OPTIMAL SELECTION OF THE MOST RELIABLE DESIGN WITH A RECIPROCAL WEIBULL DEGRADATION RATE

Abstract

ABSTRACT In industries, the manufacturer of a product is usually confronted with the problem of selecting the most reliable design among several competing designs for some parts (or components) at the research and development stage. It is not easy to conduct such a selection work for highly reliable products, since there are few (or even no) failures can be obtained by using traditional life tests or accelerated life tests. In such a case, degradation tests will be effective techniques to assess the products' reliability, if degradation measures relating to reliability can be taken over time. Several experimental elements (e.g., the inspection frequency, the sample size and the termination time) are closely related to the experimental cost and the selection precision of a selection experiment. The main purpose of this paper is to deal with the optimal design problem of selecting the most reliable design with a reciprocal Weibull degradation rate. First, an intuitively appealing selection rule is proposed. Then, under the constraint that the selection precision is not lower than a pre-determined level, the optimal experimental setting (including the sample size, inspection frequency, and the termination time needed by the selection rule for each of competing designs) is determined such that the total experimental cost is minimized. An example is provided to illustrate the proposed method. Finally, we also compare the optimal experimental settings obtained in the present paper with those for a lognormal degradation rate studied in Yu [26].