Model-Based & Product Line Testing

The domain of research "Model-based testing and Software Product Line testing" focuses on model-based testing of software in general on model-based Software Product Line testing.

Software Product Lines


Software Product Lines are an approach to improve reusability of software in a large number of products that share a common set of features.
Usually, more than a million different products can be derived from one Software Product Line. Miscellaneous approaches exist dedicated to Software Product Line testing. The challenging problem is that it is not feasible to test all products individually neither during nor after the development process. To approach this problem we currently focus on two major research fields:

1. Generating a representative set of products. Instead of testing each product individually only this set needs to be tested to reveal faults within a Software Product Line. We currently work on different strategies to find a minimal set, which is an NP-complete problem.

2. Extending Feature Models, which are frequently used to describe the commonalities and variabilities of a Software Product Line, for testing purposes. We aim to automatically generate test cases from the Feature Model.

Model-based Testing


Our main focus on model-based testing is to address the whole test design problem from choosing input values, generating sequences of operation calls to generating executable test cases. Therefore, we need a suitable model (e.g. statecharts) of the System Under Test (SUT) behaviour that include oracle information. To generate tests with oracles, the test generator must know enough about the expected behaviour of the SUT to be able to predict or check the output values. Furthermore, the model must be small in relation to the size of the SUT so that it is not to costly to produce, but it must be detailed enough to accurately describe the characteristics that we want to test.

Current Projects

feasiPLe is a project funded by the BMBF. The goal of the feasiPLe research project is to evaluate existing approaches for Software Product Line development and combining it with aspect-oriented and model-driven software development to a practical useful solution. We are responsible for one of eight workpackages focusing on the development of test methods for Software Product Lines.
Partners in this project are the Technische Universität Darmstadt, the Technische Universität Dresden, pure-systems and the SAP AG.

Completed Projects

In 2000 and 2001, the CTT (Configurable Classification Tree Transformer) project investigated model based generation of input data for classification tree based testing. This project was supported by the DaimlerChrysler AG, Research and Technology, Frankfurt.

Contact

Dr.-Ing. Harald Cichos

Dr.-Ing. Sebastian Oster

Metamodellierung & Modelltransformationen

Modellgetriebene Softwareentwicklung

Publications

Related Publications

Kontakt

Technische Universität Darmstadt

Institut für Datentechnik

Fachgebiet Echtzeitsysteme

Prof. Dr. rer. nat. Andy Schürr

Geb. S3|06 (3. Stock, Raum 313)

Merckstr. 25

64283 Darmstadt

+49 6151 16-6940
+49 6151 16-6942


A A A | Drucken | Impressum | Sitemap | Suche | Kontakt
Zum SeitenanfangZum Seitenanfang