Test Suite Minimization Through PSO

Regression testing is a critical stage that ensures the software changes are not affecting existing functionality. Ideally, conducting regression tests after each software change is essential. However, the exhaustive repetition of full tests for every alteration proves time-consuming, expensive, and demands significant resources. Test suite minimization serves as a solution by eliminating redundant test cases. This research presents an advanced approach to eliminate redundant test cases effectively. The proposed method focuses on identifying a concise yet effective set of test cases that align with requirements and efficiently uncover faults. This streamlined approach not only reduces testing expenses but also contributes to a more efficient testing process. The study demonstrates its effectiveness in minimizing the test suite by employing the Particle Swarm Optimization (PSO) algorithm with varied inertia weight strategies. Results from simulations indicate the success of this approach in reducing the test suite size while maintaining or even improving code coverage. Comparative analysis against Genetic Algorithms (GA) and Artificial Bee Colony (ABC) methods underscores the superior performance of PSO, specifically with tailored inertia weight strategies. This research significantly contributes to optimizing software testing methodologies, ultimately enhancing fault detection efficiency in software development.