![]() |
Search Options | |||
Index | Site Map | FAQ | Facility Info | Reading Rm | New | Help | Glossary | Contact Us | ![]() |
May 23, 2000 The Honorable Richard A. Meserve Dear Chairman Meserve:
During the 472nd meeting of the Advisory Committee on Reactor Safeguards, May 11-13, 2000, we completed our review of SECY-00-0053, "NRC Program on Human Performance in Nuclear Power Plant Safety." Our Subcommittee on Human Factors reviewed this matter on March 15, 2000 . During our review, we had the benefit of discussions with representatives of the NRC staff. Observations and Recommendations
Discussion The current Program provides a good step toward the coordination of the staff's activities in this very important area. We expect that the Program will evolve as the results of the current efforts are obtained and new activities are identified. The staff reviewed the activities of other agencies in this area and performed an analysis of operating experience. Preliminary results from this analysis indicate that human performance has been an important contributor to the large majority of significant events. Latent conditions such as failure to fix known problems, inadequate attention to organizational learning, and inadequate maintenance practices figured prominently in these events. Latent conditions, in fact, outweighed active human performance errors by four to one. What remains to be done is to evaluate the significance of these observations in a probabilistic risk assessment context. The challenge here is not so much the evaluation of the risk significance of individual human errors but, rather, the significance of latent conditions that may lead to poor human performance and the potential for common-cause failures. We agree with Commissioner Merrifield's observation in his speech at the Regulatory Information Conference that "...we and our licensees must continue to wage an aggressive campaign against the buildup of latent conditions and we simply must not forget to worry." The results of the staff's analysis of operating experience indicated that a major part of latent conditions stems from organizational deficiencies. Such organizational issues constitute an important element of what the International Nuclear Safety Advisory Group has called the "safety culture" of the plant (INSAG-4, 1991). The agency's two major regulatory initiatives are to risk inform the regulations and to revise the reactor oversight process. We believe that an important consideration in determining which projects should be included in the Program is the degree to which these projects support the agency's initiatives. For example, it is not clear to us what needs of the RROP or ATHEANA will be satisfied by the results of the project on control room design or the Halden simulation experiments. Risk informing the regulations requires models for human reliability analysis (HRA). There are two broad areas of human performance activities that are included in HRA: activities before an initiating event and activities after an initiating event. The HRA for routine pre-initiator activities relies largely on the methods described in the Human Reliability Handbook. As we stated in our letter on ATHEANA dated December 15, 1999, an important omission in the analysis of pre-initiator activities is the failure to investigate how human errors during normal operations would initiate a plant event. We believe insights derived from the analysis of operating experience, as well as recent models of human error, can be used to perform this investigation. The major HRA project for post-initiator activities is ATHEANA. There needs to be more coordination between the ATHEANA project and the analysis of operating experience project. The ATHEANA data needs should be inputs to the data analysis project and the findings from the analysis could suggest improvements to the ATHEANA model. For example, the kinds of latent conditions identified by the analysis of operating experience should be significant inputs to the identification of the error-forcing context that is an important element of ATHEANA. The revised reactor oversight process defines three cross-cutting issues: human performance, safety-conscious work environment, and problem identification and resolution. An assumption in the oversight process is that the impact of cross-cutting issues on plant safety will be reflected in the plant performance indicators and the baseline inspection findings. This is an untested assumption. The proposed activity to characterize the extent to which human performance is captured in the revised reactor oversight process will test this assumption and should be pursued. We look forward to hearing from the staff on the results obtained from this Program.
References
|