Is the use of a sample size of SQRT(n)+1 an appropriate alternative to AQL? And what is the statistical basis? Is their any reference to a standard that applies?

Applicable situation is testing 50 bottles of chemicals for identity. No rejects acceptable so AQL not appropriate.

The problem with acceptance numbers greater than zero that are used with most AQl sampling plans can be eliminated with a c = 0 sampling plan. Take a look at the Nicholas Squegila Zero Acceptance Number Sampling Plans. It is available from ASQ Press in Milwaukee, WI.

Thank you Rick for your link to zero acceptance plans.
I found more to my question on the Dr W.A.Taylor web-site www.variation.com (http://www.variation.com) (look at FAQ).

EDITOR NOTE: Hot linked web url

[This message has been edited by Marc Smith (edited 17 December 2000).]