roy e.


  • roy e. replied to the topic Sample Size Question in the forum General 14 years, 1 month ago

    Yianni… as other posters said, first step is to define the tolerance and type of analysis. Once you have define those parameters, if your analysis is by attributes, pass/do not pass, an easy way to sample is using the Military Standard tables (in this case 105E). You can choose according with your AQL and several sampling plans (from very…[Read more]

  • As other have noted in this thread I do not think the real concern is about the number of data point, but about the need for a 6 level experiment. I agree on revisiting the logic behind the 6 levels to attempt simplifying to 2 or 3.

  • roy e. replied to the topic Box-Cox Transformation in the forum General 17 years, 9 months ago

    Melissa I would caution you against the used of back-transformed parameters directly.  If you data is not normal on the first place, What would be the interest of describing it with parameters typicaly used for normal distributed populations?
    Having said that, you can calculate the mean and median directly with your data, without transforming.  B…[Read more]

  • Thanks for answering guys.  I’ll look into BMP site.  Although I agree with Pepe, I believe as BB one of my responsibilities is to look for opportunities to maximize the valu of the organizational capability we have built around SS.  If my company is moving into SCOr, why not integrate both? Shouldn’t I, as BB, look for answers and options to pr…[Read more]

  • roy e. replied to the topic Simple Question…? in the forum General 17 years, 11 months ago

    Gabriel very informative response.  I learn a lot from the way you would present teh data.  Thanks a lot!

  • roy e. replied to the topic TRIZ in the forum General 18 years, 3 months ago

    Geoff Thanks a lot.  The matrix is very helpful.  We are not using TRIZ yet… but I want to get as informed as possible just to be prepared.  In your experience, how much TRIZ is used for non-technological problems?  Even though I work on a manufacturing business, most of the projects we have within out SS effort are non-technological.

  • roy e. replied to the topic TRIZ in the forum General 18 years, 3 months ago

    Geoff this is a great summary, thank you!. Would you know where can get a copy of the contradiction matrix you mentioned?  Thanks again

  • roy e. replied to the topic seis sigma in the forum General 18 years, 7 months ago

    Hola Diego.  Andrea ya resumió los aspectgos más importantes, así que dejame ver si puedo agregar algunos detalles.
    Seis Sigma nace en Motorola en los años 80 y crecre con la fundación del Six Sigma Institute (Allied Signal, IBM, TI).  El caso de implementación más “famoso” (sin entrar en discusión de bueno o malo)  ha sido General Electric,…[Read more]

  • roy e. replied to the topic Mislabeled Specimens in the forum General 18 years, 8 months ago

    I did a DMAIC on a labeling process for food products. In a nutshell we went from 3.2 to 5.6 sigma (about 20ppm).  Critical defects (mismatched, wrong info) were virtually eliminated, or at least robust process controls were established as part of the PMCS to capture them on early stages.  Most of our current defecst are missing lables d…[Read more]