Unifying and incorporating functional and non functional requirements in datawarehouse conceptual design

Abstract

Data warehouse (DW) architectures, which are at the core of many new generation information systems, are designed to provide decision makers and professional analysts with intuitive and performing means to explore and get intelligent data from large heterogeneous databases. However, the data quality is not the only concern; requirements related for example to security, performance and operability are also vital. Most existing approaches for conducting DW projects deal with the functional and nonfunctional requirements in a complete separate way. The lack of a global approach that gather concurrently both aspects from the requirements analysis to the conceptual design causes that many DW projects fail to meet their goals in satisfying users' needs and expectations. This work is the first step towards the development of a single framework that encompass a unified approach that allows us to get design information for the future datawarehouse multidimensional model on the basis of both functional and non-functional requirements. We will present the methods we adopted to achieve this goal and the enhancement we have made by integrating three additional aspects in the requirements analysis workflow, namely organizational, decisional and business process views. We select the telecom industry for our case study to demonstrate the viability of our approach.