ESTIMATION OF CURRENT RELIABILITY IN A DUANE-BASED RELIABILITY-GROWTHMODEL

In this article, I investigate the statistical inference: concerning t he current (field-stage) reliability of a reliability growth model. Th e model, assuming a step-intensity structure, evolves from the physica l consideration of the Duane learning-curve property and incorporates the effect of a test-analyze-and-fix program that is typically underta ken in a developmental testing program. Both exact and large-sample di stributional results are derived for the maximum likelihood and the le ast squares estimators of the current intensity. Under the assumption that the step-intensity model represents the reality, I provide an ass essment of the extent of ''misspecification'' when the widely used pow er law process model is fit to the failure data of a system experienci ng recurrent failures. Extensive simulation results are carried out to supplement the theoretical findings. An illustration with a dataset i s provided as a demonstration of the application of the inference resu lts.