Digital twins are not the big magic bag that holds the solution to all the problems of our future. Jantien Stoter, Professor of 3D Geoinformation at the Delft University of Technology, pleads for a realistic approach to the concept. Because properly understood, the 'digital multiples' offer plenty of potential.
It was not love at first sight that overcame Jantien Stoter when she first came into contact with the concept of the digital twin about ten years ago, since the meaning of the concept is fuzzy, the concept promises sky-high ambitions and at the same time the newly introduced concept covers challenges that have been studied in the geo-information domain for decades. But today, she says, she definitely sees the value. Because properly understood, there is great potential in digital twins.
For the professor of 3D geoinformation leading the Section Urban Data Science at the Dutch Technical University of Delft, the concept was too vague, too ambitious, too pretentious. While the waves around the new star in the geoinformation world were running high, Jantien Stoter kept her feet on the ground.
„We are facing huge challenges with regard to the climate or energy supply, and people want to tell me that smart cities and the digital twin will solve all our problems. But it's not that simple," says the geoscientist.
Why the scepticism? Stoter's most important point is that models are always abstractions at a certain scale and at a certain timestamp, so simulations based on models should not be confused with prophecies about our future. Digital twins as a big black box into which all kinds of data flow and that outputs out clear, uniform statements that can be used by everyone - including the general public - simply do not exist. Therefore, the term digital twin is also a little misleading.
The large digital model as an instant image of every moment of reality will probably not exist. Instead, Stoter advocates the idea of "digital multiples". For each application and task, there would be adapted "twins", which in the best case would be brought together, synchronised and aligned.
Another criticism of the concept, in Stoter's eyes, is that it obscures what has already been achieved in terms of 3D modelling. "We have been working with digital representations of our environment for a long time and also integrate dynamic information, for example through sensor data," says the professor. The example of GIS and BIM may illustrate what Stoter means. After long intensive discussions and studies on the subject, she says, it is now clear that GIS and BIM should keep their different view of reality. They are two different models. If you want to use data from GIS in BIM or vice versa, you have to be aware of the differences in order to bring them meaningfully together. If the time factor and "what-if" scenarios are added, it must be clear that these are abstractions that produce different results depending on the number and type of underlying variables, the accuracy of the input data and also the mathematical models used. Simulations of noise, wind or environmental pollution, for example, are so complex that they need special hardware to produce useful results and domain experts to interpret the results.
Despite all the criticism, Professor Stoter fully embraces the concept today. It makes communication easier for her, because while she used to have difficulties explaining the importance of 3D city models, geodata and standardisation, she now runs into open doors. "We as a geo-information community should welcome the concept, but not disregard the remaining challenges," Stoter says. Among these, she counts the issue of standardisation of data and processes and the ability to go from small pilots to large scale implementations. For this to happen, she says, the digital twin needs to move beyond the purely technical level. The digital twin must be understood by organisations as the basis on which processes run. Therefore, agreements must be made on how data is exchanged – also in public-private collaborations -, who is responsible for it, who tracks it. "Data is like other assets, for example like a road, about which you know where it comes from, where it is supposed to go, what it is made of and who pays for the construction and maintenance," says Stoter.
As an example of a digital twin application, Jantien Stoter mentions the joint project with the National Institute for Public Health and the Environment in the Netherlands, which is responsible for noise assessment methods. At the 3D Geoinformation research group, the large-scale, two-dimensional topographic map of the Netherlands was merged with the third dimension from LIDAR surveys to create a 3D noise model. It is now available as an open data set provided by the Dutch Kadaster that can directly be used by noise simulations software that implements the official noise assessment method of the Health Service.
Another example is the digital building permit process. Today's building permit issuance is mainly a manual, document-based process which suffers from low accuracy, low transparency and low efficiency. This leads to delays and errors in planning, design and construction. The 3D Geoinformation research group has run several projects on the digitalisation of building permit checks by the integration between BIM and Geo. Moreover, in a recently awarded Horizon Europe project, 19 partners across Europe will develop an innovative kit of both methodological and technical tools to digitise building permitting and automated compliance checks. The consortium consists of a multidisciplinary team covering GIS, BIM, municipal processes and planning, data integration and standardisation (OGC and buildingSMART). The authorities will thus be able to provide information and issue building permits faster and in a more transparent way. "Digital Permit Checks offers less scope for misunderstandings, but as with all other digital twins, it needs experts and communication between all stakeholders to balance the outcomes and make the optimal decisions.“
Even if it wasn't love at first sight, today Jantien Stoter knows the value of digital twins. Properly understood, that is, with less ambition and an awareness that the results can serve everyone, they are extremely useful. Fortunately, the professor has welcomed the digital twins and fallen in love with the "digital multiples".
Text: Monika Rech-Heider on behalf of INTERGEO® 2022, Editors for INTERGEO Newsroom
Copyright: INTERGEO®
Further Informationen can be found at the "3D geoinformation research group" of TU Delft: 3D.BK.tudelft.nl