AOAC Gluten Quantitative Validation Guidance-Round 1(Nov 2023)

Generating Fractional Factorial Designs in R 457 You can generate additional fractional factorial designs in R using the following commands. The example 458 here is for a 2 5-1 design, which has a level V resolution – this is the same example shown in Table 4: 459 >install.packages(“FrF2”) 460 >library(FrF2) 461 >fivefactors <- FrF2(nfactors = 5, resolution = 5, randomize = FALSE) 462 [In this code, “fivefactors” is a name you make up to describe the table we are trying to 463 generate, nfactors is the number of factors you are varying, and resolution is your chosen 464 resolution (you will want to keep it at 4 or 5 to avoid having main effects confounded with each 465 other).] 466 >summary(fivefactors) 467 [the summary command gives you the results of the analysis, shown below] 468 Call: 469 FrF2(nfactors = 5, resolution = 5, randomize = FALSE)

470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500

Experimental design of type FrF2

16 runs

Factor settings (scale ends):

A B C D E

1 -1 -1 -1 -1 -1 2 1 1 1 1 1

Design generating information:

$legend

[1] A=A B=B C=C D=D E=E

$generators [1] E=ABCD

Alias structure:

[[1]]

[1] no aliasing among main effects and 2fis

The design itself:

A

B

C

D

E

1 2 3 4 5 6 7

-1

-1 -1

-1 -1 -1 -1

-1 -1 -1 -1 -1 -1 -1

1

NOTE: In this result display, -1 indicates the lower level of the factor, and 1 indicates the

1

-1 -1

-1

1 1

1

1

higher level of the factor.

-1

-1 -1

1 1 1

-1

Compare to Table 4.

1

1 1

-1

1