Trouble with Spec Limits in JMP 18

[#1]

I want to show my upper spec limits in my C-Chart. JMP 17 allowed me to do this, but spec limits are "greyed out" in JMP 18. Any way to work around this? Go back to JMP 17?

[#2]

From JMP18 help

https://www.jmp.com/support/help/en/18.0/#page/jmp/options-panel-and-rightclick-chart-options.shtml#

https://www.jmp.com/support/help/en/18.0/#page/jmp/shewhart-control-charts-for-attributes.shtml#

And in JMP17

Not going to suggest this but depending on your data and if you really wish to have spec limits in control chart, you might have to use Variable control chart

vs

[#3]

could you post a picture of your graph and the greyed out menu item.

I'm not sure where that is in a c-chart

[#4]

@Byron_JMP Most likely these are disabled (for a reason I would think as this is count data and has different use?). We also posted in very similar times and I added some links from JMP Help. According to the label in the menu it is a new in JMP18, so I'm not sure where it was in JMP17.

Script to replicate (change Sigma to get C chart)

Names Default To Here(1);
dt = Open("$SAMPLE_DATA/Quality Control/Shirts.jmp");

obj = dt << Control Chart Builder(
	Show Two Shewhart Charts(0),
	Class(Shewhart Attribute),
	Variables(Subgroup(:Box), Y(:"# Defects"n), n Trials(:Box Size)),
	Chart(Points(Statistic("Count")), Limits(Sigma("Binomial")))
);

And for Shewhart Variables

[#5]

Hi All,

Thank you for the clarification and potential solutions.  JMP 17 allowed me to add the spec limits with Poisson counted attribute charts but JMP 18 doesn't allow this - at least not easily.  Interesting to compare the two versions side by side.  I prefer to use C-Charts but I guess I could simply use the variable charts and manually set the control limits to correspond to a poisson distribution.   Appreciate the rapid feedback and suggestions!!

[#6]

Pardon my comments, and please ignore me if my comments upset you, but spec limits have nothing to do with the purpose of control charts and should not be placed on control charts as they confuse the purpose of control charts (See Shewhart and Wheeler).  

Please read Wheeler and Neave:

"Shewhart’s Charts and the Probability Approach"

[#7]

I guess my response would be that depends on what you are using the control charts for.  There are situations where it can be helpful to see both the control and spec limits together.  That said, thanks for pointing the article out to me.  I will read it when I get the chance.

[#8]

Let me clarify.  You can do whatever charts/plots you want and add spec limits.  Just not control charts.  Control charts are designed to answer specific questions.

1. Which components of variation have the greatest leverage?  Ultimately you want to use them to compare sources of variation.  Where should you work to improve the process?  This is the purpose of the X-bar chart.  The spec limits are a function of the within subgroup variation (A2R-bar).  The variation in the dots plotted against the control limits are representative of the between subgroup sources of variation (a function of the x's varying at the sampling frequency).  In fact, the dots are biased to the between sources because the within sources are averaged.   The X-bar chart does NOT assess stability (neither does the Individual chart).  It can't as it is a comparison chart.  You can't assess stability of one component by comparing it another component, you must compare the same component to itself over time. Shewhart was brilliant.

2. Is the basis for comparison consistent/stable?  Before you can make the comparison, however, you must make sure the basis of comparison is consistent/stable.  This is the purpose of range and moving range charts.  If the basis is unstable or inconsistent, what good is the comparison?