An event-driven causal simulation applied to geoforecasting analysis

A qualitative-quantitative formalism centered around the process paradigm is presented. To model the causal interactions among the parameters of a system, an influence graph is used. The exploitation of the formalism consists of a causal simulation driven by dynamic occurrence of interesting events. The results of the simulation consists of the approximate evolution over time of the parameters. Because time is central in the reasoning used, evolution is explicitly represented over time of the parameters by piecewise linear functions. This approach has been applied to geoforecasting analysis to simulate plausible scenarios of natural evolution over time of deep radioactive waste-disposal sites