This month marks the seventh anniversary of the Florida International University pedestrian bridge collapse in 2018, when a fatal design flaw resulted in the deaths of six people.
Although bridge collapses in the U.S. occur on average once every year or two, lives will be lost each time and the public will ask, “How could this happen?” But thanks to new AI-based digital twin research at the University of Florida, those questions could be asked far less frequently.
“We are always asking the question, “How can we utilize existing technology to lead to a better quality of life?'” said Aaron Costin, Ph.D., an associate professor in UF’s M.E. Rinker, Sr. School of Construction Management. “Saving lives is the aim of what we try to do when researching infrastructure.”
Building a digital infrastructure
Costin has been working alongside Alireza Adibfar (Ph.D.) to develop a groundbreaking digital twin framework, using artificial intelligence that enables real-time monitoring and efficient decision-making related to the maintenance, operation and management of bridges.
About 46,100 of the 617,000 bridges in the United States, or 7.5%, are considered structurally deficient and in poor condition, according to the American Society of Civil Engineers. And traditional bridge inspections rely on manual assessments, which can be hazardous, time-consuming and subject to human error.
That is where the digital twin framework comes into play. It serves as a virtual representation of a real bridge, continuously updated with real-time and historical data to provide an accurate depiction of its past and present conditions. More importantly, this technology can simulate potential futures, predicting structural issues before they arise.
Costin compares this to the character of J.A.R.V.I.S. in the film “Iron Man”—the AI assistant that monitors Tony Stark’s suit and provides real-time diagnostics when it is damaged.
“When Tony Stark pulls up the Iron Man suit in a virtual display, he can inspect its physical status, rotate it and interact with it while talking to J.A.R.V.I.S.,” Costin said.
“Similarly, when Tony is flying and gets hit, J.A.R.V.I.S. informs him that his suit has 80% functionality left in his arm. Essentially, that’s what a digital twin framework is. It enables real-time interaction between the physical and virtual worlds.”
Preventing bridge collapses
From an engineer to an architect to a pedestrian, everyone sees a bridge differently. All these perspectives need to be somehow captured in a digital twin model to represent reality.
This pioneering reality from the Rinker School combines weigh stations, bridge sensors and AI-powered technologies to improve data collection and provide a clearer picture of a bridge’s structural health. Additionally, it allows for remote monitoring of bridge conditions, reducing the need for physical inspections and lowering the risks associated with on-site work.
The technology is not just for future bridges, however. Even existing bridges can be retrofitted with monitoring systems to detect and display critical structural indicators.
“We are moving toward a digital infrastructure where everything could be a digital twin,” Costin said. “Whether it’s the entire UF campus or the roads around Gainesville, all of them could have pieces of data working together so you can monitor them for safety.”
University of Florida
Citation:
To prevent bridge collapses, researchers use AI-based digital twin technology (2025, March 21)
retrieved 21 March 2025
from https://techxplore.com/news/2025-03-bridge-collapses-ai-based-digital.html
This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.
