Integrating Symmetries And Symbolic Enabling Test For Efficient Simulation Of SWNs

Abstract

Abstract-(Stochastic) Colored Petri Nets are a formalism widely used to specify and analyze distributed discrete-event systems. Determining the set of transition color instances enabled in a given marking is a basic task affecting analysis techniques based on state-space exploration, model-checking, and especially discreteevent simulation (the latter is an interesting alternative when analytical solutions are unfeasible due to statespace explosion). An algebraic approach to enabling test has been recently proposed as kernel for efficient state-space exploration of SWN, a Stochastic CPN flavor retaining expressive power. The approach is symbolic because it directly manipulates arc functions by means of rewriting rules. What makes interesting the SWN formalism is the ability of exploiting behavioral system's symmetries, thanks to structured color annotations. This paper illustrates in a semi-formal way how the symmetry-based technique typical of SWN can be profitably combined with the SWN symbolic enabling test, giving rise to a fully symbolic simulation kernel. An application example to a workflow model is presented.

Efficient simulation of Stochastic Well-Formed Nets through symmetry exploitation

“…Petri Nets simulation can take advantage of several optimizations: in fact the FEL update required after each state change, may be efficiently performed by taking advantage of the model structure (and of the structural conflict and causal connection relations between transitions); moreover, a considerable saving in time may derive from a properly designed enabling test for (colored) transition instances (taking advantage of the specific structure of the SWN arc functions) (Capra 2007;Liu and Heiner 2010). Last but not least, the peculiar characteristics of the SWN formalism which favors the automatic exploitation of the model behavioral symmetries, may lead to a substantial saving in the FEL management.…”