In this paper, we propose a Markov chain-based error propagation model to analyze the reliability of component-based software systems and take measures to make the critical components safer. Because it is difficult to test the whole component-based system, we apply an error propagation model to evaluate the reliability of the system with parameters obtained by preliminary data from existing components and integration testing from two connected components. The main parameters required in our Markov model are the error probability for each component, the error tolerance probability, and the error propagation probability for every two connected components. Our model is applied to compute the reliability of the system, find the most suspicious component during debugging, and protect the critical components. Finally, we simulate the process of these three applications using three different systems on MATLAB.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.