From: Stan Hilliard
Date: 12 Sep 2001
Time: 01:54:01
Greetings Bart,
> YOU SAID -- To determine a convenient quality level in production a contractor too can examine the OC-curves. Because of cost-efficiency reasons, he doesn't want to have to correct too many "rejected lots" afterwards. In case he has only one lot to deliver, he could take the risk to lower his production quality. For example: He could choose the quality level that has a Pa of 70%. In a small series of 3 consecutive lots he has only 34% (0.7 X 0.7 X 0.7) chance that all lots should be accepted.
REMARK -- Maybe a new philosophy is called for here. Hold the contractor responsible for the quality of the population rather than the sample. Specify RQL = LQ = Upper Confidence Limit and the consumers risk (beta). Let the contractor pick any AQL he/she chooses. This should assure that -- for any lot that passes the plan -- the confidence limit requirement will be met. If you are still concerned that the contractor will try to slip something past, you can reduce the consumers risk to a safe point:
Beta = 0.1 --->0.01 ---> 0.001 ---> ... 0.000001
> YOU SAID -- Do you offer a tool to calculate the risk in treating an isolated lot as lot in a series of lots?
The only reason to think of a lot as in a series of lots is to apply switching rules. Sequential sampling can serve the same purpose of adapting the plan to lot quality without using switching rules.
www.samplingplans.com/switchingrules.htm
TP105 produces sampling plans equivalent to ISO 8422 but without being limited to a specific set of oc curves.
Stan Hilliard