From a risk assessment as well as risk management perspective, human reliability (HR) is generally viewed as quite important. Hence, there are considerable expectations associated with the potential benefits of integrating HR considerations into system reliability and system safety (e.g. probabilistic safety assessment, PSA) studies. The analytical dilemma which has emerged from extensive applications over a period of about 20 years relates to the fact that there is no consensus HR methodology, nor a generally agreed acceptable HR parameter data base supporting the HR methods. Despite attempts to structure and standardize the HR analysis task, there continues to be considerable variability in the assessments of the HR contribution to plant safety.
Against this background, the HR methodology that is in general use for PSA applications is discussed. The approaches covered are those which have been tested by extensive application in PSA studies. The paper lists some of the available methods that can be used to provide input to the accident sequence delineation and quantification. Some of the theoretical problems are also presented that can result in erroneous conclusions and recommendations when certain analysis procedures are used. Finally, some recommendations are provided regarding assurance of quality in HR analyses.