A particle swarm optimization–based heuristic to optimize the configuration of artificial barriers for the mitigation of lava flow risk

Lava flows are recurring and widespread hazards that affect areas around active volcanoes, having the potential to cause significant social and economic loss. The ongoing demographic congestion around volcanoes increases the potential risk and leads to a growing demand for faster and more accurate systems to safeguard the population. The main mitigation action for slowing down and possibly diverting lava flows is the building of artificial barriers, that can limit their destructive effects and reduce losses. Here we present a Particle Swarm Optimization algorithm for the configuration of artificial barriers, in terms of location and geometric features. The goal is to minimize the lava flow impact based on the spatial distribution of exposed elements, using the physics-based MAGFLOW model to run the lava flow scenarios for each barrier configuration. Our algorithm has been tested on Etna (Italy), showing how it can effectively safeguard the threatened areas, diverting lava away from them.