ML20217D687
| ML20217D687 | |
| Person / Time | |
|---|---|
| Issue date: | 01/28/1998 |
| From: | Diaz N NRC COMMISSION (OCM) |
| To: | Hoyle J NRC OFFICE OF THE SECRETARY (SECY) |
| Shared Package | |
| ML20217D663 | List: |
| References | |
| SECY-97-205-C, NUDOCS 9803300162 | |
| Download: ML20217D687 (4) | |
Text
~
NOTATION VOTE RESPONSE SHEET TO:
John C. Hoyle, Secretary FROM:
COMMISSIONER DIAZ
SUBJECT:
SECY-97-205 -INTEGRATION AND EVALUATION OF RESULTS FROM RECENT LESSONS-LEARNED.
REVIEWS Approved x/in part Disapproved x/in part Abstain Not Participating Request Discussion COMMENTS:
See attached.
SIGNATURE O
/
Release Vote / x /
h M,\\99B DATE V WithholdVote /
/
Entered on "AS" Yes
- No
!!8 88M: 74882*
CORRESPONDENCE PDR 98035co)bL
~. ',
COMMISSIONER DIAZ' COMMENTS ON SECY-97-205 In discharging our respective duties and responsibilities to ensure a safe nuclear industry, licensees are responsible for the safe operation of nuclear power plants, while NRC is responsible for maintaining an effective and efficient regulatory regime that ensures adequate protection of the public health and safety. Presently, the best way to accomplish this is to couple state-of-the-art risk technology with licensees' operating technologies and experience. The need to couple these technologies, i.e., risk assessment with plant experience and system know-how with focus on safety, leads me to propose the following actions.
The issuance of Revision 1 to GL 91-18 was well executed and serves the agency well. Option 5 has the elements that I consider necessary to improve 50.59 in a systematic manner: " Option 5 is an incremental and progressive transition toward increased use of risk-informed decisionmaking." Based on this fundamental staff statement, and due to the different times for
~
implementation of the elements included in Option 5, I have opted to vote on and discuss its essential elements individually, yet grouped under "immediate" and "long term" categories.
The underlying rationale rests on my belief that this is an opportunity to enhance the NRC regulatory structure with risk information, and to actually do so with finality. Such actions will be conducive to better and more coherent regulations and will improve the objective and efficient implementation of NRC regulations with enhanced nuclear safety and overall net reduction in costs to the American people.
Immediate Fnhancemente to the Existino Repdatnry Framework I approve the three (3) remaining specific short term actions proposed by the staff under Option 5, as follows:
- 1. I approve the rulemaking enhancement to 50.59 (as described in Option 1), focused on clarifying and simplifying the current 50.59, and to incorporate knowledge gained from risk insights, as appropriate. The term "unreviewed safety question" (USQ) should be changed to a more exact term consistent with the intent of the rule. In order to establish realistic and safe limits allowing licensees to make changes under the rule without prior NRC approval, the staff -
should eliminate the de facto "zero-increase" criteria and allow " minimal" increases in the probability of occunence or the consequences of an accident. Similarly, licensees should be allowed to make changes without prior NRC approval when the proposed change results in the.
creation of an accident or malfunction of a different type that has " minimal" safety impact or
" minimal" reduction in the margin of safety. In addition, as recommended by the ACRS, the staff should complete NRC's endorsement of NEl %-07, consistent and concurrent with the above improvements. The staff should consider NEI's approach to " acceptance limit" and
" margin of safety" as a way to eliminate varying interpretations of licensee's safety analysis results subject to present acceptance limits. Finally, the staff should establish pertinent guidance 1
and documentation requirements to ensure integrity of the adequate protection of the health and safety envelope.
- 2. I approve the enhancements to 50.7_l(e) to clarify the scope and methods needed to update safety analysis reports, including rulemaking if necessary. In this regard, safety and risk significance should be used to support the determination and prioritization of the information to be revised or added to the Updated Safety Analysis Reports (USAR), as described in Option 2. I recommend the staff use three categories to differentiate among SSCs with high, intermediate, and low risk significance. Licensee attention, regulatory oversight and update of the USAR -
should be commensurate with risk significance. In addition, I support the use of risk information to aid in the removal ofinaccurate, obsolete, or unnecessary information. The SSCs with minimal risk significance should be considered for deletion from the USAR. This resulting new risk-informed USAR would become the de facto short term scope of 50.59.
