Design variable analysis and generation for performance-based parametric modeling in architecture

Nathan Brown, Caitlin T. Mueller

Research output: Contribution to journalArticle

Abstract

Many architectural designers recognize the potential of parametric models as a worthwhile approach to performance-driven design. A variety of performance simulations are now possible within computational design environments, and the framework of design space exploration allows users to generate and navigate various possibilities while considering both qualitative and quantitative feedback. At the same time, it can be difficult to formulate a parametric design space in a way that leads to compelling solutions and does not limit flexibility. This article proposes and tests the extension of machine learning and data analysis techniques to early problem setup in order to interrogate, modify, relate, transform, and automatically generate design variables for architectural investigations. Through analysis of two case studies involving structure and daylight, this article demonstrates initial workflows for determining variable importance, finding overall control sliders that relate directly to performance and automatically generating meaningful variables for specific typologies.

Original languageEnglish (US)
Pages (from-to)36-52
Number of pages17
JournalInternational Journal of Architectural Computing
Volume17
Issue number1
DOIs
StatePublished - Mar 1 2019

Fingerprint

Learning systems
Feedback

All Science Journal Classification (ASJC) codes

  • Building and Construction
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design

Cite this

@article{f274b06bf2ad4045b3f4e38940fdc706,
title = "Design variable analysis and generation for performance-based parametric modeling in architecture",
abstract = "Many architectural designers recognize the potential of parametric models as a worthwhile approach to performance-driven design. A variety of performance simulations are now possible within computational design environments, and the framework of design space exploration allows users to generate and navigate various possibilities while considering both qualitative and quantitative feedback. At the same time, it can be difficult to formulate a parametric design space in a way that leads to compelling solutions and does not limit flexibility. This article proposes and tests the extension of machine learning and data analysis techniques to early problem setup in order to interrogate, modify, relate, transform, and automatically generate design variables for architectural investigations. Through analysis of two case studies involving structure and daylight, this article demonstrates initial workflows for determining variable importance, finding overall control sliders that relate directly to performance and automatically generating meaningful variables for specific typologies.",
author = "Nathan Brown and Mueller, {Caitlin T.}",
year = "2019",
month = "3",
day = "1",
doi = "10.1177/1478077118799491",
language = "English (US)",
volume = "17",
pages = "36--52",
journal = "International Journal of Architectural Computing",
issn = "1478-0771",
publisher = "Multi-Science Publishing Co. Ltd",
number = "1",

}

Design variable analysis and generation for performance-based parametric modeling in architecture. / Brown, Nathan; Mueller, Caitlin T.

In: International Journal of Architectural Computing, Vol. 17, No. 1, 01.03.2019, p. 36-52.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Design variable analysis and generation for performance-based parametric modeling in architecture

AU - Brown, Nathan

AU - Mueller, Caitlin T.

PY - 2019/3/1

Y1 - 2019/3/1

N2 - Many architectural designers recognize the potential of parametric models as a worthwhile approach to performance-driven design. A variety of performance simulations are now possible within computational design environments, and the framework of design space exploration allows users to generate and navigate various possibilities while considering both qualitative and quantitative feedback. At the same time, it can be difficult to formulate a parametric design space in a way that leads to compelling solutions and does not limit flexibility. This article proposes and tests the extension of machine learning and data analysis techniques to early problem setup in order to interrogate, modify, relate, transform, and automatically generate design variables for architectural investigations. Through analysis of two case studies involving structure and daylight, this article demonstrates initial workflows for determining variable importance, finding overall control sliders that relate directly to performance and automatically generating meaningful variables for specific typologies.

AB - Many architectural designers recognize the potential of parametric models as a worthwhile approach to performance-driven design. A variety of performance simulations are now possible within computational design environments, and the framework of design space exploration allows users to generate and navigate various possibilities while considering both qualitative and quantitative feedback. At the same time, it can be difficult to formulate a parametric design space in a way that leads to compelling solutions and does not limit flexibility. This article proposes and tests the extension of machine learning and data analysis techniques to early problem setup in order to interrogate, modify, relate, transform, and automatically generate design variables for architectural investigations. Through analysis of two case studies involving structure and daylight, this article demonstrates initial workflows for determining variable importance, finding overall control sliders that relate directly to performance and automatically generating meaningful variables for specific typologies.

UR - http://www.scopus.com/inward/record.url?scp=85058644419&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058644419&partnerID=8YFLogxK

U2 - 10.1177/1478077118799491

DO - 10.1177/1478077118799491

M3 - Article

AN - SCOPUS:85058644419

VL - 17

SP - 36

EP - 52

JO - International Journal of Architectural Computing

JF - International Journal of Architectural Computing

SN - 1478-0771

IS - 1

ER -