System reliability is an important qualitative attribute of systems.System reliability analysis based on models may find problems and solve them at the early phase of development.UML profile for MARTE is an extension of UML in the domain of real-time and embedded systems.We presented a method for reliability prediction of MARTE models.The MARTE model considerd by the method includes a use case diagram,a deployment diagram and a set of activity dia-grams.A MARTE model is transformed into a network of Markov decision process,which is then analyzed by the model checking tool PRISM.We obtained the estimation of system reliability by analyzing the resulting model.A case study demonstrates that,by analyzing models with various resource reliability,our method can further reveal the impact of the reliability of each resource on the system.