Sampling
Six Sigma – iSixSigma › Forums › Old Forums › General › Sampling
- This topic has 5 replies, 5 voices, and was last updated 20 years, 10 months ago by
john beaudoin.
-
AuthorPosts
-
August 10, 2001 at 4:00 am #67981
Brett G.Participant@Brett-G.Include @Brett-G. in your post and this person will
be notified via email.Kelly:
Here is a discussion that helped me:
https://www.isixsigma.com/forum/showmessage.asp?messageID=2534
Hope it helps you as well…..0August 10, 2001 at 4:00 am #27641
Kelly BrigmanParticipant@Kelly-BrigmanInclude @Kelly-Brigman in your post and this person will
be notified via email.I am trying to determine the frequency of random samples that our operators need to do in an eight hour period. Our production number is 40,000 per week and we are a five day, 3 shift operation.
Thanks…0August 13, 2001 at 4:00 am #68019You are producing ~ 333 pieces each hour. It seems like you will want to know how often to sample and how many samples to take. You will also want to know the confidence level you have in discovering defects. Correct?
You should also consider the cost of sampling / data collection in comparison to the cost of defects. If data collection is inexpensive and easy, and the cost of a defect is expensive, you will likely want to take many samples.
I agree with Brett, read the discussions of a previous posting.0August 13, 2001 at 4:00 am #68020
Ged BryantParticipant@Ged-BryantInclude @Ged-Bryant in your post and this person will
be notified via email.If you are targeting zero defects, you will need a 4% sample, if there are 20 or more discernable characteristics. Establish a lot system, shift, tool change, operator change etc. Reject on one defect per lot.
30 characteristics will not yield Zero.
10 or less characteristics can use 2%.
If you have an AQL you will never yield zero.0August 13, 2001 at 4:00 am #68021
john beaudoinParticipant@john-beaudoinInclude @john-beaudoin in your post and this person will
be notified via email.I spent a half hour this morning reading all of the messages in that discussion and 10% of it actually had some usefull knowledge. 90% of it contained worthless opinion.
Sample Size to me is like the Holy Grail of 6 Sigma. We purchased an AQL Inspection Manual which references MIL-STD-105E and ANSI/ASQC Z1.4. These are military and goverment standards for sampling and acceptance testing.
This came from INFO, P.O. Box 58, Stillriver, MA 01467. Phone 978-456-3848.
The important thing to note about sampling, is that if you continually sample and find no defects, you can sample less and less. If you find defects, then you must sample more and more. The guide helps with this. You can select lot size, your desired AQL level and the tables will tell you how many defects are acceptable in you sample to move you to the proper inspection level.0August 13, 2001 at 4:00 am #68024
john beaudoinParticipant@john-beaudoinInclude @john-beaudoin in your post and this person will
be notified via email.See my reply to Brett G. on the web site.
0 -
AuthorPosts
The forum ‘General’ is closed to new topics and replies.