Over the past year you, like me, may have noticed the emergence of a new phrase within the built environment, ‘Digital Twin’. In the late naughties, we had organizations feign their commitment to sustainability (Greenwash). In the early teens, we had organizations feign their commitment to collaboration (BIMwash). Now in these late teens, we have organizations feigning their commitment to cyber-physical systems (Twinwash). So let’s set the record straight shall we?
What is a digital twin?
Put simply, a digital twin is a digital representation of a physical asset used to support decision making, that maintains a relation to the physical asset. For example, my:
- Facebook profile is part of a personal digital twin; and
- Asset Information Model of @TyDigidol is part of the digital twin of my house.
Similar to how you update your Facebook to show off a new haircut, in the built environment The ISO 19650 Series describes how a change to a physical asset (known as a trigger-event) prompts a change to the asset information. The physical drives the delivery of information.
However, digital twins take it a step further; allowing the digital to help make decisions which influence the physical. For example, if everyone comments brilliantly about your new haircut, you may decide to keep it and go back to the same barber. In the built environment, sensor information can be used to collect information about an asset’s use and its performance. The digital drives physical interventions.
How can the digital drive physical interventions?
Quite simply, though insight gained from information.
For example, I have a Nest Thermostat in my living room. It is a physical device that inputs sensor information into a digital twin of my home’s thermal conditions. Through the use of IFTTT, a free web-service, I have connected that same information to a Twitter account that will notify me when my living room temperature reaches above 25°.
Following this insight, I can decide on whether I should undertake an intervention such as opening a window or, if a trend is seen, install some Brise Soleil (#Fancy). Without needing user input, I am receiving notifications from my digital twin about my physical asset.
Does the digital need to duplicate the physical?
No, the digital twin only needs to represent the physical asset relevant to the decisions you want to inform; they can be use case specific! A simple decision (do I need to open the window?) needs very little information, a more complex decision (do I need to upgrade the local utility infrastructure to support an increased demand?) needs a lot of information. The frequency, detail and degree of information will depend on the decisions being made. For example, internal temperature information might need to be captured frequently (once per minute?), but geospatial information in relation to building settlement is unlikely needed as frequently (once per month?). You may have also noticed that I haven’t mentioned geometric (3D) models at all. Depending on the type of information you need, a digital twin may not need geometric models and will likely look very different when compared to the physical asset.
And that’s it
If you want to find out more, the UK’s Digital Framework Task Group (DFTG) is currently undertaking a lot of research into the concept of digital twins and how they can fully support decision making within the built environment. They have already outlined the key concepts within the Gemini Principles as well as a Roadmap of the work they plan to undertake, both of which are well worth a look!
Note: If you have any comments about ‘Twin Transparency’, then please let me know either on Twitter or by commenting below.