We are currently in the Fourth Industrial Revolution, also known as Industry 4.0. Technology changes every day, and with the development of the Internet of Things (IoT), Artificial Intelligence (AI), and Virtual Reality (VR), the lines between the physical, digital, and biological worlds have been blurred. One of the latest technologies that continues to connect the digital and physical world in incredible ways is a digital twin. While the concept of a digital twin has been around for over 10 years in the manufacturing industry, less than 5% of the AEC industry has adopted this technology.
Photography from: AEC Magazine
So, what exactly is a digital twin? In the context of architecture, a digital twin is a dynamic, digital replica of a building and all of its associated data. It is an accurate 3D model of a building equipped with any data associated with it, including the building’s construction history to real-time data transmitted from sensors. There are four main components of a digital twin: the Internet of Things, Extended Reality, Cloud Computing, and Artificial Intelligence.
IoT: IoT is the main technology used in digital twin technology. Buildings with digital twin models are equipped with IoT sensors that superimpose real-time sensor data on a 3D model. Through the IoT, data such as weather, temperature, or number of people in a building is collected from the physical world and sent to the virtual model. Application Programming Interface connects cloud data to the virtual object.
Extended Reality: Extended Reality is a tool for the visualization and simulation of spaces. Digital Twins are often created from laser scans, creating interactive spaces that simulate the environment. In some cases, VR headsets can be used to simulate being in the space itself. Users can interact with digital twin models using VR and get a better idea of what the model will physically look like without being there themselves.
Cloud Computing: Cloud Computing is the efficient storing and accessing of data over the internet. Massive amounts of data are generated for digital twins and Cloud Computing provides a way to store and sort all of that data.
Artificial Intelligence: Through machine-learning algorithms, digital twin software can take the data associated with a model and make predictions about energy usage, cost, and when systems might fail. These predictions can help buildings operate more efficiently and help designers understand the lifespan of their buildings and building systems.
In order to create a digital twin of a building, a digital model must be created first. The digital model can either be modeled using any traditional CAD program such as Revit, or it can be created using LiDAR technology where a laser scanning camera such as a Matterport Pro camera generates a point cloud of a space. Once the digital model is created, sensors are installed in the physical system. These sensors can capture temperature, pressure, smoke and gas levels, and any other necessary data. After the sensors are installed in the physical building, the model can be linked with digital twin software and the sensors can be modeled in corresponding locations on the digital model. Then, the data can be transmitted in real-time via the IoT and analyzed using AI. Manual data can also be entered into the digital model by users.
One exemplar digital twin application in the architecture industry can be seen in the UCSF Precision Cancer Medicine Building. This 7-story, 180,000 square foot addition was designed using a unique data-first approach. Instead of starting with the skeleton of the building, the designers started with the engineering systems and specific hospital equipment to meet the precise conditions required by UCSF. The engineers and architects built a digital twin using IBM Maximo software, uploading their Revit model and adding the necessary data. The resulting model, nicknamed BIM4FM, was a fully coordinated architecture and engineering effort. Not only was the design of the building much more efficient, but the maintenance process of the building was improved as well. Once it had opened, the hospital café had a significant leak through a 7-story fire-rated chase. The engineers on the job used the digital twin to isolate the leak and come to a complete diagnosis within 90 minutes. They were then able to repair the leak within 2-3 hours without shutting down the room that was the source of the leak or the café. Without the digital twin model, this process could have taken 2-3 days. Digital twins can help detect problems like this as well as predict other potential failures in MEP systems using AI.
Photography from: IBM
While this technology seems too good to be true, it is not out of reach. Design Collaborative already uses Revit to construct their models. Paired with IBM Maximo, an affordable software for all of its benefits, it is possible to generate our own digital twins. With digital twin technology, architects and engineers would be able to further improve their coordination efforts, reduce site visits, simulate building systems, and save money. Digital twins have the ability to revolutionize the AEC industry.