Spatial OLAP modeling: an overview base on spatial objects changing over time

Enabling the decision making process to support spatial queries is not a trivial task. This task becomes even harder when using geographic information systems (GIS) with data warehouse (DW) because these two technologies are in general used separately. In general, a GIS solution handles spatial data without considering time constrains or without requiring the analyses of geometric shapes evolving over time. Temporal maps further raise difficulties on indexing issues and on the associated query mechanisms. On the other hand, a typical DW operates with non-spatial data for different time periods. It also does not support spatial data types, such as point, lines, and polygons. In this paper, we propose a multidimensional spatiotemporal data model to enable spatial analysis, in a context of evolving specifications. The proposed data model addresses the problem of spatial and temporal data integration by providing information to facilitate semantic interoperability and data analysis in a spatial DW that uniformly handles all types of data. Using a practical example in the field of land parcels, we evaluate the implementation of the model.