Model-Based Strategies for Reducing the Complexity of Statistically Generated Test Suites

The main purpose of this paper is to show how model-based techniques are used to efficiently control the generation of less complex test suites. By directed adjusting specific probability values in the usage profile of a Markov chain usage model it is relatively easy to generate abstract test suites for different user classes and test purposes in an automated approach. A stepwise refinement process for hierarchical Markov chain usage models and choosing appropriate test generation, respectively selection strategies can reduce the complexity of the resulting test suite significantly. By using proper tools, like the TestUS Testplayer even less experienced test engineers will be able to efficiently generate abstract test cases and to graphically assess quality characteristics of different test suites.