Experimental design and optimization - Response surfaces and metamodels

5 important questions on Experimental design and optimization - Response surfaces and metamodels

Does effect insignificant imply corresponding regression to be insignificant and vice versa?

No this does not imply.
--> Do regression directly on raw simulation data

Multiple metamodels can be fit by using the same set of simulation runs if?

I All factors that are significant for any of the responses are included in the design
II all responses are recorded for each run

Formulate a test of quality of approximate response surface of metamodel

1) Simulate system at center point
2) Construct (paired t) confidence interval for differences between simulated values and estimated values from metamodel
3) if CI does not contain 0m difference in statistically significant
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

What is the central Composite Design (CCD)

augment the existing design points with 4 axial points:
2^k design points
1 center point
2k axial points

What is the goal of space-filling design

Spread design points "uniformly" over the experimental region

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo