CumInCAD is a Cumulative Index about publications in Computer Aided Architectural Design
supported by the sibling associations ACADIA, CAADRIA, eCAADe, SIGraDi, ASCAAD and CAAD futures

id sigradi2006_e011c
authors Narahara, Taro and Terzidis, Kostas
year 2006
title Optimal Distribution of Architecture Programs with Multiple-constraint Genetic Algorithm
source SIGraDi 2006 - [Proceedings of the 10th Iberoamerican Congress of Digital Graphics] Santiago de Chile - Chile 21-23 November 2006, pp. 299-303
summary A genetic algorithm (GA) is a search technique for optimizing or solving a problem based on evolutionary biology, using terms and processes such as genomes, chromosomes, cross-over, mutation, or selection. The evolution starts from a population of completely random individuals and happens in generations. In each generation, the fitness of the whole population is evaluated, multiple individuals are stochastically selected from the current population (based on their fitness), modified (mutated or recombined) to form a new population, which becomes current in the next iteration of the algorithm. In architecture, GAs are of special interest mainly because of their ability to address a problem offering a multiplicity of possible solutions. Contrary to other algorithms where the objective is to accommodate a manually conceived diagram, GAs are emergent procedures that evolve over time through multiple attempt cycles (i.e. generations) and therefore offer a bottom-up approach to design. In addition, by using the computational power of computers they can resolve complex interactions between multiple factors and under multiple constraints offering solutions that occasionally surprise the designer. One of the main problems in architecture today is the quantity of the information and the level of complexity involved in most building projects. As globalization and economic development has started to arise at unprecedented levels, the need for large urban developments have become commonplace. Housing projects for a few hundreds to thousands of people have started to emerge over large urban areas. In such cases, the old paradigm for housing design was the development of high rises that served as stacking devices for multiple family housing units. Such a direction was unfortunately the only way to address excessive complexity using manual design skills mainly because it was simple to conceive but also simple to construct. The unfortunate nature of this approach lies rather in the uniformity, similarity, and invariability that these projects express in comparison to individuality, discreteness, and identity that human beings and families manifest. One of the main areas of complexity that could benefit architecture is in housing projects. In these projects there is a typology of residential units that need to be combined in various schemes that will fulfill multiple functional, environmental, and economic constraints. In this paper, the design of a 200-unit residential complex on a corner of two streets in an urban context was investigated as a case study. Recent advancement in tectonics and structural engineering enables the realization of buildings in mega scales and starts to introduce another layer of complexity into the building programs. Conventional design methods relying on the preconceived knowledge based approaches are no longer reliable. Beyond the certain quantitative factors and the complexity of the problems, search occasionally enters into the unpredictable domain of the human perception. Computational approaches to design allows us to go through thousands of iterations in a second and find the solution sets beyond the reach of designers’ intuitive search spaces. Genetic Algorithm can be a potential derivative for finding optimum design solution from indeterminate search spaces constrained by multi dimensional factors.
keywords Genetic Algorithm; Housing Design; Multiple-constraint
series SIGRADI
email narahara@mit.edu
full text file.pdf (357,750 bytes)
references Content-type: text/plain
details citation check to select
100%; open Caldas, Luisa G. and Norford, Leslie K. (1999) Find in CUMINCAD A Genetic Algorithm Tool for Design Optimization , Media and Design Process [ACADIA ‘99] Salt Lake City 29-31 October 1999, pp. 260-271
100%; open Elezkurtaj, Tomor and Franck, Georg (1999) Find in CUMINCAD Genetic Algorithms in Support of Creative Architectural Design , Architectural Computing from Turing to 2000 [eCAADe Conference Proceedings] Liverpool (UK) 15-17 September 1999, pp. 645-651
100%; open Holland, J. (1992) Find in CUMINCAD Genetic Algorithms , Scientific America, July 1992
100%; open Jakimowicz, A., Barrallo, J. and Guedes, E.M. (1997) Find in CUMINCAD Spatial Computer Abstraction: From Intuition to Genetic Algorithms , CAAD Futures 1997 [Conference Proceedings] München (Germany), 4-6 August 1997, pp. 917-926
100%; open Narahara T. and Terzidis K. (2006) Find in CUMINCAD Multipleconstraint Genetic Algorithm in Housing Design , ACADIA 2006 [International Conference, Synthetic Landscapes | Digital Exchange] Louisville (USA) 12-15 October 2006
100%; open Pereira, A.G., Munda, G. and Paruccini, M. (1994) Find in CUMINCAD Generating Alternatives For Siting Using Genetic Algorithms and Multiple Criteria Decision Techniques , Second Design and Decision Support Systems in Architecture & Urban Planning (Vaals, the Netherlands), August 15-19, 1994
100%; open Snijder, H.P.S. (1994) Find in CUMINCAD The Use of Genetic Algorithms in Spatial Optimisation Problems , Second Design and Decision Support Systems in Architecture & Urban Planning (Vaals, the Netherlands), August 15-19, 1994

last changed 2016/03/10 08:55
HOMELOGIN (you are user _anon_99449 from group guest) Works Powered by SciX Open Publishing Services 1.002