iSixSigma

DOE-attribute

Six Sigma – iSixSigma Forums Old Forums General DOE-attribute

Viewing 13 posts - 1 through 13 (of 13 total)
  • Author
    Posts
  • #52944

    Tinoco
    Participant

    I want to find out how to analyze a DOE when you have attribute (pass/fail) response.
     
    For example, I’m doing a DOE on laser marking the parts. First phase was a screening  just to get an acceptable laser mark. Some pieces burned so they were ‘fail’, some pieces didn’t mark at all and those were also ‘fail’. So I have a table that has all these passes and fails but it doesn’t tell me how to adjust my parameters to get where I want to go.

    0
    #186917

    …thanapost
    Participant

    So you want to build a reliable optimization model using experimental data which consists of a binary Dependent  Variable with some combo of factors/covariates for your Independent Variables?
    I am no expert, but in your shoes I would consider going back to ‘design’ and finding a way to use a continuous response…you’ll get much greater precision in your model and the analysis will be straight-forward…..otherwise, all I can think offer is a binary logistical regression approach…good luck…and verfiy.

    0
    #186921

    Robert Butler
    Participant

      It sounds to me like you have a continuous variable that you are forcing into an attribute response.  You said, “Some pieces burned so they were ‘fail’, some pieces didn’t mark at all and those were also ‘fail’ In other words things are going from burning to not burning and everything in between. Somewhere in there is the correct amount for marking – how is this correct amount measured – depth of mark, measure of char, or????
     

    0
    #186922

    StuW
    Member

    Compute the percentage pass (or fail) for each cell, combination of -1 and +1 factors, in your design.   Use that result as the dependent variable in the analysis.   The factor level combinations are your set of independent variables in the matrix. 
    Several points, some of which were covered with the previous responses.   If you have too few units in each treatment combination, you may not see any discrimination in the results, or if the results are too close to zero or 100%, you may well need a transformation applied in order to analyze the data appropriately. 
    If the type of failure is also a concern, no mark versus burned, than you may need multiple response variables to address that concern, also.  A Pareto of the failure types by treatment combinations might also be of use before going too much further.

    0
    #187687

    anupam
    Participant

    Where R U going ?
    For DOE you need atleast two parameters for which the result is measured. In your case result is Pass/Fail.
    But where are parameters ?
    Temperature, Intensity of Laser, Time of Exposure etc.
     

    0
    #187688

    Remi
    Participant

    Hai Anthony,
    Ask your experts to give a score on the quality of the mark on a scale of 1 to 10: a “5” is considered perfect marking; a “1” is for ‘burned’ and a “10” is for ‘no mark’. In this way you have a ranking of the laser marking from ‘too much’ (1) through ‘good’ (5) until ‘too little’ (10).Use this score as the Y of your DoE analysis and optimize to a Target  of  “5” (you don’t want the maximum or minimum Y). Don’t forget to check the residuals.
    For finding the perfect optimal setting this method probably is too crude but for a screening it will work all right.Maybe you even can finetune with decimal points if all products are between “4” and “6”.
    Good luck, Remi

    0
    #187690

    Thejasvi
    Member

    Hi
    Your “Y” is following a binomial distribution.
    First, for each factor-level combination find our the no. of success and failures. Then caluclate “p” which is success/failure. Then take square-root of p. For this square-root of p, take the trignometric function “Sine inverse”. Now use this Sine inverse root p value as your “Y” and carry out DOE (more like a one-way ANOVA).
    Do let me know your results.
    All the best!!!!

    0
    #187691

    Thejasvi
    Member

    Hi
    Your “Y” is following a binomial distribution.
    First, for each factor-level combination find our the no. of success and failures. Then caluclate “p” which is success/failure. Then take square-root of p. For this square-root of p, take the trignometric function “Sine inverse”. Now use this Sine inverse root p value as your “Y” and carry out DOE (more like a one-way ANOVA).
    Do let me know your results.
    All the best!!!!

    0
    #187692

    bbusa
    Participant

    Bizarre!

    0
    #187707

    roarty
    Participant

    Hi Thejasvi,
    Can you explain why you recommend this series of steps.

    0
    #187721

    Thejasvi
    Member

    Hi
    These steps are carried out to meet the properties of normal distribution (remember, we are transforming a binomial distribution into a normal distribution)

    0
    #187722

    bbusa
    Participant

    He who is good with the hammer , thinks everything is a nail !

    0
    #187733

    Bower Chiel
    Participant

    Hi
    A few comments: –
    1.  Information on the arcsine transformation may be found at http://demonstrations.wolfram.com/TheArcsineTransformationOfABinomialRandomVariable/
    2.  An example of the use of the transformation in a DoE scenario may be found at http://www.asq.org/data/subscriptions/qp/2004/0304/qp0304million.html
     
    3.  The book by Ellis R Ott and others entitled Process Quality Control: Troubleshooting and Interpretation of Data has a chapter entitled Troubleshooting with attribute data which has lots of sound advice on factorial experimentation where the responses are based on counts.
     
    Of course, prior to any formal analysis or transformation of the data, one can often gain valuable insights by plotting the data from a factorial experiment in main effects and interaction plots.
     
    Best Wishes
     
    Bower Chiel
     

    0
Viewing 13 posts - 1 through 13 (of 13 total)

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