UW–Madison Industrial and Systems Engineering Researchers Lead AI & Systems Integration for Major DOE Fusion Award

Researchers in the Department of Industrial and Systems Engineering (ISyE) at the University of Wisconsin–Madison are transforming the landscape of fusion innovation. This work is a cornerstone of the multi-institution Fusion Innovative Research Engine (FIRE) Project, an initiative funded by the U.S. Department of Energy and led at UW–Madison. The project aims to accelerate the development of technologies needed for commercial fusion power.

While fusion research has traditionally been driven by physics and nuclear engineering, UW-ISyE researchers are introducing Model-Based Systems Engineering (MBSE) and artificial intelligence (AI) approaches to integrate experimental efforts, manage system complexity, and improve decision-making across large-scale testing programs. Led by Prof. Hantang Qin, the MBSE/AI team is bringing advanced industrial engineering methodologies into nuclear fusion in ways that could transform how next-generation fusion systems are designed, tested, and deployed. These contributions support the development of a volumetric neutron source for the establishment and validation of integrated blanket technology in prototypical environments.

One of the most technically demanding challenges in fusion reactors is the design of the blanket, a component that surrounds the reactor core and must simultaneously extract heat, breed tritium, and shield sensitive components from radiation. To address this challenge, the UW-ISyE team is developing physics-informed machine learning models to guide the design of experiments used to manufacture advanced graded materials for the blanket unit. These models combine fundamental physics principles with machine learning to identify optimal processing windows that produce high-quality, structurally robust materials capable of withstanding the extreme conditions of a fusion reactor.

Another recent effort from the UW-ISyE group focuses on helping non-nuclear researchers engage with complex fusion experiments, including the Wisconsin HTS Axisymmetric Mirror (WHAM) device, a major platform used to study plasma stability and confinement. The team is developing large language model (LLM)–based tools that guide scientists and engineers through experimental data, documentation, and operational knowledge related to the WHAM system. By making specialized plasma physics knowledge easier to access and interpret, these tools can accelerate data analysis and strengthen collaboration across interdisciplinary research teams

Using model-based systems engineering frameworks, the team helps coordinate data and knowledge generated across the project’s experimental thrust groups. By integrating experimental results, operational constraints, and engineering requirements into a unified decision-support framework, this approach helps researchers better leverage testing campaigns and accelerate the transition from laboratory experiments to deployable fusion technologies.

The integration of systems engineering, AI, and nuclear fusion marks a significant shift in the development of large-scale energy technologies. By combining MBSE, AI-driven experimentation, and intelligent knowledge systems, UW–Madison ISyE researchers are creating a new paradigm that could accelerate the path toward abundant, reliable, and carbon-free energy.

Cover photo: MBSE Team (Left to right): Hasnaa Ouidadi (Postdoc Scholar), Kaibo Liu (Professor), Ying Fu (PhD Student), Xuepeng Jiang (PhD student), Hantang Qin (Associate Professor), Shiyu Zhou (Professor)