Integrating Probability with Time and Shared-Variable Concurrency

Complex software systems typically involve features like time, concurrency and probability, where probabilistic computations play an increasing role. It is challenging to formalize languages comprising all these features. In this paper, we integrate probability, time and concurrency in one single model, where the concurrency feature is modelled using shared-variable based communication. The probability feature is represented by a probabilistic nondeterministic choice, probabilistic guarded choice and a probabilistic version of parallel composition. We formalize an operational semantics for such an integration. Based on this model we define a bisimulation relation, from which an observational equivalence between probabilistic programs is investigated and a collection of algebraic laws are explored. We also implement a prototype of the operational semantics to animate the execution of probabilistic programs