Stochastic Decision Systems
About
We develop stochastic and data-driven decision models for systems operating under uncertainty. Our research focuses on sequential decision-making, stochastic optimization, dynamic programming, and probabilistic modeling to design policies that improve outcomes in complex healthcare environments. Applications include treatment planning, disease management, patient adherence, and personalized interventions.
Projects
We develop stochastic decision models to design personalized interventions that prevent disease progression before adverse health outcomes occur. Our research focuses on identifying when to collect patient information, when to intervene, and which intervention is most appropriate for each individual over time. Applications include determining how frequently cholesterol should be monitored to support timely medication decisions and designing six-month intervention policies that maximize quality-adjusted life years (QALYs).
We develop decision models for selecting which patients should receive interventions when healthcare resources are limited and patient information changes over time. Our research combines stochastic modeling, dynamic prediction, and optimization to prioritize individuals who are most likely to benefit from an intervention at a given decision point. The goal is to support fair, timely, and effective allocation of preventive and adherence-improving interventions in healthcare systems.