Disease surveillance is vital for monitoring outbreaks and designing timely public health interventions. However, especially in developing contexts, disease surveillance efforts are constrained by challenges of data scarcity. In this work, we discuss the deployment of system dynamics simulation models to aid in local disease surveillance programs in the Philippines. More specifically, we propose that (a) available time series records of disease incidence can be used to initialize simulation models with high accuracy and interpretability, and (b) virtual experiments can be used to test various what-if scenarios in designing potential interventions. Experiments with three years of data on dengue fever in the Western Visayas region illustrate our proposed framework as deployed on the FASSSTER platform. We conclude by outlining challenges and potential directions for future work.