An automated technique for risk-based test case generation and prioritization

In practice, available testing budgets limit the number of test cases that can be executed. Thus, a representative subset of all possible test cases must be chosen to guarantee adequate coverage of a test object. In risk-based testing, the probability of a fault and the damage that this fault can cause when leading to a failure is considered for test case prioritization. Existing approaches for risk-based testing provide guidelines for deriving test cases. However, those guidelines lack the level of detail and precision needed for automation. In this contribution, we introduce the risk-based testing technique RiteDAP, which automatically generates system test cases from activity diagrams and prioritizes those test cases based on risk. The results of applying the technique to a practical example are presented and the ability of different prioritization strategies to uncover faults is evaluated.