Markov models (MM) are widely used in dependability assessment of complex safety-critical systems, such as NPP I&C system. The main computational difficulties in case of using MM are model size and stiffness, which pose a problems in its construction, storage and solution. Selection of the solution approach and method, based on analysis of such MM features as stiffness and complexity, increases the assessment accuracy. Result of such analysis helps in making decision between direct and indirect research techniques and set of software packages (SP) to provide the high accurate assessment results.
This paper presents the case study for safety assessment of NPP I&Cs. This is a two-channel FPGA-based Reactor Trip System with three parallel tracks on voting logic “2-out-of-3” in each channel. Several solution techniques and SPs were used to analyze and describe the ways in which main modelling risks can be avoid. Analysis of case study results using different SP allows to formulate few application problems: importance of usability-oriented SP selection in case of solving complex MM; achieving an accurate result for stiff MM; support the results verification to ensure the needed level of confidence.