Choose Language Hide Translation Bar
Level III

How to script process capability analysis with best fit

I create script to generate the process capability. Below script works fine with the normal distribution (default). How can I assign the best fit to the calculation ?


dt_raw = data table ("raw_by_stage");
ColList = dt_spec<< get column names (numeric);
obj = dt_raw << manage spec limits (Y(eval(ColList)), load from limits table (dt_spec), save to column properties(1));
obj << close window;

obj = dt_raw << process capability( process variables(eval(Collist) ), Individual Detail Reports( 0 ), Capability Box Plots( 0 ), Overall Sigma Summary Report( 1 ), Goal Plot( 0 ), Capability Index Plot( 0 ), );

Re: How to script process capability analysis with best fit

One of the fundamental tenets of process capability is that the process is known to be stable. A stable process is predictable. The distribution of the stable process should not change and should be determined from the new data each time capability is evaluated.


So I would recommend determining the distribution model at the beginning after it is demonstrated to be stable, and then use the same distribution model in the future. I would not use the 'best fit' approach every time at future evaluations.

Learn it once, use it forever!
Article Labels