Year
1985
Abstract
Variables inspection can be used in the attribute mode, if an assumed falsification of reported data is sufficiently small so as to escape detection with the attribute tester, i.e., a variable tester can be used in order to make a qualitative statement. The efficiency of such a procedure can be determined if the statistical properties of the tester are known. - In this paper first, for one class of material data the probability of detection is determined as function of the total number of item data to be verified, inspector's sample size, fraction of falsified data and false alarm probability, both for drawing without and with replacement. For the latter case and given total falsification, the minimal, i.e. guaranteed probability of detection is determined. Second, for several classes of material data it is shown under some simplifying assumptions in which way the available inspection effort has to be distributed optimally on the various classes. Finally, some remarks are made about the use of two different measurement techniques for the verification of data of one class of material.