FUTURE TRENDS AND CONCLUSIONS

In this chapter, we have discussed the requirements that an ideal conceptual multidimensional model should fulfill. These are suggested by general information system modeling principles and by the specific characteristics of OLAP applications. Starting from these requirements, we have presented a simple conceptual multidimensional data model, called MD, which can be used to describe the basic aspects of a business application in a way that is easy to understand and independent of the criteria for actual data organization in the various systems. With this model, we have tried to capture both the conceptual means used in business applications to describe information and the core of the various multidimensional data models proposed in the scientific literature or adopted by commercial systems. The model relies on two principal, agreed-upon concepts: the dimension and the data cube. A dimension represents a business perspective under which data analysis is to be performed and is organized in a hierarchy of levels. The levels of a dimension correspond to different ways of grouping dimension members. A data cube represents the factual data on which the analysis is focused and associates measures with coordinates, defined over a set of dimension levels. Using these concepts as a reference, we have summarized the general features that a multidimensional conceptual model should support and mentioned the various multidimensional models which have been proposed.

Clearly, much work remains to be done in this area. First of all, the use of conceptual data models still has difficulties to overcome in the applicative area, and the research community should clearly demonstrate the benefits to be gained by adopting them. Moreover, with such a proliferation of data models, a commonly accepted formalism is strongly advisable. This is fundamental for support of interoperability and standardization. Another problem that still needs to be solved is the definition of an effective and general methodology for the development of OLAP applications, an important aspect which has received little attention (Cabibbo & Torlone, 1998a; Golfarelli, Maio, & Rizzi, 1998; Kimball, 1996). This would also lead to the development of CASE tools which, in contrast to the present situation, were not strictly related to a specific OLAP system. Devising a common standard declarative language is also of high importance, and the use of a conceptual multidimensional model (independent of the underlying physical model) could give useful results in the area of logical optimization and caching rules (in order to exploit the possibility of reusing existing data cubes for the computation of new ones). Finally, there are a number of specific problems, such as the characterization of summarizability, for which a definitive solution has not yet been given.