An Innovative Model of Reliability—The Pseudo-Entropic Model

Abstract

There is an entire field of literature on reliability models. Building reliability models is most frequently done by starting from the failure rate (or hazard rate). Creating a reliability model starts with the specific type of product behavior over time, so there are model classes for the specifics of different product categories: electronic, mechanical products, mixed systems etc. The aim of this paper was to develop a statistical model which would allow the study of the durability of products, and particularly, in the present case, of electrical switches. The procedure has a broad range of applicability and can be extended to whole categories of products that have components both in motion, and therefore subject to wear, and also that bear additional stress. In our case, an electrical switch involves the shock of the electrical contact, which additionally requires the constituent material. This article started from an indicator similar to entropy (an entropy-like transformation) that was developed according to the stated purpose of the paper. We believe that the appropriate name for the proposed indicator is pseudo-entropic transformation, as we operated with the derivative of g(t), which is basically a probability density. The model developed herein is original and, from a practical point of view, it is convenient for treating and developing analytical and practical applications for classes of products subjected, during use, to a process of wear, degradation, and damage by use.