The approach to multifactorial safety assessment of critical infrastructure based on application of dynamic Bayesian belief network (DBBN) is suggested in the paper. Application of DBBBN allows taking into consideration the complicated dynamic of accident propagation, changes of critical infrastructure states and systems mutual influence as well. Also DBBN allows considering the set of external and internal factors which determine CI safety level. Thus, DBBN is suggested as a basis for multifactorial safety assessment of CI. The application of the approach proposed in this paper helps to increase the accuracy of CI safety indexes and ground the measures to decrease the risks of CI –related technological catastrophes. Beside, the application of this approach increases the scope of probabilistic safety assessment (PSA) due to accounting of subjective probabilities taken from experts. As an example of DBBN application Fukushima NPP accident is considered.
critical infrastructure, multifactorial safety assessment, dynamic Bayesian belief network