The thesis addresses the spatial aspect of locating emergency shelters in urban areas. Its aim is to develop a simple method that serves urban planners as a tool for analyzing the current state and supporting data-driven decisions about suitable locations for emergency shelters. The method enables an assessment of population coverage and promotes fair and spatially optimized planning of shelter facilities. It takes into account spatial limitations, existing infrastructure, and identified needs for shelter capacities based on population density. The proposed method relies on the use of open spatial data, which, in modern planning practices, supports decision-making by offering insights into spatial patterns of vulnerability. Cities act as generators of heterogeneous spatial data, which can be used as a rich informational foundation for spatial risk analyses. Analytical technologies open new possibilities for informed planning and placement of critical and protective infrastructure in urban environments. Unlike traditional approaches, data-driven methods and artificial intelligence techniques can simultaneously integrate and process large volumes of data from multiple sources. The method makes use of publicly available spatial data layers from various sources and is designed for implementation within the open-source geographic information system QGIS. It includes the selection and integration of spatial layers and their analysis within the QGIS graphical modeler, resulting in georeferenced location proposals for new shelter facilities. The method is transferable and applicable to urban centers of the second level, as defined by the Spatial Development Strategy of Slovenia (SPRS). It supports spatial planning efforts aimed at increasing urban resilience to natural or other emergencies and improving the city's response in crisis situations.
|
