Proving the CAPA Was Effective Part 2!
On our last email we discussed Ppk calculations between an old process and an improved process.
In this email we will show how to demonstrate that the new Ppk value is in no way related to the previous validation!
For reference the Ppk values:
To prove this we will use a formula to calculate the confidence Interval of the Original Process, if the new Ppk is not within this confidence Interval then it means it is a brand new process!
The formula is the following:
To obtain N-1 we simply take how many samples we took and subtract 1. Since we took 30 samples then N-1 = 29.
For the Chi-square section we can use Minitab. We click on Graph > Probability Distribution Plot. Enter View Probability, select Chi-Square, enter 29 in degrees of freedom, then shaded area, select both tails.
领英推荐
After clicking on Ok you will obtain the following plot. The value for the lower confidence Interval portion is 16.05 and the value for the upper portion is 45.72.
From here we simply divide these values by 29, and then take the square root of these results. This value is multiplied by the Ppk value and Presto! We have our confidence Intervals.
Take a look at the following Excel table for reference:
In other words the original process has a confidence Interval Ppk from 0.56 to 0.95.?
The Ppk value from the new process is 1.90 and is outside of the range. This provides documented evidence that the new Ppk value is a brand-new process with a better process capability.
All corrective and preventive actions are effective!
If you felt lost at any point of this explanation just send us an email! We are happy to help!
This doesn't even cover?the surface of all we teach at Engineering Masters, to join visit our site!