CAMBRIDGE, MASSACHUSETTS — Dean Price joined the Massachusetts Institute of Technology faculty in September 2025 as an assistant professor in the Department of Nuclear Science and Engineering and the Atlantic Richfield Career Development Professor in Energy Studies. Price is developing artificial intelligence-driven multiphysics models for nuclear reactor design and safety analysis.

"Nuclear energy has been a tremendous part of our nation's energy infrastructure for the past 60 years, and the number of people who maintain that infrastructure is incredibly small," Price said. "By becoming a nuclear engineer, you become one of a select number of people responsible for carbon-free energy generation in the United States."

The United States has 94 operating nuclear reactors, more than any other country, and those reactors collectively provide nearly 20 percent of the nation's electricity. Multiphysics modeling — the simultaneous analysis of interacting physical processes inside a reactor core rather than studying them in isolation — is well established for large light water reactors with capacities on the order of 1,000 megawatts. Methods for modeling small modular reactors, with capacities ranging from roughly 20 to 300 megawatts, and microreactors rated at 1 to 20 megawatts are far less advanced. Only a very small number of such reactors are operating today. Small modular reactors and microreactors have the potential to produce power more cheaply and more safely and offer greater flexibility in power output and size.

The approach involves coupling processes such as neutronics — the movement of neutrons within a reactor core that causes nuclear fission and generates power — with thermal hydraulics, which involves cooling the reactor to extract heat. A multiphysics simulation analyzing the interaction of these processes can show how heat removal affects neutron behavior, since hotter fuel is less likely to cause fission. These simulations can require supercomputers to solve or approximate coupled, nonlinear equations.

Price is exploring AI approaches that could reduce that computational burden by bypassing nonlinear differential equations. When provided with sufficient data, machine-learning models can help understand relationships between key physical processes without solving nonlinear differential equations. Machine-learning models can also infer fuel temperature and three-dimensional temperature distribution in a reactor core from its power level. "If you ever want to change your power level, or do anything with the reactor, the temperature of the fuel is a critical input that you need to know," Price said.

"Multiphysics modeling allows us to correlate the fission neutronics processes with a thermal property, temperature. That, in turn, can help us predict how the reactor will behave under different conditions," Price said. "By really pinning down those relationships, we can make better design decisions in the early stages."

Price is also investigating applications where AI may help design novel types of nuclear reactors. According to Price, AI would augment established safety procedures developed over the past 50 years rather than directly interfacing with safety-critical systems. "We could then rely on the safety frameworks developed over the past 50 years to carry out a safety analysis of the proposed design," Price said. "In this way, AI will not be directly interfacing with anything that is safety-critical."

Before arriving at MIT, Price studied the safety of steel and concrete casks used to store spent reactor fuel rods as an undergraduate at the University of Illinois Urbana-Champaign. His analysis indicated that storing spent fuel rods in such casks after they had cooled in water tanks for several years is quite safe, though the question of long-term disposal of spent reactor fuel casks remains open in the United States. He began graduate studies at the University of Michigan in 2020, where his research focused on multiphysics modeling of interacting physical processes in a nuclear reactor core.