data with masked failure cause. The generalized maximum likelihood estimator of the joint survival function of the failure time and the failure cause is studied under mixed case interval censorship and random partition masking. Strong consistency in the L ₁(μ)-topology is established for some finite measure μ which is derived from the joint censoring and masking distribution. Under additional regularity assumptions we also establish the strong consistencies in the topologies of weak convergence, point-wise convergence, and uniform convergence.     

Bayesian reliability when system and subsystem failure data are obtained in the same time period

Previously, Bayesian anomaly was reported for estimating reliability when subsystem failure data and system failure data were obtained from the same time period. As a result, a practical method for mitigating Bayesian anomaly was developed. In the first part of this paper, however, we show that the Bayesian anomaly can be avoided as long as the same failure information is incorporated in the model. In the second part of this paper, we consider a problem of estimating the Bayesian reliability when the failure count data on subsystems and systems are obtained from the same time period. We show that Bayesian anomaly does not exist when using the multinomial distribution with the Dirichlet prior distribution. A numerical example is given to compare the proposed method with the previous methods.     

Parametric Modeling for Survival with Competing Risks and Masked Failure Causes

We consider a life testing situation in which systems are subject to failure from independent competing risks. Following a failure, immediate (stage-1) procedures are used in an attempt to reach a definitive diagnosis. If these procedures fail to result in a diagnosis, this phenomenon is called masking. Stage-2 procedures, such as failure analysis or autopsy, provide definitive diagnosis for a sample of the masked cases. We show how stage-1 and stage-2 information can be combined to provide statistical inference about (a) survival functions of the individual risks, (b) the proportions of failures associated with individual risks and (c) probability, for a specified masked case, that each of the masked competing risks is responsible for the failure. Our development is based on parametric distributional assumptions and the special case for which the failure times for the competing risks have a Weibull distribution is discussed in detail.     

Optimal number of components in a load-sharing system for maximizing reliability

A $k$-out-of-$n$:G load sharing system is a cluster of $n$ components designed to withstand a certain amount of load in field operation, working only if no fewer than $k$ components work. Previous research on a load sharing system has focused on predicting the time-independent reliability from the stress–strength model or estimating the unknown parameters of the time-dependent reliability for a given load sharing rule. Differently, in this paper, we consider the problem of determining the optimal $n$ to maximize the reliability of both $n$-out-of-$n$:G and ($n$–$1$)-out-of-$n$:G load sharing systems. Since the load of each component decreases in $n$, the proportional hazard model is employed to relate the component failure rate with the load, assuming that the components, which have exponential distributions for given loads, are independent of each other. We then derive a sufficient condition under which a smaller number of components each withstanding a high load is preferred to a larger number of components each withstanding a small load. A numerical example is given for the rocket propulsion system to illustrate the result.     

Assessing component reliability using lifetime data from systems

ABSTRACT

In this paper, we introduce a competing risks model for the lifetimes of components that differs from the classical competing risks models by the fact that it is not directly observable which component has failed. We propose two statistical methods for estimating the reliability of components from failure data on a system. Our methods are applied to simulated failure data, in order to illustrate the performance of the methods.     

Bayesian estimation of reliability using time-to-failure distribution of parametric degradation models

This paper focusses on computing the Bayesian reliability of components whose performance characteristics (degradation – fatigue and cracks) are observed during a specified period of time. Depending upon the nature of degradation data collected, we fit a monotone increasing or decreasing function for the data. Since the components are supposed to have different lifetimes, the rate of degradation is assumed to be a random variable. At a critical level of degradation, the time to failure distribution is obtained. The exponential and power degradation models are studied and exponential density function is assumed for the random variable representing the rate of degradation. The maximum likelihood estimator and Bayesian estimator of the parameter of exponential density function, predictive distribution, hierarchical Bayes approach and robustness of the posterior mean are presented. The Gibbs sampling algorithm is used to obtain the Bayesian estimates of the parameter. Illustrations are provided for the train wheel degradation data.     

Early inference on reliability of upgraded automotive components by using past data and technical information

When a new product is the result of design and/or process improvements introduced in its predecessors, then the past failure data and the expert technical knowledge constitute a valuable source of information that can lead to a more accurate reliability estimate of the upgraded product. This paper proposes a Bayesian procedure to formalize the prior information available about the failure probability of an upgraded automotive component. The elicitation process makes use of the failure data of the past product, the designer information on the effectiveness of planned design/process modifications, information on actual working conditions of the upgraded component and, for outsourced components, technical knowledge on the effect of possible cost reductions. By using the proposed procedure, more accurate estimates of the failure probability can arise. The number of failed items in a future population of vehicles is also predicted to measure the effect of a possible extension of the warranty period. Finally, the proposed procedure was applied to a case study and its feasibility in supporting reliability estimation is illustrated.     

On Non-parametric Estimation of the Survival Function with Competing Risks

The non-parametric maximum likelihood estimators (MLEs) are derived for survival functions associated with individual risks or system components in a reliability framework. Lifetimes are observed for systems that contain one or more of those components. Analogous to a competing risks model, the system is assumed to fail upon the first instance of any component failure; i.e. the system is configured in series. For any given risk or component type, the asymptotic distribution is shown to depend explicitly on the unknown survival function of the other risks, as well as the censoring distribution. Survival functions with increasing failure rate are investigated as a special case. The order restricted MLE is shown to be consistent under mild assumptions of the underlying component lifetime distributions.