cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
  • New to JMP? Join us Sept. 23-24 for the Early User Edition of Discovery Summit, tailor-made for new users. Register now for free!
  • Use World Cup data to build models, explore spatial relationships, and create informative visualizations in JMP. Register. July 17, 2 pm US Eastern Time.
  • Your voice matters! Tell us how you prefer to receive JMP updates, so we can tailor our communication to your needs. Take short survey.

Discussions

Solve problems, and share tips and tricks with other JMP users.
Choose Language Hide Translation Bar
Gadsdon
Level III

SPC: Run test 2

The customised run test2, can be customised from 7 to 11, say you set it to 7 and you have 7 + another 11 points all below the centerline why does the process screen function.  Why does it

1) Record 11 as the number of alarms, where in fact there is only 1 single stepchange
2) Surely it shoud just record 1, as there is only 1 stepchange that affect 7+11 points.  You are not goin to ivestigate all 18 items you are going to look for what changed once
3) So if there are 2 stepchanges, withthe current setup you potentially will have in excess of say 20 alarms where in fact the process has only changed twice

Summary
For run-rule screening, overlapping alarm counts should not be interpreted as independent special cause events. A sustained sequence of points on one side of the centreline represents a single process signal, regardless of how many overlapping rule windows are detected. The appropriate investigation unit is the run or step change, not each individual alarm generated within the run. Therefore, alarm counts should be consolidated into unique special cause events defined by the start and end of each continuous non-random sequence.

0 REPLIES 0

Recommended Articles