3 Conceptual Models for Spatio-temporal Applications

Improved support for modeling information systems involving time-varying, georeferenced information, termed spatio-temporal information, has been a longterm user requirement in a variety of areas, such as cadastral systems that capture the histories of landparcels, routing systems computing possible routes of vehicles, and weather forecasting systems. This chapter concerns the conceptual database design phase for such spatio-temporal information systems and presents two models, namely the spatio-temporal Entity Relationship (ER) Model and the Extended spatio-temporal Unified Modeling Language (UML) as proposed in [33,34] and [26], respectively. The conceptual design phase focuses on expressing application requirements without the use of computer metaphors. The design should be understandable to the user and complete, so that it can be translated into the logical phase that follows without any further user input. Popular conceptual models include the ER model [6], IFO [2], OMT [30], and UML [17]. For conventional administrative systems, exemplified by the “supplier-supplies-parts” paradigm, the available modeling notations and techniques that support the conceptual and logical modeling phases are mature and adequate. However, this is not the case in non-standard systems managing spatio-temporal, multimedia, VLSI, image, and voice data. Rather, these lead to new and unmet requirements for modeling techniques. The basic rationale behind the work presented here is to introduce new modeling techniques, based on minimal extensions of existing models, developed to accommodate the peculiarities of the combined spatial and temporal information. The ER Model and the UML have been extended as prototypical examples for this purpose. First, we present the fundamental aspects of the spatio-temporal domain, covering concepts such as objects, properties, and relationships. Based on these, in the ER approach, we present a small set of constructs aimed at improving the ability to conveniently model spatio-temporal information at the conceptual level. These constructs may be included in a wide range of existing conceptual data models, improving their modeling capabilities without fundamentally changing the models. We incorporate the proposed modeling constructs into the ER model ([5]), resulting in the semantically richer Spatio-Temporal ER (STER) model [33].