Hung-Po Cheng didn’t just come to the University of Utah to earn an engineering Ph.D.—he came to create a real-world impact by making the roads we travel every day just a little safer. With a passion for data-driven safety solutions and a vision for smarter cities, Hung-Po is using artificial intelligence to improve traffic safety.

“I chose the University of Utah for its outstanding transportation engineering program and its emphasis on cutting-edge AI research,” he says. “It’s exciting to work on projects that bring together real-world transportation challenges and innovative technology.”

Under the mentorship of Associate Professor Xiaoyue Cathy Liu, Hung-Po is leading research that uses knowledge graphs and large language models (LLMs) to predict high-risk traffic accident hotspots. His goal? To develop tools that help cities prevent crashes before they happen—saving lives through smarter planning and decision-making.

Hung-Po’s work recently earned him a $2,000 Student Mini-Project Award from the Center for Smart Transportation (CST). This award supports student-led research and publications, empowering scholars like Hung-Po to take charge of their projects and publish as first authors. The funding will help advance a publication titled "Enhancing Traffic Safety Analysis through Knowledge Graphs and LLM Integration."

This project aligns with CST’s Focus Area 4: Smart Cities, Big Data & Innovative Technologies to Improve Transportation System Efficiency, and will be submitted to the Journal of Transport Geography.

Hung-Po’s research interests include:

• AI-driven frameworks to predict and mitigate high-risk traffic accident hotspots
• Multimodal AI pipelines using vision-language models (VLMs) for deeper transportation safety insights

With the CST award and the support of the University of Utah, Hung-Po is setting a new standard for how artificial intelligence can be applied in transportation engineering—making the roads we travel every day just a little safer.