While battle cries resound about AI, there is another revolutionary concept related to new technologies that we have not given as much attention to: the so-called “digital twins”.
Definition of digital twin
This issue of digital twins is quite new and, although the origin is often attributed to NASA, all indications are that the first reference we have to the use of the term dates back to 2002 when a renowned academic named Michael Grieves coined the term:
“A digital twin is a virtual representation of a physical entity or system that is updated with real-time data and uses simulations, machine learning and reasoning to aid in decision making.”
Michael Grieves hit the nail on the head about the definition, even though his field was product lifecycle management (PLM) systems. His work has been instrumental in connecting virtual simulation technology with manufacturing and industrial processes, laying the foundation for Industry 4.0.
But this field is not the only one to which the digital twin concept is adapted. As if it were an order of magnitude scale, the digital twin applies to different levels. Let’s take a broad look at the layers at which this concept operates, what technological tools are required in each case, and what its main uses are.
The digital twin of a historical object
Let’s start with the lowest scale, the human scale we could say. We can establish digital twins of objects from very small to a large sculpture, for example. As my field is Digital Humanities, let’s take the example of a Greek vase. Its digital twin would be an exact replica of the figure in three dimensions. For this we can use two techniques: photogrammetry and 3D scanning. Photogrammetry starts with a series of photographs of the object from multiple points of view and with algorithmic software is able to create a three-dimensional file with the geometry and texture of the object. If we talk about 3D laser scanning what we will obtain is a cloud of high precision points that give us back the figure, but without texture, which can be added later.
And what do we do with this 3D historical object? Well, we can upload it to a web repository (Sketchfab is the most famous, although Europeana is catching on) and share it with the world. It can be used in digital exhibitions or serve as a prop for a period film shot on a virtual set. But the most ambitious use is to run this virtual object through an artificial intelligence trained in its subject matter and it will be able to offer us the dating of the object, its provenance and, if I dare say it, even the meaning of the paintings on its surface. These are just two examples, but there are many more uses.
Digital Twin of a Historic Building
Let’s go one step further, to a historic building. The scale has already increased. Level somewhat higher than human but the digitizing procedure has things in common with the previous one. Again we can use photogrammetry and 3D scanner (technically called LiDAR). Photogrammetry requires this time the use of drones to fly over the building and capture its facades and roofs.
The second, complementary alternative, which can also be used with drones, is LiDAR. LiDAR is a device that allows to determine the distance from a laser emitter to an object or surface using a pulsed laser beam that determines the distance to the object and measuring the delay time between the emission of the pulse. With this it generates points that it collects and then expresses the set of them located three-dimensionally.
But here it is not enough to upload it as an object to 3D repositories (which is also possible), but HBIM (Historic Building Information Modeling) technology can be applied. This is a digital methodology that focuses on documentation, management and preservation applied to historic buildings through advanced digital tools. It combines the 3D model generated with geographic information systems with planimetry, thus creating an accurate and detailed representation of the architectural heritage. This approach allows the integration of historical, architectural and structural data, through sensors, for example, facilitating both the analysis and conservation of building structures that may be at risk or possess significant cultural value. Just look at the Notre Dame fire. Had there been no three-dimensional survey, the cathedral’s recovery would have been much more imperfect.
But this technology also allows various uses, each one more interesting than the last. On the one hand, the simulation of structural and functional processes, such as electrical systems, ventilation or evacuation, adapting the needs and current regulations to historical buildings. It also allows the historical record of restorations and modifications as well as the real-time visualization of human or energy traffic inside the building through sensors. It can also serve as a predictive tool for preventive maintenance.
A case to mention is the recent scanning of the Monastery of San Lorenzo de El Escorial, a feat performed by ACRE Surveying Solutions and ATC Proyecta who have topographically surveyed this treasure of humanity with more than four thousand points of data collection to generate a point cloud that was then converted to HBIM and from where you can extract all kinds of floor plans, elevations, sections … El Escorial naked.
But this is not the end of the uses of this twin. This project, funded by the Recovery, Transformation and Resilience Plan within the Spain Audiovisual Hub program, will make the digital assets generated freely available to national and international production companies, facilitating their integration into films, series or video games. There I leave it.
Digital Twin of a Factory
At this next level of the ladder, we do not so much consider reproducing existing factories as modeling the facilities a priori, which, through artificial intelligence, allows the simulation of the production chain, process optimization and cost minimization, all prior to their construction. An example of this was discussed in After when we talked about the company NVIDIA and the application of both its hardware (data processing cards) and its Omniverse software for the electric vehicle plant in Debrecen, Hungary, which is due to open this year. Thanks to the powerful tool that is Omniverse, the BMW team has been able to integrate data into high-precision, high-performance models, connect its various software tools, and enable real-time collaboration between multiple users in different locations. All with a clear goal: to refine and perfect every step of the process to achieve unprecedented optimization in their new factory.
