How to calculate Cpk
Six Sigma – iSixSigma › Forums › Old Forums › General › How to calculate Cpk
 This topic has 8 replies, 7 voices, and was last updated 14 years, 10 months ago by Omashi Sabachi.

AuthorPosts

July 2, 2007 at 2:12 pm #47433
I have some confusion in calculating Cpk of one of my grinding process. It is mentioned in the drawing that surface roughness should not be less than 0.6 Rz. Now we have minimum limit here but of course if one gets better than that like 0.5, 0.4 etc, then it is more better. So in this case upper specs limit is not mentioned. Now in the formula of Cpk, we have USL and LSL. My question is, we have lower Specs but not upper. What to do now, how to calculate Cpk.
Thanks for any comments.
AL0July 2, 2007 at 2:33 pm #158116
Omashi SabachiParticipant@OmashiSabachi Include @OmashiSabachi in your post and this person will
be notified via email.Try to use the formula
(Xbar) – LSL divided by 3s (SD) (it could be used when the specification is a minimum tolerance (unilateral tolerance)0July 2, 2007 at 3:14 pm #158120
MakiwariParticipant@Makiwari Include @Makiwari in your post and this person will
be notified via email.How many fomulae does Cpk have? Is it one or two? Why does it have two?
Now that you’ve thought about that how do you interpret Cp?
What other measures can you use?0July 2, 2007 at 6:13 pm #158124hi Omashi – wanted to understand here and as well try to answer Mak….
There are two formulas to find out (in my easier terminology) outliers from USL and LSL.
Per me, Omashi has given the formula for LSL cos u have specified that there is only LSL identified in your process and there is no USL.
As far as I remember, formula for USL is USLx bar divided by 3SD.
Omashi am I correct?0July 2, 2007 at 7:24 pm #158128
Omashi SabachiParticipant@OmashiSabachi Include @OmashiSabachi in your post and this person will
be notified via email.You calculate both figures ,but you consider the smaller one only
0July 9, 2007 at 2:26 pm #158380Allen,
If you have a onesided lower spec, then Cpk=(XbarLSL)/3s.
0July 9, 2007 at 5:45 pm #158390Just to kick this horse one more time, a simpler definition for Cpk would be to use the distance between your mean and nearest specifcation limit divided by (3) short term standard deviations (ie remember to subgroup). Also, you will need to consider stability and normality.
0July 12, 2007 at 2:24 pm #158527
k.bhadrayyaParticipant@k.bhadrayya Include @k.bhadrayya in your post and this person will
be notified via email.Dear Allen
Go with the formula given by Mr.Omashi. . That is correct.
K.Bhadrayya
0July 12, 2007 at 9:06 pm #158559
Omashi SabachiParticipant@OmashiSabachi Include @OmashiSabachi in your post and this person will
be notified via email.Thank You
0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.