631151_en-US

liliconcom

Binning by Spec limits

If I have a column with spec limits defined as column properties, is there a "one-click" solution *) to bin the data into 3 groups:

x < LSL

LSL <= x <= USL

x > USL

Data Blending and CleanupData Exploration and VisualizationRe: Binning by Spec limits

So, there is no single click solution?
I also checked the Custom Binning menu. 

Several options are listed, but none to use the spec limits:

 

Re: Binning by Spec limits

I was able to create a grouping based upon the binning of the Spec Limits

Using the Levels option, and specifying Create Transform Column 

I simply created the following formula to create the binning

and it worked as desired.

In my example, I used the Semiconductor Capability sample data table, and the PNP3 column.  It has outliers, however, it has no outliers that are less than the LSL value, so I manually changed one row to have a low value, just to test the formula.

specs = :PNP3 << get property( "spec limits" );
If(
 :PNP3 < specs["LSL"], -1,
 specs["LSL"] <= :PNP3 <= specs["USL"], 0,
 1
);

 

 

Re: Binning by Spec limits

Thanks @txnelson for the JSL example

The disadvantage: a new column is added.

 

If I do this for all column, it gets quite crowded.

 

I hope a similar functionality is directly implemented in Jmp - but couldn't find it.

neither here:

nor here:

 

Is it hidden behind a Shift/Alt/Ctrl Click?

Re: Binning by Spec limits

Here is an example of one way to handle this

Names Default To Here( 1 );

dt = // Open Data Table: semiconductor capability.jmp
// → Data Table( "semiconductor capability" )
Open( "$SAMPLE_DATA/semiconductor capability.jmp" );

dt << New Column( "bin",
 formula(
  As Constant( specs = :PNP3 << get property( "spec limits" ) );
  If( specs["LSL"] <= :PNP3 <= specs["USL"],
   1,
   0
  );
 )
);