Care as needed: Tailored response in healthcare operations & humanitarian logistics

I investigate how a tailored operational response can help healthcare operations and humanitarian logistics cope with uncertainty and the heterogeneity of required services. The first essay focuses on hospital readmissions. Empirical predictions of time to readmission are integrated with optimization models to design post-discharge monitoring plans. This results in early detection of potential readmissions and sufficient system capacity allocation to administer these optimal treatment plans. I transform an intractable monitoring plan optimization with stochastic discharges and health state evolution into a weakly-coupled network flow model. Using this approach on two large inpatient datasets, I show that optimal readmission prediction and monitoring plans, tailored to an individual patient's risk profile, can mitigate 40%-70% of readmissions before they generate an emergency readmission. My second essay builds on the first and models how payment policies impact hospital readmissions. I consider fee-for-service and bundled payment plans in addition to Hospital Readmissions Reduction Program (HRRP) penalties. I analyze how groups in the healthcare provider system (hospitals, physicians, post-discharge follow-up care providers) would adjust their readmission avoidance procedures depending on the payment policy and if any misalignment of incentives between healthcare providers will occur. I also show how these payment policies impact patient throughput, waiting times, and overall access to care. The third essay studies vehicle supply chain design for humanitarian logistics with a dynamic hub location model. When considering the entire disaster management cycle, the results show that keeping a lean centralized hub configuration with the option for temporary hubs in mega disaster locations can reduce overall supply chain costs in the long term. I also show that it is possible to structure a supply chain to take advantage of earmarked funding operationally, which eases the complexity of using these funds.