Engineers have, by necessity, a practical nature. This stands in contrast to scientists, who focus on the complete understanding of a subject of interest. For example, scientists study the mating habits of constipated scorpions or the cognitive impact of video games on children. Engineers, in contrast, are focused on solutions that are practical, optimized, and always useful.
That’s why we want to define digital twins in engineering, rather than scientific, style. We think digital twins are described too theoretically. As an alternative, we are going to break down the concept into smaller components. We’re going to tell you how we’re making digital twins and how we’re using them.
Syska Hennessy’s Definition of a Digital Twin
A digital twin is a representation of a real-world object that is used to link to other, related data.
Let’s break it down:
- Representation. It can be complicated to explain how something is represented. For instance, an object can be represented with a photorealistic mesh and texture, a mesh created by a manufacturer, or even just a geolocated bounding box. All of these representations can meet the needs of a contractor, but what does the client need? That depends on how the digital twin is created, deployed, managed, and maintained. Our definition emphasizes digital abstraction: the process of taking away or removing characteristics from an object in order to reduce it to a set of essential characteristics that are easy to navigate/represent.
- Real-world Object. An object in design software isn’t a real object until it gets built or manufactured. It’s that simple. An HVAC cooling tower in Autodesk Revit isn’t a digital twin of the cooling tower until it represents a real-world object because everything about the cooling tower continues to have the ability to change, which ruins the context of linked data. The data linked to a digital twin is more valid because the context – known as typing to a computer scientist — is connected to a real-world object.
- Link to Data. The ability to link data is the core value and utility of a digital twin. Whether the actual data is contained within the same system as the digital twin or hosted outside is immaterial. The ad-hoc, flexible ability to link data to digital twins is what makes them valuable. Consider how the HTML link tag has evolved over time. The value of an HTML link tag is that it can take you to a different web address, link to a position in the same document, or do a ton of other actions. HTML link tags are the foundation of internet commerce in 2022 because they use the context of the tagged object and link to other data. Digital twins are like hyperlinks for 3D data.
How is Syska Hennessy Using Digital Twins?
Syska Hennessy considers our tech stack to be a big competitive advantage. (In fact, we think it’s second only to our people!) So, we don’t want to say too much… but we can say that our engagement with digital twins focuses on applying this emerging technology to industry challenges:
- Use Digital Twins for Improved Project TurnoverSyska Hennessy is engaged to define and engineer building systems, which are then constructed by others. Digital twins improve outcomes for our clients when our intellectual work flows into project turnover assets. This can take many shapes, but we think engineering is a requirement for turnover data. (Get a hold of us if you want a specific example!)
- Align Project Digital Twins with Our Client’s Operational UsesAs we said above, representation of real-world assets can take many shapes, which will depend on the client’s use of the digital asset. For example, you’d be surprised by the value of a PDF or working model with each telecom data port labeled and linked to our client’s ERP system for managing a network. Now, imagine this same information in augmented reality. This same ideology can then be applied to all MEPT trades, electrical panels, plumbing valves, mechanical VAVs, etc. The possibilities are extensive.
- Be Our Own Customer FirstAll projects get to a point where certain data becomes irrefutable and fixed. When this occurs we can use digital twin technology to cross the barrier from design into operational analysis. Normalizing the data pulled at a granular trade level vs. using designed/expected data provides valuable insight. Categorizing this data and then processing through some clever machine-learning algorithms results in optimized design solutions that we can recommend. The machine-learning algorithms can be as simple as thermostat placement and as complicated as condenser water-control reset strategy. We can’t wait to show you more!
As always, discussion is the most valuable part of communication, and we welcome your thoughts on how we’re defining digital twins in our part of the AEC industry. Please don’t hesitate to weigh in and offer definitions of your own.