Towards an Intelligent Built Environment: Digital Building Logbooks and AI Systems for Managing and Reasoning over BIM Data

The digital transformation of the built environment increasingly depends on the ability to manage, integrate, and reason over heterogeneous building data. Although Building Information Modeling (BIM) and standards such as IFC enable structured data exchange, their complexity still limits accessibility and reuse across disciplines. In parallel, Artificial Intelligence (AI) offers new opportunities to make this information more usable and actionable throughout a building’s lifecycle. This thesis advances the vision of an Intelligent Built Environment by presenting two complementary research directions: Digital Building Logbooks (DBLs) for data management and AI systems for intelligent reasoning on building information. The first part introduces a DBL, a modular and interoperable architecture ensuring data traceability, integrity, and accessibility over time. Developed within the Horizon Europe CHRONICLE project, the proposed DBL acts as a data sink, aggregating heterogeneous information from multiple sources through standardized APIs and persistent storage mechanisms. It supports data versioning, validation, and visualization via user-oriented interfaces, establishing a robust foundation for lifecycle information management and interoperability with external systems such as Common Data Environments and Digital Twins. The second part presents two AI-driven systems that enhance accessibility and reasoning on BIM data. Both systems start from BIM models, and build upon them to enable different forms of intelligent interaction. The first, ASK-BIM, allows natural language querying of building data by converting BIM models into knowledge graphs and leveraging Large Language Models to translate user requests into SPARQL queries. The second, a spatial reasoning and AI-based rule-checking system, converts BIM-derived geometric into a PostgreSQL database extended with PostGIS, where a set of dedicated spatial queries captures directional relations, such as in front of, behind, and others, between building elements. An AI module then exploits these queries to evaluate compliance with health and safety requirements, detecting both spatial and non-spatial violations within 3D building scenarios. Together, these contributions show how DBLs and AI reasoning systems address complementary challenges in building data accessibility and analysis, paving the way for more intelligent, interoperable, and data-driven decision-making in the built environment.

Grantor dc:publisher

Università degli Studi di Cagliari

Year dc:date

2026

Author dc:creator

• IBBA, ANDREA

Contributors dc:contributor

• RIBONI, DANIELE
• REFORGIATO RECUPERO, DIEGO ANGELO GAETANO

• Settore INF/01 - Informatica

Statement dc:rights

• info:eu-repo/semantics/openAccess

Language dc:language

eng

Handle dc:identifier

https://hdl.handle.net/11584/477587

OAI identifier oai:identifier

oai:iris.unica.it:11584/477587