Process Screening - Grouping by date results in unformatted date

Question

I've been doing a historical analysis using the process screening tool and breaking my data down into YearQuarters to check quarterly performance.    Problem is that the date format gets lost in the process screening template, so all dates are just a long number (e.g "3.38e+9").    When I export the results + graphs to an additional table, the date format comes back, which is great since 90% of my work is being done on the exported results of the process screening tool, but when i want to deep dive my data i can't really use the numeric date format.

If I remember correctly you can directly control the table within process screening as a table, but I'm a bit clueless on how I might go about that.

Any thoughts on this?

Also, as an aside, the process screening platform + graphlets is by far one of the coolest things I have worked with in JMP 15.1.  It is simply insanely powerful what I can do with this.

Evan_Morris · Answer

Ok, so using the Consumer Prices sample data, create a formula column converting the Date -> Year Quarter

Then you can run the following script

Process Screening(
Y( :Price ),
Grouping( :Series, :Year Quarter ),
Control Chart Type( "Indiv and MR" )
);

You'll see that the nested group for Year Quarter uses the seconds since midnight result. It doesn't appear to do this for Year.

txnelson · Answer

I assume the long number you are seeing is just the number of seconds since midnight, January 1, 1904, which is what JMP dates are based upon.  So, setting a format for what you are seeing is what needs to be done to correct the issue.  However, it is not clear to me, exactly where the long numbers are being displayed.  Could you please provide an image or an actual output of the item you are seeing, to aid in debugging the issue?