The domain of software engineering involves applying engineering principles and methods to create software systems, focusing on producing high-quality software efficiently and cost-effectively. The implementation of software testing is necessary in order to actualize these commitments. The comparative analysis between manual testing and automated testing reveals that the latter, employing automatically generated test cases, surpasses the former in terms of efficiency and effectiveness. As the size of test suites increases, the execution of all test cases within the suite becomes increasingly laborious and time-consuming, rendering it impractical to execute every single test case. Several different methodologies have been suggested to tackle the challenge of minimizing test suites. However, the no-free-lunch (NFL) theorem and its NP-completeness reveal that no single methodology can ensure an optimal-sized collection of test suites. To address redundant test cases within test suites, the authors introduce SSTSM, a novel algorithm for test suite minimization. This algorithm, outlined in the study, is a black box technique leveraging the fault-exposing potential of test cases alongside a nature-inspired metaheuristic called Squirrel Search (SSTSM). Additionally, three other reduction techniques are explored. The study compares the performance of Particle Swarm Optimization (PSO), Bat Search Algorithm (BA or BAT), and Genetic Algorithm (GA) against the proposed approach. The comparison is conducted using the same input dataset, and the evaluation criteria include the size of the reduced set (RS), reduction factor (RF), and execution cost. The experimental findings indicate that SSTSM outperforms PSO, BA or BAT, and GA in all comparison projects.