Deriving Symbolic Ordinary Differential Equations From Stochastic Symmetric Nets Without Unfolding

This paper concerns the quantitative evaluation of Stochastic Symmetric Nets (SSN) by means of a fluid approximation technique particularly suited to analyse systems with a huge state space. In particular a new efficient approach is proposed to derive the deterministic process approximating the original stochastic process through a system of Ordinary Differential Equations (ODE). The intrinsic symmetry of SSN models is exploited to significantly reduce the size of the ODE system while a symbolic calculus operating on the SSN arc functions is employed to derive such system efficiently, avoiding the complete unfolding of the SSN model into a Stochastic Petri Net (SPN).