The deconstruction of a building marks the end of its life cycle. At the same time, this phase offers many opportunities for the acquisition and reuse of used building materials. During the deconstruction phase, planning and execution are carried out using a variety of different software applications. Despite the use of digital 3D models according to the Building Information Modeling (BIM) method, there is no interoperability along the value chain in the deconstruction of a building. To counteract the lack of interoperability in the deconstruction phase of buildings, the article develops an ontology for the formalization of knowledge within the deconstruction phase and the usability of heterogeneous data sets in connection with the 3D BIM model. After assessing the status quo on the topic of BIM in deconstruction and ontologies in building construction, the systematic development of the deconstruction ontology (DCO) and its validation on a demonstrator is conducted. The result is the presentation of the basic usability of ontologies in conjunction with graphical databases and digital 3D models in the deconstruction phase. This ensures interoperability along the value chain and counteracts data loss.

In recent decades, the condition of many bridge structures has deteriorated and the need for maintenance measures has increased. Until now, these maintenance measures have mainly been carried out manually and reactively. The use of digital 3D models based on Building Information Modelling (BIM) can remedy this situation and create the basis for predictive maintenance management. While the generation of 3D models of new bridge structures is simple, the digitization of existing structures can be a complex process. This article provides an overview of the state of the scientific practice with regard to the procedures, technologies and data used to generate 3D models of existing bridge structures using the BIM method. Using a systematic literature analysis, scientific databases are searched for suitable literature and analysed with predefined filtering parameters. The results provide a uniform understanding of the current status quo of the digitisation of existing bridge structures and show existing degrees of digitisation and automation.

For a detailed analysis of damage to bridge structures, it is necessary to apply diagnostic methods. The results of these procedures are often complex results represented in heterogeneous and incompatible data formats. These are sometimes difficult to integrate with 3D Building Information Modelling (BIM) models. This article examines two system- and data-level integration methods for linking heterogeneous building diagnostic data with 3D digital bridge models in Industry Foundation Classes (IFC) format. At the beginning of the article, an ontology for non-destructive diagnostic methods (SODIA ontology) is developed and linked to other bridge ontologies. The first approach consists of transforming the diagnostic data and the data of the IFC model into the data schema of the Resource Description Framework (RDF), linking them together and integrating them into a graphical database (data integration). The second approach investigates system integration using information containers according to the ISO linking of ISO 21597. By validating the two approaches, it can be demonstrated that both methods provide a way to link heterogeneous diagnostic data to a 3D bridge model and achieve interoperability in maintenance management. The results can improve the use of data by bridge asset managers in the operation and maintenance management of bridges. At the same time, they serve as a basis for possible data analyses that lead to predictions about damage progression and changes in condition.

In light of the climate protection goals of the Paris Agreement, optimizing the sustainability of planning processes is becoming increasingly important. Building Information Modeling (BIM) centralizes planning information for interdisciplinary evaluation, enabling sustainable decision-making. This paper presents a systematic review of BIM-based decision support approaches for building sustainability. Following vom Brocke’s five-phase model and the PRISMA 2020 standard, 70 studies were analyzed to identify current methods, their respective strengths and limitations, and future research needs. The findings reveal a highly dynamic but fragmented field of research. Assessment-Based Optimization and multi-criteria decision-making (MCDM) methods dominate. However, the holistic integration of ecological, economic and social indicators remains rare, with social sustainability receiving the least attention. Most approaches rely on proprietary BIM environments, while open BIM applications and interoperable data standards remain underdeveloped. Standardized data sources, such as Environmental Product Declarations (EPDs), are well established for ecological assessments, but are largely lacking for the economic and social dimensions. The review highlights the urgent need for interoperable data formats, standardized evaluation methods, and accessible databases to enable scalable and comparable BIM-based sustainability optimizations. Advancing these foundations will be essential for achieving consistent, holistic sustainability optimization in the construction industry.

Bridge structures are considered complex and significant. Accordingly, the knowledge of the engineering domain of bridge construction and related specialist areas is multidimensional and highly specific. Sometimes this knowledge is explicitly documented in standards, technical regulations, or information sheets. At other times, it resides implicitly in the expertise of the specialists involved. Ontologies are used to structure and formalize such domain knowledge, but creating them is resource-intensive and requires specialized expertise. Large language models (LLMs) offer one way to automate ontology creation through their natural language processing capabilities. This article examines LLMs’ ability to generate ontologies in the specialized field of structural non-destructive testing (NDT) in bridge construction. Four different LLM-based approaches are employed. The results are compared with a previously created human-generated ontology and subsequently evaluated by external experts. Experts rate the human-developed SODIA ontology highest, with an average score of 3.44 out of 5 points. Only the ChatGPT 4.0-created ontology performed similarly well, with a score of 3.3 out of 5.00. All other LLM-based ontologies with ratings below 3.0 are of minor quality. These results underscore the potential and constraints of using LLMs to structure and formalize engineering domain knowledge into ontologies.