A micro-level representation is shown to be a useful feature of spatially explicit disaster simulation models. Executable simulation models for answering policy questions were designed and implemented for a flood management case and for a disease transmission case, which is underway. The flood simulation model differs from earlier natural disaster simulation models in several respects. It represents the geographical location and the economic strength of each household explicitly. It is also equipped with a graphical user interface, making it possible to interactively design policies and test the outcomes of imposing these. If policy options are compared, the simulation results can automatically be transformed into decision trees. The flood simulation model shows that a micro representation makes it possible to investigate the distributional effects of policy changes. Novel features of the disease transmission model include the use of (anonymized) data on real individuals, the large number of model objects representing close to nine million individuals, the explicit representation of persons and places, and the inclusion of important parts of the social structure. The disease transmission model shows that the incorporation of social structure allows for a more realistic representation of disease spread than models that assume homogenous mixing. It is possible to perform experiments of great policy relevance, such as investigating the initial growth of an epidemic on a real network. Together, the two executable micro models demonstrate the usefulness of such models for policy evaluation in the area of disaster management.
Submitted to Stockholm University in partial fulfilment of the requirements for the degree of Doctor of Philosophy