process capability
Six Sigma – iSixSigma › Forums › Old Forums › General › process capability
 This topic has 7 replies, 7 voices, and was last updated 20 years, 3 months ago by TCJ.

AuthorPosts

May 8, 2002 at 10:55 pm #29405
MicheleParticipant@Michele Include @Michele in your post and this person will
be notified via email.What process capability level does the 6 sigma program try to achieve (optimally)?
0May 8, 2002 at 11:50 pm #75281
CordobaParticipant@alejandro Include @alejandro in your post and this person will
be notified via email.Cpk =1.67 means 6 sigma level=3.4 PPM
Regards0May 9, 2002 at 12:35 am #75282CpK is the smaller of (USLXbar)/3 sigma or (LSL Xbar)/3 sigma. More simply: (Closest spec limit – Xbar)/3 sigma is the calculation for CpK. In any case, the distance from the closest spec linit to the process mean is divided by 3 sigma. Therefore, if this smallest distance has 6 standard deviations divided by 3, then a 6 sigma process is equal to 6/3, or a CpK of 2.00. A CpK of 1.67 is 5/3=1.67. This is a 5 sigma process, not a 6 sigma process.
0May 9, 2002 at 2:00 am #75284
GabrielParticipant@Gabriel Include @Gabriel in your post and this person will
be notified via email.I am not an expert in SS at all, but from what I could get (mainly from this forum) I understand that 6 sigma level assumes a process shift (in the long term) of 1.5 sigma from the target (center of spec.). This fact of the process shift was largely discussed in previous threads, and I won’t add nothing about it.
But what I can say is that:
– The name “six sigma” comes from assuming that the process variation contained within the average value +/ 6S is fully within specification. At least, this is Cp=2 and Cpk=2. This leads to much less than 3.4 PPM (I dont remember now but think it was less that 1 part per BILLON)
– Then, somehow we get to the 1.5 sigma shift. So we have Cp=2 (like before) but now Cpk=1.67. This leads to the famous 3.4 PPM. I am not saying that I agree with the 1.5 sigma shift, but this is how it is axplained that a process with a Cpk=1.67 (average 4.5 sigma away from the closest limit) is said to be a 6S process (still, because Cp=2, the width of the spec is equal to the width of the process +/6S, but shifted).
– Cpk=1.67 is not equal to 3.4 PPM, unless you have a Cp=2. So it is not correct (or not complete) to say, as in the forst response, that 6 sigma is Cpk=1.67 and that this is 3.4 PPM. In fact, a process with a Cpk=1.67 and a Cp=1.67 will double the PPM to 6.8, and will not be considered a 6S process. The spec limits would be cincident to the process average +/ 5S.0May 9, 2002 at 4:16 am #75285
Ashok SarkarParticipant@AshokSarkar Include @AshokSarkar in your post and this person will
be notified via email.This has reference to so many discusssion on SIx Sigma. We may conclude in the following way;
1. If Cpk=1.5 and Cp=2 ( Shift of average is 1.5 sigma) then ppm is 3.4 and this will be true for one side specification (offcouse Cp is not applicable). This is we understand as six sigma level and this means at worst case process will deliver 3.4 ppm defective product (May be we use the word opportunity also).
2. If shift is more or less than 1.5 sigma then the ppm level changes.
Regards
0May 9, 2002 at 11:18 am #75294
GabrielParticipant@Gabriel Include @Gabriel in your post and this person will
be notified via email.My previous post is full of mistakes and if I could I would delete it. It seems that my brain was on strike yesterday. Forget (and forgive)everything I said about 5 sigma and Cpk=1.67
Six sigma level is taken as 3.4 PPM what, for a normal distribuited process, is equivalent to Cpk=1.5 and Cp=2. This a process where its +/6S distribution width is equal to the spec width, but the process average is shifted 1.5S from the center of the spec (i.e the process average is 4.5S from the closer spec limit and 7.5S form the other limit). Note that Cpk=1.5 alone is not enough to get this 3.4 PPM. With Cp=Cpk=1.5 you double this value to 6.8 PPM.0May 9, 2002 at 11:52 am #75297Cpk =2.0
0May 9, 2002 at 2:27 pm #75320The goal of the “6 sigma program” is to eliminate defects (0 defects). Is that practical? Of course not; therefore, we strive for incremental improvement i.e. take a 2sigma process to 3 sigma, 4 to 5, etc. The definition of a 6 sigma process as mentioned earlier (Cp=2, Cpk = 1.5). The process potential Cp is 2 defects per billion and the capability with the inclusion of the1.5 sigma shift, Cpk = 1.5, is 3.4 defects per million.
0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.