Research

Research Vision

In engineering systems, degradation, performance, and operational decisions evolve together: degradation limits what a system can do, and operating decisions in turn shape how it degrades. My long-term vision is to enable systems that anticipate how degradation changes what they can reliably accomplish and adapt operations in response. Toward this vision, I pursue three directions: 1) Learning system capability from heterogeneous data, including sensor signals, events, and operator logs. 2) Predicting capability evolution under different decisions with digital twins and uncertainty quantification. 3) Optimizing operations as capability evolves, with reinforcement learning and mathematical programming. I study these across physical scales, from material and component degradation to asset- and system-level performance, so that decisions at every scale inform how degradation and capability evolve. I apply these ideas in smart manufacturing, logistics robotics, and fusion and nuclear energy systems.

Potential Funding Sources

I will seek funding from NSF (AMDO, FRR, and CPS), ONR (Science of Autonomy), DoD (CBM+), and DOE (NEUP and FES).