Harrison M. Wadsworth

The Effects of Class Inspection Under MIL-STD-105D

A discussion of the effect of grouping defects by classes while using MIL-STD-105D is presented both in terms of acceptance probabilities and cost. Proposals are made for establishing more realistic AQL values dependent on the number of defects being gr..

A Modified Beattie Procedure Process Monitoring

Beattie (1962) devised a continuous process-monitoring procedure that does not require 100% inspection. The procedure is a composite of two cumulative sum procedures. He considered the case in which the variate under observation is discrete. Here we extend Beatties procedure to monitor a process when the observed variate is normally distributed. Additionally, we show how this procedure can be used to monitor two-sided variations of the parameter under observation. We use Walds approximating expressions for the sequential probability ratio test to obtain an approximate operating characteristic (OC) curve for the procedure. We also evaluate the accuracy of the OC-curve expression and use it to develop design guidelines. Two examples demonstrate the simplicity of the design rules.

Equivalency of alternative cotton dust samplers to the vertical elutriator.

The OSHA standard for measuring cotton dust permits the use of samplers which have been demonstrated to be equivalent to the vertical elutriator (VE). The standard does not include enough criteria for the equivalency of an alternative device to be determined. Furthermore, the impact of the inherent variability of the VE on the ability to establish equivalency is not addressed in the standard. We propose a protocol which will be statistically sound and, by means of which, the equivalency of any other device to the VE may be determined. We also assess the impact of the variability of the VE on the probability of demonstrating the equivalence of another VE.