A practitioner’s guide to integrating model-based testing in a project’s tool landscape

• Bernhard Peischl
• Rudolf Ramler
• Vitalina Turlo
• Stefan Mohaci
• Valery Safronau
• Tobias Walter

Software testing is a knowledge-intensive task that is driven by the information flow generated from the creativity, motivation and cooperation of the involved people. Repetitive mechanical activities, manual steps in automated processes, an unaligned infrastructure and disruptive workflows create painful blockers for this information flow. As some test-related tools have their own proprietary data-storage approach, the resulting isolation makes sharing information increasingly difficult, and tools that are not aligned properly may lead to unneeded complexity and manual test-related activities [Hüttermann 2011]. Hence, the creation of test cases alongside the corresponding test data is increasingly supported by model-based testing (MBT) tools. MBT promises (1) the detection of faults at an early stage in the development cycle and (2) reduces the maintenance effort of test scripts as the test cases are deduced from the model. Mature processes [van Ewijk 2011] and a smooth integration of the underlying tools [Turlo & Safronau 2011] allow test engineers to get rid of mechanical (testrelated) activities and focus their time on the development of models for the generation of test cases.