How can I Show Limit Labels on Control Chart for SPEC Limits

[#1]

Hi Folks. 

On the Control Chart Builder window, there is a setting under Limits called "Show Limit Labels". This option shows the numerical value of the calculated control limits. Is there a way to do the same function for user set specification limits? 

Here is what the default option shows:

Here is what I would like for it to show

In the example above, I manually typed in the actual USL and LSL values through right clicking -> axis setting updates. 

It would be so great if that could be automated. 

Much appreciated!

[#2]

I believe the only way to do this is through JSL using the Add Graphics Script() function.  

[#3]

Hi Jim,

Thanks for the input. I'm still green to using JSL but I'll look into it. If you are aware of any guidelines to use JSL in a similar case feel free to share!

-Naz

[#4]

Save your manually annotated example and then open up the saved JSL and it should give you an example of how to do it.

[#5]

A different issue is what is the reason for placing specification limits on a control chart? You can only maybe talk me into placing spec limits on one and only one type of control chart...an individuals chart...otherwise it's generally considered a bad practice to put spec limits on control charts.

[#6]

Hi P_Bartell,Preview

I'm not familiar with that general understanding. I'd love to hear your thoughts behind it.

In my case, I am reporting out a conformance report for one of our products. For the report, we want to see how the statistically calculated control limits compare with currently set internal release limits. The "specification" limit is not a real spec here. 

Thanks,

Naz

[#7]

Spec limits should never be associated with control chart method. Specifications are independently derived and have very little (or nothing) to do with the actual variation in a process. Shewhart's control chart method is a means of assessing components of variation in a process. There are two charts used, a range chart and an average chart. The range chart assesses the consistency of the within subgroup components of variation and answers the question; Are the within subgroup components of variation (x's changing within subgroup) consistent? If not, the investigator should seek to understand why. If those sources do exhibit consistency, then those sources can be compared to the other components of variation in the study. That comparison is done with an average chart. The average chart answers the question, Which components of variation have greater leverage? The within subgroup (as a function of the x's changing within subgroup as depicted by the control limits) or the between subgroup ( a function of the x's changing between subgroup.) None of these questions have anything to do with specifications. In fact, specifications on these charts confuses the interpretation.

[#8]

Hi Statman,

I appreciate your write up. As I noted I am referring to my reference lines as "specification limits" as a matter of aligning with JMPs terminology. 

My request is strictly pragmatic in nature. I'm looking for an automated/streamlined method to show what values are entered for "USL" and "LSL" reference line in the parameter being charted. 

Whether or not USL/LSL reference lines should even be used for control charts is a lively discuss but not suited for my specific request. 

I'm sure there's any number of useful reasons a user might want to show a reference line with the label and set value written out for clarity. 

[#9]

There is no need for me to answer your question asking me for "...your thoughts behind it." @Mark_Bailey and @statman have stated the case quite clearly. I concur 100% with their replies and have nothing further to add...or detract.

[#10]

There is a place where the natural variation, the focus of control charts and their limits, are brought together with the specifications: process potential and capability. Assuming your process has achieved a state of 'in control,' you can use one of the JMP tools for a capability analysis.