Cumincad : CUMINCAD Papers : Search Results

Hits 1 to 2 of 2

_id	caadria2016_177
id	caadria2016_177
authors	Wortmann, Thomas and Giacomo Nannicini
year	2016
title	Black-Box Optimisation Methods for Architectural Design
source	Living Systems and Micro-Utopias: Towards Continuous Designing, Proceedings of the 21st International Conference on Computer-Aided Architectural Design Research in Asia (CAADRIA 2016) / Melbourne 30 March–2 April 2016, pp. 177-186
doi	https://doi.org/10.52842/conf.caadria.2016.177
summary	Black-box optimization methods play an important role in automated design space exploration, but to-date have not been sys- tematically compared on problems from architectural design optimiza- tion. This paper presents a quantitative comparison of the three major types of black-box optimization: metaheuristics, direct search, and model-based methods. We compare the performance of one repre- sentative algorithm of each type (including a genetic algorithm) on four performance-based design problems of varying complexity and characteristics. Our results show that metaheuristics are greatly out- performed whenever evaluating tens of thousands of design candi- dates is not an option, and suggest direct search and model-based methods as viable and more efficient alternatives.
keywords	Black-box optimization; simulation; direct search; surrogate models; genetic algorithms
series	CAADRIA
email
last changed	2022/06/07 07:57

_id	ecaade2016_072
id	ecaade2016_072
authors	Wortmann, Thomas
year	2016
title	Surveying Design Spaces with Performance Maps - A Multivariate Visualization Method for Parametric Design and Architectural Design Optimization
source	Herneoja, Aulikki; Toni Österlund and Piia Markkanen (eds.), Complexity & Simplicity - Proceedings of the 34th eCAADe Conference - Volume 2, University of Oulu, Oulu, Finland, 22-26 August 2016, pp. 239-248
doi	https://doi.org/10.52842/conf.ecaade.2016.2.239
wos	WOS:000402064400023
summary	This paper presents a novel method to visualize high dimensional parametric design spaces with applications in computational design space exploration. Specifically, the visualization method presented here supports the understanding of design problems in architectural design optimization by allowing designers to move between a high dimensional design space and a low dimensional "performance map". This performance map displays the characteristics of the fitness landscape, develops designers' intuitions about the relationships between design parameters and performance, allows designers to examine promising design variants and delineates promising areas for further design exploration.
keywords	Fitness Landscape; Design Space Exploration; Multivariate Visualization; Optimization; Star Coordinates
series	eCAADe
email
last changed	2022/06/07 07:57

No more hits.