Supporting decision making for spatiotemporal phenomena

Abstract

Multidimensional structures are used in OLAP technology to aggregate and format data with the goal of optimizing responses to users' queries. These structures have been widely used in applications that deal with discrete dimensions. In this paper we present a brief survey of the different multidimensional models and propose a model that supports both discrete and continuous dimensions with more concentration on the latter. We first define continuous fields. Then we present our model, which is based on discrete and continuous basic cubes. By applying spatial and temporal interpolation functions to a sample of data of the discrete basic cube a continuous basic cube is constructed. Hypercubes are built by applying aggregation functions to basic cubes. Two classes of operations associated with continuous fields are also defined.