Generative AI for Sustainable Architectural Design Optimization
Scientific Journal of Technology · 21.05.2025 · 7 (5) · 96-106
DOI: 10.54691/czaf4b21 ↗Kurz gefasst
Fachartikel in der Zeitschrift Scientific Journal of Technology, online erschienen am 21.05.2025. Verfasst von Lin Yang. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 29 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
This paper examines recent academic research on applying generative AI and computational methods to optimize sustainable architectural design. Buildings account for roughly one-third of global carbon emissions and energy use, especially via HVAC systems. Early-stage design optimization is therefore crucial to improve efficiency and reduce environmental impact. Generative design including evolutionary and parametric algorithms has emerged as a key approach to exploring diverse design options for sustainability. More recently, advanced AI (e.g. machine learning surrogates, GANs, diffusion models) are being incorporated to accelerate and enrich design optimization. This paper systematically survey peer-reviewed literature (2020–2024), focusing on methods (evolutionary algorithms, ML surrogates, GANs), application domains (energy, daylighting, carbon), and performance outcomes. Key findings include evidence that generative design methods can substantially improve energy and thermal performance (e.g. reported reductions of 23–28% and that genetic algorithms remain widely used. Emerging themes include multi-objective optimization (balancing energy, comfort, cost) and use of deep learning (surrogate models, GAN-based generation) to expand design search. We highlight agreements and gaps: consensus on benefits of generative optimization for sustainability, but challenges remain in model explainability, data requirements, and transferability across contexts. The paper identifies that most studies use quantitative simulation-optimization loops, whereas few provide holistic frameworks and conclude that generative AI holds great potential for sustainable architecture design, but future work must address data/ethical challenges and integrate multi-scale, user-involved approaches.
Zitieren
APA
Yang, L. (2025). Generative AI for Sustainable Architectural Design Optimization. Scientific Journal of Technology, 7(5), 96-106. https://doi.org/10.54691/czaf4b21
Vancouver
Yang L. Generative AI for Sustainable Architectural Design Optimization. Scientific Journal of Technology. 2025;7(5):96-106. doi:10.54691/czaf4b21
BibTeX
@article{Yang2025,
author = {Yang, Lin},
title = {Generative AI for Sustainable Architectural Design Optimization},
journal = {Scientific Journal of Technology},
year = {2025},
volume = {7},
number = {5},
pages = {96--106},
doi = {10.54691/czaf4b21}
}
EUVIMED