A Markov Chain and Likelihood-Based Model Approach for Automated Test Case Generation, Validation and Prioritization: Theory and Application

This research presents a model-based systems engineering approach for the creation of system behavioral models and the automation of test case generation. Although recent advances in systems engineering (SE) approaches for requirement generation have significantly improved system reliability and quality, system validation and verification still remain an arduous task. Manual generation of the test cases is a major barrier to system validation activities. With the growing complexities of modern-world systems, automatic generation and execution of the test case is desirable. In this research, Markov chains and Markov modulated processes are used to model the system behavior. Due to monetary and time constraints faced by the systems engineers and project managers in completing the projects, test case prioritization is another desirable attribute of a test suite. To address this issue we extend the presented model-based approach with a novel likelihood-based prioritization scheme for testing the most used system features and trajectories first. The effectiveness of the presented model-based approach is discussed and demonstrated in the context of two case studies, a simulation case study and a web application usage data case study. Beyond system validation activities, the presented approach can also be extended to the field of requirement gathering and regression testing.