Performance-Driven Design of a Residential Unit with the Use of Genetic Algorithms

This study examines and integrates the principles of multi-objective optimization, space syntax, and graph analysis in the context of Artificial Intelligence (AI) applications for architecture and construction, particularly automated space planning. The primary purpose of the study is to determine the efficacy of various advanced strategies for generating a wide range of design options. Building codes, customer requirements, and other architectural constraints must be addressed while developing optimal solutions. After selecting the parameters to deal with, the process aims to thoroughly study design possibilities while balancing features in competing positions. The process adopts space syntax and graph analysis to visually represent the functional connections between various spaces and users’ movements within the building. The analyses are carried out within the Grasshopper platform using the Wallacei and Termite Nest plug-ins. Wallacei employs multi-objective evolutionary ptimization methods to tackle conflicting design issues, whereas Termite Nest develops design possibilities combining space syntax and graph analysis. This study aims to demonstrate the limitations and the potential of computerized architectural space planning by showing how these advanced algorithms can generate optimized design solutions that meet particular requirements.