Eliciting and structuring business indicators in data warehouse requirements engineering

Traditionally, data warehouse requirements engineering is oriented towards determining the information contents of the warehouse to be. This has resulted in a de‐emphasis of the functional perspective of data warehouses. Consequently, it is difficult to specify functions needed for computing business indicators. Our approach aims to elicit needed business indicators from organizational decision makers. Thereafter, indicator hierarchies are built. Then we associate functions with business indicators of the hierarchy. These functions are visualized as use case diagrams. To do this, we extend these diagrams to allow for actor aggregation in addition to actor specialization. Further, we introduce the ‘estimated from’ relationship between use cases, in addition to the ‘extend’ and ‘include’ relationships of UML. We illustrate our proposals with an example.