[Federal Register Volume 65, Number 43 (Friday, March 3, 2000)]
[Proposed Rules]
[Pages 11488-11505]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 00-5016]


=======================================================================
-----------------------------------------------------------------------

NUCLEAR REGULATORY COMMISSION

10 CFR Parts 21, 50, 52, 54 and 100

RIN 3150-AG42


Risk-Informing Special Treatment Requirements

AGENCY: Nuclear Regulatory Commission.

ACTION: Advance notice of proposed rulemaking.

-----------------------------------------------------------------------

SUMMARY: The Nuclear Regulatory Commission (NRC) is considering 
promulgating new regulations that would provide an alternative risk-
informed approach for special treatment requirements in the current 
regulations. This action is a result of the Commission's continuing 
efforts to risk-inform its regulations. The NRC invites comments, 
advice, and recommendations from interested parties on the contemplated 
approach for this rulemaking.

DATES: Comment period expires May 17, 2000. Comments received after 
this date will be considered if it is practical to do so, but the 
Commission is able to ensure consideration only for comments received 
on or before this date.

ADDRESSES: Send comments to: The Secretary of the Commission, U.S. 
Nuclear Regulatory Commission, Washington, DC 20555-0001, Attention: 
Rulemakings and Adjudications Staff. Deliver comments to: 11555 
Rockville Pike, Rockville, Maryland, between 7:30 a.m. and 4:15 p.m. on 
Federal workdays.
    You may also provide comments via the NRC's interactive rulemaking 
website through the NRC's home page (http://ruleforum.llnl.gov). This 
site provides the capability to upload comments as files (any format) 
if your web browser supports that function. For information about the 
interactive rulemaking website, contact Ms. Carol Gallagher, (301) 415-
5905; e-mail [email protected].
    Copies of comments received may be examined at the NRC Public 
Document Room, 2120 L Street NW. (Lower Level), Washington, DC.

FOR FURTHER INFORMATION CONTACT: Thomas A. Bergman, Office of Nuclear 
Reactor Regulation, U.S. Nuclear Regulatory Commission, Washington, DC 
20555-0001; telephone: (301) 415-1021; e-mail: [email protected].

SUPPLEMENTARY INFORMATION:
I. Background
II. Rulemaking Plan
    A. Vision.
    B. Strategies.
    C. Objectives.
    D. Selection of Candidate Rules.

[[Page 11489]]

    E. Rulemaking Alternatives.
    1. Define New Term.
    2. Redefine Current Terms.
    3. Issue New Rule.
    4. Comprehensive vs. Phased Rulemaking.
    F. Implementation.
    1. New Appendix vs. Regulatory Guide.
    2. Additional Guidance.
    G. Pilot Plant Program.
    H. South Texas Exemption Request.
    I. Schedule.
III. Specific Proposal
    A. Approach.
    B. New Rule for Part 50.
    C. New Appendix to Part 50.
IV. Issues
    A. Selective Implementation.
    B. Impact on Other Regulations.
    C. Need for Prior NRC Review.
    D. Identification and Control of Attributes Requiring Special 
Treatment.
V. Specific Questions
    A. Approach.
    B. Screening.
    C. Categorization Methodology.
    D. Pilot Plant Program.
    E. Identification and Control of Special Treatment Attributes.
    F. Selective Implementation.
    G. Impact on Other Regulations.
    H. Need for Prior NRC Review.

I. Background

    On August 16, 1995 (60 FR 42622), the Commission published a policy 
statement entitled ``Use of Probabilistic Risk Assessment (PRA) Methods 
in Nuclear Regulatory Activities.'' Since then, the Commission has 
issued guidance \1\ on the use of risk information for reactor license 
amendments. This guidance is currently being used in processing license 
amendment applications that use risk information as part of their 
technical justification. However, fundamental reactor regulations 
remain largely deterministic. In addition, in meetings between the 
Commission and various stakeholders, a concern was expressed that the 
NRC is not placing enough emphasis on risk-informing its reactor 
requirements with the results of risk assessments. The Commission's 
current reactor regulatory framework (based largely upon design-basis 
events rather than on core-damage-accident scenarios) results in 
reasonable assurance of adequate protection to public health and safety 
but, in some cases, also results in unnecessary regulatory burden. In a 
staff requirements memorandum (SRM) dated September 14, 1998, the 
Commission requested the NRC staff to present a set of options to make 
the requirements in the Commission's regulations risk-informed. The 
Commission expects that making the regulations risk-informed would 
result in a reduction of unnecessary regulatory burden while 
maintaining safety because there will be a better focus of the NRC's 
and industry's resources on the more safety significant structures, 
systems, and components (SSCs) and, therefore, address the expressed 
concern.
---------------------------------------------------------------------------

    \1\ To date, this guidance includes Standard Review Plan (SRP) 
Chapter 19 and related Regulatory Guide (RG) 1.174 on risk-informed 
decision making; SRP Section 3.9.7 and related RG 1.175 on risk-
informed inservice testing; SRP Section 16.1 and related RG 1.177 on 
risk-informed technical specifications; RG 1.176 on risk-informed 
graded quality assurance; and SRP Section 3.9.8 and related RG 1.178 
on risk-informed inservice inspection.
---------------------------------------------------------------------------

