iSixSigma

How to Detemine DOE Levels

Six Sigma – iSixSigma Forums General Forums Tools & Templates How to Detemine DOE Levels

Viewing 2 posts - 1 through 2 (of 2 total)
  • Author
    Posts
  • #54245

    Steve
    Guest

    Here is the data I have:

    Trial Staff sz Order Method Software Rm temp Results (Y) (patient wait time measured in minutes)
    1 1 1 -1 -1 -1 9
    2 -1 -1 -1 1 1 7
    3 1 1 1 1 -1 25
    4 -1 -1 1 1 -1 28
    5 -1 1 1 -1 1 26
    6 1 -1 -1 1 1 8
    7 1 -1 1 -1 -1 28
    8 -1 1 -1 -1 1 6

    Staff size – 16.75
    Staff size +. 17.50

    Order – 17.75
    Order + 16.50

    Method – 7.50
    Method + 26.75

    Software – 12.00
    Software + 17.00

    Room Temp – 22.50
    Room Temp + 11.75

    Which two factors are significant and what are the optimum settings for these factors?

    Method & Room Temp

    My question: I believe my DOE is correct except your I can’t figure out the optimum settings for method and room temp. I need two settings. Settings are the levels you used in your experiment.

    0
    #194417

    Robert Butler
    Participant

    This sort of sounds like a homework problem so I’m going to be a bit vague with my answer. However, I’ll provide some hints with respect to what you should do.

    1. You should re-check your analysis – there is only one variable that is significant. If you had plotted your data prior to running the regression (something you should ALWAYS do) the significance of that variable and the lack of significance of the other variables would have been apparent.

    2. If you run the analysis on either the coded variables or on the raw variables and run either backward elimination or forward selection with replacement (stepwise) regression the analysis converges to that same variable and it has a P-value <.0001. The last variable to be eliminated using backward elimination has a P-value prior to elimination is .21 – no significance at all.

    Thus, based on the data provided, there is only one variable that is significant. If you have to choose another variable besides the significant one (say, the last variable to be eliminated using backward elimination)then the "optimum" setting of that insignificant variable (indeed of all of the insignificant variables) would be the setting that incurred the least cost/effort/time/etc.

    0
Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.