Genetic Evolutionary Algorithm in Architecture: Analyzing the Methods of Genetic Evolutionary Algorithms in the Process of Reproducing the Aesthetic Values of Architectural Design
Subject Areas : Special
Seyed Aliyar Ebrahimi Vafayei
1
,
Mahdi Khakzand
2
,
Mohammad Hasan Talebian
3
,
Mohammad Behzadpour
4
,
Fahimeh Motazedian
5
1 - Ph.D. student, Department of Architecture, Pardis Branch, Islamic Azad University, Tehran, Iran
2 - Department of Architecture, Faculty of Architecture and Environmental Design, University of Science and Technology, Tehran, Iran
3 - Department of Architecture, Faculty of Fine Arts, University of Tehran, Tehran, Iran
4 - Educational Institute of Raja Non-Profit University, Qazvin
5 - Assistant Professor, Department of Architecture, Pardis Branch, Islamic Azad University, Tehran, Iran
Keywords: Architectural design, Genetic algorithm, Evolutionary algorithm, Aesthetic,
Abstract :
This study examines the application of the Genetic Algorithm (GA) in the architectural design process with a focus on reproducing aesthetic values such as balance, proportion, harmony, rhythm, and innovation. In today’s world, where the boundaries between art, science, and technology—especially with the advent of artificial intelligence—are increasingly converging, the use of computational tools like genetic algorithms in architectural design has become an inevitable necessity. The fundamental issue addressed in this paper is the challenge of understanding and reproducing the subjective and aesthetic components in architectural design—concepts that are typically considered intuitively and non-quantitatively in traditional design processes. However, genetic algorithms possess the capability to convert these concepts into measurable and quantifiable parameters that can be input into a computational system in the form of a fitness function. Accordingly, the paper’s hypothesis is that by properly designing chromosomes, defining an appropriate objective function, and applying genetic operators such as mutation and crossover, it is possible to generate forms that embody architectural aesthetic values. The research method is qualitative and based on content analysis. In this approach, the authors analyze seven design case studies based on genetic algorithms, examining various stages of architectural design using this algorithm. These stages include encoding forms as digital chromosomes, defining the objective function based on aesthetic criteria, executing an iterative process of generation production, selecting the best forms, and progressively improving design quality.