1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
Dismiss Notice
You must be a registered member in order to post messages and view/download attached files in this forum.
Click here to register.

4 factors, 8 runs. Fractional Factorial Design or Plackett-Burman?

Discussion in 'DOE - Design of Experiments' started by Marcel, Mar 19, 2019.

  1. Marcel

    Marcel New Member

    Joined:
    Mar 19, 2019
    Messages:
    4
    Likes Received:
    0
    Trophy Points:
    1
    Hello everyone,

    My colleague did a design with 4 factors and 8 runs and said its a Plackett-Burman Design with resolution III.

    A picture of his design is attached.

    I read a little bit about DoE now and in my opinion its a resolution IV design and I would call it rather a Fractional Factorial Design than a Plackett Burman design. Am I right or is he right?

    Thanks,
    Marcel
     

    Attached File(s): 1. Scan for viruses before using. 2. Report any 'bad' files by reporting this post. 3. Use at your own Risk.:

    Marcel, Mar 19, 2019
    #1
  2. Miner

    Miner Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    560
    Likes Received:
    474
    Trophy Points:
    62
    Location:
    Greater Milwaukee USA
    This design appears to have been randomized, and is nearly impossible to tell. Can you provide the design in the standard order with the columns in the standard design sequence?

    I attached a Plackett-Burman design in standard order. The fact that the design you provided has 7 columns of which only 4 were used is a strong indicator that it probably is either a PB design or a Taguchi design. Fractional factorial design do not typically show unused columns.

    PB8.png
     
    Last edited: Mar 19, 2019
    Miner, Mar 19, 2019
    #2
  3. Marcel

    Marcel New Member

    Joined:
    Mar 19, 2019
    Messages:
    4
    Likes Received:
    0
    Trophy Points:
    1
    Thanks for the quick answer!

    1.) Unfortunately, I only have this table. Is this design really randomized? Because for each row the signs are moved one position to the left. I thought, this is how you create a Plackett-Burman design.

    2.) Why I came up with the idea, the plan has resolution IV. Wikipedia says:
    "If N is a power of 2, however, the resulting design is identical to a fractional factorial design, so Plackett–Burman designs are mostly used when N is a multiple of 4 but not a power of 2 (i.e. N = 12, 20, 24, 28, 36 …)"

    So according to Wikipedia, it should be identical to a fractional factorial design then? And when I looked up fractional factorial designs with 4 factors and 8 runs I got a resolution of IV. This is what confuses me.
     
    Marcel, Mar 19, 2019
    #3
  4. Miner

    Miner Moderator Staff Member

    Joined:
    Jul 30, 2015
    Messages:
    560
    Likes Received:
    474
    Trophy Points:
    62
    Location:
    Greater Milwaukee USA
    It is a Plackett-Burman design, but the columns are out of sequence from the standard design. PB designs are considered resolution III because there is aliasing between factors and 2-way interactions. However, a true resolution III design, such as you would get from a highly fractionated factorial is 100% aliasing. In a PB design, the aliasing is not 100%, but is only partially aliased, so it really falls somewhere between resolution III and IV.
     
    Miner, Mar 19, 2019
    #4