Application of the modified APELL methodology coupled to the use of Bayesian network for external emergency plans

Resumo

The issue of the risk of the activities on the facilities belonging to the nuclear fuel cycle has occupied a prominent position in discussions about the facilities' security, not only of their employees, but also to the population living in their surroundings. This work presents a case study based on the perception of the local population to verify the existence of a risk communication between a nuclear facility and the local population (structured questionnaire) in order to analyze their social perceptions with regard to the nuclear facility itself. The questionnaire was developed with a view to be included in communication programs for local emergencies, should they surpass the limits of their respective facilities, using a modified APELL (Awareness and Preparedness for Emergencies at Local Level) methodology, as the facility in question does not currently use questionnaires in its mobilization plans. The questionnaire with their answers was modeled using Bayesian Networks in order to identify the district improvements regarding the credibility of the facility installation before the population. The calculations were performed using the Netica^TM program and their results have shown that this is an appropriate approach as it improves the reliability of the questionnaire automatically generating predictions or decisions.