Efficient Test Case Prioritization in Software Testing Using DistilRoBERTa for Fault Detection Optimization

The very critical phase in SDLC is software testing, where application reliability, security, and efficiency are ensured. However, increasing complexity in software has made traditional test case prioritization (TCP) methods difficult, with regards to high execution time and computational overhead. The existing approaches such as Genetic Algorithms (GA) are highly computationally expensive, and the adaptability of test cases with new evolvement cannot really be integrated with the processes. This study proposes an artificial intelligence-based approach with DistilRoBERTa for test case prioritization to improve fault detection and optimized test execution. Unlike traditional methods, DistilRoBERTa uses deep learning to analyze semantic and historical defect data of the test cases to intelligently prioritize. The proposed method achieves 93% test case coverage (against 90% in GA), 90% execution efficiency (against 85%), and 96% reliability (against 95%) while significantly reducing computational overhead to 53% (against 70%). All these aspects, therefore, make the results much more scalable, efficient, and adaptable as compared to software testing. An edge over competitive heuristic-based TCP methods is that the proposed model offers faster execution coupled with minimal resource consumption—the perfect environment for extensive testing. Management of test cases proves to be one of the important tasks since there are a large number of test cases in software. This paper develops an automated Intelligent test case prioritization process. A centralized intellectual resource is established through the complete understanding of test cases, their interdependencies, requirement analysis, defect analysis, and processing of information for prioritization. The application of the resulting development would open new horizons to the evolution of intelligent testing.