Special Issue on AI Approaches to Architectural Design

This special issue presents state-of-the-art AI approaches to architectural design and new reflections on the relationship between human cognition and built environment. The computational approaches prior to 2015 mainly stemmed from optimisation, rule-based systems, and generative systems. They unintentionally inherited a Cartesian dichotomy between the human mind and the world. The rise of deep learning and foundation models over the last decade has reshaped the landscape of design thinking and digital tools. Engaging and learning the world, instead of symbolic modelling, has become the focus of cutting-edge research. Then architects may define themselves through the mirror of AI.

Architectural design, once formulated as ‘wicked problem’, is undoubtedly a very complex task. The GAN model, graph network, diffusion models, and transformers have tackled the design problem in unprecedented ways. The scope of AI-driven approaches is also extended to construction intelligence, embodied cognition, and knowledge mining, among many others. However, understanding and manipulating high-level concepts of architecture based on data is still a work in progress.

Topics covered

These include, but are not limited to:

• Embodied cognition of built environment from AI perspectives
• Novel applications of AI, especially Multimodal Large Language Models, to architectural design
• AI innovations in the construction, structure, and material of architecture
• Higher-level processing of building representations, such as BIM files, CAD drawings, and mesh

Important deadline

• Submission deadline: 8 November 2024

Submission instructions

Please read the Guide for Authors before submitting. All articles should be submitted online; please select SI: AI Approaches to Architectural Design on submission.

Articles accepted for this special issue will be published free of Article Processing Charges (APC). For more information about the Journal's APC, please visit this page.

Editor

Prof. Peng Tang, Southeast University, China. Email: tangpeng@seu.edu.cn