Modeling Service Composition Using Priced Probabilistic Process Algebra

Process algebras are a set of formal languages that are suitable to describe concurrent and composite systems including services composition. Nowadays, though process algebras have been effectively exploited for modeling and verifying functional aspects of services composition, modeling non-functional aspects of services composition with process algebras has been ignored. Cost is one of the most important non-functional aspects of services composition, but existing process algebras have not cost modeling capability. We propose Priced Probabilistic Process Algebra by extending existing probabilistic process algebra with cost. We model services composition with the algebra and demonstrate that services composition can be controlled by Markov decision process with target function of optimal cost.