An spanning cluster-avoiding process (SCA) is an Explosive percolation model based on classifying bonds between those that facilitate the creation of the spanning-cluster, and those that don't, and preferentially selecting those that don't. They are similar to Achlioptas processes (-edge processes). However, they don't require the candidate edges to be chosen at random between any pair of nodes, and instead the candidate edges can belong to a predetermined underlying network, common a hypercubic lattice. They are capable of showing discontinuous transitions, for certain choices of the number of candidate edges chosen per step
The most common spanning cluster-avoiding process (introduced here) starts by considering a finite hypercubic lattice in dimensions of size and unoccupied bonds. Then, inspired by the best-of-m model (see Tricritical Point in Explosive Percolation), the rule of the mode is as follows as follows:
Getting the Jump on Explosive Percolation
Avoiding a Spanning Cluster in Percolation Models
These models were introduced to clarify the order of the transition in explosive percolation processes in Euclidean lattices, which had been studied numerically before: Explosive Growth in Biased Dynamic Percolation on Two-Dimensional Regular Lattice Networks – Scaling behavior of explosive percolation on the square lattice.
Extensive numerical simulations and theoretical results have shown that the explosive transition in SCA model in the thermodynamic limit, can be either discontinuous or continuous depending on dimension the number of potential bonds (see here, here, and Two Types of Discontinuous Percolation Transitions in Cluster Merging Processes).