    In SECY-98-300, ``Options for Risk-Informed Revisions to 10 CFR 
part 50--``Domestic Licensing of Production and Utilization 
Facilities,''' dated December 23, 1998, the NRC staff proposed three 
high-level options for making the NRC's regulations risk-informed. In 
an SRM dated June 8, 1999, the Commission approved the NRC staff's 
recommendations.
    One of the options presented in SECY-98-300 was to make special 
treatment requirements (e.g., quality assurance, environmental 
qualifications, technical specifications, reporting) risk-informed. 
Special treatment as used here may be defined as--

    Current requirements imposed on structures, systems, and 
components (SSCs) that go beyond industry-established requirements 
for equipment classified as ``commercial grade'' that provide 
additional confidence that the equipment is capable of meeting its 
functional requirements under design basis conditions. These 
additional special treatment requirements include additional design 
considerations, qualification, change control, documentation, 
reporting, maintenance, testing, surveillance, and quality assurance 
requirements.

    This definition does not encompass functional design requirements; 
that is, an SSCs functional design requirement is not considered a 
special treatment requirement. This definition applies, hereafter, when 
the term ``special treatment'' is used.
    This advance notice of proposed rulemaking presents the approaches 
that the Commission is contemplating to risk-inform special treatment 
requirements. Several public meetings have been held to obtain comments 
on the NRC's efforts related to this task. Comments and suggestions 
obtained from these meetings have been incorporated, to the extent 
possible, into these approaches.

II. Rulemaking Plan

A. Vision

    Develop alternative regulations in 10 CFR part 50 (and other 
applicable parts) that would modify the requirements for special 
treatment to focus on those SSCs that have been identified as important 
to protect public health and safety by using a risk-informed approach.

B. Strategies

    Increase the use of risk-informed approaches to modify the special 
treatment requirements imposed on SSCs under existing Part 50 
requirements (and those of other applicable parts).
    Maintain overall safety provided by the existing Part 50 while 
reducing unnecessary burden associated with these requirements for 
licensee operational and licensing activities and for NRC oversight and 
licensing activities.
    Risk-inform the special treatment requirements imposed on SSCs 
under Part 50 (and other applicable parts) in a manner that encourages 
public participation and results in public confidence in the product 
and process.

C. Objectives

    Establish the criteria for acceptable methods for determining the 
SSCs that require special treatment in the regulations of Part 50. 
These criteria should be sufficiently clear and robust such that if a 
licensee's program meets the criteria there is not a need for prior NRC 
review and approval of the plant-specific program.
    Assign priorities to the rules to be modified, taking into 
consideration the maintenance of safety, the reduction of unnecessary 
burden for industry, the effect on NRC efficiency and effectiveness, 
public confidence, and the complexity of modifying each rule.
    Ensure that the categorization process has been evaluated under a 
pilot program to verify that the requirements and their associated 
guidance can be implemented by industry, and that the results of 
licensee implementation provide reasonable assurance that public health 
and safety is maintained.
    Issue a proposed rule for the initial set of rules to be modified 
within 1 year of the Commission's approval of the rulemaking plan, and 
a final rule within 1 year of the completion of the associated pilot 
program.
    The proposed risk-informed regulatory alternatives should reduce 
unnecessary burden so that licensees with more than 10 years remaining 
on their license would find it beneficial to voluntarily implement the 
risk-informed alternative requirements.

D. Selection of Candidate Rules

    The Commission believes that the set of rules to be considered in 
this effort must be identified early so that rule-specific issues can 
be identified and

[[Page 11490]]

addressed. Also, because implementation of any rules resulting from 
this effort is optional, the Commission does not intend to expend 
resources to modify rules that industry does not expect to implement, 
unless the modifications are necessary to maintain safety. However, the 
Commission notes that the set of rules included in this effort should 
be chosen such that implementation of the rules will require little or 
no exemptions. Therefore, rules that may require exemptions before a 
licensee can implement changes in other rules (e.g., 10 CFR 50.59) 
should be considered in this rulemaking effort.
    The NRC has developed and applied a systematic approach to identify 
the rules that should be included in this rulemaking effort. A scoping 
review of all the regulations in 10 CFR parts 21, 50, 52, 54, and 100 
identified a set of potential candidate rules that could be included. 
Screening criteria and a logic for applying these criteria were then 
developed to identify the subset of rules to which risk-informed 
changes can be made consistent with the intent of this effort. The 
screening criteria were based on the following elements: Maintaining 
safety, improving NRC staff efficiency and effectiveness, reducing 
unnecessary regulatory burden, and increasing public confidence. In 
addition, and because this effort is focused solely on special 
treatment requirements, the NRC limited its selection to those rules 
that include special treatment requirements. Rules which would have to 
be modified in order to efficiently implement other rules included this 
effort were also included. The criteria and logic were then applied to 
the set of potential candidate rules identified by the scoping review. 
The screening process and results are illustrated in Figure 2. The 
results of the evaluations of the rules against each of the screening 
criteria are presented in the attached Table. As a result of this 
screening process, the NRC has identified the following candidate rules 
for inclusion in this effort:

