Model-based System Reliability Analysis by using Monte Carlo Methods

With the increasement of integrity and complexity of aircraft systems, it is difficult to evaluate the impacts of the component failure modes on the systems. In this paper, a method for system reliability analysis of large and complex systems with multiple failure modes is proposed by combining the Monte Carlo (MC) method and model-based technology. The MATLAB/Simulink language is used to create the nominal model. And the model extension is obtained by injecting failure modes based on the nominal model. The extended system model is used to observe and analyze the behaviors and performances of the complex systems in the presence of different faults. Performance metrics are used to evaluate system effects caused by component failures. A procedure for system reliability evaluation based on the MC method is given, which can be applied to the reliability evaluation of a system. The method proposed is insensitive to the dimensionality of problems and can be used to the reliability evaluation of large and complex systems. The system response with fault injection can be analyzed to determine the effect of component failures or their combinations in system reliability analysis, which can avoid the dependence on the subjective judgment and experience of analysts. Furthermore, it can help improve the systems development. A case study is given to illustrate our proposed method.