Model Checking Functional and Performability Properties of Stochastic Fluid Models

Model checking of stochastic processes has been introduced to verify functional as well as performability properties of formally specified systems. In this paper, model checking for a wide class of stochastic fluid models (SFMs) is considered. We present a branching time temporal logic, which is similar to the continuous stochastic logic (CSL), for expressing real-time probabilistic properties of SFMs. The model checking problem for this logic can be tackled through repetitive application of transient and steady state analysis of a modified version of the SFM under study. The complexity of the analysis techniques developed for SFMs depends on the size of the state space. We present techniques that allow to reduce the state space in the solution of model checking problems. A case study illustrates the logic and the model checking procedure.