Abstract
A hybrid technique based on Cellular Automata and Genetic Algorithms has been used for modeling lava flows on Mt Etna volcano (Italy). In particular, a Parallel Master-Slave Genetic Algorithm has been applied to evolve a two dimensional Cellular Automatamodel for "aa"-type lava flows, typical of the considered study area. Specifically, the 2001 Etnean Nicolosi case study has been considered for model calibration, while a subsequent validation phase has been performed by considering further events occurred in the same area. Results have confirmed both the goodness of the cellular automata simulation model and of the genetic algorithm employed for its calibration. An application concerning hazard evaluation has been subsequently performed on a vast area of the volcano, regarding different types of vulnerability assessments. Results have pointed out the validity of the methodology and its usefulness for Civil Defence purposes.