Reducing the Discard of MBT Test Cases using Distance Functions
Thomaz Diniz, Everton L. G. Alves, Anderson G.F. Silva, Wilkerson L. Andrade · 2019
Model-Based Testing (MBT) is used for generating test suites from system models. However, as software evolves, its models tend to be updated, which often leads to obsolete test cases that are discarded. Test case discard can be very costly since essential data, such as execution history, are lost. In this paper, we investigate the use of distance functions to help to reduce the discard of MBT tests. For that, we ran a series of empirical studies using artifacts from industrial systems, and we analyzed how ten distance functions can classify the impact of MBT-centred use case edits. Our results showed that distance functions are effective for identifying low impact edits that lead to test cases that can be updated with little effort. Moreover, we found the optimal configuration for each distance function. Finally, we ran a case study that showed that, by using distance functions, we could reduce the discard of test cases by 15%.