10 CFR part 50--Sections 50.34, 50.36, 50.44, 50.48, 50.49, 50.54, 
50.55, 50.55a, 50.59, 50.65, 50.71, 50.72, and 50.73
10 CFR part 50--Appendix A (GDCs 1, 2, 3, 4, 37, 40, 42, 43, 45, and 
46), Appendix B, Appendix J, Appendix R, and Appendix S
10 CFR part 21, 52, 54, 100, and Appendix A to Part 100

E. Rulemaking Alternatives

    The NRC has evaluated alternatives to rulemaking and has concluded 
that, if sufficient industry interest exists, rulemaking is the most 
effective tool for implementing the type of generic changes encompassed 
by this effort. If sufficient interest does not exist, review and 
approval of a limited number of exemptions under 10 CFR 50.12 would be 
more efficient. Assuming industry interest does exist as has been 
indicated in public meetings, the NRC has evaluated several rulemaking 
alternatives to accomplish this task. These alternatives are discussed 
below.
1. Define New Term
    This alternative would entail the definition of a new term in 10 
CFR 50.2 (e.g., ``safety-significant'') that describes, for the 
purposes of special treatment requirements, which SSCs are safety-
significant and, therefore, need to be within the scope of the special 
treatment requirements. This new term would then be incorporated into 
each rule that contains special treatment requirements to allow 
licensees to voluntarily revise the scope of SSCs that are subject to 
special treatment requirements. To determine which SSCs are safety 
significant, the Commission would issue a new Part 50 appendix that 
contains the requirements governing the categorization of SSCs 
consistent with the new term defined in Sec. 50.2. Alternatively, the 
Commission could issue a regulatory guide that contains the SSC 
categorization guidance.
    Regulatory treatment requirements in addition to the special 
treatment requirements currently in the regulations may be necessary as 
a result of the risk categorization processes. These additional 
requirements would have to be added to the regulations and, therefore, 
additional changes to each affected rule may be required to ensure that 
the new regulatory treatment requirements are appropriately captured in 
the regulations. Because this alternative would result in duplicate 
changes to multiple rules, the NRC did not choose this alternative.
2. Redefine Current Terms
    This alternative would expand the definition of the term ``safety-
related'' in 10 CFR 50.2, or as an alternative, define the term 
``important to safety'' such that the redefined term would contain a 
portion that allows special treatment requirements to be risk-informed. 
Licensees could then elect to risk-inform the scope of SSCs that are 
subject to special treatment in all the applicable rules. This approach 
would expand the definitions of the current terms (which reside in the 
existing rules) so there is no need to add new terms to the governing 
regulations. However, a significant effort would be required to review 
all the regulations to ensure that the Commission has not 
unintentionally revised any non-special treatment rules and to make 
appropriate changes to preclude such occurrences. In a similar fashion 
to the ``new term'' approach, this approach would also need to be 
supplemented with either a new Part 50 appendix that contains the 
requirements governing the risk-informed categorization of SSCs, or a 
regulatory guide that contains the SSC categorization guidance.
    This alternative would introduce unnecessary complications and 
confusion in the application of the terms at plants that choose to 
implement the new scope for a subset of the special treatment 
requirements covered in this effort, or for some systems and not 
others. Such a situation would result in the use of similar language 
with different meanings in the licensee's licensing basis documents and 
in the associated plant implementation documents. Furthermore, 
regulatory treatment requirements, in addition to those currently in 
the regulations, may be necessary as a result of the risk 
categorization processes. These requirements would have to be added to 
the regulations. Therefore, changes to other rules may still be 
required. The NRC did not choose this alternative.
3. Issue New Rule
    This approach entails the development of a new rule that would be 
added to Part 50. The rule would ``list'' the provisions that contain 
special treatment requirements that may have their scope risk-informed 
in accordance with the methodology requirements contained in either a 
new appendix that would also be added to Part 50, or in guidance 
contained in a regulatory guide (similar to above two alternatives in 
this respect). In addition to identifying which rules can be risk-
informed for special treatment, the new rule would address rule 
specific issues resulting from this effort and contain new requirements 
concerning the type of regulatory treatment that SSCs would receive.
    The NRC believes that this alternative is the simplest and most 
efficient regulatory approach because it appears to not require 
defining new terms which in turn requires subsequent revisions to each 
affected rule. In addition, this alternative has the benefit of 
integrating all the affected special treatment requirements into one 
rule which would make it easier for licensees and the NRC to implement. 
Therefore, the NRC has decided to proceed with this alternative.

