Transposing a data table with multiple columns for same label

[#1]

Hello!

I have a data table with multiple wafers, the tools they pass at a certain step.  Sometimes a wafer may hit multiple tools at the same step.  I am losing toolsets using:

TrSWC = SWC << Transpose(
	columns( :Column ), 
	Transpose selected rows only( 1 ),
	By( :Wafer ),
	Label( :Label ),
	Output Table( "Transpose_Table" )
);	

where SWC=

Wafer	Label	Column
1	X	A
1	X	B
1	Z	C

TrSWC =

Wafer	Label	X	Z
1	Column	B	C

I lose the 'A' tool that wafer 1 hit at label 'X'

Thanks for any help!

-Tim

[#2]

Hi,

Two low tech solutions:
1) Omit the Label variable in the transpose parameters. Limitation: the columns will be labelled by ROW numbers
2) Create a dummy label column by concatenating the Label and Column (:Label || "," || :Column) and use it as Label in the transpose parameters

There is a chance that neither suggestions will meet your expectations, therefore let me know if there are any additional information you could share.

[#3]

Thanks, Thierry.

Both of those will have my Labels mixed(I have many wafers/labels/columns).

I'm thinking of iterating through the table and changing any duplicated label with a different column.  this would give me a column 'Xa' as well as 'X' and not mix things up.  I didn't really want to go that route, though :).

-Tim

[#4]

It would help to know what you want the transposed data table to look like in the case of the wafer hitting two tools in one step.

In your example, where do you want the A and the B to end up? Both in the X column? An X prime column? A new row?

[#5]

Jeff,

I would prefer it end up in the X column.  I have currently written a loop to change X to X' if it is duplicated, which works for me when I transpose.  I will end up putting the 'A' that is in column X' into the X column.

I was hoping there was an easier solution than something like that.

Thanks!

-Tim

[#6 · Accepted solution]

@timjnixon ,

This can be done with a Table Split, once an additional column that acounts for the number of tools used for that step is created.

The table is attached.  The formula and Split script are embedded in the table and cane be found below

Split script embedded in the table. Of coursethis can be done with the UI Table>Split

Current Data Table()  << Split(
	Split By( :Step ),
	Split( :Tool ),
	Group( :Wafer, :step count ),
	Remaining Columns( Drop All ),
	Sort by Column Property
)

 The resulting table will look like this

If this is not what you want you can also split by step and count

Data Table( "WaferStepSplit" ) << Split(
	Split By( :Step, :step count ),
	Split( :Tool ),
	Group( :Wafer ),
	Sort by Column Property
)

And this creates a table like the one  below

[#7]

@timjnixon ,

This is just an FYI.  I don't know your task, but suspecting a tool usage or anaomaly detection type analysis.

Check if your version of JMP contains the analysis platform Process History. I think you need JMP PRO, version 14 or later. The analysis platform is still experiment.

It can be useful looking for tool usage  (number of times a tool touched that wafer) and tool, operation, time conditions that might indicate what is causing a change in yield or defects or other metric.  It still is a bit limited, but for those two conditions, I have found it useful and it does not require a lot of table rstructuring to use it.

This is just an FYI. 

[#8]

gzmorgan0,

Thank you very much.  Quick and efficient :).

-Tim