BHGPT (Built Heritage Generative Pretrained Transformers) is an AI-driven framework designed to improve knowledge retrieval and management in the built heritage context. While traditional methods struggle with heterogeneous sources, inconsistencies, and evolving historical data, this study explores customized GPT, Fine-Tuned BHGPT, and an AI Assistant with Retrieval-Augmented Generation (RAG) to enhance the accessibility and interpretability of heritage information. This study adopts a domain knowledge first and operational approach, in which discipline-specific concepts and terminology guide the organization of information and the structuring of model outputs. The approach begins with a customized GPT model and prompt engineering to test the model's response behavior. After that, fine-tuning the model on specific datasets enhances domain-specific accuracy. Finally, the AI Assistant with RAG integrates structured HBIM data and unstructured archival sources, enabling dynamic querying and cross-referencing of historical and architectural knowledge. The framework is tested on the Sanctuary of Hercules and the Former Segrè Papermill in Tivoli, evaluating its performance across historical evolution, architectural aspects, and interdisciplinary knowledge. Results indicate that Fine-Tuned BHGPT significantly improves site-specific knowledge extraction, while AI Assistant with RAG provides the most flexible and adaptive responses by linking multiple data sources. However, its accuracy is dependent on data availability and retrieval mechanisms. Overall, the contribution clarifies the utility of the BHGPT framework for heritage professionals, enhancing accessibility through natural language querying, producing structured and interoperable outputs, and supporting transparent, source-grounded interpretation.

Classical and industrial archaeologies are a complex cultural field where singularity and uniqueness are expressed through past memory evidence and identity recognition. In order to obtain these values acknowledgement, it is compelling to highlight the material and intangible knowledge by using suitable ICT tools capable of handling complexity and managing large sets of heterogeneous data usually subjected to changes, different interpretations, inconsistencies and sometimes uncertainty. Although the HBIM method has been largely used in the past years, it shows significant limits when dealing with large and heterogeneous information requiring the introduction of advanced methods and tools. In this context, this study presents an approach to the architectural heritage and historical manufacturing activity representation based on integrating the HBIM process with a structured knowledge base, demonstrated through its application to the Sanctuary of Hercules and the former Segrè Papermill case study. The work develops an ontology-based system using existing ontologies for the three domains of interest: architectural artefact, cultural heritage and industrial processes directly connected with the informative model. The intent is to give an overall support system for the complex semantics formalization of these assets to aid the interpretation, intervention and valorization activities

Building Information Modelling (BIM) and Life Cycle Assessment (LCA) are increasingly integrated to support environmental assessment in the built environment. However, applications to existing and heritage buildings remain fragmented, and the contribution of Heritage Building Information Modelling (HBIM) to Life Cycle Assessment is still poorly defined. This study presents a systematic review of 41 peer-reviewed publications retrieved from Scopus and reported according to the PRISMA 2020 guidelines, combining bibliometric, thematic and workflow-oriented analyses. Studies were selected according to predefined eligibility criteria based on relevance to HBIM-LCA integration. Three main BIM–LCA workflow typologies are identified: manual/export-based, plug-in-based and IFC-enabled interoperability approaches. While these workflows have progressively improved automation and data exchange, their application to existing and heritage buildings remains largely scenario-driven and focused on comparing refurbishment, retrofit, adaptive reuse and reconstruction alternatives. The review shows that HBIM distinctive contribution does not lie in environmental calculation methods, but in its ability to structure knowledge related to existing assets, including conservation state, intervention history, information provenance, uncertainty and life cycle transformations. However, current HBIM information structures remain only partially aligned with LCA requirements, particularly regarding service life, end-of-life scenarios and knowledge management. Semantic technologies emerge as a promising pathway towards knowledge-level interoperability. Limitations are related to the heterogeneity of the reviewed studies and the predominance of qualitative evidence. Digital-Green Heritage Workflows integrating HBIM, LCA and semantic technologies enable more transparent, life cycle-oriented and sustainable heritage conservation.