scientific article; zbMATH DE number 4113806

From MaRDI portal
Publication:4729207





[For the entire collection see Zbl 0676.00016.] When inference concerning the life length of highly reliable items is required, often the test is constructed under more severe stress conditions. Such a test is called an overstress or accelerated life test (ALT) and is usually undertaken to save the time and cost of testing. Here we present a Bayesian perspective on accelerated life testing and present the initial results of two different directions of research. By adopting a Bayesian perspective of ALT, we enable the incorporation of relevant prior information into the inference process as well as provide a coherent basis for making inferences from such tests. In section 2 we define the ALT problem and introduce the necessary notation. In section 3 we discuss the application of the Kalman filter (KF) to ALT, under three testing scenarios. In section 4 we discuss the use of the beta and Dirichlet distributions for modeling the failure rates of the items tested at the different stress levels.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4729207)