A New Multidimensional Design Method Based on Meta Model Assistance
Nazih Selmoune, Zaïa Alimazighi · International Journal of Knowledge-Based Organizations · 2016
Data warehouses have become over the last two decades an essential element in the area of economic intelligence. Although the basic concepts related to this technology are now more standardized (multidimensional model, OLAP technology...), the design method issue remains in discussion, since no proposed method does seem attractive enough to be used rigorously in the real case studies. Indeed, most of the 'Data Warehouse projects' simply follow some design guidelines from the most informal methods. In this paper the authors propose an approach based on the use of the sources metadata to provide help in the multidimensional design. Their approach falls into the category of supply-driven methods, and attempts to cover the various stages of the lifecycle of data warehouses design. The proposed method is associated with a MultiDimensional Design Assistance environment (M2DA) to support the automatic and semi-automatic steps, providing the expected forms of assistance, and managing the interaction with the designer.