Design and Implementation of Enterprise Spatial Data Warehouse

Traditional enterprise data warehouse, an integral part of decision support systems (DSS), provides a variety of granularity data to satisfy requirements of different users. It is estimated that about 80% of the information stored in data warehouse is geo-spatial related. However, traditional data warehouse cannot efficiently process spatial data. With the increasing amount of spatial data stored in spatial databases, how to utilize these spatial data is becoming a critical issue of data warehouse. In this paper, we focus on designing and implementing the enterprise spatial data warehouse for spatial decision-making. We propose three methods of building enterprise spatial data warehouse, and extend traditional enterprise data warehouse model into spatial multidimensional data model, which consists of both spatial and non-spatial dimensions and/or measures. Spatial index with the pre-aggregated results is built on spatial dimension and use the groupings of the index to define a hierarchy. Methods for computation of spatial measure are studied. Extended enterprise spatial data warehouse can accelerate spatial OLAP operations and support the spatial data analysis for decision-making support purposed.