Model-based testing

Taking test automation to the next level

Systematic testing is important to ensure software quality. Manual testing is expensive, error-prone and very time-consuming. Therefore test automation is a key consideration in improving test efficiency. Currently, however, only manually crafted test scripts are automatically executed. 

Model-based testing takes test automation to the next level. Models, increasingly used in model-driven product design, enable software tests to be automatically generated. Thousands of tests can be generated and automatically executed, and the outcome analysed for errors without any human intervention. 

Newly developed techniques extend model-based testing to real-time and data-intensive systems such as testing a wireless sensor network’s synchronisation protocol.  These model-based techniques significantly improve the quality, efficiency and coverage of software tests, making the entire test process faster, cheaper, less error-prone and easier to maintain.

Model-based testing

Method details


Method description

  1. Develop a model of the system’s required behaviour
  2. Develop a test infrastructure for automatic test execution
  3. Generate tests cases, including selected inputs and expected outputs, from the model using a model-based test tool
  4. Execute the tests automatically, and analyse the outcome
  5. After changes or modifications to the system, adapt the model and automatically generate the tests again

Research project

Partners

Industrial partners:

  • Axini
  • Chess
  • Philips Healthcare

Academic partners:

  • Radboud University Nijmegen
  • University of Twente

Contact

Business manager:

Frans Beenker

Telephone:

+31 (0)88-866 5420

Email:

frans.beenker@tno.nl