[[Page 11491]]

4. Comprehensive vs. Phased Rulemaking
    The NRC considered whether it should proceed with a comprehensive 
rulemaking covering all special treatment requirements or a phased 
approach. The NRC's objective is to proceed with a comprehensive 
rulemaking. However, the NRC recognizes that this approach may prove 
problematic. Because of the uniqueness of the special treatment 
requirements, the potentially different effects that may result from 
modifying these requirements, and the inconsistencies that currently 
exist between the various special treatment requirements, the NRC notes 
that the comprehensive rulemaking approach would be a large and complex 
task. The comprehensive rulemaking approach appears to have a greater 
potential for delay because of the time required to review each of the 
affected requirements and the potential for issues to arise that can 
have impacts on the schedule. A comprehensive rulemaking must address 
all affected requirements and issues before the rulemaking may be 
completed. Consequently, this might delay implementation of some rules 
due to complications with others. If complications do arise, the NRC 
may elect to proceed with a phased approach that allows the NRC to 
issue some revised rules while continuing to address issues that arise 
on others.

F. Implementation

1. New Appendix vs. Regulatory Guide
    Each of the alternatives discussed in Section E include either the 
development of a new Appendix to Part 50 or the issuance of a 
regulatory guide that would contain the requirements governing the 
categorization of SSCs. The NRC has considered these two alternatives 
(a new appendix vs. a regulatory guide) and concluded that a new 
appendix approach is preferred because it would provide a more stable 
and predictable regulatory framework. Such a framework should result in 
the least burden on NRC and industry resources both from the standpoint 
of any prior NRC review that is required and from the standpoint of the 
staff's inspection of this task. If an appendix can be constructed that 
when implemented by licensees yields consistent, objective, 
enforceable, and inspectable results, then this regulatory approach 
should allow for implementation of the resulting risk-informed special 
treatment requirements with little or no NRC review. On the other hand, 
putting categorization guidance into a regulatory guide would require 
that the staff review and approve licensee submittals prior to 
implementation because of the flexibility inherent in a regulatory 
guide. The NRC expects the pilot plant program to enable it to 
determine if development of an appendix in lieu of a regulatory guide 
is sufficient to support a no prior NRC review regulatory approach. If 
the pilot plant program reveals that development of the appendix does 
not minimize the need for NRC review, the NRC will reconsider whether 
an appendix remains the best approach.
2. Additional Guidance
    In addition to either an appendix or a regulatory guide, the 
Nuclear Energy Institutes (NEI) has indicated that it will submit an 
implementing document for this effort. The NRC intends to review this 
implementing document. The objective of this review will be to reach 
agreement with NEI concerning the implementation of risk-informed 
special treatment, and to be able to endorse the NEI guidance in a 
regulatory guide. Consequently, the Commission does not currently plan 
to develop draft regulatory guidance to implement this rulemaking. 
Additional NRC efforts would be required to update current regulatory 
guides that address the current SSC categorization approach, as 
appropriate.

G. Pilot Plant Program

    The Commission believes that the pilot plant program is an 
essential component of this rulemaking effort. The purpose of this 
program would be to demonstrate the viability of the requirements 
contained in the resulting rule and appendix before final rulemaking 
and the viability of the proposed NEI guidance for the implementation 
of the resulting rule and appendix. The program will also help the NRC 
identify the special treatment requirements that industry believes 
should be addressed.
    The most important aspect of the pilot plant program will be to 
demonstrate the viability of risk categorization processes to establish 
alternative risk-informed special treatment requirements. These 
processes must be based on the requirements in the resulting rule and 
appendix in order to provide meaningful feedback on the rulemaking 
effort. In addition, the categorization processes must be evaluated 
against the set of special treatment requirements they are applied to 
so that critical attributes are appropriately evaluated. The 
categorization processes must also be applied to a variety of plant 
systems, including mechanical (active and passive), fluid, and 
electrical systems, and safety-related and nonsafety-related systems, 
so that technical aspects of the categorization processes and their 
implementation can be thoroughly exercised. The Commission may 
explicitly exclude any attributes that are not exercised by the pilot 
plant program from consideration in this effort.
    The pilot plant program must be integrated with the rulemaking 
plan. It must agree on overall and plant-specific schedules and the 
rules to be piloted. Pilot plant program participants must commit to 
meet the resulting rulemaking requirements and proposed NEI guidance 
for categorization and implementation. In addition, pilot program 
submittals should address how design basis functions will be preserved 
when special treatment for safety-related SSCs is reduced as a result 
of the risk categorization processes. The discussion should address how 
these SSCs will be treated by the licensee's design control and 
corrective action programs. Similarly, licensees should discuss how 
critical attributes identified by the risk categorization processes 
will be identified and controlled. This applies to safety-related and 
non-safety-related SSCs that are found to be significant as a result of 
the risk categorization processes. The processes established should be 
capable of reflecting changes to the facility and categorizing new and 
modified equipment as these changes are made.

H. South Texas Exemption Request

    In addition to the pilot plant program, the Commission notes that 
South Texas Project Nuclear Operating Company has submitted an 
extensive exemption request related to a number of special treatment 
requirements. This submittal was developed before initiation of this 
effort, and so was not coordinated with the development of the 
rulemaking plan. Presently, the NRC expects to complete review of this 
submittal before the proposed rulemaking stage of the effort would 
begin. The NRC believes that, if approved, the South Texas exemption 
request will serve as a proof-of-concept prototype which will provide 
useful information and experience when the rulemaking for this effort 
is developed.

