Probability does not exist: some considerations on seismic risk mitigation

The aim of this paper is to develop some considerations on seismic risk mitigation, with particular reference to: 1) the role of deterministic and probabilistic methods and the importance of their integration in hazard assessment; 2) the problem of the comparison between alternative probabilistic models, stressing the importance of paying more attention to the believability of the models used than to the validation of the results. A possible procedure to check the believability of the model is suggested; 3) the decision process when dealing with events characterized by low probability of occurrence and by catastrophic consequences. It is important to stress the need for a clear distinction among the three components of the process: the modelling (risk evaluation with the related uncertainties), the decision procedure (comparison of risk levels, uncertainties, aims, costs, and benefits connected with possible alternative decisions), and finally the decision aimed at risk mitigation; 4) clear explanation of the decision procedure, with the modelling containing parameters (or quantities) that can support the decision. They are referred to here as decision parameters. The risk mitigation procedure should imply that all steps of the decision process must be clearly explained to all persons involved, including the population.