A challenging research issue in analyzing probabilistic real-time systems is to model the tasks composing the system and the resource provided to the system. In this paper we propose a solution based on a probabilistic componentbased model that abstracts the functional and non-functional requirements of real-time components. The obtained interfaces encode timing requirements and probability information of the component in a probabilistic version of the real-time calculus. Besides, it has been derived probabilistic scheduling and compositional guarantees to provide real-time analyses of probabilistic real-time systems. Finally, a test case illustrates the potentialities of the proposed model and its applicability in a large variety of problems within the probabilistic real-time scenario.