| _id |
5007 |
| authors |
Elezkurtaj, Tomor and Franck, Georg |
| year |
1999 |
| title |
Genetic Algorithms in Support of Creative Architectural Design |
| source |
Architectural Computing from Turing to 2000 [eCAADe Conference Proceedings / ISBN 0-9523687-5-7] Liverpool (UK) 15-17 September 1999, pp. 645-651 |
| doi |
https://doi.org/10.52842/conf.ecaade.1999.645
|
| summary |
The functions supported by commercial CAAD software are drawing, construction and presentation. Up to now few programs supporting the creative part of architectural problem solving have become available. The grand hopes of symbolic AI to program creative architectural design have been disappointing. In the meantime, methods called referred to as New AI have become available. Such methods includegenetic algorithms (GA). But GA, though successfully applied in other fields of engineering, still waits to be applied broadly in architectural design. A main problem lies in defining function in architecture. It is much harder to define the function of a building than that of a machine. Without specifying the function of the artifact, the fitness function of the design variants participating in the survival game of artificial evolution remains undetermined. It is impossible to fully specify the fitness function of architecture. The approach presented is one of circumventing a full specification through dividing labor between the GA software and its user. The fitness function of architectural ground plans is typically defined in terms only of the proportions of the room to be accommodated and certain topological relations between them. The rest is left to the human designer who interactively intervenes in the evolution game as displayed on the screen. |
| keywords |
Genetic Algorithms, Creative Architectural Design |
| series |
eCAADe |
| email |
|
| last changed |
2022/06/07 07:55 |
