Considering variability modeling in requirements engineering becomes increasingly relevant for developing, maintaining and evolving requirements specifications. This is particularly the case when dealing with
- large requirements specifications and heterogeneous stakeholder interests (i.e. when not all requirements need to be reviewed by all stakeholders, but are relevant to particular stakeholders only)
- and software product lines (i.e. when a family of related software products is developed, a significant commonality between products exists and the variability is systematically organized).
In such cases language support for an explicit specification of variability and tool support for an efficient generation of tailored requirements specifications is crucial. Dedicated concepts and tools can further raise the efficiency of creating, visualizing, browsing and evolving of such product line specifications.
In the SPREBA research project we are developing new concepts and techniques to support an efficient description, visualization and handling of variability in requirements. Our work builds on integrated modeling of requirements (see e.g. ADORA), compositional modeling of variability (i.e. aspect-oriented modeling) and automatically analyzing variability constraints (e.g. hierarchies or other constraints between variable features). Most existing languages and approaches still exhibit inherent weaknesses. Most notably, these include information scattering of requirements modeling that defines a variable feature over multiple diagrams, only little automation support when documenting variable features and only limited automation support when reasoning about variability constraints and feature impact. The novel ideas we have developed within the SPREBA project can significantly ease the creation of product-line requirements models and improve the understandability of variability and its impact on any potential products. Further, these concepts enable an efficient and intuitive derivation of products and they can guarantee that all created and derived products are correct by construction - to a certain extent.
For further reading please find slides of a recent introductory presentation here (PDF).
Stoiber, R., M. Glinz (2010). Feature Unweaving: Efficient Variability Extraction and Specification for Emerging Software Product Lines. Proceedings of IWSPM'10: Fourth International Workshop on Software Product Management. At RE'10. IEEE Computer Society. Held in Sydney, Australia. [pdf] [doi] [slides (pdf)]
Stoiber, R., M. Glinz (2010). Supporting Stepwise, Incremental Product Derivation in Product Line Requirements Engineering. Proceedings of VaMoS'10: Fourth International Workshop on Variability Modelling of Software-intensive Systems. SSE, University of Duisburg-Essen. Held in Linz, Austria. [pdf] [proceedings] [slides (pdf)]
Stoiber, R., M. Glinz (2009).
Modeling and Managing Tacit Product Line Requirements Knowledge.
Proceedings of MaRK'09: Second International Workshop on Managing Requirements Engineering Knowledge.
At RE'09. IEEE Computer Society. Held in Atlanta, USA.
Find an full list of SPREBA-related publications here.
Bachelor and Master Theses
Stefan Schadauer (Master's Thesis): Industrial Evaluation of the SPREBA Method to Software Product Line Requirements Engineering. [pdf]. Ongoing. In cooperation with JKU Linz and InsideAX GmbH in Linz.
Urs Zoller (Bachelor's Thesis): A Comparison of three different Concepts for Requirements Modeling of Software Product Lines - A Case Study-based Investigation. [pdf]. Completed. Written in German.
Michael Jehle (Master's Thesis): Feature Unweaving - Semi-Automated Aspect Extraction in Product Line Requirements Engineering. [pdf]. Completed.
Anil Kandrical (Master's Thesis): Graphical Weaving of Aspects in Product Line Requirements Engineering. [pdf]. Completed.
Find a complete list and the written thesis documents here.
- Prof. Dr. Martin Glinz (project leader)
- Reinhard Stoiber
SNF (Personen- und Projektförderung) and Universität Zürich.
SPREBA in the SNF project database [1, 2] and the UZH research database.