Professor Sara Shashaani
- Core subject area: Stochastic optimisation, derivative-free optimisation, robust decision-making under uncertainty, digital twins
Research Interests: Sara Shashaani is Associate Professor and Bowman Faculty Scholar in the Fitts Department of Industrial and Systems Engineering at North Carolina State University, and Oliver Smithies Visiting Lecturer at Balliol for Michaelmas Term 2026. Her work is about using mathematical programming and applied probability for making good decisions with computer simulations of problems that may be very nonlinear but also where the true values are unknown and only accessible with a simulation model that is random, where each run gives a different answer. She designs algorithms that decide how much simulation effort to spend, and where to spend it, to find the best decisions reliably. She co-developed a family of adaptive-sampling algorithms, called ASTRO-DF, for optimising noisy simulation models for zeroth-order and nonconvex problem, and the SimOpt open-source library for testing simulation-optimisation solvers. She is working with Professor Coralia Cartis’s group on scaling these methods to high dimensions with random subspaces during her stay. At Balliol, she is giving two public lectures on digital twins, on 11 and 18 November.