Flagship Project BAU DNS

As part of the BAU-DNS flagship project, an automated building data capture system is being developed, based on a handheld sensor prototype that combines a laser scanner and a camera system, specifically designed for efficient and accurate building surveys.

A key focus is the semantic enrichment of captured data to enable detailed building descriptions. An integrated real-time AI engine analyzes and semantically annotates the data, structuring it for easier interpretation and further processing.

User interaction is also integral: real-time feedback and manual data input capabilities allow for interactive and complete building documentation. Continuous completeness checks during the survey process ensure that no relevant details are omitted.

Captured data is made available for Building Information Modeling (BIM) to support planning processes. This structured information feeds into a BIM model, enabling architects, engineers, and planners to base decisions on accurate, comprehensive data.

The development of a production-oriented, modular digital twin that supports resource-efficient and cost-effective planning, implementation, fabrication, and operation in building renovation forms the core of the information model. Based on requirements analyses, the basic information model and methods for the production-based, modular digital twin are being defined. In addition, concepts for model federation and semantic integration across various phases of circular building renovation are being developed.

A project-specific data space provides structured building data for further use, including an AI-based system that automatically generates BIM models from survey data. A web visualization tool is also being implemented to provide customizable views of buildings and their continuously enriched information and process data.

In parallel, image- and model-based AI processes are being developed to achieve greater process efficiency through comprehensive digitalization and automation. In close cooperation with those responsible for planning, production, and operation, suitable AI frameworks are selected, training data is prepared, and the AI networks are configured and validated.

A cloud architecture is being developed for the technical implementation of the digital twin infrastructure. This supports both fast, event-driven data processing directly at the measuring devices and more precise, computationally intensive batch processing in the cloud. This allows different requirements to be covered depending on data usage.

AI methods are also integrated into the infrastructure as service-based smart tools and apps and controlled via a workflow management system.

The objective is to develop a comprehensive modular kit for industrially manufactured façade modules, demonstrated via a prototype (mock-up). A key component is the creation of a material and component catalog providing a structured overview of available options for materials and building components. This register enables the targeted selection and combination of elements for the building envelope design.

A fully digital planning workflow is being developed to cover the entire construction process from design to execution, enhancing efficiency and minimizing errors. Integrated planning promotes seamless collaboration among all stakeholders involved in the construction process.

This subproject serves as a hub that integrates outcomes from other subprojects into a cohesive System Construction Kit. It ensures that new research results are continuously integrated and synergies between the subprojects are optimally exploited.

A key research focus is on the research and development of prefabrication-based renovation methods. Various fast and efficient installation techniques for modular façade systems are being investigated. Standardization and industrialization aim to save time and costs while improving construction quality. The project merges innovation, digital workflows, and industrial standards to promote sustainable, efficient building renovation.

On the one hand, automation of prefabricated module production is being investigated. The materials used, the required dimensions, and the range of variants play a decisive role, as they enable different automation options. A very high degree of automation can already be achieved in the area of hybrid element construction, for example. A detailed analysis has already been carried out for the first façade module developed in the project.

The second focus is on assisted or automated installation of modules on-site. The goal is to reduce the costs of installing prefabricated elements, in particular by reducing the number of employees required and the costs for setting up and renting scaffolding. All planning and installation data is compiled into so-called LOIN tables (Level of Information Need) as part of the information model.

Requirements for prefabrication and installation were captured using the Axiomatic Design Method and structured in morphological boxes that represent different solution spectra. This ensures the connection to the development of the System Construction Kit.

The workstream is also closely linked to integrated construction processes. Reference processes, such as those for external thermal insulation composite systems (ETICS), were analyzed, and key performance indicators (KPIs) were developed to evaluate and benchmark existing and new façade module systems.

This subproject focuses on improving the sustainability of materials and systems compared to the status quo by focusing on increasing circularity and carbon neutrality. A comprehensive material review evaluated both conventional and novel materials, such as aerogels and mycelium-based products, for their suitability in modular prefabrication. Functionality, geometric adaptability, and application flexibility of the materials were key evaluation criteria. A market-available insulation material not yet used in modular systems was identified, enabling investigation of system openness to new materials and functionalities.

Based on the research, a commercially available insulation material was identified that has not yet been used in modules of this type. This enables the system's openness to new materials and their functionalities to be investigated.

The findings significantly contribute to the overall innovation and sustainability impact of the BAU-DNS project