On 21 June, three speakers - from the academic world, the private sector and the local authorities - addressed the topic of Digital Twins. The participants in this last Smart Lunch of the 2021-2022 season received clear explanations of this dynamic technology, its use, its benefits, its applications and its future!
To familiarize us with the concept of digital twin, Mr Gilles-Antoine Nys, a doctoral student in geomatics at ULiège, announced that the concepts of digital twin, on the one hand, and Smart City, on the other, should be clearly distinguished. If a Smart City is a space where traditional models and services are optimised through the use of digital solutions, which have benefits for citizens and businesses, the smart city is more about governance and a goal in itself, while digital twins are the technical answer to this governance issue.
Also noting that several visions of the digital twin coexist, Mr Nys presented the definition of the multinational IBM, which states that a digital twin is a virtual representation of an object or system that spans its lifecycle, is updated from "real-time" (or close to real-time) data and uses simulation, machine learning and reasoning to help decision making. The twin will in fact approximate, represent as well as possible, a reality in the real world.
With regard to the "Urban (or City) Digital Twin", no clear definition or basic building blocks really exist yet at the national and international levels. The definition of the Geomatics Unit of ULiège, to which Mr Nys belongs, specifies that it is a system of systems; the first system (a geographic information system) is a base that integrates other kinds of systems.
As with any spatial data infrastructure, different levels co-exist and it is these different levels that characterise the digital twin. Each part of this infrastructure will be interoperable so that the basic system integrates as many systems as possible, these systems communicate with each other, and we end up with a decision-making process.
Different phases are at work. First, the import of all the subsystems into the system, then the modelling and, in a third phase, the definition of the applications, which are layers/systems that are added to the basic system. There is also a maintenance and storage phase. Lastly, once everything is integrated, it must be made available to all other users and/or citizens, through the phase of advertising the structured information.
Mr Nys finally indicated that the use of the term Digital Twin is currently rather inappropriate, as a "real" Digital Twin should be able to start from the digital city model/digital world and have an influence on the real world [actuator] (as opposed to the Digital Shadow [sensors, for example]). What we will end up with in a horizon of ten or fifteen years.
For once, the inspiration for the local authorities came from France. Christelle Gibon, Head of Innovative Projects at Rennes Métropole, explained the approach of the Breton métropole regarding digital twins. For Rennes, the digital twin makes it possible to describe, represent, analyse and model urban functions.
The desire was to approach the territory systemically, by adding data (mobility, air quality, demography, etc.) to a 3D model. This dataset has been combined with exploitation and cooperation tools to assist with decision making.
This notion of digital twin is in fact the result of 20 years of work during which a 3D model of the territory was used for general public purposes (territorial communication) and we then progressively moved on to forms of urban consultation. Ms Gibon said she realised that the 3D model was a trigger for debate that encouraged citizen engagement.
From this came urban walks, fun devices to introduce young people (RennesCraft) to urban planning, etc. The model also promoted the economic development and attractiveness of the territory (during visits by delegations, trade fairs, etc.) as well as decision making through the modelling of urban functions (noise modelling, simulation of the 100-year flood, etc.) and the simulation of the effects of public policies so they can be adjusted if necessary.
In parallel with this base - the 3D model - there was a context of opening up territorial data to Rennes Métropole. In 2010, open data was launched in Rennes Métropole. The 2016 arrival of the Law for a Digital Republic, which introduced the notion of data of general interest [In addition to public data, there is private data that has an environmental, social and economic interest and this law launched the idea of opening up this data of general interest] directed the thinking of Rennes Métropole on the organisation of this data of territorial interest based around a Metropolitan Public Data Service. The objective was to share this territorial data to enhance the various projects and the digital twin.
