focus is on the investigation of the main effects and interactions can be disregarded
[40]. However, a mixing of the effects can occur [38, 41].
2.2 Optimization Designs
In order to maximize a response, the levels of the influencing factors are optimized in
so-called optimization designs. Therefore, the most known designs, like the central
composite, Box-Behnken, optimal, and space-filling designs, are briefly introduced
in the following.
F Factor A
Factor C
Factor B
----+
+-+-+
+++
-++
-+2 3 -factorial
2 3-1 -fractional factorial
Central Composite Face Centered
Central Composite Inscribed
Box-Behnken
D-Optimal
Latin-Hypercube-Sample
Central Composite Circumscribed
A
B
C
D
E
F
G
H
Fig. 1 Geometrical representation for screening (a, b) and optimization designs (c–h) and optimization designs with three factors (Factor a, Factor b, and Factor c). Dots represent the recommended
experiments. The gray dots are the star points, and the black dots are the central points. All designs
are examined at two levels (+ and -)
Digital Twins and Their Role in Model-Assisted Design of Experiments
35