Abstract
In general, regression testing plays an important role in software testing which can offer help to minimize test suite size or rearrange the test cases. In software regression testing, the code coverage of target program is a crucial factor while we perform test case reduction and prioritization. Modified Condition/Decision Coverage (MC/DC) is one of the most strict and high-accuracy criterion in code coverage. In previous research, Hayhurst et al proposed a method to implement the MC/DC criterion. However, there might be some flaws when using Hayhurst’s method to select test cases. In this study, we will give some examples to indicate the importance of some test cases which are considered as redundant test cases. Then we describe how to enhance Hayhurst’s method with a greedy-based algorithm. Finally, we perform an experiment with four target programs from three different sites to verify our proposed method, and further prove the construct, internal and external validity of our experiment.