Team harnessing big data to make bridges safer
A revolutionary new way to share, link and analyze critical information about the nation’s infrastructure is in development, thanks to a $1 million grant from the National Science Foundation (NSF) to a research team led by CEE associate professor Nora El-Gohary. The Civil Infrastructure Systems Open Knowledge Network (CIS-OKN) promises to enable safer, more efficient and more cost-effective construction, operation and maintenance of U.S. infrastructure by enabling open access, integration, and analysis of the massive amount of data that exists about the condition of infrastructure – but which has been widely inaccessible and fragmented until now.
The project is being funded under an NSF pilot program called the Convergence Accelerator that, NSF writes, aims to “accelerate use-inspired, convergence research in areas of national importance” by funding multidisciplinary teams from academia, industry and government agencies. The goal is a fast track from research to prototypes to implementation. The project was featured in July at the NSF Convergence Accelerator’s virtual event, NSF CA Expo 2020.
U.S. infrastructure as a whole has been given a grade of D+ by the American Society of Civil Engineers. A wealth of data exists about infrastructure conditions, El-Gohary said, but it’s too disconnected and heterogeneous to be useful. Without it, though, decision-makers can’t make informed decisions about prioritization, maintenance and rehabilitation of infrastructure systems. The team’s goal is to develop and test a cyberinfrastructure system for locating, accessing, integrating and analyzing data about civil infrastructure from multiple sources and in heterogeneous formats to support Artificial Intelligence (AI) and decision making.
For the team’s initial proof-of-concept project, they will focus on the nation’s bridges, El-Gohary said. More than 6,000 U.S. bridges are structurally deficient, and 188 million trips a day are taken over structurally deficient bridges, she said.
“We have a lot of data – structured inspection data, weather data, traffic data, sensor data, socioeconomic data – and we also have a lot of unstructured data that have been hard to digest – inspection reports, inspection images, accident reports, etc.,” El-Gohary said. “We can learn from these data to better predict deterioration and failure, as well as to learn how to maintain our bridges in a better way in terms of safety, cost and socioeconomic impacts. This project will make this happen.”
In addition to El-Gohary, the Illinois team includes CEE professors Ximing Cai and Yanfeng Ouyang, Computer Science professor ChengXiang Zhai, Urban and Regional Planning professor Arnab Chakraborty and Jong Lee of the National Center for Supercomputing Applications. Collaborators from other institutions include researchers in civil engineering, computer science, data analytics, urban planning and social sciences from the University of Illinois, the University of Southern California, Purdue University, Carnegie Mellon University and Arizona State University. Thirty-two partners and collaborators from transportation centers, state and federal transportation agencies, contractors, consultants and major technology development companies are also involved.
The result, said El-Gohary, will be an open knowledge network that links the physical elements with the relevant data and data sources and provides data-driven AI analytics that support innovative solutions and decision-making for “safer, more resilient and more sustainable urban infrastructure.”
Top photo: iStock.com/Lawrence Sawyer