A digital twin is a digital copy of a physical asset, person or system. It receives real-time information from the real thing so that it can replicate its performance – and then explore ways of improving upon it. This improvement information can be passed on to the asset directly for self-management, or to decision makers or users. With this capacity for learning and continuous improvement, it’s clear that digital twins can play a critical role in making our infrastructure more intelligent.

Digital twins have actually been around for decades; any time we use digital technology to model elements of the real world, we are building digital twins. In a digital form, computing power can be applied to increase intelligence. In the 1950’s, in the bid to win the Space Race, digital twins were used to train astronauts, understand how spacecraft and materials would perform in space, identify how all of the systems would work to support missions, and test a range of other scenarios — all using the digital technology of the time. Today, digital twins are increasingly pervasive; by 2020, more than half of manufacturers with annual revenues in excess of $5 billion will have launched at least one digital twin initiative.

The basic building blocks of a digital twin are:

1. The asset being ‘twinned’
2. Basic inventory data
3. Existing information (e.g., Cloud, Crowd Sourcing, Spatial, Databases, Structured and Non-Structured data, and Enterprise Systems)
4. Real-time information sensors (situated on the real assets to provide updated information on status and performance)
5. Communications/Internet of Things (IOT) (to retrieve and share information. The faster the communications, the more real-time the information, the faster adjustments can be made, and the faster performance can be improved)
6. Data management and analytical capabilities (Cloud, Big Data, BIM, Machine Learning and Artificial Intelligence)
7. Visualization representation, for better understanding of the asset, the situation, the prediction, and the recommended solution (Virtual, Augmented, and Mixed reality, and 3Dto6D Models, where the elements beyond 3D incrementally cover the following dimensions: Scheduling & planning, Cost, and Operation)

At a high level, the digital twin is created by scanning an existing asset or designing a new one, converting it to a 3D image, and connecting it to relevant static and dynamic information sources.

With new technologies, the digital twins of today can make our infrastructure systems more intelligent than ever before. To become more intelligent, infrastructure shifts from having a mere replica of itself, to being able to proactively manage itself. It also has to recognize its performance as part of an entire ecosystem of assets, users, and environment. The image below shows how the Internet of Things, Big Data, Artificial Intelligence, Spatial Networks and other technologies have come together to enable the creation of a modern, highly intelligent digital twin — one that has more context for greater situational analysis, greater ability to diagnose a situation, can predict with more accuracy, and can introduce faster self-corrections to the actual asset for superior performance.

Digital twins are only now taking off in popularity, because a handful of macro trends in society are acting to greatly simplify the process and reduce the costs involved. The expenses that were once associated with creating and accessing the information required to build elaborate and detailed digital twins could be prohibitively expensive, but with the advent of 5G and other technology advancements, we can now generate, store and move more data more quickly and cheaply than ever before. With this greater intelligence, predicting changes in society, climate, nature and the built environment facilitates better decision making that will prepare and protect Canadians.

Today’s digital twin can be used to predict and improve the flow of people and goods; to manage water supply and demand; to predict the lifespan of a bridge; to assess the vitality of neighbourhoods; and even to analyze human and planetary health, as shown in the image below.

Digital twins are not a magical crystal ball that will tell us what the future will look like for certain — but they do give us the means to model and predict a variety of different potential future scenarios. They can help us ensure that what we design and build today has resilience against a number of different future scenarios that might come to fruition. A range of different predictions for rising sea levels, for example, provides tremendous insight into how current coastal communities might be impacted and also provides valuable input for planning and designing future use of coastal lands.

Digital twins can help us better envision, design, build, operate, maintain and manage virtually everything in the built world. With increasing intelligence, they offer us increasing opportunity to expand how we contribute to a sustainable and resilient future for the built environments that we design.