cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Choose Language Hide Translation Bar

Control Charts with Nelson rule n°3

Hello everyone, 

 

I hope you are doing well.

 

I have a question about the control chart concerning the Nelson rule n°3 when there is a phase in the control chart.

Is there any way for the Nelson rule n°3 to be activate even there is a break with the phases of the control chart (see screenshot)?

PersuasionCamel_1-1667823511292.png

In fact there are more than 9 points in a row that are upper the mean and the Nelson rule n°3 is not activated. Is there any way to activate it across the phases of the control chart?

 

Thank you for your help and have a nice day,

 

Jean

4 REPLIES 4
statman
Super User

Re: Control Charts with Nelson rule n°3

Jean,

I obviously don't have the data table to see what you did or how you treated the chart, but there is something wrong.  The green line (I assume the average) is not centered within the control limits.  I'm not sure how you got this, but this should be resolved before discussing the Western Electric Rules.

 

Once you have that figured out, after you create the IMR chart, if you select the red triangle next to the chart and scroll down to Tests, you can turn any test on or off.  Or you can set it in preferences (Preferences>Platforms>Control Chart)

"All models are wrong, some are useful" G.E.P. Box

Re: Control Charts with Nelson rule n°3

Hi,

 

In fact these data are simulated to present my problem (having > 9 data points in a row higher than the mean) :)

I activated the rule number 3. Without any phase applied to the control chart, the rule number 3 is activated. However, if a phase is applied to the control chart, the rule is not activated (see screenshots):  

PersuasionCamel_0-1667834151899.png

 

PersuasionCamel_4-1667834202195.png

My question is: is there any way to have a phase in my control chart and the rule number 3 activated. My understanding is that with a phase applied to the control chart, there is a discontinuity between the 2 phases (past and current) and the rule n°3 is not activated.

statman
Super User

Re: Control Charts with Nelson rule n°3

Sorry, I do not deal with simulated data, but perhaps, since it isn't real data, you could kindly attach the JMP data table and "we" can have a look at the issue or possible remedy.  It appears you do not have 6 in a row within each phase.  And, again the mean is not possible with the control limits shown.

 

Test 3

Six points in a row steadily increasing or decreasing (anywhere on the chart)

 

Perhaps you want rule 2:

Test 2

Nine points in a row in a single (upper or lower) side of Zone C or beyond

Detects a shift in the process mean.

 

For a description of the rules:

https://www.jmp.com/support/help/en/16.2/?os=mac&source=application#page/jmp/options-panel-and-right...

 

"All models are wrong, some are useful" G.E.P. Box

Re: Control Charts with Nelson rule n°3

When a Phase variable is specified, the counts for each test are reset at the start of each phase.  This information can be found in the documentation on the following page:

Options Panel and Right-Click Chart Options (jmp.com)