Is there a different calculation of Cpk for unilateral tolerances
Example Cpk = USL( or LSL)-XBar- / 3 Sigma
Cpk= upper spec limit ( Or Lower Spec Limit) minus Xbar divided by 3-sigma
Should unilateral tolerance Cpk be calculated by using a 6 -sigma divisor?
There is no difference in the Cpk formula if the process has a two-sided spec or a one-sided spec. The Cpk formula has always been :
|Closest Spec – Xbar| / 3 sigma.
The only potential concern on a unilateral spec is the estimate for sigma. Many of the process that have a unilateral spec are not normal due to physical bounds (e.g. cycle time can’t go below zero). In many of the Six Sigma initiatives today, they teach transforming the data and using the transformation to calculate the capability value. I would suggest in many cases, it is simply easier to use the DPMO calculation and back into the Cpk value if you need to report it.
The forum ‘General’ is closed to new topics and replies.