XBar Chart not Calculating or Displaying UCL and LCL Correctly

[#1]

Hi, I'm a JMP 10 user. I can't figure out why my XBar charts are not calculating or displaying the UCL's and LCL's correctly. Here's my script and a snip-it of what the charts look like. I'm not sure if something strange is going on with my data set or my script. Thanks. 

Control Chart(
	SendToByGroup( {:Category == "A"} ),
	Sample Label( :Name( "Date" ) ),
	KSigma( 3 ),
	Chart Col( :Process, XBar, R ),
	By( :Category ),
	SendToByGroup(
		{:Category == "A"},
		SendToReport(
			Dispatch(
				{"Variables Control Chart Category=A", "XBar of Process"},
				"2",
				ScaleBox,
				{Min( 9.79 ), Max( 10.21 ), Inc( 0.05 ), Minor Ticks( 0 ),
				Add Ref Line( 9.9, Solid, "Dark Blue", "Lower Spec" ),
				Add Ref Line( 10.1, Solid, "Dark Blue", "Upper Spec" ),
				Rotated Labels( "Horizontal" )}
			)
		)
	),
	SendToByGroup(
		{:Category == "B"},
		SendToReport(
			Dispatch(
				{"Variables Control Chart Category=B", "XBar of Process"},
				"2",
				ScaleBox,
				{Min( 9.65 ), Max( 10.25 ), Inc( 0.1 ), Minor Ticks( 0 ),
				Add Ref Line( 9.9, Solid, "Dark Blue", "Lower Spec" ),
				Add Ref Line( 10.1, Solid, "Dark Blue", "Upper Spec" ),
				Rotated Labels( "Horizontal" )}
			)
		)
	),
	SendToByGroup(
		{:Category == "C"},
		SendToReport(
			Dispatch(
				{"Variables Control Chart Category=C", "XBar of Process"},
				"2",
				ScaleBox,
				{Min( 9.945 ), Max( 10.055 ), Inc( 0.01 ), Minor Ticks( 0 ),
				Add Ref Line( 9.97, Solid, "Dark Blue", "Lower Spec" ),
				Add Ref Line( 10.03, Solid, "Dark Blue", "Upper Spec" ),
				Rotated Labels( "Horizontal" )}
			)
		)
	)
);

[#2]

I'd be happy to try to help, but first some questions:

Why are you not letting JMP just create the control charts (or are you showing the JMP script when you use the menu?)?

Could you attach the data set?

I can't tell from the attached pictures why you conclude the limits are wrong?  

Are A, B & C three different process variables (as JMP calls them)?

I actually don't see the values for the CL's on a couple of your pics.

At first glance it appears you do not have a constant subgroup size.  I can't tellwhether you are using R-bar/d2 or Std Dev to calculate limits.  

[#3]

Hi Statman, thanks for looking into this! Here are the answers to your questions:

• The data table has 100,000 rows. 86,000 rows are excluded. The XBar chart kept including the excluded data- not sure why. I decided to use JSL to create a subset and delete the excluded rows. The control chart script shown is JMP generated.
• The CL values are not showing up on the graphs where they are not pictured. That's why I said that the UCL's and LCL's aren't displaying correctly. Or am I mistaken and the CL values aren't always supposed to be present? If that's the case, how can I find out what those values are? As you stated, the limits might very well be correct.
• A, B & C are subgroups for the process variable.
• I do not have a constant subgroup size.

[#4 · Accepted solution]

The legacy control charts have very specific rules for hide and exclude.  Below are some excerpts from the documentation.

"

All rows of the sample are excluded before creating the chart.

Sample is not included in the calculation of the limits, but it appears on the graph.

Sample is excluded after creating the chart.

Sample is included in the calculation of the limits, and it appears in the graph. Nothing changes on the output by excluding a sample with the graph open.

All rows of the sample are both excluded and hidden before creating the chart.

Sample is not included in the calculation of the limits, and it does not appear on the graph.

All rows of the sample are both excluded and hidden after creating the chart.

Sample is included in the calculation of the limits, but does not appear on the graph. The sample marker disappears from the graph, the sample label still appears on the axis, but limits remain the same.

Hide and Exclude operate only on the row state of the first observation in the sample. For example, if the second observation in the sample is hidden, but the first observation is not hidden, the sample still appears on the chart.

Note: Excluded rows in Presummarize charts are excluded from calculations, regardless of which position they are within a sample.

• An exception to the exclude/hide rule: Both hidden and excluded rows are included in the count of points for Tests for Special Causes. An excluded row can be labeled with a special cause flag. A hidden point cannot be labeled. If the flag for a Tests for Special Causes is on a hidden point, it will not appear in the chart.

"

Your control limits are not constant because you do not have a constant sample size.  The control limits are not displayed on the graph because they are not constant.  What would be displayed?  The last UCL and LCL, the first UCL and LCL, or some value in between?  To see the actual values, click on the red triangle at the top and select Save Summaries.

[#5]

Thank you tonya_mauldin0 for the explanations! It sounds like my XBar chart is working correctly and now I know how to get the actual CL values.