How to Generate Structured Gauge R&R Reports Across Multiple Tests (with Summary Tables)

[#1]

Hi JMP Community,

Please see the attached image below for reference. I'm using JMp18.2.0. I attached the example file below as well. 

I'm running Gauge R&R studies using JMP with the following setup:

• Operator1 is mapped to DIB_id

• Operator2 is mapped to Site

• I have 8 parts used for each test

• The Part ID column is used as the last nesting level (as shown in the "Variance Components for Gage R&R" section)

The image shows the full Gauge R&R output I am interested in:

1. The Measurement Source Table with variance, percent contribution, and tolerance-based ratios.

2. The Summary Statistics and Gage R&R Summary table.

3. The Variance Component Table (using the last column, Part ID).

I need to generate this same report format across hundreds of tests, each with different test names.

Is there a way to automate this analysis such that JMP produces a structured report, preferably in tabular format, where each test name (e.g., DummyTest, RealTest1, etc.) is shown as a row, and all associated summary statistics (from the 3 table sections mentioned above) are shown as columns?

Ideally, I would like to batch process these and export the summary into a single summary data table (or Excel file) with columns like:

• Test Name

• EV (%)

• AV (%)

• R&R (%)

• Part Variation (%)

• % Tolerance

• NDC

• etc.

Any JSL-based automation or built-in workflows would be extremely helpful.

Thanks in advance!

[#2]

You can calculate gage for multiple measurements at the same time and then use make combined data table to get the result table. BUT depending how your limits are set how to do this, might differ slightly (you might want to disable Show Gauge R&R Specification Dialog from preferences)

[#3]

Thanks @jthi. It works! What I did, I performed the gauge RR for like ~2 tests(smaller set to make sure everything works). Then, I made the combined table and the combined table works perfectly. The problem here is when I tried to do a larger set ~100 tests, it takes a long time to calculate the gauge RR.

Is it possible to do so with a JSL code that makes all the necessary calculations in the back and produces a combined table Any tips would be appreciated!  Thanks!

[#4]

You can definitely use JSL to automate gage calculation and I have done it multiple times. I have used few different methods and I would try them in this order:

1. Calculate gage for all parameters at the same time and just let JMP do its work
2. Calculate it one by one and combine using JSL
3. Use different platform (Fit Model) for slightly different results (can replicate the way Minitab calculates variance) but will require extra calculations

If you can provide small example dataset, I community can most likely create demo JSL for you. 

[#5]

I think this is still mostly valid but you might have slightly better options in newer versions of JMP

How to get Repeatability, Reproducibility and Gauge R&R Variation values for each parameter in a tab... 

Names Default To Here(1);

dt = Open("$SAMPLE_DATA/Semiconductor Capability.jmp");

var = dt << Variability Chart(
	Y(:NPN1, :PNP1),
	Model("Main Effect"),
	X(:SITE),
	Historical Sigma(0),
	Invisible
);
var << "Gauge R&R"n();
wait(0);

result_ref = Report(var)[OutlineBox("Gauge R&R")];
wait(0);

dt_result = result_ref[TableBox(1)] << Make Combined Data Table;

var << close window;

From the resulting table you can do some renaming if necessary, drop rows and then split the table. You can also make second combined table for other results