4.10.6.6 Lab Statistical Methods

[Marc]

Can anyone give me an example? Will someone please explain this to me? I'm lost here. HELP!

4.10.6.6 "Laboratory Statistical Techniques"

"Appropriate statistical techniques should be applied to verification activities whose deliverables are data."

Of course, this is from QS9000 - so should means shall....

[This message has been edited by Marc Smith (edited 02-07-99).]

[Marc]

Also see <a href="http://www.qs9000.com/ubb/Forum10/HTML/000022.html"> this thread</a>.

Marc,
I don't know the complete answer to this one. What I am currently doing is in my cal labs. I am running X-Bar & R charts on data from calibration of my transfer standards as a check to see that the Calibration Service does not shoot me some bogus readings. Of course this also shows trending for such things as wear. If I have to change a calibration service supplier for my standards it is also a benchmark for comparison to the new data from the new lab.

In my product test lab I have on occasion done histograms and capability plots when the data is otherwise difficult to understand, due to quanity of data collected. Do you think I am on the right track?

[Don Winton]

Do you think I am on the right track?

I am not being a QS expert, but I agree. From the description, when data is a deliverable, there may be a requirement to use statistical techniques to determine if the data is capable of meeting capability.

Marc, does this and my other post help any?

Regards,
Don

[Marc]

Yup - helps a lot. I appreciate the info. I still have some thinking to here, though.