Student award at TRB

Rutgers doctoral student in Civil and Environmental Engineering, Bingyan Cui, was presented the First Place award at TRB’s Annual Meeting in January.

The contest encourages undergraduate and graduate students to explore research areas and challenges in pavement engineering through analysis of long-term performance data.

Rutgers Engineering student Bingyan Cui was selected as the first-place winner in the USDOT Federal Highway Administration’s (FHWA) 2023 Long-Term Infrastructure Performance (LTIP) Student Data Analysis Contest.

Bingyan is a doctoral student advised by Professor Hao Wang in the Department of Civil and Environmental Engineering at the Rutgers University School of Engineering.

From sea level rise to increased flooding and severe storms, the impacts of climate change expose transportation infrastructure to new stressors that are making it difficult to reliably predict the life-cycle performance of the nation’s roads, tunnels, and bridges.

The contest paper, “Predicting Pavement Deterioration under Climate Change Uncertainty,” developed a promising approach for incorporating climate uncertainty into pavement performance prediction—a critical consideration for asset managers who are responsible for planning maintenance activities and allocating funds to infrastructure projects.

Bingyan was recognized by FHWA as the contest winner at the 2024 Transportation Research Board (TRB) Annual Meeting this January.

Collaborating with Professor Wang, also an affiliated researcher at the Center for Advanced Infrastructure and Transportation (CAIT), Bingyan used Long-Term Pavement Performance data from 1989 to 2021 to develop an improved pavement performance prediction model incorporating climate uncertainty quantification into machine learning models.

Specifically, the impact of climate change on International Roughness Index (IRI) and rut depth was analyzed across four different regions. She found increases in IRI and rut depth were more significant using the developed model compared to using historical climate data alone across the regions. Under future climate projections, IRI changes can range from 12 to 17 percent, while increases in rut depth may exceed 40 percent with variations in climate regions, as compared to those predictions using historic climate data.

“Relying only on historical climate data underestimated pavement deterioration. This project highlights the importance of analyzing and using different data streams to make more informed infrastructure management decisions,” Professor Wang said. “I am glad to see that Bingyan uses the power of machine learning in her research to support and design strong, durable pavements that can stand up to climate change.”

The FHWA’s LTIP Data Analysis Contest is designed to introduce future transportation professionals to data, providing an opportunity to engage with quality performance metrics, apply information through appropriate research methods, and derive recommendations that are data-driven and implementable.