AbstractPerusal of quality- and reliability-engineering literature indicates some confusion over the meaning accelerated life testing (ALT), highly (HALT), stress screening (HASS), auditing (HASA). In addition, there is a significant conflict between as part an iterative process finding removing defects means estimating or predicting product reliability. We review basics these methods describe how they relate to statistical for estimation prediction reliability growth. also outline potential...

Abstract The exponential distribution is inadequate as a failure time model for most components; however, under certain conditions (in particular, that component rates are small and mutually independent, failed components immediately replaced or perfectly repaired), it applicable to complex repairable systems with large numbers of in series, regardless distributions, shown by Drenick 1960. This result implies system behavior may become simpler more added. We review necessary the present some...

One common challenge of modeling intersection related crash data is the high proportion sites with zero crashes. Extensive research has been done on appropriate methods to handle excess zeroes. There some reluctance use zero-inflated models in traffic safety literature. The primary purpose this paper evaluate determine if they are a suitable method for counts. An approach model selection choose that best accomplishes objectives rather than attempting discover true underlying generating...

To successfully determine if a system meets some reliability standard, typically assurance tests are conducted on the full system. This proven approach is extremely effective. However, it can be cost prohibitive when expensive. Previous work has incorporated component and subsystem data to supplement prior distributions before test. Although this helpful, we propose an additional advance of allowing as part We show that by augmenting with tests, in cases, reduce For decomposed into...

Modern complex engineering systems often present the analyst with a mix of data types that can be used for reliability prediction: system test results, lifetime from unit tests components, and subsystem data, all which may have predictive value lifetime. We hierarchical nonparametric framework, using Dirichlet processes, in time-to-event distributions estimated sample or derived based on physical failure mechanisms. By applying Bayesian methodology, framework incorporate prior information,...

Semi–Markov processes (SMPs) provide a rich framework for many real–world problems. However, owing to difficulty in implementing practical solutions they are rarely used with their full capability. The theory of SMPs is quite mature but was mainly developed at time when computational resources were not widely available. With the exception some simplest cases, inherently numerical, and have been underutilised by practitioners because applications. This paper demonstrates methods needed...

We discourage the use of a diagnostic for normality: interquartile range divided by standard deviation. This statistic has been suggested in several introductory statistics books as method to assess normality. Through simulation, we explore rate at which this converges its asymptotic normal distribution, and actual size tests based on distribution sample sizes. show that there are nonnormal distributions from cannot detect difference. Additionally, power test normality is quite poor when...

The Weibull distribution has long been a popular choice for modeling lifetime data of various mechanical and biological phenomena when the associated hazard rate function is constant or monotone increasing decreasing. However, nonmonotone functions are common in reliability survivability contexts where system may undergo an initial "burn-in" prior to periods useful life eventual wear out. In these scenarios, can only model portion "bathtub" curve but incapable adequately entire failure...

Chronic kidney disease (CKD) affects many lives and has a large impact on health systems around the world. To better understand predict costs for insurance plan people with CKD in United States, we built new model of their individual costs. Our is first to explicitly both stage transition process distribution given those stages. Additionally, it incorporates numerous covariates comorbidities. We applied models two rich datasets, one commercial other Medicare fee-for-service, totaling about...

Test planners have long sought the ability to incorporate results of highly accelerated life testing (HALT) into an early estimate system reliability. While case studies attest effectiveness HALT in producing reliable products, capability translate test's limited failure data a meaningful measure reliability improvement remains elusive. Further, review quality and literature indicates that confusion exists over what defines how differs from quantitative methods. Despite many authors making...

Statistical flowgraph models have proven useful for analysis and modeling of complex systems viewed as multistate processes that lead to outcomes such degraded operation or failure. This article provides an engineering-oriented introduction statistical models: system representation, setting up a model, parameter estimation, solution the model (using either frequentist Bayesian approach), interpretation outputs. The method is illustrated with piping reliability in nuclear power plant,...

KEY POINTIncluding censored observations, when estimating correlations, has a nontrivial impact on the analysis.

All too often statisticians do not have access to raw experimental data. These scenarios require additional methodology properly account for the missing information. In this article, we demonstrate a technique analyzing averages of lifetime data collected at various conditions that provides inference factor effects. To handle these summaries, use some numerical techniques calculate probability density function average independent and identically distributed lognormal random variables. We...

Probability density estimation is a central task in statistics. Copula-based models provide great deal of flexibility modelling multivariate distributions, allowing for the specifications marginal distributions separately from dependence structure (copula) that links them to form joint distribution. Choosing class copula not trivial and its misspecification can lead wrong conclusions. We introduce novel random Bernstein functions, studied support behavior posterior The proposal based on...

We propose the attraction Indian buffet distribution (AIBD), a for binary feature matrices influenced by pairwise similarity information. Binary are used in Bayesian models to uncover latent variables (i.e., features) that explain observed data. The process (IBP) is popular exchangeable prior matrices. In presence of additional information, however, exchangeability assumption not reasonable or desirable. AIBD can incorporate yet it preserves many properties IBP, including total number...

Although exchangeable processes from Bayesian nonparametrics have been used as a generating mechanism for random partition models, we deviate this paradigm to explicitly incorporate clustering information in the formulation our model. Our shrinkage distribution takes any and shrinks its probability mass toward an anchor partition. We show how provides framework model hierarchically-dependent temporally-dependent partitions. The parameters control degree of dependence, accommodating at...

We present a Bayesian nonparametric system reliability model which scales well and provides great deal of flexibility in modeling. The approach naturally handles the disparate amounts component subsystem data that may exist. However, traditional models are quite computationally complex, relying on MCMC techniques. Our utilizes conjugate properties beta-Stacy process, is fundamental building block our model. These individual linked together using method moments estimation approach. This fast,...