Reducing MBT test cases discard with distance function at TJPB

Abstract

The main gain of Model-based Testing (MBT) is the generation of robust test suites with minimal effort. However, as the software evolves along with its models, test suites often need to be generated again, resulting on test case discards and the loss of historical data. To address this issue, prior research has proposed a strategy to reduce the discard of MBT test cases, specifically those derived from CLARET use case models, employing distance functions to identify test cases that have undergone minimal impact due to changes and can be easily updated instead of being discarded. In this work, we report a case study in which we applied this approach to an industrial system developed at the TJPB (Court of Justice of the State of Paraíba). At least 7% of the test cases that would be discarded could be reused, which reduces information losses and improved the traceability of the project issues.