I. Schedule

    The NRC has developed a schedule covering the following activities 
which influence this rulemaking: (1) The South Texas exemption request, 
(2) development and issuance of this advanced notice of proposed 
rulemaking, (3) the pilot plant program,

[[Page 11492]]

(4) NRC review of the NEI implementation guidance, (5) development and 
issuance of the proposed rulemaking, and (6) development and issuance 
of the final rulemaking. The NRC estimates that a final rule can be 
issued by March of 2002. This rulemaking includes milestones that 
depend significantly on NEI to develop implementation guidance and 
pilot plant program participants to develop and implement 
categorization processes.

III. Specific Proposal

A. Approach

    To effect the described changes, the Commission is considering an 
approach that consists of issuing a new rule (10 CFR 50.69) and a new 
appendix (Appendix T to 10 CFR part 50). The new rule and appendix 
would allow licensees, for purposes of special treatment requirements, 
to categorize SSCs with regard to their importance to plant safety. The 
result of such a rulemaking, when combined with the current 
deterministic design basis, would result in SSCs being classified in 
two different manners. One would be consistent with the safety-related/
non-safety-related philosophy that exists today for the deterministic 
design basis. The other would be consistent with a risk-informed 
philosophy. A graphical depiction of the results of the contemplated 
changes is illustrated in Figure 1. The figure is only intended to 
provide a conceptual understanding of the new SSC categorization 
process. The NRC's thinking on this matter is continuing to evolve. The 
NRC will explore the idea of more than two levels of safety 
significance. The NRC is requesting stakeholder feedback on the safety 
significance categories in question C.3 of Section V of this notice. 
The figure depicts the current safety-related versus nonsafety-related 
SSC categorization scheme on the horizontal axis with an overlay of the 
new risk-informed categorization on the vertical axis. The risk-
informed categorization would group SSCs into one of the four boxes.
[GRAPHIC] [TIFF OMITTED] TP03MR00.002

    Box 1 of Figure 1 contains safety-related SSCs that a risk-informed 
categorization process concludes are significant contributors to plant 
safety. These SSCs are termed risk-informed safety class 1 (RISC-1) 
SSCs. SSCs in this box would continue to be subject to the current 
special treatment requirements. In addition, it is possible that some 
of these SSCs may have some additional requirements concerning 
reliability and availability if attributes that cause the SSC to be 
safety significant are not sufficiently controlled by current special 
treatment requirements. However, the NRC is not currently aware of any 
examples of this situation.
    Box 2 of Figure 1 depicts the SSCs that are nonsafety-related, and 
that the risk-informed categorization concludes make a significant 
contribution to plant safety. These SSCs are termed RISC-2 SSCs. 
Examples of RISC-2 SSCs could include the station blackout emergency 
diesel generator, the startup feedwater pump for pressurized water 
reactors (PWRs), and SSCs used for ``feed and bleed'' operations at 
PWRs. For RISC-2 SSCs, there will probably need to be requirements to 
maintain the reliability and availability of the SSCs consistent with 
the PRA. It is currently envisioned that the new rule would contain the 
requirements regarding reliability and availability of RISC-1 and RISC-
2 SSCs.
    Box 3 of Figure 1 depicts the currently safety-related SSCs that a 
risk-informed categorization process determines are not significant 
contributors to plant safety. These SSCs are termed RISC-3 SSCs. The 
rulemaking effort would revise 10 CFR part 50 to contain alternative 
requirements such that RISC-3 SSCs would no longer be subject to the 
current special treatment requirements. For RISC-3 SSCs, it is not the 
intent of this rulemaking to allow such SSCs to be removed from the 
facility or to have their functional capability lost. Instead, the 
RISC-3 SSCs will need to receive sufficient regulatory treatment such 
that these SSCs are still expected to meet functional requirements, 
albeit at a reduced level of assurance. The NRC may determine that this 
level of assurance can be provided by licensees' commercial grade 
treatment programs. It is envisioned that the new rule would contain 
the regulatory treatments requirements for RISC-3 SSCs (e.g., the new 
rule may require commercial treatment for RISC-3 SSCs).
    Box 4 of Figure 1 depicts SSCs that are nonsafety-related and 
continue to be categorized as not being significant contributors to 
plant safety. These SSCs are out of scope of both the current special 
treatment regulations and of the new rule. The functional performance 
of these SSCs would be controlled under

[[Page 11493]]

the licensee's commercial grade program (no change from the current 
requirements).

B. New Rule for Part 50

    The Commission expects that the new rule that would (1) identify 
the special treatment requirements in the current regulations whose 
scope could be modified consistent with the requirements resulting from 
this effort, (2) address rule-specific issues that arise as a result of 
the new scope by, for example, specifying, on a rule-by-rule basis, the 
applicability of the new scope, (3) specify all additional regulatory 
requirements that would result from this effort, and (4) reference the 
new appendix as providing the requirements governing the categorization 
of SSCs.

C. New Appendix to Part 50

    The Commission expects that the new appendix would contain the 
elements discussed below. The discussion consists of NRC expectations 
of the SSC categorization process and is not presented as proposed rule 
language. When finalized, the appendix would establish minimum 
requirements for the process and decision criteria for use in the 
categorization of SSCs into two groups--those that have safety 
significance and those that have low safety significance. This is 
consistent with the process to categorize SSCs into RISC classes as 
discussed above in which the safety significant and low safety 
significant categorization in used in the vertical axis.

Appendix T to Part 50

Categorization of SSCs Into Risk-Informed Safety Classes

    The principal activity required for the categorization of 
structures, systems and components (SSCs) into risk-informed safety 
classes is the categorization of the SSCs according to safety 
significance. Treatment requirements for SSCs will be dependent on 
this safety classification. This appendix establishes minimum 
requirements for the process and decision criteria for use in the 
categorization of SSCs.

Process for Categorization

    The determination of safety significance of SSCs must be 
performed as part of an integrated decision-making process which 
uses both risk insights and traditional engineering insights. In 
categorizing SSCs, it must be demonstrated that the defense-in-depth 
philosophy is maintained, that sufficient safety margin is 
maintained, and that increases in risk (if any) are small.
    To accomplish these objectives, the process to categorize SSCs 
should consist of the following elements:
    (1) Identification of current treatment requirements for SSCs.
    (2) Assessment of the capability of the plant-specific 
Probabilistic Risk Assessment (PRA) to support the categorization 
process.
    (3) Use of the PRA to determine the relative importance of 
modeled SSCs to accident prevention and mitigation.
    (4) Use of an integrated decision-making panel (IDP) to 
determine the safety significance of SSCs. The categorization of 
SSCs as either safety significant or low safety significant must 
include considerations of:
    a. Results of the PRA importance evaluation.
    b. Deterministic and other traditional engineering analyses.
    c. Maintenance of the defense-in-depth philosophy.
    d. Maintenance of safety margins.

    (5) Evaluation of the change in risk resulting from 
reclassifying SSCs.
    a. Determination of treatment requirements for SSCs based on 
their initial safety significance categorization.
    b. Evaluation of the overall change in plant risk as a result of 
changes in treatment requirements, and readjustment (if necessary) 
of the categorization of SSCs based on this estimation of change in 
risk.

    (6) Documentation of the process and the decision criteria used 
for the categorization of SSCs.
    (7) Monitoring of the impact of the change in treatment 
requirements.
    The remainder of this appendix discusses requirements and 
decision criteria for the above elements in more detail.

Requirements and Decision Criteria

Element (1): Identification of Current Treatment Requirements for SSCs

    All safety-related as well as non-safety-related SSCs in the 
plant are within the scope of this categorization process. For each 
SSC where changes to the treatment requirements are considered, 
current requirements must be identified and documented so that the 
effect of the changes can be more easily understood.

Element (2): Assessment of the Capability of the PRA to Support the 
Categorization Process

    PRA scope. At a minimum, a PRA modeling the internal initiating 
events at full power operations must be used for SSC importance 
analysis and determination of change in risk from the application. 
The PRA must be capable of quantifying core damage frequency (CDF) 
and large early release frequency (LERF). When categorizing SSCs, 
the licensee shall also consider external event initiators, as well 
as the shutdown and low-power modes of operation, either by PRA 
modeling or by the integrated decision-making process. Element 
(4)(b) discusses the requirements for cases when PRA modeling is not 
available.
    PRA quality. The PRA should conform to the consensus ASME/ANS 
PRA Standard documents as endorsed by the NRC. In addition to the 
technical requirements, the PRA shall conform to the requirements in 
the areas of documentation, configuration control, quality 
assurance, and peer review. Where elements of the Standard are not 
met, justification of why these elements are not important to the 
results must be documented and available for NRC review.
    PRA updates. The PRA must reflect the as-built and as-operated 
plant. When used for SSC categorization, and as long as regulatory 
requirements are being dictated by this categorization, the PRA must 
be updated on a periodic basis, that is, annually or within six 
months after each refueling outage provided the interval between 
successive updates does not exceed 24 months. These updates are 
mandatory before implementation of changes to plant design or 
procedures if these changes affect the categorization of SSCs. A PRA 
update is also required upon receipt of new PRA information which 
would invalidate the results of the categorization process. Upon the 
completion of the PRA update, the SSC categorization shall be 
revisited in accordance with Elements 3 through 5 of this process 
with a focus on the impact of the changes on SSC categorization.

