Modeling Inhomogeneous Markov Process in Smart Component Methodology

Abstract

Dynamic reliability analysis methods can account for the interactions present between physical process, control systems’ hardware and software,  and human actions in reliability analysis of dynamic systems. They provide scope for high fidelity in reliability modeling. A smart component-based methodology is developed recently to serve as a generic method for dynamic reliability analysis while solving existing challenges of dynamic reliability analysis, such as state space explosion, easy system structuring. The method is based on object-oriented representation of the dynamic systems’ structure and interactions, and Monte Carlo simulation for reliability simulation. The method can account for the dynamics generated from the above-mentioned interactions. In addition to that, modeling and demonstration of the aging and wear in processes through time-dependent reliability parameters is needed. In this paper, we demonstrate time-dependent reliability parameters in the framework of smart component methodology (SCM) using inhomogeneous Markov process. The generality of SCM for inclusion of Weibull distributed failure rates and various repair schemes is validated with example systems and the reliability results from the literature, and numerical and fault tree methods. An acceleration scheme is also implemented within the SCM framework and results are found to be consistent.