Howard University civil engineering assistant professor Sanjib Sharma, Ph.D., and his College of Engineering and Architecture research team have conducted studies on how AI can help address urgent infrastructure and public health challenges in response to increasing climate change crises.
According to the National Resources Defense Council, communities across the country face growing challenges from aging infrastructure, contaminated water systems, and increasingly severe urban flooding. Addressing these risks requires new tools that can detect hazards earlier and guide smarter decisions.
The CEA researchers focus on integrating AI, big data, and high performance computing to improve predictions of extreme events, such as floods, droughts, and heatwaves. Two of their studies have been published in Nature Scientific Reports.
The first published study, Explainable machine-learning-based predictions of blood lead levels and school drinking water contamination among children: a case study in Washington D.C., was led by Dylan Darling (BSCE ’24), who graduated from the civil engineering program just last semester.
Explainable machine learning is used in the study to predict lead contamination risk in drinking water systems. Lead in drinking water poses a serious health risk. No level of lead exposure is considered safe. The United States has an estimated over 9 million service lines containing lead.
Reflecting on his research experience, Darling shared that the project taught him how data science and engineering can work together to tackle real public health challenges and generate information that helps communities stay safe.
“Seeing an undergraduate student lead research published in a high-impact journal is truly inspiring,” said Sharma. “It reflects the strong dedication our undergraduates bring to meaningful research that advances both scientific understanding and community well-being.”
The use of AI is a powerful way to pinpoint the location with the greatest risk of lead contamination. This information helps to make informed decisions on infrastructure investment.
The second study, Ensemble learning for enhancing critical infrastructure resilience to urban flooding, was led by Yogesh Bhattarai, Howard University civil engineering Ph.D. student. The research focuses on improving predictions of urban flooding.
“Our goal was to build models that can support real-time decision making,” said Bhattarai. “Urban flooding is highly localized and fast-moving, and traditional flood models often miss those details. By integrating crowd-sourced data with AI models, we’re able to generate more accurate, actionable information for communities and emergency managers.”
AI can overcome the limits of coarse flood models and outdated hazard maps by providing street-level predictions in real time. These advances give emergency managers and city planners sharper, faster insights, making communities better prepared for extreme weather.
Sara Kamanmalek, Ph.D., Howard University civil engineering assistant professor, and Vijay Chaudhary, Ph.D., Howard University computer science lecturer, contributed to the research studies as co-authors.
This research reinforces Howard University’s efforts to drive meaningful, responsible AI innovation.
