Impact of Sampling Technique Selection on the Creation of Response Surface Models 2004-01-3134
This paper evaluates and compares a variety of sampling techniques, including both classical and modern Designs of Experiments, to create a more structured approach to selecting the most apt DoE for a specific type of problem. Six different designs are investigated through a design analysis for a notional commercial aircraft. The appropriateness of each sampling technique is determined based on a number of criteria, including code execution time, independent variable correlation, and distribution of data points throughout the design space. Additionally, the resulting models are evaluated using a systematic procedure for checking quality to quantify the accuracy and predictive capability of a given model.
Citation: Barros, P., Kirby, M., and Mavris, D., "Impact of Sampling Technique Selection on the Creation of Response Surface Models," SAE Technical Paper 2004-01-3134, 2004, https://doi.org/10.4271/2004-01-3134. Download Citation
Author(s):
Peter A. Barros, Michelle R. Kirby, Dimitri N. Mavris
Affiliated:
Aerospace System Design Laboratory (ASDL), School of Aerospace Engineering, Georgia Institute of Technology
Pages: 14
Event:
World Aviation Congress & Exposition
ISSN:
0148-7191
e-ISSN:
2688-3627
Also in:
SAE 2004 Transactions Journal of Aerospace-V113-1
Related Topics:
Commercial aircraft
SAE MOBILUS
Subscribers can view annotate, and download all of SAE's content.
Learn More »
