Extending the skill test for disease diagnosis.

For diagnostic tests, we present an extension to the skill plot introduced by Briggs and Zaretski (Biometrics 2008; 64:250-261). The method is motivated by diagnostic measures for osteopetrosis in a study summarized by Hans et al. (The Lancet 1996; 348:511-514). Diagnostic test accuracy is typically defined using the area (or partial area) under the receiver operator characteristic (ROC) curve. If partial area is used, the resulting statistic can be highly subjective because the focus region of the ROC curve corresponds to a set of low false-positive rates that are chosen by the experimenter. This paper introduces a more objective diagnostic test for which the focus region depends on a skill score, which in turn depends on the loss functions associated with misdiagnosis. More specifically, the skill-based diagnostic test serves as a more objective version of the nonparametric test introduced by Dodd and Pepe (Biometrics 2003; 59:614-623).

Estimating Load-Sharing Properties in a Dynamic Reliability System.

An estimator for the load share parameters in an equal load-share model is derived based on observing k-component parallel systems of identical components that have a continuous distribution function F (.) and failure rate r(.). In an equal load share model, after the first of k components fails, failure rates for the remaining components change from r(t) to gamma(1)r(t), then to gamma(2)r(t) after the next failure, and so on. On the basis of observations on n independent and identical systems, a semiparametric estimator of the component baseline cumulative hazard function R = - log(1 - F) is presented, and its asymptotic limit process is established to be a Gaussian process. The effect of estimation of the load-share parameters is considered in the derivation of the limiting process. Potential applications can be found in diverse areas, including materials testing, software reliability and power plant safety assessment.

Reliability estimation based on system data with an unknown load share rule.

We consider a multicomponent load-sharing system in which the failure rate of a given component depends on the set of working components at any given time. Such systems can arise in software reliability models and in multivariate failure-time models in biostatistics, for example. A load-share rule dictates how stress or load is redistributed to the surviving components after a component fails within the system. In this paper, we assume the load share rule is unknown and derive methods for statistical inference on load-share parameters based on maximum likelihood. Components with (individual) constant failure rates are observed in two environments: (1) the system load is distributed evenly among the working components, and (2) we assume only the load for each working component increases when other components in the system fail. Tests for these special load-share models are investigated.

Estimating distributions with increasing failure rate in an imperfect repair model.

A failed system is repaired minimally if after failure, it is restored to the working condition of an identical system of the same age. We extend the nonparametric maximum likelihood estimator (MLE) of a system's lifetime distribution function to test units that are known to have an increasing failure rate. Such items comprise a significant portion of working components in industry. The order-restricted MLE is shown to be consistent. Similar results hold for the Brown-Proschan imperfect repair model, which dictates that a failed component is repaired perfectly with some unknown probability, and is otherwise repaired minimally. The estimators derived are motivated and illustrated by failure data in the nuclear industry. Failure times for groups of emergency diesel generators and motor-driven pumps are analyzed using the order-restricted methods. The order-restricted estimators are consistent and show distinct differences from the ordinary MLEs. Simulation results suggest significant improvement in reliability estimation is available in many cases when component failure data exhibit the IFR property.