ELICTING INFORMATION REQUIREMENTS FOR DATA WAREHOUSES

Abstract

Data Warehouse support in terms of Requirements Engineering models and techniques has been extensively provided for operational level of decision making. However, it is increasingly being recognized that there are other forms of decision making that exist in an organization. These are strategic in nature and ‘above’ operational decision making. This thesis addresses the issue of providing decision making support to both strategic and operational decision making in the same DW system. The solution starts by defining two broad categories of decisions for which decision support is needed, one for policy enforcement rule (PER) formulation decisions and the other for operational decisions. Both kinds of decisions are structured based on a generic decision meta-model developed here. The process starts by developing two Data Warehouses, one for policy enforcement rules and the other for operational decisions. In order to identify the needed information for supporting decision making, a set of generic techniques for eliciting information is proposed. This information is stored in the DW. Again, the structure of information for the two DWs is based on a generic information meta-model developed here. The two DWs are integrated upstream in the requirements engineering phase using an integration life cycle proposed in this thesis. It is argued that there is a need for integration following the problems of inconsistency and loss of business control that can occur. This is due to common information and differing refresh times between the two Data Warehouses. Further, three tools were developed to provide computer support for arriving at information for (a) PER, (b) operational decisions and (c) integrating information. This process was validated using AYUSH policies.