Cpk and Ppk

[#1]

This is a good place to start: Statistical_process_control for the basic terms and definitions.

I highly recommend going through the Quality Methods module here:  online-statistics-course.html

Cpk is calculated using an estimate of sigma that is based on the moving range, the difference between consecutive points.

Ppk is calculated using an estimate of sigma that is based on differences from the mean. (aka Stand Deviation)

Cpk is short term variation (between consecutive points) Ppk is long term or Overall.

[#2]

Do we have to try and guess the question ?  I'll try:

"what is one of the worse interventions in the history of process capability analysis"

[#3]

@David_Burnham: I had to chuckle. Back in the day when teaching basic quality and statistical methods I used to say, “The two most abused statistics in Statistics are R^2 and Cp/Cpk. Buy me a beer at the bar tonight and sit back and listen to my rant.”

[#4]

Ha, I was wondering if someone was going to comment on this thread. @David_Burnham. great question.  I used to take data sets from clients from which they were reporting a Cpk of 2.  Using their data and using simple enumerative statistics to estimate confidence intervals around the mean and standard deviations they used in their Cpk calculation, I could get Cpk of 0.2 - 6.0.  How useless Cpk is (or any of those capability ratios)!

[#5]

I guess I’m OK with curious people that want to fully understand the math behind these sorts of statistics. But for decision making, process understanding and communicating and sharing the data, conclusions and recommendations in the process capability space,,,,

Simply put there are three rules for successful data analysis:

1. plot the data
2. Plot The Data
3. PLOT THE DATA

There are so many useful data visualization techniques in JMP’s Process Capability and Process Screening platforms, only those suffering a severe case of ‘mononumerosis’ would bother with capability indices.

[#6]

Fully agree to points 1, 2 and 3.

Perhaps a point 4 could be "plot the data IN CONTEXT" (e.g. using the time sequence, making use of the context to make the plot more insightful e.g. plotting the results from different production shifts in different colours, which is so easily done in JMP)