iSixSigma

Control Charts

Six Sigma – iSixSigma Forums Old Forums General Control Charts

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • #51159

    newbie
    Participant

    Does the basic tenet of “we would expect OOC points to occur only 3/1000” change as you invoke more rules into the chart (ie If I check every westing house rule in MTB vs. simply using “any point more than +/- 3 std dev from the mean)?
    Thanks!

    0
    #176882

    Chris Seider
    Participant

    Yes….
    Consider the case of 7 points consecutively above or below the center line.  Randomly you would only expect that to occur (1/2)**7 times so that would add to the 0.27% for the 1 point outside the 3 sigma control lines. 
    Keep in mind that you are referring to the alpha risk of assuming the process has changed when it has not.  It does NOT refer to the number of times you would expect a special cause to occur, one cannot predict special causes very well!

    0
    #176884

    Mikel
    Member

    Chris’s answer is right.BTW, they are Western Electric rules, not Westinghouse rules.

    0
    #176889

    newbie
    Participant

    Good deal..thank you very much!

    0
    #176891

    newbie
    Participant

    Stan, Chris….thanks so much!

    0
    #176893

    newbie
    Participant

    …that’s embarassing…thanks Stan….

    0
Viewing 6 posts - 1 through 6 (of 6 total)

The forum ‘General’ is closed to new topics and replies.