Authors:
Lars Burmester
1
and
Matthias Goeken
2
Affiliations:
1
Institute of Management Information Systems, Philipps-University of Marburg, Germany
;
2
HfB – Business School of Finance and Management, Germany
Keyword(s):
Data Warehouse, Multidimensional Modelling, Process Model, Modelling Framework.
Related
Ontology
Subjects/Areas/Topics:
Data Warehouses and OLAP
;
Databases and Information Systems Integration
;
Enterprise Information Systems
;
Information Engineering Methodologies
;
Information Systems Analysis and Specification
;
Methodologies, Processes and Platforms
;
Model-Driven Software Development
;
Modeling Concepts and Information Integration Tools
;
Modeling Formalisms, Languages and Notations
;
Requirements Analysis And Management
;
Software Engineering
;
Systems Engineering
Abstract:
The paper describes a method for data warehouse development. One critical success factor of data warehouse development is determining information requirements. Hence, the method focuses on gathering of requirements and information needs of the users first. An extended data warehouse architecture and a technique for decomposition of the system serve as a developing framework. On the one hand this framework is used to define releases (builds) of the system, which is indispensable for an incremental development process. On the other hand it defines intermediate and final work products (artifacts) that are produced and used during further development stages. Starting with information requirements elicitation, each increment is realized through a series of data models which successively are transformed from conceptual to logical level. These logical data models are then used for implementation as well as for the modelling of ETL processes.