Interview with Iryna Osadcha

Iryna Osadcha, a PhD student and Junior Researcher for the SmartWins project, conducted research on updating Digital Twins in construction for her article titled “Geometric parameter updating in digital twin of built assets: A systematic literature review.” In an interview conducted by Paulius Spūdys, Iryna Osadcha explains the concept of Digital Twins and their importance in accurately reflecting building data and enabling continuous monitoring.

Paulius Spūdys: You wrote an article about how Digital Twins are being updated in the construction industry. Could you start by explaining what a Digital Twin is and how it differs from a Building Information Model (BIM)?

Iryna Osadcha: BIM and Digital Twin are both digital technologies to help construction professionals manage building data, yet serve different purposes in the built environment. Designed for collaboration between architects and engineers, BIM helps visualise the building and plan construction activities, from laying the foundation to installing electrical wiring and HVAC systems. But while 3D BIM models are a powerful tool for visualising the design and construction of an asset, Digital Twins take things to the next level by enabling virtual interaction with this asset and representing how people interact with the built environment during its lifetime.
Digital Twins enable tracking and managing physical objects through a digital model, including real-time monitoring and simulation-based forecasting to predict and plan behaviour and future states.
Digital Twins are the creation of a digital representation of a physical object that accurately captures its characteristics and adapts to changes in its environment for optimal performance. It is not a new concept. It is already widely used in industries such as aerospace, industrial manufacturing, robotics, automotive, wind energy, telecommunications, healthcare and others, which use object prototyping to improve outcomes.

PS: Why is it so important to update Digital Twins during the life of the building?

IO: This may seem obvious at first. But updating Digital Twins is such a complex task because buildings are very sophisticated structures. As a result, only those elements that serve the purpose assigned to the tool will be updated. There is also the involvement of multiple stakeholders and unpredictable factors.
Updating the Digital Twins can accurately reflect the current state of the building, including any deviations from the original design. For example, if a building experiences structural change or damage, an updated Digital Twin can help identify the extent of the problem and support informed repair and maintenance strategies.
Additionally, updating the Digital Twin will enable continuous monitoring of building performance and health. Updating the Digital Twin ensures everyone has access to the latest and most accurate data, fostering collaboration and streamlining workflows. This ensures that the Digital Twin remains a reliable and valuable tool throughout the lifecycle of the building.

PS: What are the main findings of your research?

IO: In my article, I conducted a systematic literature review of 56 articles to shed light on the updating of Digital Twins geometry for built assets in the construction industry. I observed that while there is a growing interest in this topic in recent publications, there remains a lack of comprehensive coverage. Most existing studies tend to focus on specific aspects or processes related to virtual modelling or data processing.
One of the unique aspects of my research lies in its specific focus on the geometry of Digital Twins within the construction industry, distinguishing it from other industries where DTs are also applied. This allowed to establish a foundation for a unified maintenance method for virtual model geometry that can be widely utilised in construction. By addressing the current state of the industry, I aimed to identify crucial problem areas and develop a practical and scalable method for updating Digital Twins geometry in the future.
The complexity of building geometry poses a challenge when accurately representing and integrating it into a Digital Twin. The process of updating the digital twin geometry involves various stages, including data collection, data processing, and modelling. Each stage has its own methods, challenges, and considerations that need to be addressed.

PS: Considering the complexities involved, what criteria did your research reveal for updating building geometry data in Digital Twins?

IO: I highlighted the significance of establishing clear criteria for updating building geometry data. This ensures that the digital twin accurately represents the building and that any updates are performed in a timely and efficient manner. The decision to update the geometry data can be based on factors such as the building life cycle stage or the specific requirements of the building type.
I also showed that various technologies and methods are used to update the geometry of the digital twin, including laser scanning, UAV/photogrammetry, and total stations. However, these methods are not universally applicable or scalable to all types of buildings and structures. Further research is needed to develop a general solution applicable to a wider range of structures.

PS: What are the main steps involved in updating Digital Twins geometry and the challenges at each stage?

IO: Updating Digital Twin geometry involves three main steps: Data Collection, Data Processing, and Modelling. In Data Collection, various methods like UAV/photogrammetry or laser scanning can be used. Challenges arise in scalability or applicability to different building types. Data Processing includes registering data into the Digital Twin coordinate system and filtering out noise. Challenges exist in object recognition and modelling, choosing appropriate methods, and merging existing BIM models. Modelling requires enriching geometric shapes with semantic information and establishing relationships between structural elements. Challenges involve interoperability, merging models, and ensuring efficient data handling and analysis.

PS: Based on your findings, what key areas should future research focus on?

IO: Future research should focus on specific case studies of well-defined building types and life cycle stages that can be generalised. Improving object detection, segmentation, and classification in raw data is crucial for accurate Digital Twin representation. Developing interoperable data and unified software solutions to facilitate efficient DT updating is essential. Addressing these areas will advance the field and overcome current challenges in updating DT geometry in construction.
Democratising the use of Digital Twins in the construction industry is critical to decarbonisation efforts. By enabling collaboration, promoting sustainable practices and optimising energy performance, digital twins allow stakeholders to make informed decisions that reduce carbon emissions and create a more sustainable built environment.