- 3. I approve the enforcement policy changes allowing more discretion in determining the severity level of violations involving USQs, including maintaining the Enforcement Panel to ensure consistent treatment of 50.59 violations, and with due consideration ofimmediate changes to the rule.
Long Term Risk-Informed Fnhancements to the Existing Regulatnry Framework The principal driving force for reviewing and re-issuing 50.59 and 50.71(c) is the pressing need to stabilize the regulatory process at levels consistent with NRC's commitment to establish effective and efficient regulation. Specifically, the Commission has directed the staff to employ progressively risk-informed state-of-the-art methodologies via the PRA Policy Statement, the PRA Implementation Plan, and Direction-Setting Issue (DSI)-12. The results must be rules and regulations that are consistent with each other and maintain or improve safety while ' simplifying, defining, and reducing burden on licensees and the NRC staff. I believe that, today, risk information technology is capable of accomplishing these objectives. In fact, the staff and the ACRS have recently stated that the NRC is capable of permeating its regulations with risk-informed enhancements, as directed by the Commission. I also believe that it is not practical to implement a few selected risk-informed mies within a limited resource infrastructure; the entire regulatory fabric must be infused with risk information, and patchwork regulation eliminated, for processes to be effectively and economically implemented.
I vote for the establishment of a realistic, accountable program to convert 10 CFR 50 entirely to risk-informed regulation, with front-end emphasis on 50.59 and 50.71(c), as a necessary first step to achieve finality in employing risk technology in NRC regulations. Appendices A and B constitute fundamental cornerstones of reactor safety, licensing and operation, and they should be revised concurrently with 50.59. Section 50.65 should serve as a solid starting point for risk-informed conversions because it has already established broad categories of SSCs according to risk significance, including safety-related and non-safety-related SSCs. The risk-informed Part 50 will then be consistent with the NRC's PRA Policy Statement and the Maintenance Rule.
2
The staff should continue to work with the industry and use quantitative risk metrics to complement deterministic measures in defining as licensing commitments only those docketed issues with risk significance. Results from SECY-97-036, Short-Term Action 4, should be used to help define the licensing commitments. The staff efforts to define commitments related to the 50.59 scope should be closely coordinated with the effort in the Process Improvement Plan to identify and track licensing commitments. The staff should consider development of a new
' Current Licensing Basis (CLB) specific to nuclear power plants that would encompass a risk-informed safety analysis report and those commitments determined by the NRC to be permanent and impacting on plant safety. Temporary plant configuration changes, prxedures, and commitmenta can be monitored through the Maintenance Rule. This risk-informed safety analysis report would be similar to the operating safety analysis report (OSAR) I proposed in my vote for SECY-97-036, containing only those SSCs with risk significance, with priorities given to risk critical SSCs. This new, risk-informed CLB will then define the new scope for the appropriate regulatory processes, including 50.59. I do agree with Chairman Jackson that
" current licensing basis" should replace " safety analysis report" in the 50.59 rule, but as part of the long term enhancements. The systematic use of risk information, flowing from the Maintenance Rule through the PRA implementation plan and into other rules and regulations, would promote plant safety and reduce regulatory costs.
The staff should prepare a detailed plan for the conversion of Part 50 to risk-informed regulation where risk insights are used to provide additional metrics to the robust deterministic processes founded in defense-in-depth. Experience gained from the current PRA pilot applications and the "immediate" risk-informed enhancements discussed earlier should be considered in developing the plan. This plan, including a milestone schedule to achieve Part 50 conversion, should be presented for Commission approval no later than June 30,1998, for possible implementation in the first quarter of the year 2000.
NRC has, for more than two decades, sustained efforts to bring risk technology, including PRA, into its regulatory processes. I believe that PRA technology, in particular, has matured and its regulatory applications have been accelerated in the past two years. We should delay no longer in reaping the benefits of applying risk information in nuclear regulations. A risk-informed Part 50 will put teeth into the PRA Implementation Plan such that nuclear safety.will be enhanced and the nation will be able to realize the benefits.
+
3
)