Control chart, alarm and filter

Question

Hi,

I have a strange behavior when I want to filter a control chart with Alarm (Test2, 3, etc).
imagine you have the following control chart. My alarms are active and it is ok!

But what if now I filter?
I have less data so Test 2 (9 consecutive point on one side), or 3 (6 Increasing/decreasing points) should appear less. it is not true, it appears as if we have all the data!!!!

it is really an unexpected behavior!!!

any workaround???? -> Except split the data and use the column switcher which generates me another abnormal behavior (see my following post!!!)

Thank a lot,

Florence

tonya_mauldin · Answer

After more discussion on this topic with Florence, it appears that JMP can already give her what she wants.  A sample label column needs to be added and the Show Excluded Region option needs to be turned off. This will cause Test 2 to not be flagged and the gaps can still be seen in the data.

Control Chart Builder(
       Show Excluded Region( 0 ),
       Variables( Subgroup( :Sample ), Y( :indice de fluidité ) ),
       Chart(
              Position( 1 ),
              Warnings(
                     Test 1( 1 ),
                     Test 2( 1 ),
                     Test 3( 1 ),
                     Test 4( 1 ),
                     Test 5( 1 ),
                     Test 6( 1 ),
                     Test 7( 1 ),
                     Test 8( 1 )
              )
       ),
       Local Data Filter(
              Add Filter( columns( :Carrière ), Where( :Carrière == "Kuanga B" ) )
       )
)

ian_jmp · Answer

I'm not sure if this is a voting process, but if it is, I'd like to register a vote in favour of the option that Tonya mentions immediately above.

In so far as one of the primary use cases of CCB is to assess the degree of control one currently has over a process before making a chart live, and that the runs rules are an attempt to codify non-random behaviour, having the ability to easily decide if an omnibus chart or more granular charts are most appropriate (via local data filtering) seems a very useful feature.

tonya_mauldin · Answer

This is by design.  Please take a look at the documentation.

https://www.jmp.com/support/help/en/16.0/?os=win&source=application&utm_source=helpmenu&utm_medium=application#page/jmp/excluded-and-hidden-samples.shtml#

Specifically "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."

There has been some discussion on adding a preference so the user has the ability to alter this behavior, but this functionality has not yet been added.