Research
My research develops methods that help decision makers learn, adapt, and act when data are uncertain, incomplete, costly, or evolving over time.
I study how AI and statistical learning can support decision-making in business and service systems. My methodological foundation includes Bayesian optimization, adaptive experimental design, uncertainty quantification, machine learning, mathematical modeling, and simulation. I am especially interested in how these tools can be translated into decision support systems that are useful for managerial and organizational contexts.
Many important decisions are sequential: a decision is made, outcomes are observed, and the next decision should adapt. My work studies methods that learn from feedback and update decisions under uncertainty.
Predictions are most useful when decision makers understand how much uncertainty remains and how that uncertainty affects action. I work on modeling frameworks that quantify uncertainty and turn it into decision-relevant insight.
Simulation enables decision makers to test assumptions, examine scenarios, and evaluate robustness before implementing decisions in the real world. This theme connects my teaching and research in business analytics.
When measurement is expensive or limited, the design of data collection becomes a decision problem. My prior work develops optimal and adaptive design strategies for learning efficiently from dynamic data.
Collaboration interests
I welcome academic and applied collaborations involving adaptive decision-making, simulation-based analytics, customer and market response, supply chain uncertainty, predictive modeling, and decision support systems.