Data Warehouse requirements engineering has been extensively looked at from the ENDS perspective of the Business Motivation Model, in terms of goals the system to-be wants to achieve. The authors propose that the MEANS perspective of this Model can drive the requirements engineering process. MEANS are organized into business policies and ‘policy enforcement rules'. Starting from policies expressed in a higher order logic, the authors propose an approach to formulate policy enforcement rules. That subset of the set of formulated policy enforcement rules which is most appropriate for the business is to be selected. For this, the information relevant to the rules is to be kept in the Data Warehouse. The authors technique picks up the components of the policy enforcement rule to elicit the information that has a bearing on its selection. The elicited information is represented as an ER diagram. The authors rely on existing methodologies to convert an ER form into star schemas. The authors use the medical domain to illustrate our methodology.
