Bayesian techniques to reduce the sample size in automotive electronics attribute testing

This paper discusses the application of Bayesian techniques to the determination of sample sizes required for an attribute test of a product in order to demonstrate a target reliability with a specified confidence. The method is based on analyzing statistical data on similar products and incorporating them into a Bayesian prior distribution for the unknown reliability. A mixture prior obtained by combining a beta prior with the uniform rectangular prior (representing the unknown content of the new product design) is discussed. The suggested method can significantly lower sample sizes for attribute tests and thus reduce cost, time, and resources currently being spent on reliability demonstration testing. A numerical example at the end of the paper illustrates the method.