Is there a Standard defining what a valid sampling plan is?

[Dawn]

I am looking for info on standards stating what makes for a valid sampling plan. Does ANSI have this, and if so can you tell me how to get it?
Thanks!

 

[Lassitude]

I don't know about any such ANSI standard. Maybe you should post a copy of this in the Statistics forum. Sorry - can't help you.

 

[Roger Eastin]

There are several sampling "standards" that you can get. I guess it depends on what you are looking for. MIL-STD-105 E was a standard for years (I don't know if it has been superseded or not). If this is still available, ANSI would have it. You can check its availability at //global.ihs.com. At any rate, ASQ has several publications that will give you valid sampling plans. What is your application?

 

[Dawn]

I need to prove to a QS auditor that a 3 pc. check is a statistcally valid sampling plan. I also need to prove to him that a six pierce final for any amount of pieces is a valid sampling plan. I'm not sure I can do this. He told me to get the standard from Ansi. They have it all figured out for you. I can't seem to find anything. Thanks!!!

 

[TheOtherMe]

I'm sure others will have some direct stuff, but here are some threads on sampling:

There is www.qs9000.com/ubb/Forum27/HTML/000008.html and there is www.qs9000.com/ubb/Forum10/HTML/000002.html and there is www.qs9000.com/ubb/Forum10/HTML/000006.html and there is www.qs9000.com/ubb/Forum10/HTML/000009.html

This a 'meaty' one: www.qs9000.com/ubb/Forum10/HTML/000027.html
To finish off, there is www.qs9000.com/ubb/Forum10/HTML/000038.html and there is
and last, but not least, is: and there is www.qs9000.com/ubb/Forum27/HTML/000004.html

[This message has been edited by TheOtherMe (edited 01 August 1999).]

 

[Roger Eastin]

Marc has given you some good threads here. For what you want, you could even use a "plug and chug" formula to determine validity, but you need some basic information like alpha and beta risks, fraction defective, etc. It would be helpful (and this may be in one of the threads that Marc has given you) to have a brief description of what type of plan you need (attribute, variable, acceptance number - sounds like c=0 for you, risk you are willing to accept) and see which plan "fits" you. For instance, there are several c=0 plans depending on the type of process you are taking your samples from. Then you can order the appropriate plan from the website that I listed above.