User-Centric Spatial DataWarehousing: A Survey of Requirements & Approaches

Data warehouses have traditionally been used to analyse large multidimensional datasets and provide enterprise decision support. With an increased availability of spatial data in recent years, several new strategies have been proposed to enable their integration into data warehouses and OLAP systems and perform complex analysis on them. One strategy to achieve this is to integrate existing Geographic Information Systems (GIS) (and their geo-spatial analysis tools) with OLAP and relational DBMSs. Another strategy has been to leverage existing spatial capabilities of databases to perform spatial OLAP analysis. In either scenario, several new challenges emerge related to the handling of spatial data in a multidimensional context and user-friendly data analysis. In this article, we review existing modelling strategies for spatial data warehousing at all three levels: conceptual, logical and implementation. We gather the most essential requirements for handling spatial data in data warehouses and use insights from spatial databases and GIS systems to design a “metaframework” that would enable a user-centric modelling of spatial data warehouses. Our strategy is to keep the user as the focal point in the analysis process and to achieve a clear abstraction of the data for all stakeholders in the system. This lays the foundation for a conceptual model based on multidimensional abstract data types (ADTs) and operations to support complex spatial data in data warehouses.