In recent years, the concept of digital twins has gained significant attention as a disruptive technology with vast potential across various industries. This revolutionary concept bridges the physical and digital worlds, transforming industries and unlocking new possibilities.
At its core, a digital twin is a virtual representation of a physical asset, process, or system. It encompasses a rich amalgamation of real-time data, sophisticated modelling techniques, and simulation capabilities.
Integrating the physical and digital realms, digital twins provide valuable insights, facilitate predictive analysis, and enable informed decision-making. Their potential spans various sectors, from manufacturing and healthcare to urban planning and infrastructure management.
At the GWF 2023 in May, I moderated two panels on Digital Twins, which elaborated on their role in infrastructure industry and digital city planning.
Though the discussions were quite insightful and pertinent, I noticed some confusion and discrepancy in terms of terminology used, particularly regarding digital twin, a digital model, and a digital shadow. These are all related concepts but with different meanings and applications.
For more clarity, I offer an explanation of each term –
Digital Twin
As mentioned above, it is a virtual representation of a physical object, system, or process. As the name suggests, it is a real-time, digital counterpart that simulates the behaviour, characteristics, and interactions of its physical counterpart. Digital twins are typically created using data from sensors, Internet of Things (IoT) devices, and other sources to provide a realistic and dynamic model of the physical object or system.
They are used for monitoring, analysis, optimization, and predictive purposes. Digital twins enable organizations to gain insights, perform simulations, and make informed decisions about the physical entity they represent.
Digital Model
A digital model refers to a computerized representation of an object, system, or concept. It can be a simplified or detailed representation that captures the essential features and properties of the subject being modelled. Digital models can take various forms, such as 3D models, mathematical models, simulations, or graphical representations. They are often used in design, engineering, architecture, and other industries to visualize and analyse concepts, structures, or processes.
Unlike digital twins, digital models may not have a direct connection or real-time synchronization with the physical entity they represent.
Digital Shadow
A digital shadow refers to the digital footprint or data trail that is left behind by individuals, organizations, or systems as a result of their online activities. It encompasses the data generated through various online interactions, such as browsing history, social media posts, online purchases, and other digital transactions.
Digital shadows are used for profiling, tracking, and analysing behaviour patterns, preferences, and trends. They are often employed in areas like marketing, cybersecurity, and personalized services to understand and predict user behaviour, provide targeted recommendations, or detect anomalies.
In the nutshell, while there is some overlap in these concepts, the key differences lie in their purposes and applications.
Different Connotations
My maiden takeaway at GWF was that a common definition of a digital twin doesn’t exist, or better put, there are multiple definitions.
Carsten Ronsdorf, Strategic Proposition Manager of Ordnance Survey, asked to vote about two unusual definitions: “Digital Twins are primarily about 3D technology/data as well as immersive experiences” and “Digital Twins are primarily about providing the right data to solve a problem.” However, the audience was divided.
Eric DesRoche, Director, Infrastructure Business Strategy, AEC Design Solutions of Autodesk, recognized that digital twins are disrupting the status quo in the infrastructure industry, and they should be used for design, planning, building, operating, and monitoring in a circular way. It is noteworthy that on the Autodesk website there isn’t a specific definition of what a digital twin is, but five types of digital twins are described:
Level 1: Descriptive twin is a visual replica with live, editable design and construction data, including 3D models and BIM
Level 2: Informative twin uses increased integration with sensors and operations data for insights at any given time
Level 3: Predictive twin captures real-time data, contextual data, and analytics to identify potential issues
Level 4: Comprehensive twin leverages advanced modelling and simulation for potential future scenarios, as well as prescriptive analytics and recommendations
Level 5: Autonomous twin has the ability to learn and make decisions through artificial intelligence, while using advanced algorithms for simulation and 3D visualization).
Bespoke Definitions
Most geospatial companies provide their own definition clearly influenced by their “technological” perspective. So, it is not surprising to see more emphasis on data collection, 3D and augmented reality, data analytics, and other aspects when talking with different stakeholders.
Efforts towards standardization are underway. For example, on December 3, 2020, the Digital Twin Consortium released the following definition:
“A digital twin is a virtual representation of real-world entities and processes, synchronized at a specified frequency and fidelity.
- Digital twin systems transform business by accelerating holistic understanding, optimal decision-making, and effective action.
- Digital twins use real-time and historical data to represent the past and present and simulate predicted futures.
- Digital twins are motivated by outcomes, tailored to use cases, powered by integration, built on data, guided by domain knowledge, and implemented in IT/OT systems.”
When looking at specific sectors, standardization efforts become clearer. For example, in the urban sector, Claudius Lieven, Head of Stadtwerkstaat, Hamburg Department of Urban Development and Housing presented the initiative “DIN SPEC 91607 Digital Twin for cities and communities” to create a (national, i.e., German) standard for transferring the concept of digital twins to urban space, including the presentation and description of application scenarios, data access and visualization methods, and the use of available standards.
