RE: Tom Ryan's reply to the SPC theory point

["DeRoeck, Rich" <RDeRoeck@alphaind.com>]

That's a nice story...but as with most legends probably neither true or
accurate. I'm not for accepting/buying any defects but my point is that many
folks (i.e. auditors) think a c=0 plan allows for no defects in the lot. It
isn't so, as my example clearly demonstrates. Even if c=0, other things DO
matter. As to 100% inspection, I don't have the resources or staff to
implement such a strategy which, in any case, does not guarantee 100%
conforming product (as many studies have shown). I have been working with my
suppliers (mostly in Asia) to implement effective process controls (EMP
studies, Control Plans, Measurement Flowcharts, OCAPs) but haven't had much
success in the US.

Below is a quote from the book I referenced in my post that you might find
of interest.

"...a further common misconception that the sample must not contain any
nonconforming units for acceptance of the lot to be permitted. That is, the
only valid sampling plans are those with acceptance numbers c equal to zero.
... Implied by this promotion is that a sampling plan with c=0 allows no
nonconforming units to be passed on to the customer. How silly!...

Regards,

Rich DeRoeck       

-----Original Message-----
From: GrantBlair@aol.com [mailto:GrantBlair@aol.com]

I'm reminded of the "urban legend" about a customer who specifed a quality 
level of 3 defects per thousand in his contract. When he received his first 
shipment of 1000 parts, it included 3 parts in a separate bag, along with an
apologetic letter from his Japanese supplier explaining that, although he 
didn't understand the strange request, the required number of defects were 
produced, and included, as requested.