ello everyone,
Currently, I wish to realize an isovariant CCD response surface design of experiment (2 levels for 6 factors) for the study of the extraction of a plant. Obviously, this means that I wish to estimate the effects of the different factors (Bi, Ba, Bb, Bc, Bd, Be and Bf), the interaction effects (B(ab), B(ac), B(ad), B(ae), B(af), B(bc), B(bd), B(be), B(bf), B(cd), B(ce), B(cf), B(de), B(df) and B(ef)) and the quadratic effects (Ba^2, Bb^2, Bc^2, Bd^2, Be^2 and Bf^2).
First of all, I have determined the best experimental design using the function :
library(rsm)
library(base)
library(readxl)
library(RcmdrPlugin.DoE)
library(mixexp)
ccd.pick(6, n.c = 2^5, n0.c = 1:10, blks.c = 1, n0.s = 1:10, bbr.c = 1, wbr.s = 1, bbr.s = 1, best = 10, sortby = c("agreement", "N"))
I kept the value of alpha = 2.3784 to obtain an isovariant design of experiment.I then performed it on Excel named DataDOE (in Archive PDF: Archivez et partagez facilement vos documents. Sorry but I'm new and i can't upload file). To check that the formulas I was doing on Rstudio are consistent. I made a "simulation" by inventing the effects of the different factors and the corresponding answers (top). My objective is, thanks to the Rstudio software, to find the effects of the different factors from the answers. This will be the proof that it works. You will thus find in the attached excel file, the matrix of the experimental design (middle), the matrix of experiment (bottom left) as well as the values of the different levels of each factor -alpha, -1, 0, 1, alpha (bottom right).
The imported excel file named "Matexp" is the matrix at the middle of the "DOEdata" file.
I am new to the R software and have done some research without success. Here is what I have done:
Matexp <- read_excel("C:/Users/ctaillandier/Desktop/Matexp.xlsx")
View(Matexp)
DOECod <- coded.data(Matexp, A ~ (Temp. - 50)/0, B ~ (Hydrom. - 8)/0, C ~ (Agit. - 225)/0, D ~ (Light - 4.3)/0, E ~ (Duree_Extract. - 17.5)/0, F ~ (pH - 7)/0)
as.data.frame(DOECod)
DOE1.rsm <- rsm(Y ~ FO(A, B, C, D, E, F), data=DOECod)
summary(DOE1.rsm)
DOE1I.rsmi<-update(DOE1.rsm,. ~ . + TWI(A, B, C, D, E, F))
summary(DOE1I.rsmi)
head(Matexp)
library(rsm)
DOECCD <- rsm(Y ~ SO(A, B, C, D, E, F), data = Matexp)
summary(DOECCD)
After all these lines, I can't find the values of the effects I simulated. This shows how wrong I was in my approach.
Thank you for your attention to this message. Yours sincerely,
Charli T