IEEE Standards Interpretations: IEEE Std 1076-1987, IEEE Standard VHDL Language Reference Manual
|Publication Date:||1 January 1991|
(This introduction is not part of IEEE Std 1082-1997, IEEE Guide for Incorporating Human Action Reliability Analysis for Nuclear Power Generating Stations.)
Any process that requires manual control to minimize public risk will require a high level of human reliability. This reliability can be evaluated through the systematic application of a probabilistic risk assessment (PRA). However, such an assessment requires a detailed understanding of human factors to form a complete reliability estimate.
The initial risk assessment made in the nuclear power plant industry, WASH-1400, recognized the need for a discipline of human reliability analysis (HRA), systematically incorporated within the PRA enterprise. But the methodology-both of analyzing human failure events and identifying and incorporating them appropriately in the PRA-was new, incomplete, and in several ways inadequate.
The limitations of the understanding of human reliability in the mid-1970s was vividly demonstrated by the accident at Three Mile Island (TMI). Following TMI, the Nuclear Regulatory Commission (NRC), in conjunction with the Institute of Electrical and Electronics Engineers (IEEE), immediately called for a conference on the human factor issues raised by TMI. This conference has become a series, the fifth of which was held in Monterey, California, in 1992. Parallel to this activity, Subcommittee 7, Human Factors and Control Facilities, of the Nuclear Power Engineering Committee, began discussing the standardization of the HRA technology. The PRA/HRA interface of incorporating and performing an HRA in the context of a PRA was recognized as the most mature of the efforts of HRA. A guide, the least mandating of the IEEE standards documents, was approved as an IEEE standards project in 1984.
In recent years, some convergence in the ideas and methods of HRA has occurred. Most PRAs perform HRA in a standard way, even when using approaches as diverse as the Systematic Human Action Reliability Procedure of the Electric Power Research Institute (EPRI) and the Accident Sequence Evaluation Program HRA method of the NRC.
This guide outlines the steps necessary to include human reliability in risk assessments. It is not its intent to discuss the details of HRA methods, since this technology is evolving and cannot be addressed in the needed depth in this guide. The reader is urged to review bibliography entries [B1], [B2], [B5], [B12], and [B22] for this information. Since human error has been found to be an important contributor to risk, this guide underscores the systematic integration of the HRA at the earliest stages and throughout the PRA.
It is anticipated that human reliability technology will not escape controversy in the near future. This is especially true regarding the quantification of Human Error Probabilities, which is not specifically addressed in this guide. In one sense this is a blessing, since it may ensure that the industries do not become complacent to the contribution of people and their actions to the risk spectrum of the technology they work.
This guide provides a structured framework for the incorporation of human/system interactions into probabilistic risk assessments (PRAs).