Hello JohnSrinivassan,
Based on
http://en.wikipedia.org/wiki/Process_capability_index ...
1) Your data are not even close to be normally distributed, and one of the underlying assumptions for the
process capability index is that the data are at least approximately normally distributed
2) You did not provide a mean or a std deviation for the data. Based on the histogram, the following
formulae *estimate* the mean and std deviation for your sample:
Mean: =SUMPRODUCT(A2:A110,B2:B11
0)/B111
Std Dev: {=SQRT(SUM(B2:B110*(A2:A11
0-C1)^2)/(
B111-1))}
Note that the std dev formula is an array formula, and not a normal formula. To enter an array formula, do
not enter the curly braces, and hit Ctrl+Shift+Enter instead of Enter to finish it off. Excel will then display
those braces to indicate that it's an array formula.
3) If you create Names Mean, SD, Lower, and Upper for the mean, std deviation, lower specification bound,
and upper specification bound, respectively, then these formulas get CP and CPK:
CP: =(Upper-Lower)/(6*SD)
CPK: =MIN((Upper-Mean)/(3*SD),(
Mean-Lower
)/(3*SD))
Regards,
Patrick