Strategies for Prioritizing Test Cases Generated Through Model-Based Testing Approaches

Software testing is expensive and time consuming, especially for complex software. In order to deal with the cost of testing, researchers develop Model-Based Testing (MBT). In MBT, test cases are generated automatically and a drawback is a huge generated test suite. Our research aims at studying the Test Case Prioritization problem in MBT context. So far, we already evaluated the influence of the model structure and the characteristics of the test cases that fail. Results suggest that the former does not affect significantly the performance of techniques, however, the latter indeed represents a major impact. Therefore, a worthy information in this context might be an expert who knows the crucial parts of the software, thus we propose the first version of a prioritization technique that considers hints from the expert and the distance notion in order to prioritize test cases. Evaluation and tuning of the technique are ongoing, but preliminary evaluation reveals promising results.