Data Transformation : Automatic Transform also Specs Limit in Column Properties

[#1]

Dear Community,

I am opening this new discussion, as I'm not able to find any related discussion to this.

I have input Specs Limit in the Column Properties in a column of data.

When I do a Data Transformation on this column of data, JMP will create a new column with the transformed data.

However, this transformed data column do not contain any transformed specs limit (from the original data column).

My question : Is there anyway JMP can also transformed the specs limit (in the column properties of the original data) to the new transformed data column properties?

Thanks.

Chris

[#2]

JMP does not have a built in way to transform the Spec Limits.  However, it is easily done with a simple script.  The below script creates a new column in the Semiconductor Capability data table.  The new column is a Log Transformation of column NPN1.

The script then exacts the current Spec Limits from the NPN1 column and then applies the same transform formula to them as it sets the Spec Limits in the new data column.

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( "Log NPN1", formula(log(:NPN1)));

specs = dt:NPN1 << get property("Spec Limits");

Eval(
	Eval Expr(
		dt:Log NPN1 << set property(
			"Spec limits",
			{LSL( Expr( Log( specs["LSL"] ) ) ), Target( Expr( Log( specs["Target"] ) ) ),
			USL( Expr( Log( specs["USL"] ) ) )}
		)
	)
);

[#3]

Hi Jim,

I found your code looking for a solution to a slightly different problem.  I want to extract the spec limits and then assign them to a new table, but the code fails if "LSL" or "USL" are not assigned to the column.  Does your code deal with missing spec limits?

[#4]

Most of the code in the community is intended to show the general approach - and needs some fine-tuning to be robust against any problems.

In this case, just replace 

specs["LSL"]

with

Try(specs["LSL"],.)

If there is no lower spec, the list indexing will throw an error , "Try" will catch it and return ".", i.e. missing .