Document Type



Doctor of Philosophy


Industrial Engineering

First Adviser

Thiele, Aurelie C.

Other advisers/committee members

Thiele, Aurelie C.; Defourny, Boris; Duzgun, Ruken; Paxton, Stuart; Zuluaga, Luis F.


This dissertation presents innovative applications of robust optimization for long-term decision-making. The first chapter focuses on Value Based Insurance Design (VBID) in the context of heart disease treatment. The high-level aim of VBID is to provide incentives for patients to better align their behavior with the system-level optimum of medication adherence and early (as opposed to late and more expensive) treatment. Our goal is to investigate a VBID approach with different cost-sharing parameters for low risk and high risk patients, in order to achieve a trade off for current and future costs for patients while improving patients’ life quality. The second chapter analyzes via simulation two mathematical modeling frameworks that reflect different managerial attitudes toward upside risk in the context of R&D portfolio selection. The manager seeks to allocate a development budget between low-risk, low-reward projects, called incremental projects, and high-risk, high-reward projects, called innovational projects. We study the differences in strategy and portfolio’s risk profile that arise between a risk-aware manager, who takes upside risk because he has to for the long-term competitive advantage of his company, and a risk-seeking manager, who will take as big a bet as allowed by the model. The third chapter studies hospitals’ optimal strategies of building community health program portfolio in order to achieve the maximum potential benefits under a worst case benefit tolerance level. Our model incorporates the fact that hospitals might have tolerances for upside and downside deviation and thus different uncertainty budgets for upside risk and downside risk and analyzes how key parameters influence the optimal portfolio and implement our approach in a numerical example with promising and insightful results.