[ | Next | Previous | Up ]

Re: Sampling for Destructive Tests

From: Stan Hilliard
Date: 10/14/2005
Time: 10:23:35 PM

Comments

The fact that the test is destructive does not effect the statistics. But you would want to have the most efficient decision rule possible so as not to waste product.

The best way to evaluate the probabilities for an acceptance plan is with the oc curve.

www.samplingplans.com/usingoccurves.htm

The probabilities will not look good with only 4 or 5 parts per lot. Attribute Sample Planner (TP105 v3.0 $245) will do a lot of analysis for designing a plan. It is not described on the website yet but the version it replaces, v1.22, is described here:

www.samplingplans.com/programtp105.htm


Last changed: November 20, 2007