Ontologies and knowledge graphs provide a semantic approach to modelling engineering knowledge and making it machine-interpretable. They enable the integration of heterogeneous data sources, such as building information models (BIM), databases, technical standards, sensor data, and documents, thereby creating a reliable knowledge base for digital engineering.

At the same time, recent advances in artificial intelligence—particularly generative AI and large language models (LLMs)—are transforming engineering workflows by supporting design, knowledge retrieval, code compliance checking, and decision-making. Combined with ontologies and knowledge graphs, these technologies can provide more accurate, explainable, and trustworthy AI systems grounded in engineering knowledge.

Therefore, we aim to create a platform for knowledge exchange on AI- and ontology-supported design of concrete structures. The special session will bring together researchers and practitioners to present recent advances and discuss how AI, generative AI, ontologies, and knowledge graphs can support the design, construction, maintenance, and lifecycle management of concrete structures.

This special session focuses on reinforced concrete heritage as a field for training, knowledge transfer, and professional capacity building. Complementing existing fib 2027 sessions on assessment, rehabilitation, adaptive reuse, and conservation, it ex-plores how engineers, architects, technicians, and practitioners can be trained to re-sponsibly understand and intervene on existing reinforced concrete structures.

Contributions may address university courses, VET training modules, interdisciplinary and innovative teaching approaches, pilot projects, field surveys, advanced diagnostic investigations, structural assessment exercises, conservation-oriented design, digital documentation, and practice-based learning. The session welcomes experiences that bridge academic education, vocational training, professional practice, heritage institu-tions, and collaboration with building owners, public authorities, and other relevant stakeholders.

Its goal is to discuss how RC heritage can become a shared pedagogical and profes-sional platform, linking historical knowledge, structural safety, sustainability, and inter-vention ethics. The expected outcome is a clearer agenda for competence develop-ment in the sustainable transformation of existing concrete structures.

Digital twins are evolving from data visualization platforms towards intelligent decision-support systems for bridge lifecycle management. A key challenge is the integration of reliable prognostic capabilities that combine physics-based deterioration models, structural mechanics, monitoring data and artificial intelligence to predict future structural performance. This special session presents recent advances in predictive digital twins for bridges, with a particular focus on integrating hybrid prognostic models into operational asset management workflows. Topics include

• deterioration and durability modeling,
• structural safety assessment,
• structural health monitoring,
• machine learning for anomaly detection and prognosis,
• uncertainty quantification,
• sensor data fusion,
• hybrid modelling approaches,
• validation using real bridge data,
• and practical implementation within digital twin environments.

The session aims to connect researchers and practitioners working on the next generation of digital twins that enable condition forecasting, risk-informed maintenance planning and data-driven decision making for resilient and sustainable bridge infrastructure.

Distributed fiber optic sensing (DFOS) has become a powerful technology for structural health monitoring (SHM), enabling continuous, high-resolution measurements of parameters such as strain, temperature, and vibration over extended lengths. Its ability to capture global and local structural behavior using a single optical fiber makes DFOS well-suited for monitoring complex structures. This special session brings together experts to present recent advances, experiences, and applications of DFOS in SHM. Topics will include deployments of DFOS in various structures, sensor integration and installation strategies, integration of DFOS data within existing SHM frameworks, and case studies demonstrating the benefits and challenges of DFOS in real-world applications. The session will also explore novel signal processing and data analysis techniques to extract information about structural behavior and support decision-making. The overall objective of this session is to present the latest advances in distributed fiber optic sensing and discuss their potential for structural health monitoring.

This special session provides a comprehensive overview of the development of high-speed railway bridges (HSRB), reflecting the work of the fib Working Party 1.1.1 “Bridges for High-Speed Trains”. The presentations will cover the full lifecycle of HSRB:

• review past projects, focusing on structural health monitoring (SHM) of existing bridges and key lessons learned,
• ongoing projects,
• current design guidelines,
• and concepts and emerging research directions.

The aim is to present an integrated narrative that connects historical experience with current practice and future innovation. By bringing together insights from design, monitoring, and research, the session offers valuable perspectives for advancing the performance, safety, and sustainability of HSRB systems in the context of increasing demands on high-speed rail infrastructure.

Vision-based sensing has emerged as an efficient and economical approach for monitoring the condition and performance of concrete structures. Recent developments in computer vision, artificial intelligence, and imaging technologies have significantly improved the capability to detect structural damage, quantify deformation, identify dynamic characteristics, and support condition assessment. This special session focuses on the latest developments in vision-based damage detection and structural health monitoring of concrete infrastructure. Topics include crack/corrosion detection and characterization, displacement and strain measurement, vibration-based monitoring, digital image correlation (DIC), UAV-assisted inspection, deep learning, multi-modal sensing, and long-term structural monitoring. Contributions covering laboratory validation, field implementation, benchmark datasets, and practical engineering applications are especially encouraged. The session aims to provide an international forum for exchanging innovative methodologies and promoting the integration of computer vision into next-generation monitoring and maintenance of concrete infrastructure.

Concrete is a heterogeneous material whose mechanical behaviour is strongly influenced by its internal structure, damage evolution and crack development. Advanced imaging techniques provide valuable tools to observe and quantify these processes with high spatial and temporal resolution as well as high accuracy. This session focuses on experimental methods such as digital image correlation (DIC), computed tomography (CT), high-speed camera systems, ultrasonic tomography and three-dimensional image or volume correlation in the context of concrete material testing. These techniques enable the characterization of deformation fields, crack initiation and propagation, internal defects, pore structures and damage mechanisms under static, cyclic and dynamic loading conditions. By combining mechanical testing with modern imaging methods, a deeper understanding of the relationship between material composition, microstructure and structural performance can be achieved. The session aims to bring together recent developments, methodological challenges and practical applications of advanced imaging techniques for the characterization of concrete materials.

Large language models (LLMs) are rapidly creating opportunities across the full structural engineering lifecycle, from design, analysis and code navigation to assessment, inspection, monitoring, rehabilitation and infrastructure management. This special session invites contributions that investigate how LLMs and AI-assisted workflows can support engineering tasks while preserving technical correctness, validation, traceability and human responsibility. Topics may include LLM-supported design and verification, access to standards and fib knowledge, retrieval-augmented generation, decision support, BIM and digital twin integration, inspection and monitoring workflows, maintenance planning, benchmarking, reproducibility, failure modes, and safety-critical limitations. The session welcomes academic, industrial and public-sector contributions, with particular interest in case studies and critical analyses that clarify both the benefits and the risks of using LLMs in structural engineering practice.