Tests of Increased Severity

Publisher Website
Google Scholar
Abstract
Certain manufactured items must be produced of such quality that only a very small fraction will fail when used. In problems of this sort standard sampling plans are impractical because of the excessively large sample sizes required. In this paper we consider the statistical questions involved in developing an efficient sampling plan for such problems, and recommend a procedure which has been found very useful in biological assay and in the testing of explosives. There is good reason to believe that the same statistical techniques can also be used in many other fields in which one is interested in studying the response of objects tested as a function of the strength of the stimuli to which they are subjected.

This publication has 0 references indexed in Scilit: