Safety and security are two of the most important requirements of the nuclear industry. In the event of a potential problem, the consequences can have serious implications for the public and the environment. Measures should be taken against various hazards and threats by analyzing possible realistic scenarios. Therefore, probabilistic risk assessment is one of the necessary technologies to achieving safe and secure nuclear facilities. In the study, a limited scope probabilistic risk assessment was made for a possible terrorist attack against a generic small modular reactor (SMR). A possible attack threat was selected to develop scenarios by following a probabilistic risk assessment approach.
In the scenarios created, terrorists have to pass all physical barriers that security guards protect. Thus, the decisions and actions of the security guards directly affect the result of the attack. To analyze these events, a human reliability assessment (HRA) was employed. In the first study, each security guard’s decision-making process was analyzed using the Standardized Plant Analysis Risk Human Reliability Assessment (SPAR-H) method. The purpose of its use in this study is to verify the SPAR-H method’s applicability for security applications.
In this paper, we give the likelihoods of each security guard making a decision and taking action to prevent terrorists from passing obtained using the SPAR-H method. Besides, event tree and fault tree analyses were performed using the SAPHIRE PRA software.
Finally, since the current HRA methods were designed for control room operators, we introduce a new model-based HRA methodology applicable for security guards to be used in physical security PRAs.