Combinatorial testing is a black-box testing technique, which detects failures in the system under test. It reduces the number of test cases compared to exhaustive testing by applying combination strategies. Many generation algorithms and combinatorial testing tools have already been developed.

Recent studies have focused on the integration of combinatorial testing into automated testing frameworks with the goal to contribute for its widespread adaptation. By the development of those frameworks the research has tended to focus on the integration and automation of all steps of the combinatorial process and the overall extensibility of the developed frameworks. Rather less attention has been paid to the area of user interaction, which is required for the modeling of the test inputs and the development of test scripts. An easy and intuitive way for modeling of the parameter space and the combinatorial tests for the system under test could have a significant impact over the usability of a framework.

This thesis introduces an approach for optimized combinatorial input and test modeling in Java. It presents some of the limitations and problems of the modeling process and proposes solutions for them. The focus is on the development of a modeling interface and the adaptation of the testing process to the specifics of the combinatorial testing. The identified ideas are implemented as proof of concept in the combinatorial testing framework coffee4j.

Info

• • • Language: German, English
      • Supervisor: Konrad Fögen (E-Mail)
      • coffee4j on Github
      • ROBUSTA – Combinatorial Robustness Testing