Drawing the Line: Basin Boundaries in Safe Petri Nets

Attractors of network dynamics represent the long-term behaviours of the modelled system. Understanding the basin of an attrac-tor, comprising all those states from which the evolution will eventually lead into that attractor, is therefore crucial for understanding the response and differentiation capabilities of a dynamical system. Building on our previous results [2] allowing to find attractors via Petri net Un-foldings, we exploit further the unfolding technique for a backward exploration of the state space, starting from a known attractor, and show how all strong or weak basins of attractions can be explicitly computed.