Element (3): Determination of Relative Importance of SSCs Using the PRA

    Relative importances of SSCs modeled in the PRA should be 
determined using PRA importance measures. The results of this 
process together with results of sensitivity studies will be used as 
inputs to the integrated decision-making process for the 
categorization of SSCs.
    Risk metrics and importance measures. SSC importances must be 
determined based on both CDF and LERF. Importance measures should be 
chosen such that results can provide the IDP with information on the 
relative contribution of an SSC to total risk. Examples of 
importance measures that can accomplish this are the Fussell-Vesely 
(F-V) importance and the Risk Reduction Worth (RRW) importance. 
Importance measures should also be used to provide the IDP with 
information on the safety margin available should an SSC fail to 
function. The Risk Achievement Worth (RAW) importance and the 
Birnbaum importance are example measures that are suitable for this 
purpose.
    Screening criteria. Importance measures do not directly relate 
to changes in the absolute value of risk. Therefore, the criteria 
for categorizing SSCs into the safety significant and the low safety 
significant categories shall be based on an assessment of the 
overall impact of SSC re-categorization and a comparison of this 
impact to the acceptance criteria for changes in CDF and LERF, see 
Element (5)(b). However, in the initial screening stages, an SSC 
with F-V 0.005 based on either CDF or LERF, and RAW 2 based on 
either CDF or LERF can be considered as potentially low safety 
significant. Elements 4 and 5 must be carried out to confirm the low 
safety significance of these SSCs.
    Truncation limit. The truncation value used for PRA model 
quantification must be set to a value that is sufficiently low so 
that the resultant minimal cutsets contain the significant 
contributors to risk and that at least 95 percent of the CDF and 
LERF is captured in the final solution.
    Sensitivity analyses. The sensitivity of SSC importances to 
uncertainties in the parameter values for component availability/
reliability and human error probabilities should be

[[Page 11494]]

evaluated. Results of these sensitivity analyses should be provided 
to the IDP for deliberation.
    Combining models for different initiating events and plant 
operating modes. The PRA models for external initiating events 
(e.g., events initiated by fires or earthquakes), and for low power 
and shutdown plant operating modes may be conservative with respect 
to those for internal initiating events. Use of conservative models 
can influence the calculation of importance measures by moving more 
SSCs into the low safety significance category. Therefore, when PRA 
models for external event initiators and for the low power and 
shutdown modes of operation are available, the importance measures 
shall be evaluated for each analysis separately, as well as 
integrally. Results of the analyses should be provided to the IDP 
for deliberation.

Element (4): SSC Categorization by the Integrated Decision-Making Panel

    An integrated decision-making panel, for example, an Expert 
Panel similar to the one used in implementing 10 CFR 50.65, must be 
used to determine the safety significance of SSCs. The 
categorization of SSCs as either safety significant or low safety 
significant must consider: results of the PRA importance analysis; 
deterministic and other traditional engineering analyses; 
maintenance of the defense-in-depth philosophy; and maintenance of 
safety margins. Elements (4)(a) through (4)(d) describe these 
requirements in more detail. Element (6) describes the requirements 
of the IDP process, and the documentation required of this process.

Element (4)(a): Use of PRA Insights

    Results of the PRA importance analysis, including results from 
sensitivity studies, and results from the external initiating events 
and the low power and shutdown modes of operation when available, 
should form the initial inputs to the categorization process:
    (i) For screening, an SSC with F-V  0.005 based on either CDF or 
LERF, and RAW  2 based on either CDF or LERF can be considered as 
potentially low safety significant.
    (ii) Results of sensitivity analyses shall be used to show that 
SSC categorization will not change for the expected range of values 
of SSC reliability/availability and human error probabilities.
    (iii) When PRA models are available, the importance measures for 
external event initiators and for the low power and shutdown mode of 
operation shall be evaluated for each analysis separately, as well 
as integrally, and only when an SSC is low safety significant for 
each of these analyses will it be assigned to the low safety 
significant category.
    Application of the above guidelines will yield a list of SSCs 
that are determined to be safety significant by the PRA. These SSCs 
shall not be re-categorized as low safety significant by the IDP 
process.

Verification of Low Safety Significance for SSCs Implicitly Modeled 
in the PRA

    For SSCs which have not been identified as safety significant by 
PRA importance measures, the IDP must verify that these SSCs are not 
implicitly depended upon in the PRA. The IDP must determine if:
    (i) Failure of the SSC will significantly increase the frequency 
of an initiating event, including those initiating events originally 
screened out in the PRA.
    (ii) Failure of the SSC will fail a safety function, including 
SSCs that are assumed to be inherently reliable in the PRA (e.g., 
piping and tanks) and those that may not be explicitly modeled 
(e.g., room cooling systems, and instrumentation and control 
systems).
    (iii) The SSC supports operator actions credited in the PRA.
    (iv) Failure of the SSC will result in failure of safety 
significant SSCs (e.g., through spatial interactions).
    If any of the above conditions are true, the IDP should use a 
qualitative evaluation process to determine the impact of relaxing 
requirements on SSC reliability and performance. This evaluation 
should include identifying those failure modes for which the failure 
rate may increase, and those for which detection could become more 
difficult. The IDP can justify low safety significance of the SSC by 
demonstrating one or more of the following:
     The reclassification is consistent with the defense-in-
depth philosophy and sufficient safety margin is maintained.
     Relaxing the requirements will have minimal impact on 
the failure rate increase.
     Historical data show that these failure modes are 
unlikely to occur.
     Such failure modes can be detected in a timely fashion.

