Title: A Mutation-Based Approach for Testing Atlas Model Transformation Specifications


Abstract:

Model transformations play a critical role in Model-Driven Software Engineering. Faults in model transformations will produce defective models, and eventually propagate to other artifacts. Existing literature on model transformation verification and validation has considered coverage based techniques. Mutation testing can effectively detect inadequacies in coverage based test suites. To support the mutation testing of model transformations, this study proposes a set of mutation operators for the Atlas Transformation Language (ATL) and an automated tool, MuATL, for mutant generation. The effectiveness of the proposed operators is evaluated using two model transformation programs, implemented in ATL. The first transforms Use Case Map scenario specifications to UML Activity Diagrams, and the second refactors UML use case models. The results show that the proposed operators successfully detect inadequacies in coverage based test suites.