Research on Automatic Test Case Generation Technology Based on Finite State Machines and Function Block Diagrams

Software testing is an important means for ensuring software quality. The current main approach to software testing design involves manually writing test cases according to software requirements, which is inefficient and can hardly guarantee design quality. To deal with these challenges, this paper proposes a model-based testing design methodology. Following the modeling based on finite state machines and function block diagrams, this study explored a transformation algorithm that converts function block diagrams into finite state machines and an automatic test case generation algorithm. Additionally, a model-based testing design platform was established, which was subsequently verified and applied to software testing for train traction control systems. The verification results demonstrated the enhanced efficiency achieved by this platform, compared to manual writing. Furthermore, the test cases generated perfectly aligned with modified condition decision coverage (MC/DC), highlighting the quality assurance in test case generation.