We meet Sabine Sint in the meeting room of the Building Physics Research Department, which should really be called the Planning Room. The walls are covered with flipcharts detailing the projects for the coming years, because Sabine Sint works here as a project manager. She originally comes from the field of business informatics, where she is currently working on her PhD. She came to building physics for a programming job and grew into project management. We talk about maintaining the diverse data landscape in her department, about building simulations and digital twins, and how important documentation is for secure, sustainable data sharing.

Building automation and its simulation

“Building automation is essential during the planning stage, and there must be a way to specify these control algorithms in the planning stage, independently of the platform. This means that, regardless of who the executing company is, I can try it out once in a simulation. And that's where we have automated the translation of this platform-independent code into specific simulation code and specific execution code.”

Sabine Sint tells us about one of her most recent projects, in which she worked with measurement data from the Research Unit of Automation Systems test laboratory, among other things. Here, a ventilation system was operated in summer and winter modes. The aim was to check whether the automatically generated code behaved in the same way as in the simulation. Depending on the temperature deviation, the system should heat or cool. In building automation, concepts such as “demand-based ventilation” and other control mechanisms are often outlined. However, the specific specification and programming of these controls often remain unclear and are only specified in later phases of implementation, which leads to errors or untapped potential.

Sabine's research group is therefore striving to enable a continuous, automated pipeline from planning and simulation to execution. The central focus of this research is to demonstrate a continuous process chain and, for example, generate the same code for control algorithms in such a way that it can be used directly in execution. This is intended to bridge the gap between planning scenarios and real-world implementation.

From digital model to digital twin

“First, there is a digital model, where I analyse the current or planned system and create a model for it. It is important to know that models always simplify and reduce, because I have a specific purpose in mind for building this model.”

Today, digital models are the starting point for almost all new construction projects, in the form of platforms where information flows together for the involved architects, structural engineers, spatial planners, etc.. Since no building or infrastructure have been constructed yet, there is no real data, so the knowledge is initially created virtually. However, as Sint emphasises, every model is only a snapshot of reality – shaped by the questions asked of the system. The next step in development leads to the so-called digital shadow, in which real measurement data flows back into the digital model to improve it and perform target/actual comparisons. Only when the automated flow works bidirectionally between the digital and real context, we speak of a digital twin. In practice, this development is already automated in certain areas, but it usually remains semi-automatic as humans are involved in checking whether the system's statements are plausible and transferable.

Data sharing with reservations

"But then there is the question of data sovereignty and data protection. That is, what information do I pass on? We are doing more and more digitally; we want to test everything in advance, check everything in advance. But that means data management itself is becoming increasingly important, requiring a much more interdisciplinary approach. Just because I am a good data manager does not necessarily mean that I have a good understanding of data from construction, architecture, physics, and mathematics."

According to Sabine Sint, the question of which and how data may be openly shared and what is better kept under lock and key is not only a question of ethics, but also of security. Detailed building data can often be traced back to individual households, for example, via energy consumption or mobility patterns. In certain cases, the publication of individual data units could, for example, allow personal data to be traced back to individuals. This makes disclosure risky in terms of data protection law. In addition, simulations can reveal potential vulnerabilities, which also makes disclosure critical from a security perspective. Therefore, only aggregated and processed data is suitable for public use. 

Sabine emphasises the need for interdisciplinary cooperation, as data management alone is not sufficient to understand and document subject-specific content appropriately. She sees increasing pressure from funding bodies such as the EU and FWF to submit data management plans, which has a positive effect on ensuring the reuse and further development of research data. At the same time, she recognises an internal need to address these aspects.

Documentation for the future

“It's a time factor, because usually when projects are completed, the next project comes along. But the old project should be cleaned up, and the data should be processed to make it easy for others to understand and use. Instead, however, things usually remain on our servers and then eventually fade into the background.“

Sabine Sint reflects on the future of data management, particularly in the context of machine-readable standards and automation through AI, and sees a growing importance for well-documented, structured and accessible data. “Programming is more fun than documentation,” she admits with a smile, but she sees it as a crucial contribution to the sustainability of research. Her team maintains a building data model that covers everything from planning to operation. But the existing documentation is mainly created when working in a team; outside the project, there is often a lack of perspectives that critically review the documentation and make it understandable. Her core goal: practical, clear research data documentation that can also be used by colleagues outside the project.

“That's why we do research, so that our findings can be applied outside our field of research. We want to contribute to climate neutrality and energy efficiency, especially in our key areas of focus.”

Contact

Sabine Sint
Real lab for Smart and Resilient Cities (in progress)
TU Wien
sabine.sint@tuwien.ac.at

Center for Research Data Management
TU Wien
research.data@tuwien.ac.at