Abstract
Software testing is an essential part of the software development life cycle. However, due to limited time and resources, extensive testing of highly configurable software is impractical. In addition, extensive testing can lead to combinatorial explosion problems, where test cases grow exponentially with the increase in software input. Because of their effectiveness in detecting errors, many researchers are turning to a sampling strategy based on input interactions, called t-way testing, where t represents the strength of the interaction. It is known to be an NP-complete problem (ie, non-deterministic polynomial time). Due to the potentially large search space generated when dealing with large input values, the process of minimizing t-way test cases is challenging. So far, many designs and strategies utilizing the t-way methods have been proposed in the literature. Recently, researchers have advocated the use of emerging of fields to call search-based software engineering (SBSE), which utilizing the metaheuristic-based search algorithm. Therefore, in this research, a new hybrid metaheuristic-based is proposed based on merging the advantages of BA and PSO. Mainly, the design and implementation of a new hybrid metaheuristic-based t-way strategy is presented called , BAPSO, based on the two well-known algorithms, namely, Bat-inspired algorithm and Particle Swarm Optimization algorithm. BAPSO strategy can contribute to the field of software testing, as it achieved comparable performance in terms of minimizing the number of test cases used for test execution.