What the Future Holds?
Can we expect a common “standard” definition for Digital Twin in the future? I don’t think so. Like BIM or Artificial Intelligence, multiple definitions will continue to co-exist in the near future.
For example, Geert De Coensel, Co-founder and CEO of Merkator, in presenting their work on Digital Twins to achieve operational excellence for telecom and utility networks, clearly highlights that they don’t need 3D for this specific business but for them it is essential to have accurate, complete, and up-to-date data.
The same concept has been expressed by Serge Lupas, CEO of Cyclomedia. He said, and “The core of a digital twin is a model. The model should be accurate and use accurate data.” This raises the issue of properly addressing the data management process and the related data governance. A strategy to ensure real-time access to all relevant data and adequate standards for data sharing has been advocated by most of the people I met.
These statements make clearer why some efforts focus on extensive data collection to support the creation of a Digital Twin. From a data provider’s perspective, a Digital Twin represents an opportunity to give more value to the data they collect. This explains initiatives to create the so called “National” Digital Twins starting from a data perspective. It seems the natural evolution of the shift from Spatial Data Infrastructures to Geospatial Knowledge Infrastructures and to Digital Data Ecosystems.
While I see value in this approach, particularly for areas where multiple applications can benefit from common accurate data (e.g., Digital Cities), I have some doubts about a data-driven strategy not accompanied by a correct business analysis.
Anthony Ruffour, Director of GIS and Digital Twins at BuroHappold Engineering has clearly insisted on a systems thinking approach.
Before starting developing a Digital Twin the following questions should be answered: How can the Digital Twin create value? How does it fit in the current workflow? Which technology is needed for its implementation? Which data should be collected?
I personally believe that the systems thinking approach is more in line with the original definition of a Digital Twin, first applied by Dr Michael Grieves, a professor at the University of Michigan, in 2002. Dr Grieves popularized the term while working on a project with the US Air Force to improve the manufacturing and maintenance of aircraft. More in general a Digital Twin should help in planning, operation, and monitoring for a specific purpose.
Authenticity
There are two other aspects that have been highlighted as particularly important: trust and the change of workflow and behaviour.
How can we trust a Digital Twin when it recommends specific actions? Like in the case of Artificial Intelligence, to trust a machine, we need to ensure that the design and operation of the Digital Twin have followed quality standards. Accurate (and unbiased) data are needed. Real-time data access should be granted. More generally, a systems engineering approach is required, including quality control and assessment of the entire system in all phases.
Some of the people I met consider that the technology is ready to be fully deployed, but the main obstacles are in the change of existing “operational” workflows (moving from the known to the unknown) and as well the change of human behaviour (moving from human decisions to decisions taken by a Digital Twin).
In my opinion, Digital Twins will be accepted when their value will be clearly demonstrated. Like in the case of in-car navigation systems, people who were initially reluctant to accept them changed their opinion when the real benefits were proven.
Use-Cases and Challenges
The success stories emerging from the digital twin initiatives presented at the GWF are nothing short of inspiring. By integrating real-time data from various sources, such as satellite imagery, sensors, and ground observations, users gained valuable insights into geospatial patterns, analysed trends, and supported informed decision-making for various applications.
The result? Improved urban planning, optimized resource management, and enhanced disaster response capabilities.
While digital twins offer immense potential, challenges remain on the path to their widespread adoption. Some issues pose significant hurdles:
- Data Integration: Gathering and integrating data from various sources can be complex and challenging. Different data formats, compatibility issues, and data security concerns may hinder the seamless integration of diverse datasets.
- Scalability: As the complexity and scale of digital twin systems increase, managing and processing massive volumes of data in real-time becomes more demanding. Ensuring the scalability and efficiency of the underlying infrastructure is crucial.
- Interoperability: Interoperability between different digital twin platforms and systems is crucial for seamless collaboration and data exchange. Standards and protocols need to be established to ensure compatibility and integration between various technologies.
- Security and Privacy: Digital twins generate and handle sensitive data, making security and privacy concerns paramount. Robust security measures, data encryption, and compliance with privacy regulations are necessary to protect both the digital twin and its physical counterpart.
Looking ahead, the future of digital twins in the geospatial industry, as supported by GWF, is promising. As technology advances, the integration of artificial intelligence, machine learning, and geospatial analytics will further enhance the capabilities of digital twins. This will enable real-time predictive analysis, remote monitoring, and data-driven decision-making for geospatial applications.
Embarking on this journey at the Geospatial World Forum has revealed the tremendous potential of digital twins to revolutionize industries and reshape our world.
By embracing digital twin concepts, GWF is paving the way for a more efficient, sustainable, and interconnected future. Together, we can harness the power of digital twins to drive positive change in the geospatial industry and beyond.
