Multiple Failure Types Research Articles

Quantification of metallic coating failure on carbon fiber reinforced plastics using acoustic emission

Carbon fiber reinforced plastic substrates were subsequently coated with an electrochemically applied nickel and an electroplated copper layer. The coating-substrate system was loaded in four-point bending and the acoustic emission from coating failure was recorded during loading. The acoustic emission signals were analyzed using pattern recognition and frequency analysis techniques. This approach yields three distinguishable types of acoustic emission signals, which are correlated to three different failure mechanisms: i) nickel cracking ii) copper cracking and iii) delamination between the two coating layers. To confirm the correlation between the types of acoustic emission signals and the respective failure mechanisms and to assess the validity of the acoustic emission method to describe mechanical failure, the micro-mechanical fracture energies released during mechanical loading were calculated based on microscopic measurements of the crack progress utilizing scanning electron microscopy and scanning acoustic microscopy. These energies were compared to the associated acoustic emission signals energy for each failure mechanism. We found the calculated micro-mechanical energy values to be proportional to the measured accumulated acoustic emission energy of the associated acoustic emission signal type. We conclude that the reported failure classification method offers the possibility to compare fracture toughness values in multilayered coatings with multiple failure types, derived solely from acoustic emission energies.

A Joint Model for Longitudinal Measurements and Survival Data in the Presence of Multiple Failure Types

In this article we study a joint model for longitudinal measurements and competing risks survival data. Our joint model provides a flexible approach to handle possible nonignorable missing data in the longitudinal measurements due to dropout. It is also an extension of previous joint models with a single failure type, offering a possible way to model informatively censored events as a competing risk. Our model consists of a linear mixed effects submodel for the longitudinal outcome and a proportional cause-specific hazards frailty submodel (Prentice et al., 1978, Biometrics 34, 541-554) for the competing risks survival data, linked together by some latent random effects. We propose to obtain the maximum likelihood estimates of the parameters by an expectation maximization (EM) algorithm and estimate their standard errors using a profile likelihood method. The developed method works well in our simulation studies and is applied to a clinical trial for the scleroderma lung disease.

Testing treatment effects in the presence of competing risks

Competing risks are often encountered in clinical research. In the presence of multiple failure types, the time to the first failure of any type is typically used as an overall measure of the clinical impact for the patients. On the other hand, use of endpoints based on the type of failure directly related to the treatment mechanism of action allows one to focus on the aspect of the disease targeted by treatment. We review the methodology commonly used for testing failure specific treatment effects. Simulation results demonstrate that the cause-specific log-rank test is robust (in the sense of preserving the nominal level of the test) and has good power properties for testing for differences in the marginal latent failure-time distributions, whereas the use of a popular cumulative incidence based approach may be problematic for this aim.

Optical layer survivability-an implementation perspective

This paper looks at several aspects of optical layer protection techniques from an implementation perspective. We discuss the factors that affect the complexity of optical protection schemes, such as supporting mesh instead of ring protection, handling low-priority traffic, and dealing with multiple types of failures. The paper also looks at how the client layer interacts with the optical layer with respect to protection, in terms of how client connections are mapped into the optical layer, and how protection schemes in both layers can work together in efficient ways. Finally, we describe several interesting optical protection implementations, focusing on the ones that are different from conventional SONET-like implementations.

Modeling Rail Fatigue Behavior with Multiple Hazards

Derailments can be caused by rail fatigue failures, which are of great concern to railroad companies. These companies, therefore, have as an objective to predict the locations and times of potential failures and to perform preventive maintenance. Understanding the relationship between fatigue failures and factors affecting fatigue is essential to estimate when and where rail fatigue failures will occur. This paper presents a statistical rail fatigue model in four stages: 1) the model describes the relationship between probability of failure and usage; 2) the model extends to cases of multiple types of failures; 3) the model is derived for discrete usage categories or time intervals with a continuous spatial representation of rail units; and 4) explanatory variables capturing geometry, maintenance, and material properties are included to explain differences in fatigue behavior among heterogeneous rail units. Using available rail fatigue data, the model is estimated. The modeling framework is useful for other multitype failure behavior and for reliability problems, such as vehicle breakdown.

A SEMIPARAMETRIC MIXTURE MODEL FOR THE ANALYSIS OF COMPETING RISKS DATA

SummaryThis paper deals with the regression analysis of failure time data when there are censoring and multiple types of failures. We propose a semiparametric generalization of a parametric mixture model of Larson & Dinse (1985), for which the marginal probabilities of the various failure types are logistic functions of the covariates. Given the type of failure, the conditional distribution of the time to failure follows a proportional hazards model. A marginal like lihood approach to estimating regression parameters is suggested, whereby the baseline hazard functions are eliminated as nuisance parameters. The Monte Carlo method is used to approximate the marginal likelihood; the resulting function is maximized easily using existing software. Some guidelines for choosing the number of Monte Carlo replications are given. Fixing the regression parameters at their estimated values, the full likelihood is maximized via an EM algorithm to estimate the baseline survivor functions. The methods suggested are illustrated using the Stanford heart transplant data.

A machine interference problem with multiple types of failures

SUMMARY A numerical method for obtaining equilibrium performance measures for a single group of N identical machines, each subject to k(>2) types of failures, is presented. The time intervals between breakdowns of a machine are exponentially distributed. The mean time between type i failures is 1/&λi , i = 1, 2, ⃛, k. The service times required to repair any type of failure can be either deterministic, exponential, hypo-exponential, or hyper-exponential random variables, with different means for different types of failures. The service discipline is a non-preemptive fixed-priority rule, with different priorities assigned to different types of failures. System performance measures, such as average time spent by the machines waiting for service, average number of idle machines and machine operator utilization are obtained through imbedded Markov chain analysis. The algorithms used to obtain these measures exploit the special structure of the one-step transition probability matrix. The sensitivity of the performance measures to the density function of the service times and the priority assignments given is examined.