November 20, 2024
According to the consensus study report Foundational Research Gaps and Future Directions for Digital Twins published in December 2023 by the National Academies of Science, Engineering and Medicine (NASEM): “A digital twin is a set of virtual information constructs that mimics the structure, context, and behavior of a natural, engineered, or social system (or system-of-systems), is dynamically updated with data from its physical twin, has a predictive capability, and informs decisions that realize value. The bidirectional interaction between the virtual and the physical is central to the digital twin.”
Karen Willcox, professor of aerospace engineering and engineering mechanics and director of the Oden Institute for Computational Engineering and Sciences at The University of Texas at Austin, knows a thing or two on the topic. Her TEDx talk, which was elevated to TED.com in September 2023, has surpassed one million views. She also chaired the NASEM consensus study. Willcox says that it's an exciting time for digital twins, while noting it’s important to separate what is currently achievable versus what is aspirational.
What impact has the NASEM digital twins report had so far?
We've already seen a lot of action! The White House created a Fast-Track Action Committee to draft a national strategic plan for digital twins. Federal agencies are responding to the report's call for cross-agency collaborations, with the National Science Foundation (NSF) hosting an Interagency Digital Twins Day with 13 different agencies represented. There have been cross-agency funding calls from NSF, the Air Force, National Institutes for Health, and the Federal Drug Administration. It is wonderful to see these emerging collaborations across domains.
Why so much interest? What's so compelling about digital twins?
Imagine the possibilities in medicine, where digital twins could combine predictive modeling with patient data to drive clinical decision-making tailored to the individual patient. Or imagine how a digital twin of planet Earth might help humanity to understand our changing climate and make better decisions about managing resources. But I have to emphasize - while we are imagining possibilities, it is also critical that we separate the aspirational from the actual. The NASEM study concluded that the publicity around digital twins currently outweighs the evidence base of success.
What role will computing advancements play in achieving these aspirational visions?
For critical digital twin applications, like those in medicine, engineering, and the Earth sciences, predictive mathematical models must form the core of the digital twin. These mathematical models encode the governing laws of nature, often represented as differential equations. Solving these equations at the scale of planet Earth or the scale of a full human goes beyond what we can do today. Increased computing power, together with advances in theory and algorithms, will advance the scale and complexity at which we can solve these models. The digital twin's bidirectional interaction between virtual and physical also drives computing needs. Advances are needed in hardware and algorithms for real-time data assimilation, digital twin updating, dynamic predictions, visualization, decision-making, and human-digital twin interactions. Many of these computations must take place in situ – in the clinic, or embedded onboard an aircraft, or out in the wilderness – which means an important role for lightweight computing at the edge.
The NASEM Report devotes an entire section to trust and validation. How do you go about gaining trust for digital twins?
Trust is paramount. There is a pressing need to advance methods that quantify and characterize uncertainties in digital twin predictions. That's a grand challenge within the field of computational science. From a practical perspective, it’s important to recognize the role of regulation – engineers have been using computational tools for decades to create complex systems like cars, aircraft, and bridges. Why is it that flying is so safe, and buildings so resilient to earthquakes? It’s people designing engineering systems within a culture of safety, together with the regulations that wrap around. Digital twins aren’t replacing that - rather, they complement it.
What are some of the digital twin efforts ongoing here at UT Austin?
So many examples! The Oden Institute is home to the Department of Energy's Mathematical Multifaceted Integrated Capability Center on the Mathematics of Digital Twins, addressing applications in manufacturing and climate. Medical digital twins are another large focus area. For example, a partnership among TACC, the Oden Institute, and MD Anderson Cancer Center.
