Theory-Lesson 8(DoE) Flashcards
What is the idea of DoE?
Methodology of how to conduct and plan experiments in order to extract the maximum amount of information in the fewest number of runs.
It’s an approach for effectively and efficiently exploring the cause and effect relationship between DVs and OFs.
One At a Time approach
Change a seperate factor at a time
DRAWBACK: I don’t have any information about the factors relationships. It gets stuck far from the optimum.
Full factorial DOE
Gives information about all the main effects, two way interactions and other higher-order interactions.
However, it requires a lot of computational power.
x DVS—> 2^x runs
Fractional factorial DOE
Generated from a full factorial DOE. The idea is to combine some inputs with other inputs (confounded effects)
example: 3 DVS :X1,X2,X3=X1*X2
However, the accuracy is less here beacause only the combined effect of X1*X2 can be studied.
Instead of 2^3 runs, we do 2^2 runs.
Design Resolution
Resolution describes the degree to which estimated main effects are confounded with estimated 2-level interactions, 3-level interactions etc.
Resolution 3:Main effects are counfouned with 2-level interactions c=a*b
Resolution 4: Two factor interactions are confounded with each other cd=ef
Resolution 5: Two factor interactions are confounded with three factor interactions ab=cd*e
Hierarchical Ordering Principle
-Lower order effects are more likely to be important than higher order effects. (x1 more important that x1*x2)
-Effects of the same order are equally likely to be important.
Effect Heredity Principle
In order for an interaction to be significant, at least one of its parents should be significant.
Quadratic Emperical model
The quadratical empirical model can refuce the number of coefficients to be calculated, reducing the computational cost
(2 DVs-> 6 coefficients
bo+b1x1+b2x2+b3x1^2+b4x2^2+b5x1x2
3DVs-> 10 coefficients)
