Statistical analysis relating to recent Federal Standard 209 (cleanrooms) revisions

Abstract

The recent revisions of Federal Standard 209 regarding cleanrooms include two statistical tests for the concentration data obtained from sampling airborne particles: a) that none of the location averages exceeds the Class Limit (CL) and b) that there be 95 percent confidence that the true mean of the location averages does not exceed the CL, based on the data making up the grand mean of averages. For sampling from a constant and uniform concentration (Poisson data) or from a concentration giving data that are normally distributed, we show the statistical implications of the 95 percent confidence requirement on the grand mean.