Using visualization tools to create kriging models

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

The effective use of trade space exploration requires the generation of a large number of feasible designs to permit investigation of the trade space to identify regions of interest and to understand the local tradeoffs that may exist within that region. This generation of a large number of feasible designs is often too computationally expensive to perform. The information present in a trade space can often be adequately represented with only a few points or examples, but it is difficult to interpret this information in small regions of the space with visualization tools when only a few observations are present. This work presents the use of kriging models to create surrogates to the original system analysis models and generate a large amount of surrogate designs to populate the trade space and perform trade space exploration. This work demonstrates the combination of the previously developed trade space visualization tool with new kriging model creation software. Given the graphical nature of the tool, a graphical approach to assess and update kriging approximations in regions of interest is presented. This approach allows users to leverage multi-dimensional data visualization tools to explore, understand, and identify regions of interest in the trade space. Once regions of interest have been identified, users can select points within these regions and invoke high fidelity simulation models, thereby reducing error in the surrogates.

Original languageEnglish (US)
Title of host publication13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010
DOIs
StatePublished - Dec 1 2010
Event13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO 2010 - Ft. Worth, TX, United States
Duration: Sep 13 2010Sep 15 2010

Publication series

Name13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010

Other

Other13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO 2010
CountryUnited States
CityFt. Worth, TX
Period9/13/109/15/10

Fingerprint

Visualization
Data visualization
Systems analysis

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering
  • Mechanical Engineering

Cite this

Stump, G. M., Martin, J. D., & Simpson, T. W. (2010). Using visualization tools to create kriging models. In 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010 (13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010). https://doi.org/10.2514/6.2010-9030
Stump, Gary Michael ; Martin, Jay Dean ; Simpson, Timothy William. / Using visualization tools to create kriging models. 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010. 2010. (13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010).
@inproceedings{be6c7fbb343941e28b4af37e7c9220d7,
title = "Using visualization tools to create kriging models",
abstract = "The effective use of trade space exploration requires the generation of a large number of feasible designs to permit investigation of the trade space to identify regions of interest and to understand the local tradeoffs that may exist within that region. This generation of a large number of feasible designs is often too computationally expensive to perform. The information present in a trade space can often be adequately represented with only a few points or examples, but it is difficult to interpret this information in small regions of the space with visualization tools when only a few observations are present. This work presents the use of kriging models to create surrogates to the original system analysis models and generate a large amount of surrogate designs to populate the trade space and perform trade space exploration. This work demonstrates the combination of the previously developed trade space visualization tool with new kriging model creation software. Given the graphical nature of the tool, a graphical approach to assess and update kriging approximations in regions of interest is presented. This approach allows users to leverage multi-dimensional data visualization tools to explore, understand, and identify regions of interest in the trade space. Once regions of interest have been identified, users can select points within these regions and invoke high fidelity simulation models, thereby reducing error in the surrogates.",
author = "Stump, {Gary Michael} and Martin, {Jay Dean} and Simpson, {Timothy William}",
year = "2010",
month = "12",
day = "1",
doi = "10.2514/6.2010-9030",
language = "English (US)",
isbn = "9781600869549",
series = "13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010",
booktitle = "13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010",

}

Stump, GM, Martin, JD & Simpson, TW 2010, Using visualization tools to create kriging models. in 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010. 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010, 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference, MAO 2010, Ft. Worth, TX, United States, 9/13/10. https://doi.org/10.2514/6.2010-9030

Using visualization tools to create kriging models. / Stump, Gary Michael; Martin, Jay Dean; Simpson, Timothy William.

13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010. 2010. (13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Using visualization tools to create kriging models

AU - Stump, Gary Michael

AU - Martin, Jay Dean

AU - Simpson, Timothy William

PY - 2010/12/1

Y1 - 2010/12/1

N2 - The effective use of trade space exploration requires the generation of a large number of feasible designs to permit investigation of the trade space to identify regions of interest and to understand the local tradeoffs that may exist within that region. This generation of a large number of feasible designs is often too computationally expensive to perform. The information present in a trade space can often be adequately represented with only a few points or examples, but it is difficult to interpret this information in small regions of the space with visualization tools when only a few observations are present. This work presents the use of kriging models to create surrogates to the original system analysis models and generate a large amount of surrogate designs to populate the trade space and perform trade space exploration. This work demonstrates the combination of the previously developed trade space visualization tool with new kriging model creation software. Given the graphical nature of the tool, a graphical approach to assess and update kriging approximations in regions of interest is presented. This approach allows users to leverage multi-dimensional data visualization tools to explore, understand, and identify regions of interest in the trade space. Once regions of interest have been identified, users can select points within these regions and invoke high fidelity simulation models, thereby reducing error in the surrogates.

AB - The effective use of trade space exploration requires the generation of a large number of feasible designs to permit investigation of the trade space to identify regions of interest and to understand the local tradeoffs that may exist within that region. This generation of a large number of feasible designs is often too computationally expensive to perform. The information present in a trade space can often be adequately represented with only a few points or examples, but it is difficult to interpret this information in small regions of the space with visualization tools when only a few observations are present. This work presents the use of kriging models to create surrogates to the original system analysis models and generate a large amount of surrogate designs to populate the trade space and perform trade space exploration. This work demonstrates the combination of the previously developed trade space visualization tool with new kriging model creation software. Given the graphical nature of the tool, a graphical approach to assess and update kriging approximations in regions of interest is presented. This approach allows users to leverage multi-dimensional data visualization tools to explore, understand, and identify regions of interest in the trade space. Once regions of interest have been identified, users can select points within these regions and invoke high fidelity simulation models, thereby reducing error in the surrogates.

UR - http://www.scopus.com/inward/record.url?scp=84880785929&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84880785929&partnerID=8YFLogxK

U2 - 10.2514/6.2010-9030

DO - 10.2514/6.2010-9030

M3 - Conference contribution

AN - SCOPUS:84880785929

SN - 9781600869549

T3 - 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010

BT - 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010

ER -

Stump GM, Martin JD, Simpson TW. Using visualization tools to create kriging models. In 13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010. 2010. (13th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference 2010). https://doi.org/10.2514/6.2010-9030