Dynamic resource allocation in anticipation of uncertain demand is a common problem faced in society today, e.g., public transportation and health infrastructure must be allocated to maximize the utility of the beneficiaries. In this talk, I will highlight some general decision-theoretic frameworks for modeling such problems and then focus on one specific case study. I will discuss the development of the ADVISER framework, which is a data-driven combinatorial optimization framework for optimizing the allocation of health interventions. This project, developed in collaboration with HelpMum, a non-profit agency based out of Nigeria, optimizes the distribution of heterogeneous interventions to increase vaccination uptake. The project was recently awarded the best paper award at IJCAI 2022.