Overview
My research focuses on the use of stochastic simulation for decision-making under uncertainty. This encompasses everything from the design and analysis of ranking-and-selection (R&S) procedures to the comparison of simulation-optimization (SO) algorithms to the development of new methods for simulation output analysis.
Exploiting Structure in SO
In certain cases, SO problems possess structural properties that can be verified analytically, e.g., a bounded, Lipschitz-continuous objective function. I am studying ways to incorporate different forms of information into algorithms to improve their efficiency or strengthen their statistical guarantees.
Benchmarking SO Algorithms
Compared to deterministic optimization algorithms, SO algorithms present additional challenges when it comes to benchmarking. I am exploring ways to evaluate and compare the finite-time performance of SO algorithms. This effort has led to a major redesign of SimOpt – a growing testbed of SO problems and solvers.
R&S Guarantees
Ranking-and-selection procedures select from among a finite set of simulated alternatives and can provide either a frequentist or Bayesian statistical guarantee. I am examining the interplay between the design of R&S procedures and the guarantees they deliver. In particular, the choice of guarantee can dictate how a procedure collects and analyzes simulation data and the kinds of assurances provided to decision-makers.
Reusing Simulation Outputs
In settings in which running simulation replications is computationally expensive, finding ways to reuse the outputs from past replications is especially appealing. “Green simulation” is one such form of reusing simulation outputs to enhance the efficiency of SO algorithms. I am investigating how reusing simulation outputs can affect the statistical properties of SO algorithms, in at times unexpected ways.