ML20196B313

From kanterella
Jump to navigation Jump to search
Responds to Re Concerns About Objective Risk Informed,Performance Based Regulation of Nuclear Power Industry in Areas of Mgt & Training
ML20196B313
Person / Time
Issue date: 11/19/1998
From: Shirley Ann Jackson, The Chairman
NRC COMMISSION (OCM)
To: Shea D
AFFILIATION NOT ASSIGNED
Shared Package
ML20196B317 List:
References
GL-88-20, NUDOCS 9812010044
Download: ML20196B313 (2)


Text

r-D C-

, y

[ k} UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, o.C. 20555-0001 g g s ,

f November 19, 1998 CHAIRMAN Mr. Daniel J. Shea 1001 Fannin, Suite 210 Houston, Texas 77002-8712

Dear Mr. Shea:

In the letter you sent me on September 29,1998, you raised some ideas and concerns about objective " risk-informed, performance-based" regulation of the nuclear power industry in the areas of management and training. You underlined the importance of using probabilistic risk assecsment (PRA) insights in conjunction with real-time accident analyses to " design" the operator's job based on "how the plant reacts" by identifying and studying risk-important " control opportunities" presented to the operators during accidents. This, you suggested, could be used to develop key performance indicators, to determine if the plant is sufficiently fault-tolerant to compensate for operator errors, and to remove the possibility of error by changing the control scheme.

Since the Three Mile Island (TMI) accident, significant strides were made in these areas '

through the development of symptom-based emergency operating procedures (EOPs), use of full-scope simulators, and improvements in operator training. The Nuclear Regulatory Commission (NRC) believes that PRA is a valuable tool that can be used by plant management to generate insights not readily available from the traditional deterministic and licensing analyses. Several important insights have been derived frorn the integrated model of plant l system behavior and operator actions used in PRAs. In response to NRC Generic Letter 88-20, which asked licensses to perform individual plant examinations (IPEs) to find potential l vulnerabilities at their plants, the U.S. nuck ~ powe p! ants performed PRAs that are being used to make the preceding activities " risk informed."

To reduce the dependence of operator response on knowledge-based behaviors during abnormal events, symptom-based EOPs were developed. These procedures are based on the

//

philosophy of symptomatic responses to an emergency operating situation and, therefore, reduce the diagnosis of an event to just responding to cues such as alaims, annunciators, and indicators. This aims at limiting the impact of the "information overload" phenomenon, ,

mentioned in your letter, by limiting the cognitive aspects of the diagnosis of an abnormal event. 'o Symptom-based EOPs were developed to provide operators wi5 guidance that covers any nuclear steam supply system (NSSS) condition that potentially threatens a barrier and that could lead to the release of radioactive material. The approach has several key features. The NSSS condition criteria for avoiding barrier failure do not require identification of the accident scenario; thus, accidents that have never been recognized as potential accidents are addressed. A few key symptoms, obtained from the control room inforniation displays with little or no operator interpretation, are sufficient to assess the NSSS condition. Operator actions are fully described and focused on symptoms whenever co rective action is needed. Each cf these features addresses opcrator "information overload" concems as well as concerns regarding 9812010044 981119 PDR COMMS NRCC CORRESPONDENCE PDR j c - - _ _

l l.

2 operator failure to diagnose an accident that " fits" available EOPs, as was the case during the TMI accident mentioned in your letter. This approach also parallels your " hands-off" suggestion by properly focusing operator actions on symptoms. If an operator makes an error, it would be reflected in the symptoms (if it is important to safety), providing feedback to the operator that l demands appropriate corrective action.

We recognize your concerns regarding operator implementation of EOPs and associated training. The post-TMI, symptom-based EOPs currently in use were developed through well-defined processes that incorporated the engineering and operational knowledge of both the industry owners groups and the individual facilities. The NRC provided oversight and guidance i

in this multidisciplinary effort, and subsequently conducted a specialinspection program for l EOPs. Following the inspection program, the NRC provided additional guidance and lessons-learned addressing the development, implementation, and maintenance of EOPs in NUREG-1358, Supplement No.1, " Lessons Learned From the Special Inspection Program for Emergency Operating Procedures." The plant-specific EOPs continue to mature with time.

Information resulting from an enhanced understanding in such areas as PRA, event dynamics, plant system and operator performance, and job task analysis continues to be incorporated into the EOPs, applied in a systems approach to training (SAT), and employed in the determination of staffing needs. Operator training includes the use of certified, plant-referenced simulators to conduct real-time training scenarios. Facilities provU . pecific written guidance regarding procedural adherence to ensure that procedures, incLding EOPs, are performed as intended, l and shift technical advisors (STAS) provide engineering expertise on shift during plant operations. .

Furthermore, the NRC encouraged the development of severe-accident management guidelines (SAMGs), which address the response to specific risk-significant scenarios for the purpose of preventing core damage or mitigating its consequences. Plant-specific SAMGs are being developed, which, in effect, extend the defense-in-depth principle to plant operating staff by extending the operating procedures well beyond the plant design basis and by making full use of existing plant equipment and operator skills to terminate severe accidents and limit l

offsite releases.

As the industry moves in the direction of " risk-informed, performance-based" regulation, open meetings and workshops involving the NRC and industry organizations, such as the various owners' groups, Nuclear Energy Institute (NEI), Institute for Nuclear Power Operations (INPO),

and Electric Power Research Institute (EPRI), are being conducted. Your participation in such public workshops, where technical and regulatory issues are discussed in the context of " risk-informed, performance-based" regulation, would provide useful feedback. Thank you for your I contribution to our effort to move the NRC in the direction of " risk-informed, performance-l based" regulation and your interest in nuclear safety.

l Sincerely, b

Shirley Ann Jackson 4