Conceptual Models and Model-Based Business Metadata to Bridge the Gap between Data Warehouses and Organizations

This work has been finished in November 2007.

Data warehouse systems are used by decision makers for performance measurement and decision support. Measures such as the number of transactions per customer or the increase of sales during a promotion are used to recognize warning signs and to decide on future investments with regard to the strategic goals of the organization.

Currently, the main focus of the data warehouse research field is on database issues. The data warehouse’s interaction with the organization and the way it supports the organization’s strategic goals have not yet been considered in depth. Conceptual models that describe the data warehouse from various viewpoints, including an outside view of the data warehouse system, its environment and expected usage, are missing. Moreover, even though the data in the data warehouse by its very nature has to be closely related to the concerns of the organization, current data warehouses also lack sufficient business meta-data that would inform users about the organizational context and implications of what they are analyzing.

This thesis targets the relationship between the data warehouse and the structure, behavior, and goals of the organization.

In order to describe this relationship, a conceptual modeling language was developed. It consists of models for describing the interdependencies between data warehouses and business processes, including so-called active data warehouse solutions; a model for identifying business objects such as customers and products in the data warehouse data model, and for constructing data models that comply to the state models of such business objects; as well as a model of data warehouse usage, which includes modeling the users, user groups, and user skill levels, the intensity with which they use the data warehouse infrastructure, temporal issues such as the required time and urgency of data access, and indicators of the relative importance of data warehouse usage.

This thesis also introduces an approach to using models to enhance the way users access the data in the data warehouse. It presents model-based business metadata, which links enterprise models such as business process models or goal models to the data model of the data warehouse through the mechanism of model weaving. A prototype illustrating how models can be weaved and used for business metadata in a business intelligence tool has been developed as part of this thesis.

 

Download the paper