Non Normal Data
Six Sigma – iSixSigma › Forums › Old Forums › General › Non Normal Data
- This topic has 5 replies, 4 voices, and was last updated 17 years, 11 months ago by
Ward.
-
AuthorPosts
-
September 14, 2004 at 6:57 pm #36874
Just looking for some direction. I have production data for 7 months on several product lines. When I look at it by product, it is not normal. When I look at it by month, some of the monthly data is normal while some is not.
Recommendations please.
0September 14, 2004 at 6:59 pm #107384Gina,
Can you tell us in more detail what you’re trying to accomplish?0September 15, 2004 at 6:15 pm #107432Matt,
To determine if my process is in control and then do process capability.
0September 16, 2004 at 1:29 am #107445First, identify all the logical subgroups. Run a control chart, evaluate outliers and identify special cause. Remove special cause were it is appropriate. Re-run with revised data. Other options to explore: 1) Can you approximate normal based on the Central Limit Theorem. 2) Transform data and run capability
0September 20, 2004 at 9:37 pm #107666TCJ,
I did identify one logical subgroup but it looks like I may need to look at others. I’ll give it a try.
Thanks,
Gina0September 20, 2004 at 11:02 pm #107672Are any items out of spec? If so, determine the percentage that fall out of spec and convert to a capability index. For instance if 10% fall out of spec, that would be a Z value of about 2.33, which when divided by 3 should be Cpk of about .78.
I do support the ideal of the logical subgrouping for the control charts, which should always be done prior to the process capability study anyway.
0 -
AuthorPosts
The forum ‘General’ is closed to new topics and replies.