Formal Specification and Probabilistic Verification of SysML Activity Diagrams

Abstract

Model-driven engineering refers to a range of engineering approaches that uses models throughout systems and software development life cycle. Towards sustaining the success in practice of model-driven engineering, we present a probabilistic verification framework supporting the analysis of SysML activity diagrams against a set of quantitative and qualitative requirements. To this end, we propose an algorithm that maps SysML activity diagrams into probabilistic models, specifically Markov decision processes, expressed in the probabilistic symbolic model-checker (PRISM) language. The generated model can be verified against a set of properties expressed in the probabilistic computation tree logic. In order to automate our approach, we developed a prototype tool that interfaces both a modeling environment and the model-checker PRISM. In order to illustrate the usability and benefit of our approach, we investigate its scalability and present a case study.