Is the normal Cpk calculation applicable for data with positive & negative values?

A

Anonaman

We have test parameters that give readings with positive and negative values, thus one of the specification window is like -1.8 to +1.8. Is the Cpk computation applicable for this case wherein the average value is being compensated due to very low negative values and very high positive values?
 

Miner

Forum Moderator
Leader
Admin
Re: Is the normal Cpk calculation applicable for data with positive & negative values

The short answer is that Cp and Cpk will work the same even if the lower and upper specs are negative or a mix of negative and positive provided you are not doing anything unusual with your data prior to the analysis.

Can you attach the data, how it is collected and a description of the process that it came from? This will allow us to provide a more detailed answer.
 

bobdoering

Stop X-bar/R Madness!!
Trusted Information Resource
Re: Is the normal Cpk calculation applicable for data with positive & negative values

Can you attach the data, how it is collected and a description of the process that it came from? This will allow us to provide a more detailed answer.

It is also handy to know what the parameter and the process is to help.
 
B

brahmaiah

Re: Is the normal Cpk calculation applicable for data with positive & negative values

We have test parameters that give readings with positive and negative values, thus one of the specification window is like -1.8 to +1.8. Is the Cpk computation applicable for this case wherein the average value is being compensated due to very low negative values and very high positive values?
The Cpk calculations will work with both positive and negative values in the same group of data. You should feed the data along with their signs(+ OR -).
V.J.Brahmaiah
 
A

Anonaman

Re: Is the normal Cpk calculation applicable for data with positive & negative values

The short answer is that Cp and Cpk will work the same even if the lower and upper specs are negative or a mix of negative and positive provided you are not doing anything unusual with your data prior to the analysis.

Can you attach the data, how it is collected and a description of the process that it came from? This will allow us to provide a more detailed answer.

:thanks:

Thanks for your prompt reply. It already helped me, i just would like to confirm actually that there is no other way to compute Cpk. Thanks again.
 
Top Bottom