iSixSigma

Censored DOE Data

Six Sigma – iSixSigma Forums Old Forums General Censored DOE Data

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #44217

    Ropp
    Participant

    I have just completed a DOE, my “y” is adhesion strength measured by pull force.  The experiement had 2 levels, one with 4 factors and the other with 3, with 3 replicates for a total of 36 runs.  In 2 of the runs the part broke before the adhesion gave way, leaving me with censored data.  I can’t very well use those parts but that of course leaves me with an unbalanced design.  Anyone run into this before?  Advice?

    0
    #141367

    Novelis_BB
    Participant

    Hello Dave:Which software are you using? Most of the statistical software can handle this problem. Since you have three replicates, effect of two missing observations often is minor. Correlation matrix will be usually helpful to determine if this might be the case. If you can copy paste your DOE here, me or someone else can help you. Note: Can you explain in detail about the experiment that you ran? Like, how many factors, each factor at what level, etc? Some of the things you mentioned did not make much sense, may be I am missing something.Novelis_BB

    0
    #141376

    BTDT
    Participant

    Dave:Keep all the data. Minitab will permit you to analyze the censored data you have under “Stat-Reliability/survival-regression with life data”. Have a look and see if that helps.Cheers, BTDT

    0
Viewing 3 posts - 1 through 3 (of 3 total)

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