Ferramenta de mapeamento do risco de degradação em edifícios através de inteligência artificial

The overarching goal of this project is the development of an AI-based tool that will revolutionize the conservation of historic buildings by providing a predictive analysis of facade degradation risks, facilitating the early detection of potential deterioration, thereby enabling timely and cost-effective interventions. To accomplish this goal, several partial objectives must be achieved. They can be grouped in two levels: BUILDING level: • Collect a diverse dataset of historical building facades, ensuring representation of different architectural styles, materials, and environmental conditions. • Inspect selected case studies to document current degradation status, including visual assessments, photographic records, and thermographic imaging. • Thoroughly document the data collection and inspection processes and protocols, ensuring transparency and reproducibility of the methodologies employed. • Establish standardized protocols for data acquisition to facilitate scalability and comparability across diverse case studies. • Conduct comprehensive laboratory and in-situ tests to analyze material properties and degradation behavior under controlled conditions. • Utilize the insights gained from lab tests to validate, refine, and improve existing degradation models. TOOL level: • From the collected data (photography and thermal images) and using computer vision, implement machine learning algorithms to recognize patterns in buildings facades, identify degradation types, severity levels, and potential contributing factors, and build robust classification models. • Develop comprehensive innovative risk assessment models that use the data from photographs, thermographic imaging, and classification results as inputs in the new degradation models. • Quantify the risk of degradation, enabling prioritization of preservation efforts, and analysis of degradation risk in the context of the entire building. • Integrate the developed tool with BIM platforms to facilitate the visualization and ensure seamless integration with existing BIM workflows and tools for broader applicability. • Ensure that the AI-system is self-improving so that it will continuously learn from new data, ensuring the tool's is always better performing in terms of reliability and accuracy. In essence, the project is about setting a new standard in the preservation of historical architecture, leveraging the latest advancements in AI to ensure our cultural heritage is well maintained. The success of this initiative will mark a significant leap forward in heritage conservation. Ultimately, the project aspires to empower heritage preservation institutions with a solution for safeguarding cultural legacies worldwide.