Our World Statistics Day conversations have been a great reminder of how much statistics can inform our lives. Do you have an example of how statistics has made a difference in your life? Share your story with the Community!
Choose Language Hide Translation Bar

Example p'-prime chart script (Constructing p'-charts or Laney p-charts)

Constructing p'-charts or Laney p-charts in JMP


The attached script will calculate the control chart sigma and generate a p-chart with the calculated sigma. The p-chart platform in JMP has a default 3 sigma set in the launch dialog. In certain cases sigma of 3 is too restrictive and larger values of sigma are prefered. The p'-chart is used for cases where the proportion of defects is low, because the denominator is very large relative to the number of defects, the control limits become very narrow and everything is out of control. The p'-charts have scaled control limits, Instead of assuming sigma =1 (KSigma=3) it is calculated from the data. This chart type uses a larger value that comes from an estimate of Sigma z, which results in wider control limits.


This script was written to calculate sigma following the same method of Laney p-charts. Other p'-chart calculators that I've seen in the past add additional columns to the data table for calculations. I used matrix functions in this script to keep the original data table tidy (the script also runs just a little faster with larger data sets.)


Background links:

Quality Digest Magazine



3/23/17 - Script Revision

I updated the script runs in JMP 12 and 13. I expect that it will run in earlier versions.

The dialog and report are simpler and cleaner.

I'm always learning new and better ways to write JSL, and often I look back at my old scripts, and wonder how they could have possibly worked in the first place. 


Also, Laura added a nice example table in the discussion board. https://community.jmp.com/t5/Discussions/Laney-P-and-U-charts-in-JMP/m-p/12249#M11676


@Byron_JMP  Is it possible find Cpk for the Laney P chart?

the key statistic for p' charts is the z-sigma.  For CPK, just assign a spec limit to the column of interest and run a distribution.  Both CPK and PPK are presented in the report along with nonconformance, which is almost better IMHO

Article Labels
Article Tags