The widely recognised definition of a digital twin was provided by Glaessegen and Stargel in 2012 as ‘an integrated multi-physics, multi-scale, probabilistic simulation of a complex product and uses the best available physical models, sensor updates, etc., to mirror the life of its corresponding twin’. Put simply it is a virtual model of a real-life object or place, linked through data feedback mechanisms in real-time. As such it is composed of three parts: the physical product, the virtual product and the connected data that ties the physical and virtual product together (Tao, et al., 2018). This has several key characteristics: the virtual space is meant to be a real-time reflection of the physical space, there is interaction and integration between the physical and virtual space, and the digital twin can update data in real time so that the virtual models can undergo continuous improvement (Tao, et al., 2018). The value of this is the ability to take a holistic view of the object or space in question whilst also enabling the user to diver deeper in specific parts of the data at the same time (Dawkins, et al., 2018).
This idea has developed in recent decades due to advancements in IT and sensor capability. This is to the extent that they are now actionable and are being used in contexts such as manufacturing, aeronautics and urban systems (Grieves & Vickers, 2017). The idea is that the digital twin is able to design, test and manufacture using the virtual version of the system to reduce failures of the physical system when it is deployed. In doing so this would reduce expenses, time and harm to the user of the system (Grieves & Vickers, 2017). Furthermore, the information gathered through the linking to the physical system can be used to explore the current state and past history of the product in question whilst also being able to predict future behaviour and performance under certain conditions. The main benefit in this sense is determining unpredictable undesirables that result from the creation of the object (Grieves & Vickers, 2017). As such, the core premise of the concept is that information is a replacement for wasted physical resources that would be used to gather such information. While information is never free to acquire or use, the assumption is that developing it this way is cheaper than the cost of the wasted resources (Grieves & Vickers, 2017).
Most of this literature covers the use of the digital twin in the case of manufacturing a product, for which the digital twin tracks the physical manifestation of the object over time and helps to improve efficiency. More recently however there have been attempts to implement the same concept in the physical urban environment, although mostly in the case of individual buildings. This can be seen in the case of UCL’s Here East Campus and the Smart Campus Demonstrator in the University of Brescia. In the latter case the focus has been tracking user behaviour, detecting indoor temperature, relative humidity, illuminance etc., to create a BIM model through which individuals could interact using an app. This allowed them to say their comfort levels which then would allow the model to learn and adapt overtime to become a dynamic IoT application (Ciribini, et al., 2017). This therefore allows for the generation of an interactive concentrated environment to improve the conditions for users and gather data for future improvements in this and other buildings. The benefit of which depends on what sensors are used, what data is gathered and how that is interacted with by these individuals.
For our purposes, the question from this becomes to what extent would this be useful as an application for a wider campus? As the example of the University of Brescia demonstrates, as they go beyond pure sensors they can then become a platform for application development which can serve many different stakeholders (Fuldauer, 2019). The applications then developed depend on the local needs and what information is available within the digital twin. Most of these therefore would seem to revolve around creating a more responsive environment that would be easier to maintain or monitoring how the current campus works towards its goals. The latter point would be useful to see whether a campus helps to facilitate interaction and the use of certain spaces. Gathering such information in a digital twin would allow for clear understanding of what is working and what isn’t and hence allow for idea generation as how to fix it. It is not only the campus that could benefit from this information but also other campuses and buildings that could learn lessons from the data generated and help to improve construction ideas and further expansion. As new technology prices continue to fall, this is becoming and more feasible and we are likely to see digital twins being developed in many different ways to aid our understanding and provide us with information that would have never had access to before.