A digital twin, as defined by the Digital Twin Consortium is:
“a virtual representation of real-world entities and processes, synchronized at a specified frequency and fidelity” using “real-time and historical data to represent the past and present and simulate predicted futures.” (1)
Although there are a variety of definitions across this growing industry, here at Implexus Lab when we produce a digital twin, we must first outline what level is best suited for our client’s needs. The way we accomplish this is by understanding to what end the twin will be utilized at this time. As technology evolves, we ensure digital twins can be adaptable to level up when needs change.
Five Levels of Digital Twins:
Here at Implexus Lab, we categorize digital twin deliverables into one of five levels:
Descriptive, Diagnostic, Predictive, Prescriptive, and Autonomous
A Descriptive twin offers historical and real-time data overlayed onto the 3-dimensional form.
The Diagnostic level is where we begin to use data analytics to showcase trends in how the space is being used, ideally diagnosing potential efficiency opportunities.
A Predictive twin builds off the diagnostic level by beginning to predict issues before they arise, based on historical data of building use and asset information. Think predictive maintenance.
The Prescriptive twin level not only diagnoses and predicts issues, but it proactively prescribes ways in which the building could be operated most efficiently. This is where we reach the “seatbelt ding” level.
Cars didn’t use to have seatbelts. And when they became mainstream beginning in the 1950’s people didn’t always wear them. Introduce the seatbelt ding - the annoying sound that results in a human behavior change towards safety. With the partnering of artificial intelligence and building operations data collected at these higher levels, the digital twin can begin to influence human behavior. In this case, instead of making your body safer in the event of an accident, a prescriptive twin can aid humans in altering building use to operate at its most efficient, saving cost and energy. It is still up to the human whether to heed or ignore the advice.
Which brings us to our final level, the Autonomous twin.
The International Energy Agency produced a report stating that “in 2021 the operation of buildings accounted for 30% of global final energy consumption and 26% of total energy sector emissions. (8% being direct emissions in buildings and 19% indirect emissions from the production of electricity and heat used in buildings.) Both energy consumption and emissions rebounded to above 2019 values, following the drop in 2020 from COVID-19 restrictions.” (3)
As we grow in the smart building industry, our hope here at Implexus Lab is that we reach a point where the partnership of building use data and artificial intelligence creates an opportunity for energy reduction. If it means removing some aspect of human choice and replaced by automated operations, we may see greatest benefit. We believe the Autonomous digital twin is ambitious at this point, but we must have that forward vision of the peak in mind as we evolve.
Comments