Home › Forums › Old Forums › General › Six Sigma – Bell Curve

This topic contains 9 replies, has 5 voices, and was last updated by Fake Stan Alert 10 years, 4 months ago.

Viewing 10 posts - 1 through 10 (of 10 total)

- AuthorPosts
- November 15, 2007 at 2:48 pm #137355
Is a Six Sigma process +/- 3 Sigma (Total of 6 standard deviations) on a bell curve or +/- 6 Sigma (total of 12 standard deviations) on a bell curve?

November 15, 2007 at 3:07 pm #137356Anoop Nair,

Six sigma is +/- 6 standard deviations between the mean of your bell curve and the nearest specification(s) for the process. I am not sure why you are referring to the bell curve. Your sigma level is simply the minimum of USL-Mean/S or Mean-LSL/S.

I hope that makes sense.

Regards,

Dr. ScottNovember 15, 2007 at 3:35 pm #137358Dear Dr. Scott – Thank you for your reply. I think I did not articulate my question very well. Assuming I have a Six Sigma process and I have an LSL and a USL, how many standard deviations would I have on either side of my mean? In other words, what would be the Z-value on each side of the mean. Thanks again for your help.

November 15, 2007 at 4:00 pm #137362Anoop Nair,

Apparently I articulated even less effectively than you did. :-)

You would need at least 6 standard deviations between your mean and either the USL or LSL. In other words, if your process is perfectly centered on the target (i.e., you process mean is the same as your target), then you would have to have 6 standard deviations of space between your mean and the USL and 6 standard deviations space between your mean and the LSL.

Does that clarify things?

Regards,

Dr. ScottNovember 15, 2007 at 9:29 pm #137394Perfect . Thank you!

November 15, 2007 at 9:43 pm #137399Anoop,

You are very welcome. Hope I helped.

Cordially,

Dr. ScottNovember 15, 2007 at 10:12 pm #137400Anoop,

While the technical definition for the sigma level of a process is the number of standard deviations that exist between the processes average and the closest of either the upper/lower specification (assuming that we can count of 6 increments before the limit) the projection of the VOP is still the normal curve at +/-3 standard deviations from the average (explaining 99.73% of the expected outcomes).

Sigma levels of a process provide a capability assessment and are integrating both VOC and VOP. Standard deviations are used to portray the expected variation of the process and are a reflection of the VOP.

Regards,

ErikNovember 16, 2007 at 6:45 am #137417Good question:the second one

November 16, 2007 at 11:04 am #137427Dear Eric – Thanks for your note. I think I understand using Sigma as representing process capability and standard deviation explaining variation in the process. The question that I had was that if we look at the bell curve having the mean equivalent to the target, how many standard deviations on either side needs to be fit to depict a Six Sigma process. I understand from Dr. Scott reply that it has to be six standard deviations on both sides (ie, a total of 12 standard deviations). Put another way, if I have the classical bell curve with mean=target, with 3 standard deviations on either side of the mean with Z=-3 coinciding with the LSL and Z=3 coinciding with the USL, explains a 3 Sigma process and not a Six Sigma process.

Thanks and Regards,

Anoop

PS: The reason for posting this question was that I had somebody looking at this classical bell curve and telling me that it represents a 6 Sigma process and my argument was that it represents a 3 Sigma process. I hope with the answers that I got from you and Dr. Scott, my understanding is right!November 16, 2007 at 12:12 pm #137429Usually both can present the 6 sigma process,but the 12 sd curve is more accurate

- AuthorPosts

Viewing 10 posts - 1 through 10 (of 10 total)

The forum ‘General’ is closed to new topics and replies.

© Copyright iSixSigma 2000-2018. User Agreement. Any reproduction or other use of content without the express written consent of iSixSigma is prohibited. More »