Hello everybody
I'm doing an experimental design (23 factorial experiment), but I need to optimize that experimental design, I haven't found the exact information to do that, maybe using Statgraphics or ANOVA
Thank you so much for your time and help
Have a nice week
I would match the factorial experiment plan with the ANOVA. However the calculation you will need to analyze a factorial made of 3 factors and 2 levels (23) might be a bit complicated. Then, I suggest you to split the 23 factorial experiments in two different 2x22 factorial experiments which might require an easier calculation. Basically, first make experiments only by changing two factors, then analyze and then start changing the levels of the third factor.
There is an option in statigraphics that will allow you to optimize the design. First, you need to analyze your data and come up with a model. After you have done that, you can use the optimization program in the software to give you the optimal combinations of factors.Keep in mind that with a 2-level design, you will most likely get one of the design points.
I would add a few center points. This will give you an idea about the variability of your system. After you add these design points, I would look at your results see if you need to run a follow up experiment. If not, then I would add some axial points. This will give you an idea about where the "true" optimal point is.
in order to do the optimization , you can use the RSM method , if your curvature is significant , otherwise , you can use simple ANOVA and first order regression model to find the optimize values.
Thank you so much to all of you. I willl use Statgraphics to do the optimization process. Next week I willl share with you my results ;)
You can read this:
http://onlinelibrary.wiley.com/doi/10.1002/ppap.201400008/full
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA