Group Testing with a Goal in Estimating the Number of Defects Under Imperfect Environmental Stress Screen Levels

Abstract

ABSTRACT A fixed sample size procedure is proposed to obtain initial and then refined estimates of defect density for Environmental Stress Screening (ESS) plans. Ideas of group testing, i.e., testing units in batches instead of individually, where each test only indicates whether the test batch contains only good units or whether it contains at least one defective, is applied to the problem of estimating the probability π of an arbitrary unit being defective. Five estimators are given with their root mean square errors or asymptotic variances. An example and tables of procedure parameters are given to illustrate the proposed method.