IEEE 1082:2017 pdf free download – IEEE Guide for Incorporating Human Reliability Analysis into Probabilistic Risk Assessments for Nuclear Power Generating Stationsand Other Nuclear Facilities
3.2.1 PRA compatibility
The HRAprocess proposed is suitable to all levels of a PRAgiven defined human failure events. If these are notdefined, then this guidance cannot be applied successfully.The risk focus of a PRA requires the quantitativeresults of an HRA to be probabilistic in nature.Applications of PRAs to risk management efforts requirethat the HRA documents in sufficient detail the analyst’s human factors considerations for the human failureevents. The PRA can have a diverse range of applications, the objectives of which may not be completelyidentified prior to the assessment. The HRA process should be flexible enough to anticipate some of the likelyapplications of the results of the HRA. For example, this may include design changes,procedure changes,training development, safety evaluations, or technical specification modification.
3.2.2 Qualitative HRA
While the approach identified in this guide supports HRA quantification as part of the PRA, it should be notedthat there is an increasing emphasis on the importance of qualitative HRA [B15], i.e.，,HRA that does notproduce a human error probability , but rather insights into the human’s role and contribution to overall systemperformance.The HRA approach in this guide supports both qualitative and quantitative aspects of HRA.For PRA, the quantitative approach should be adopted. For non-PRA applications of HRA, steps relevant toquantification should be omitted as appropriate.
3.2.3 The relationship of approach to results
Assumptions made by human reliability analysts about the relative importance of various human activitieswill influence the breadth and detail of models developed for the HRA. The data and chosen method ofquantifying human interactions will influence the specific estimates or ranges of uncertainty obtained,although there is generally good agreement between HRA methods. If results point to the need to improvethe reliability of selected systems and accompanying human interactions , these improvements should eitherbe readily identifiable from the documented HRA or should be the subject of further or different analyticalmethods that will allow improvements to be identified as described in method-specific documents. In additionto method-specific guidance, general guidance on selection of appropriate HRA approaches can be found incross-method overview documents such as [B2],[B3], and [B14].The HRAanalyst should be mindful of thiswhen considering the specific approach to be taken.
3.2.4 Matching the method to the application
Various HRA-related methods are available and being developed (e.g, cognitive approaches to human erroror approaches that address errors of commission).HRAs should be flexible enough to accommodate newfindings and model developments, while structured enough to be repeatable and traceable. HRA methodswere developed for different purposes, and they feature different strengths [B14].One emerging practice is that multiple HRA methods may be used within a single analysis to reflect the strengths of those approachesand the needs of the analysis [B7]. This guide remains method-neutral but encourages analysts to be flexible inconsidering the best HRAmethods for particular applications.
3.2.5 Limits
To be effectively incorporated into a PRA, an HRA should provide a realistic-as-possible interpretation of therole of plant personnel in accident prevention and mitigation.Accordingly, the results of an HRA should bedocumented in a format such that the basic assumptions, models, and data sources are clearly documented andthe limitations of the analysis are understandable to the user (e.g.,PRAanalyst).