This Metropolitan Public Data Service was launched in 2016 with a "public data" component (Data RM) as well as a territorial ecosystem to be organised (DatAgora), legal governance and to unite stakeholders around common rules, and, lastly, the RUDI portal, which is the operational twin of this Metropolitan service with the development of a portal for sharing territorial data for users, data producers, citizens and project developers. The data processing chain (with the questions of anonymisation and consent management it raises) has thus been studied in depth.
Then comes the question of tools: how to produce services at the end of the processing chain to promote cooperation and inform public policies for ecological and democratic transition. How can we respond to these issues? According to Ms Gibon, the visibility of the territory's data is key. Through the digital twin, Rennes wanted to make the territory's data intelligible to all stakeholders, and especially to citizens. Thanks both to data science techniques for modelling urban systems and to collaborative functions, Rennes has been able to foster cooperation and provide information to inform decisions and public policies.
Rennes Métropole has implemented an incremental approach with the initial four-year experimentation of a collaborative platform (from Dassaut Systems): the 3DEXPERIENCity Virtual Rennes. Although initially intended for the world of industry, Rennes used the platform to help the city's urban planning professions. It has provided assistance in the management and cooperation of development projects (project management, dashboard) with the underlying idea that the digital twin brings the development stakeholders together. The platform has also helped to model urban functioning and evaluate public policies and services.
In a second step, the platform was also opened up to the business ecosystem, with the knowledge that there would be a need for specific functionalities. This is why a call for projects has been launched to companies to test different inputs on this digital twin.
The current phase of this experimentation is the operational deployment and operation by use case and by phasing. There are currently three cases: the first is the solar map, through which Rennes plans to encourage citizens to fit photovoltaic panels - the second is consultation on a tram/bus project to bring systemic information to inhabitants and encourage them to give their opinion - the third is an interface on 5G which gathers information on equipment and on exposure in measured values).
The last speaker, Mr Thierry Chambon, Partnerships Manager at Loamics, focused on the contributions of the operation of a digital twin, in comparison with the traditional systems.
The main characteristic of Loamics' digital twin is that the data is bidirectional/multidirectional, meaning that the data produced feeds the digital twin, which in turn feeds other data sources and provides information on the first data source. It is this loop that is important in the digital twin.
The digital twin creates a global system of supervision of the virtual assets gathered together and enhanced with information and makes it possible to synchronise operation and decision. It is also possible to obtain many different personas, i.e. people who will act/work differently with the same data stream.
Two elements are crucial for the operation of a digital twin; first of all, the notion of asset, which can be a source of data (or several) such as the operation of a boiler room, the operation of a boiler room in a building, or even in a group of buildings. All these examples are assets, which can be combined. In a digital twin, it is possible to create an infinite number of assets and focus very specifically on certain aspects. This multi-asset view also allows the integration of AI or Machine Learning.
Another essential element is the ontology of the data (study of the link between data). This link must be able to appear in the digital twin but also be changed easily (deletion of data and/or replacement of a piece of data by another). With digital twins, we can change the ontology and study how things work.
To use a digital twin effectively, we must be able to work with the input stimuli and then choose the best configuration for each case. We can even go further and keep records of the tests performed thanks to digital twins, which guarantee continuous improvement. Digital twins are also very interesting in terms of teaching since they can instantly show the consequences of a change.
The digital twin actually creates a dynamic, multi-persona (flow manager, supervisor/decision-maker, exploitation/operations) supervision system; it dynamically creates new business datasets and interoperability in the no-code way.
Loamics is proposing a multidirectional system in which, whatever the data sources that come in, they can be analysed and/or stressed, leading to a result, which will itself return to the digital twin. This is used in Smart Cities, in industry, and in Smart Building.
According to Mr Chambon, Smart Cities data is heterogeneous and truly represents the future of the digital twin. He believes that each stakeholder in the Smart City still works too independently, but that the digital twins will help to strengthen the collaboration between Smart City stakeholders!
Would you like to take part in the Smart Lunches starting in September 2022, as a spectator or speaker? Contact Tanguy De Lestré