iSixSigma

Can you run a Randomized Block Design on an unstable process?

Six Sigma – iSixSigma Forums Old Forums General Can you run a Randomized Block Design on an unstable process?

Viewing 1 post (of 1 total)
  • Author
    Posts
  • #47354

    newbie
    Participant

    Forum, I am reading Statistics for Experimenters, 2nd ed, Box, Hunter, Hunter, pg. 154….and have a question:The text speaks of:¨In his earliest thinking about the design of experiments in the 1920s, Fisher had to discover how to run experiments on processes and systems that were never in a state of statistical control…Fishers solution to the quandary of how to run such experiments was the invention of randomized blocks…He showed it was possible to obtain results that to an adequate app. could be analyzed ¨as if¨ the usual assumptions of independence and indentical distribution were in fact true¨.Ok, so the question is, does this not violate the assumption of stability i have always heard about…ie if the process is out of control (CC or RC) then you can do nothing until you fix the special cause? Doesnt this imply you could indeed move forward using this experimental design even if your process exhibits instability? THANKS!!!

    0
Viewing 1 post (of 1 total)

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