Toward Semantic-Driven Building Data Architectures for AI-Based Architectural Design

With the rapid advancement of digitalization and artificial intelligence, architectural design faces a fundamental limitation: the dominance of geometric representations with insufficient semantic depth. This imbalance restricts the interpretability, reasoning capacity, and transferability of AI-driven design systems across the building lifecycle. Addressing this challenge, the present study explores how architectural geometry models can be semantically enhanced to support AI-based modeling and reasoning, laying the foundation for more intelligent, explainable, and integrated design processes.

Limit thinking in Geometry-Centric Architectural Models

The study critically examines the limitations of conventional geometric representations across the design–performance–construction chain. Through bibliometric analysis and an extensive literature review, it reveals how geometry-only models fail to capture intent, performance logic, construction rules, and associative knowledge. These shortcomings constrain interoperability between domains and limit the effectiveness of AI applications in generative design, performance optimization, and intelligent construction.

Classification of Semantic Enhancement Methods

A key contribution of the research is the systematic categorization of semantic enhancement techniques in architectural geometry. The study identifies three major research trajectories: data-driven geometric representation, ontology- and rule-based structured semantics, and cross-modal geometry–semantic integration. Within these trajectories, nine distinct semantic enhancement methods are classified, offering a comprehensive overview of current approaches and their respective capabilities and limitations.

Domain-Specific Semantic Requirements

The research synthesizes semantic requirements across three critical architectural domains. Generative design demands multimodal semanticization of geometry to support flexible and interpretable form generation. Building performance optimization requires bidirectional coupling and real-time feedback between performance data and geometric models. Intelligent construction relies on rule-based, executable semantics that enable automated reasoning, validation, and traceability throughout the construction process.

Spatial Semantic Information Model (SSIM)

Building on these findings, the study formally proposes the Spatial Semantic Information Model (SSIM) as a multilayer, graph-structured semantic web. Centered on a minimal semantic unit, SSIM unifies multidimensional domains—Form, Performance, Stream, Intent, and Rule—within a single computational framework. This model establishes a shared semantic foundation capable of supporting AI-driven reasoning across the entire building lifecycle.

AI-Oriented Data Architecture and Future Implications

Through graph–tensor bidirectional mapping, the proposed architecture defines critical constraints such as semantic closure, cross-domain alignment, and data governance. These conditions enable natural language–driven geometric generation, integrated performance-based construction workflows, and traceable intelligent construction processes. By providing a scalable and extensible semantic basis, the study offers theoretical support for future AI research agendas and the evolution of intelligent architectural systems.

Architecture Engineers Awards

🔗 Nominate now! 👉 https://architectureengineers.com/award-nomination/?ecategory=Awards&rcategory=Awardee 🌐 Visit: architectureengineers.com 📩 Contact: contact@architectureengineers.com Get Connected Here: ***************** Instagram :  https://www.instagram.com/architecture_engineers_awards/ Facebook :  https://www.facebook.com/profile.php?id=61576995475934 Tumblr :   https://www.tumblr.com/architectureengineersawards Pinterest :   https://in.pinterest.com/architectureengineersawards/ Blogger :   https://architectureengineers.blogspot.com/ Twitter :   https://x.com/Architectu54920 YouTube :  https://www.youtube.com/@Architechtureengineer LinkedIn :  https://www.linkedin.com/in/architecture-engineer-01a044361/