658064_de-DE

liliconcom

Hide excluded regions in Graph Builder axis (batch process data, time series and sensor data with filters)

When analyzing sensor data, it is common to filter multiple rows (e.g., removing shutdowns, weekends, certain product grades in batch processes, etc.)

This is not possible to do in JMP, while other software handles this automatically (see Grafana or another user requesting a similar feature here).  Yet, the control chart builder has this functionality, limited to run charts only.

 

Let's say I want to look at a group of events only.

I would expect to have an option to transform this view into this automatically.

In the control chart builder, you can get this plot after some configuration:

 

Lastly, an index per event will be also good to have as an option.

 

JMP needs to improve the capabilities for sensor data (not only tabular), especially now that native connectors to historians are possible.

 

Industrial applications of these visualizations can be found here, including an alignment when events have different durations:
https://arxiv.org/abs/2209.09660

 

Data Exploration and VisualizationBetreff: Ausgeschlossene Bereiche in der Graph Builder-Achse ausblenden (Batch-Prozessdaten, Zeitreihen- und Sensordaten mit Filtern)

Index pro Ereignis: über Col Rank(row(), Gruppierung) ?

Um eine virtuelle Zeilennummer der nicht fehlenden Einträge zu erhalten, kann man Col Rank() mit der neuen Option Umfang im Graph Builder kombinieren:

Names Default to Here(1);
dt = Open( "$SAMPLE_DATA/Big Class.jmp" );
Graph Builder(
	Transform Column(
		"index",
		Formula( Col Rank( row(), :"@Exclude"n, :"@Filter"n ) )
	),
	Size( 438, 290 ),
	Show Control Panel( 0 ),
	Show Legend( 0 ),
	Graph Spacing( 4 ),
	Variables( X( :index ), Y( :height ) ),
	Elements( Points( X, Y, Legend( 9 ) ), Smoother( X, Y, Legend( 10 ) ) ),
	Local Data Filter(
		Add Filter(
			columns( :age ),
			Where( :age == {13, 14, 15, 16} ),
			Display( :age, N Items( 6 ) )
		)
	)
);