PhD position on Model-Based Testing for Evidence-Driven Black-Box Checking

at  Universiteit Twente

KEY TAKEAWAYS

• 
  Hours
  32 - 40 hr.
• 
  Salary indication
  Salary gross/monthly
  based on full-time
  € 2,872 - € 3,670
• 
  Deadline
  6 Oct 2024
  Nowadays so much new and complex software is being developed that there are by far not enough specialists to test this software properly. As a result, companies need more time/resources to develop software, and their products (which we all use) may contain critical bugs. In this project, we investigate methods for automatic software testing. In test automation, model-based testing is the state-of-the-art technique, where tests are generated fully automatically from a model (typically a state-transition system), and then executed on the software system.
  In the EVI project, you will develop test generation algorithms, that generate tests from two models: a model that specifies the expected behaviour of the system, and a model that describes the actual behaviour of the system. By combining such sources of evidence, tests are better equipped for detecting software bugs, including those often missed by other tools.
  The EVI project is a collaboration between Petra van den Bos from the University of Twente (UT) and Frits Vaandrager from the Radboud University (RU), with a PhD student at each university. Marielle Stoelinga is involved as promotor of the UT PhD student.
  The theme of the EVI project is evidence-driven black-box checking, where the fields of automata learning (RU) and model-based testing (UT) are combined. The RU PhD student will develop automata learning techniques for uncovering the actual behaviour of the system. You, as the UT PhD student, will use this as input for your test generation algorithms.
  While your main focus will be on test generation, you will also investigate how specification models can be acquired. Models can be constructed from requirements on the system, that e.g. have been formulated as part of the behaviour-driven development process of the system. Furthermore, such models may also incorporate known risks on the occurrence of bugs in the system.
  Besides designing test generation algorithms, you will also implement them in a tool. With the help of this tool, you will then apply your research to relevant case studies from the industry.

Please refer the Job description for details