Today, engineers rely on conventional iterative (often manual) techniques for conceptual design. Emerging parametric models facilitate design space exploration based on quantifiable performance metrics, yet remain time-consuming and computationally expensive, leaving room for improvement. This paper provides a design exploration and explanation framework to augment the designer via a Conditional Variational Autoencoder (CVAE), which serves as a forward performance predictor as well as an inverse design generator conditioned on a set of performance requests. Hence, the CVAE overcomes the limitations of traditional iterative techniques by learning a differentiable mapping for a highly nonlinear design space, thus enabling sensitivity analysis. These methods allow for informing designers about (i) relations of the model between features and performances and (ii) structural improvements under user-defined objectives. The framework is tested on a case-study and proves its potential to serve as a future co-pilot for conceptual design studies of diverse civil structures and beyond.
%0 Journal Article
%1 BALMER2024105411
%A Balmer, Vera
%A Kuhn, Sophia V.
%A Bischof, Rafael
%A Salamanca, Luis
%A Kaufmann, Walter
%A Perez-Cruz, Fernando
%A Kraus, Michael A.
%D 2024
%J Automation in Construction
%K external
%P 105411
%R 10.1016/j.autcon.2024.105411
%T Design Space Exploration and Explanation via Conditional Variational Autoencoders in Meta-Model-Based Conceptual Design of Pedestrian Bridges
%U https://www.sciencedirect.com/science/article/pii/S092658052400147X
%V 163
%X Today, engineers rely on conventional iterative (often manual) techniques for conceptual design. Emerging parametric models facilitate design space exploration based on quantifiable performance metrics, yet remain time-consuming and computationally expensive, leaving room for improvement. This paper provides a design exploration and explanation framework to augment the designer via a Conditional Variational Autoencoder (CVAE), which serves as a forward performance predictor as well as an inverse design generator conditioned on a set of performance requests. Hence, the CVAE overcomes the limitations of traditional iterative techniques by learning a differentiable mapping for a highly nonlinear design space, thus enabling sensitivity analysis. These methods allow for informing designers about (i) relations of the model between features and performances and (ii) structural improvements under user-defined objectives. The framework is tested on a case-study and proves its potential to serve as a future co-pilot for conceptual design studies of diverse civil structures and beyond.
@article{BALMER2024105411,
abstract = {Today, engineers rely on conventional iterative (often manual) techniques for conceptual design. Emerging parametric models facilitate design space exploration based on quantifiable performance metrics, yet remain time-consuming and computationally expensive, leaving room for improvement. This paper provides a design exploration and explanation framework to augment the designer via a Conditional Variational Autoencoder (CVAE), which serves as a forward performance predictor as well as an inverse design generator conditioned on a set of performance requests. Hence, the CVAE overcomes the limitations of traditional iterative techniques by learning a differentiable mapping for a highly nonlinear design space, thus enabling sensitivity analysis. These methods allow for informing designers about (i) relations of the model between features and performances and (ii) structural improvements under user-defined objectives. The framework is tested on a case-study and proves its potential to serve as a future co-pilot for conceptual design studies of diverse civil structures and beyond.},
added-at = {2024-06-24T19:19:40.000+0200},
author = {Balmer, Vera and Kuhn, Sophia V. and Bischof, Rafael and Salamanca, Luis and Kaufmann, Walter and Perez-Cruz, Fernando and Kraus, Michael A.},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/25c1be62995264615ed128a00d8538a27/intcdc_fp2},
doi = {10.1016/j.autcon.2024.105411},
interhash = {0299a3ea4d75304dfc6bdf36a8267031},
intrahash = {5c1be62995264615ed128a00d8538a27},
issn = {0926-5805},
journal = {Automation in Construction},
keywords = {external},
pages = 105411,
timestamp = {2024-06-24T19:19:40.000+0200},
title = {Design Space Exploration and Explanation via Conditional Variational Autoencoders in Meta-Model-Based Conceptual Design of Pedestrian Bridges},
url = {https://www.sciencedirect.com/science/article/pii/S092658052400147X},
volume = 163,
year = 2024
}