Despite the high reliability of safety systems of nuclear power plants (NPP), human actions still play an important role in NPP safety. Evaluation of human reliability is therefore important for a full-scope probabilistic safety assessment (PSA) and risk analysis.
In PSA, which includes modelling of interaction of safety systems during an accident sequence, human errors are modelled together with hardware failures. Methods for the modelling of human errors and evaluation of probabilities of such errors are different compared with methods used for modelling and probability estimation of equipment failures.
The paper describes the modelling of actions that were to be performed by the Ignalina NPP operators during an accident sequence. Such modelling was applied for the PSA of the Ignalina NPP. A combination of ASEP (Accident Sequence Evaluating Procedure) and THERP (Technique for Human Error Rate Prediction) was applied. Such methodology allows evaluating the operators’ error probability at different phases of action (identification, decision-making and implementation) and enables to properly account for different factors that impact human performance (interface, alarm, indications, procedures, training, stress, time, etc.).
The human reliability analysis (HRA) presented in the paper allowed to refine the Ignalina NPP PSA model. Application of such methodology is possible in areas where operators play an important role in ensuring safety, e. g., in a new NPP or in the present and future industry of oil, gas, electricity, and transport.

Keywords: operators’ actions, human reliability analysis, probabilistic safety assessment, nuclear power plant