OptiTest+: a fuzzy analytical hierarchy process (AHP)-based approach for optimizing performance test case selection in mobile applications

Background Performance testing is a critical aspect of mobile application development, ensuring responsiveness and reliability under diverse conditions. However, traditional test case selection methods, such as random and coverage-based approaches, often fail to optimize execution while maintaining critical test coverage. Machine learning (ML)-based techniques, though promising, introduce computational overhead and lack interpretability, limiting their effectiveness in real-world testing. To address these challenges, we propose OptiTest+, a Fuzzy Analytical Hierarchy Process (FAHP) based test case prioritization framework that enhances test suite optimization for mobile performance testing. Methods OptiTest+ integrates FAHP with Five-Level Ranking Thresholds, ranking test cases based on key performance indicators such as response time, throughput, latency, and network load, critical factors in ensuring mobile application efficiency. FAHP leverages expert-driven pairwise comparisons to prioritize high-impact test cases while eliminating redundant ones. The framework was validated across ten real-world mobile applications, including Uber, GitHub, and EasyPaisa. Both manual and automated FAHP calculations were performed to ensure the accuracy of the automated tool, with results cross-verified for consistency. Results Experimental results demonstrate that OptiTest+ reduced test suite size by 35% on average, while preserving 90% of critical performance test coverage. The approach substantially reduced test execution time by 40%, optimizing testing efficiency without compromising the detection of performance bottlenecks. A comparative evaluation against random, coverage-based, Non-dominated Sorting Genetic Algorithm II (NSGA-II), and Pareto-based selection techniques confirmed that FAHP offers a superior balance between test reduction, coverage retention, and computational efficiency. Conclusion By integrating structured FAHP based decision-making with automated test prioritization, OptiTest+ provides a scalable, interpretable, and efficient alternative to traditional and ML-based selection methods. The findings demonstrate that FAHP can effectively optimize test suite selection for mobile applications, ensuring faster and more resource-efficient performance testing. Future research will explore hybrid FAHP-Machine Learning models to enhance adaptability, automate decision-making, and further improve test selection efficiency for large-scale mobile applications.