topic Apply K Sigma and extract specs in Discussions

Apply K Sigma and extract specs

Jackie_ — Fri, 14 Nov 2025 16:37:13 GMT

Hello,

I am trying to calculate and extract specs using K sigma in process capability. However, something is off, and I am stuck. Is there a way to apply the K sigma and calculate the specs after the best model fitting and save those specs into a list?

See the below pic:

Here is the jsl to plot histograms and apply the fit, I would like to know how to code the steps for applying K Sigma and extracting LSL and USL for all the cols

Names Default To Here( 1 );
dt = Open( "$SAMPLE_DATA/Semiconductor Capability.jmp" );

cols = {"PNP2", "NPN2", "PNP3", "IVP1", "PNP4"};

New Window( "",
	<<Type( "dialog" ),
	For Each( {mct, id}, cols,
		dis = Distribution(
			Uniform Scaling( 1 ),
			Stack( 1 ),
			Quantiles( 0 ),
			Summary Statistics( 0 ),
			Horizontal Layout( 1 ),
			Vertical( 0 ),
			Continuous Distribution( Column( Eval( mct ) ) )
					
			
		);
		
		dis << Fit All;
			
		scb = Report( dis )[Outline Box( "Compare Distributions" ), Table Box( 1 ), String Col Box( 1 )];
		bestdist = scb[1];
		
				eval(parse("dis <<  Fit "||  bestdist |||"(Process Capability( Set sigma multiplier for quantile spec limits( 3 )))"));
	)
);

Re: Apply K Sigma and extract specs

jthi — Fri, 14 Nov 2025 20:37:12 GMT

Names Default To Here(1);
dt = Open("$SAMPLE_DATA/Semiconductor Capability.jmp");


dist = dt << Distribution(
	Continuous Distribution(
		Column(:NPN1),
		Process Capability(0),
	)
);

dist << Fit All;

scb = Report(dist)[Outline Box("Compare Distributions"), Table Box(1), String Col Box("Distribution")];
bestfit  = scb[1];


ob = Report(dist)[OutlineBox("? " || bestfit || " ?")];
// Show(ob << get title);

(ob << get scriptable object) << Process Capability(Set sigma multiplier for quantile spec limits(5));

vals = ob[Outline Box("Process Summary"), Number Col Box(1)] << get;
lsl = vals[1];
usl = vals[2];

Re: Apply K Sigma and extract specs

shampton82 — Sun, 16 Nov 2025 20:30:03 GMT

Hey @Jackie_ ,

You can do this without coding. After you have fit all the distributions you can then go into the first columns fit, hold down control and select process capability, then select "Enter Sigma Multiplier" and click the "use calculated spec limits". when you press okay holding control it will broadcast this to all fits.

After that go to the Process Summary box of any of the fits and right click and select "Make Combined data table". this will make a data table with all the LSL and USL. You can delete the other parameters that come along with it to have just a spec limit table.

Hope this helps!