Images taken as NasaThe spacecraft crashes into the asteroid Dimorphos at around 14,000 miles per…
Xiao Liu, assistant professor of industrial engineering, received a CAREER award from the National Science Foundation. The Faculty Early Career Development (CAREER) program offers the foundation’s most prestigious awards to support early-career faculty who have the potential to serve as academic role models in research and education and advance the mission of their department or organization.
His project, titled “Domain-Aware Statistical Learning,” will help advance national competitiveness by transforming how physics and engineering domain knowledge governance is integrated into data-driven models for high-stakes applications. High-stakes engineering applications require interpretable models and explainable decisions.
While fundamental governance physics imposes critical constraints on how data should be modeled and how models can be interpreted, integrating governance physics into data-driven models becomes a critical capability that will dramatically accelerate the penetration of data science in a wide range of fields. knowledge-intensive applications, such as energy infrastructure, aviation security and manufacturing. In these environments, the old paradigm of “let the data speak for itself” is being replaced by the ability to “let the data speak based on the laws of physics and engineering.”
This project will address the development of methods to integrate data with physics-based models in three main use cases, namely environmental processes to improve the resilience of our national utilities during extreme events intensified by climate change , thermal modeling to improve energy efficiency in data center operations. and the structural dynamics to improve aviation safety in an increasingly congested airspace.
The accompanying curriculum plan aims to bridge the gaps between generalist data science education at the school and university levels and the specific needs of next-generation engineering students from diverse backgrounds. The educational plan also aims to improve data literacy among the general public by improving awareness of the growing availability of data and the ability to interpret this data through local community activities.
“The lack of explainable models and actionable insights has become the main barrier to the penetration of black-box data-driven approaches into high-stakes engineering applications,” Liu said. “The long-term vision is to break down this barrier by creating a domain-aware statistical learning paradigm that enables the direct integration of fundamental governance physics into interpretable data-driven models and nurture engineers of the next generation that apply analytical tools in domain knowledge intensive environments.”
“We are thrilled to see Dr. Liu’s work recognized by the National Science Foundation with the foundation’s most prestigious award for young teachers,” said Ed Pohl, department head. “This grant helps lay the foundation for the integration of education and research throughout his career. His work coupled with creating explainable models and actionable insights will have a significant impact in helping us better design, analyze and operate complex engineering systems in the future. We are extremely proud of him and fortunate to have him as a member of our team.”
Liu received the Rising Star Award from the College of Engineering in the spring of 2022. “It is not surprising to learn that the National Science Foundation has recognized the importance and impact of Xiao’s research with this prestigious early career,” said Kim Needy, Dean of the College of Engineering. “This funding helps cement Xiao’s status as a rising star in industrial engineering research.”