Digital Twin of a City
Here we take a qualitative leap. Scanning an entire city is no mean feat. Here I will highlight an example that touches me closely: the digital twin of the city of Madrid that is being carried out by the City Council and can be consulted in its Geoportal. But how did they achieve it? Here we start from two sources: satellite photos and airplane flights over the city.
To obtain 3D models of a building from aerial photos, the oblique projections of the photographs play a key role by tilting the axis of the camera or capture system with respect to the horizontal plane. This approach allows capturing lateral details of the building facades in addition to the top views, which enriches the three-dimensional modeling. By combining images taken from different oblique angles with techniques such as photogrammetry or LiDAR, a more accurate and complete 3D reconstruction is generated, especially in complex structures where vertical surfaces and detailed textures are crucial to the final result. Done this way for the whole city, we have a complete 3D model.
But be careful, if scanning a building is Teras of information, what happens with entire cities? Well, in the case of Madrid we have the reduction of the shape of the buildings to simple geometries, which greatly reduces the volume of information at the cost of losing resolution. But there is a much more powerful case: the digital twin of Tokyo. Hardly any news has been released, but it seems that they are going to release the digital twin of the Japanese capital with an unprecedented resolution. We’ll have to see it.
And what can we do with all this? Well, this is where the so-called smart city platforms come in, which integrate multiple layers of data ranging from information from IoT (Internet of Things) sensors distributed throughout the city to urban and meteorological databases. These platforms not only enable the efficient management of digital models, but also enable the development of specific applications, such as early warning systems for natural disasters, tools to optimize urban mobility, or solutions for energy efficiency in buildings and entire neighborhoods. In addition, smart city platforms act as a centralized point for collaboration between administrations, businesses and citizens, ensuring that data-driven decisions are transparent, inclusive and sustainable. On the other hand, the digital twin becomes an essential tool for urban planning, by projecting growth or development scenarios that anticipate needs and optimize resources, ensuring a balanced and efficient evolution of the city. But this is just the tip of the iceberg. If we begin to combine the data generated with the volumetry of the city, the applications of the twin are almost infinite.
Digital Twin of an Ecosystem
On a larger scale we have ecosystem analysis as a paradigm for environmental monitoring. The digital twin of an ecosystem can be generated through advanced tools that combine IoT sensor technologies, 3D computational modeling, artificial intelligence platforms (again our beloved AI) and geospatial data analysis systems. These tools collect and process large volumes of information in real time, creating a detailed and dynamic digital replica of the ecosystem. This digital twin allows us to simulate the complex interactions between flora, fauna and their environment, modeling biological and environmental processes that are fundamental to understanding and managing natural ecosystems.
By incorporating real-time data such as weather, soil moisture and temperature variations, the state of the ecosystem can be constantly and accurately monitored to detect critical changes. In addition, the digital twin facilitates in-depth analysis of the impact of climate change, anticipating how variations in climate may alter habitats, biological cycles and biodiversity. The ability to integrate species conservation and biodiversity data is essential to design protection strategies that ensure the viability of ecosystems and prevent the extinction of endangered species.
On the other hand, the digital twin allows simulating future scenarios, assessing how environmental policies and human activities such as deforestation, intensive fishing or land use impact the ecosystem balance. These simulations not only help to anticipate risks, but also provide a scientific basis for making informed and sustainable decisions. In this way, the digital twin of an ecosystem becomes a key tool for balancing the conservation of natural spaces with human needs, promoting a future in which nature and society coexist harmoniously.
As an example of this higher scale and to close this tour, we will cite the case of La Palma Smart Island, an ambitious project that seeks to turn the Canary Island into a global model of sustainability, innovation and efficiency through the integration of smart technologies. It has been conceived covering key areas such as natural resource management, mobility, connectivity and tourism, with the aim of improving the quality of life of the inhabitants and offering a unique experience to visitors. Through advanced digital infrastructures and technological solutions, it promotes environmental conservation, energy transition and citizen participation, consolidating La Palma as a benchmark in the development of smart islands.
In conclusion…
With this review done, we can affirm that digital twins are much more than a virtual representation; they are a bridge between the physical and digital worlds, capable of transforming the way we interact with the environment. From historical objects to entire ecosystems, their versatility lies in their ability to integrate data, simulate scenarios and optimize processes with an unprecedented level of precision. Every scale, from a factory to a city, opens up a range of possibilities for improving efficiency, preserving heritage and ensuring sustainability.
In a world where global challenges are growing, digital twins have emerged as an indispensable tool for making data-driven decisions and creating innovative solutions that benefit both humanity and the planet. This virtual mirror of reality not only allows us to better understand our environment, but also to design a smarter, more sustainable and connected future.
