[ | Next | Previous | Up ]

C=0  Small sample Size question.

From: steven.case@bradleycorp.com
Date: 11/20/2006
Time: 12:16:50 PM

Comments

I've been shown a C=0 sampling plan that can be found at:

http://www.wisc-online.com/objects/index_tj.asp?objid=QLT1804

I have issues with the smallness of the samples, for example: For a lot size of 40 and an AQL of 2.5 it yields a sample size of 5. How does that find a lot with 2.5% defectives very often?

If I plug those numbers into a hypergeometric calculator such as the one at:

http://www.stattrek.com/Tables/Hypergeometric.aspx

as follows:

Population size 40 Sample size 5 Number of successes in population 1 Number of successes in sample 0 Hypergeometric probability .875 Cumulative probability .875

Then it follows that (1-.875)x100=12.5% not close to a 95% confidence level that I would expect a plan to achieve.

So obviously I don't understand something, and I'm at a loss as to what to tell people who want to use that C=0 plan as an incoming inspection of purchased parts standard.


Last changed: November 20, 2007