Could anybody tell me about the formula on transform Cpk to DPM?

Thanks!

DPM is Defect Per Million which is same as PPM (Part Per Million).

Rgds!

I found this one with a Forums Search:

http://Elsmar.com/Forums/showthread.php?t=1943

I believe there's even a paper posted here somewhere which equates the two, if I remember correctly. I think Don Winton wrote it. Unfortunately I can't remember the name of it off hand. Maybe one of the others will remember the name of the paper Don wrote.

Assuming that you have verified the normality of the distribution before calculating the Cpk, you can determine the distance of the Upper and the lower spec from the distribution mean on the basis of the Cpk value.

Expressing these in multiples of the standard deviation will give you the Z values for Upper and Lower Spec positions on the distribution.

You can either refer to the Z tables or do normal distribution calculations to determine the fraction area under the Normal curve beyond each side of the specifications. This will give you an estimate of the ppm ( or DPM as you put it).