Element (4)(b): Use of Deterministic and Other Engineering Analyses

    For SSCs identified in Element (4)(a) as low safety significant 
by the PRA as well as those SSCs outside the scope of the PRA, the 
IDP must verify low safety significance based on deterministic and 
other engineering analyses and insights, operational experience, and 
information from licensing basis documents and design basis accident 
analyses.

Initiating Events and Plant Operating Modes not Modeled in the PRA

    When initiating events with frequencies of greater than 
10-6 per year are not modeled in the PRA, or when the low 
power and shutdown plant operating modes are not modeled, the IDP 
shall demonstrate that the relaxation of regulatory requirements 
will not unacceptably degrade plant response capability and will not 
introduce risk vulnerabilities for the unmodeled initiating events 
or plant operating modes. For these unmodeled events, the IDP 
assessment must consider whether an SSC has an impact on the plant's 
capability to:
    (i) Prevent or mitigate accident conditions;
    (ii) Reach and/or maintain safe shutdown conditions;
    (iii) Preserve the reactor coolant system pressure boundary 
integrity;
    (iv) Maintain containment integrity; and
    (v) Allow monitoring of post-accident conditions.
    In determining the importance of SSCs for each of these 
functions, the following factors must be considered:
     Safety function being satisfied by SSC operation.
     Level of redundancy existing at the plant to fulfill 
the SSC's function.
     Ability to recover from a failure of the SSC.
    sbull Performance history of the SSC.
     Use of the SSC in the Emergency Operating Procedures or 
Severe Accident Management Guidelines.
     Cumulative impacts of combinations of SSC 
unavailability which could impact an entire system or critical 
safety function.

Risk Indices Outside the Scope of the PRA

    In addition to being safety significant in terms of CDF and 
LERF, SSCs can also be safety significant in terms of other risk 
metrics. Therefore, when an SSC is not identified as safety 
significant by the PRA, the IDP must verify low safety significance 
by determining if:
    (i) The SSC is a part of a system that acts as a barrier to 
fission product release during severe accidents;
    (ii) The SSC is depended upon in the Emergency Operating 
Procedures or the Severe Accident Management Guidelines; and
    (iii) Failure of the SSC will result in unintentional releases 
of radioactive material even in the absence of severe accident 
conditions.

    If any of the above conditions are true, the IDP should use a 
qualitative evaluation process to determine the impact of relaxing 
requirements on SSC reliability and performance. This evaluation 
should include identifying those failure modes for which the failure 
rate may increase, and those for which detection could become more 
difficult. The IDP can justify low safety significance of the SSC by 
demonstrating one or more of the following:
     The reclassification is consistent with the defense-in-
depth philosophy and sufficient safety margin is maintained.
     Relaxing the requirements will have minimal impact on 
the failure rate increase.
     Historical data show that these failure modes are 
unlikely to occur.
     Such failure modes can be detected in a timely fashion.

Element (4)(c): Maintaining the Defense-in-Depth Philosophy

    When categorizing SSCs as low safety significant, the IDP must 
demonstrate that the defense-in-depth philosophy is maintained. 
Defense-in-depth is considered adequate if the overall redundancy 
and diversity among the plant's systems and barriers is sufficient 
to ensure the risk acceptance guidelines provided in Element (5)(b) 
are met, and that:
     Reasonable balance is preserved among prevention of 
core damage, prevention of containment failure or bypass, and 
mitigation of consequences of an offsite release;
     System redundancy, independence, and diversity is 
preserved commensurate with the expected frequency of challenges, 
consequences of failure of the system, and

[[Page 11495]]

associated uncertainties in determining these parameters;
     There is no over-reliance on programmatic activities 
and operator actions to compensate for weaknesses in the plant 
design; and
     Potential for common cause failures is taken into 
account.

Element (4)(d): Maintenance of Safety Margins

    When categorizing SSCs as low safety significant, the IDP shall 
demonstrate that there is sufficient safety margins to account for 
uncertainty in the engineering analysis and in the supporting data. 
Safety margin shall be incorporated when determining performance 
characteristics and parameters (e.g., component, system, and plant 
capability) or when defining mission success criteria (e.g., the 
number of system trains required to mitigate an initiating event or 
the ability of an SSC to perform in a certain environment). The 
amount of margin should depend on the uncertainty associated with 
the performance parameters in question, the availability of 
alternatives to compensate for adverse performance, and the 
consequences of failure to meet the performance goals. Demonstration 
of available safety margins shall be accomplished by use of data 
from plant operations or research studies, or by use of analyses 
using established engineering codes and standards or NRC-approved 
alternatives.

Element (5): Evaluation of the Change in Risk Resulting from 
Reclassifying SSCs

    The change in risk from reclassifying SSCs shall be quantified. 
Elements (5)(a) and (5)(b) provide the requirements for this 
quantification.

Element (5)(a): Determination of Treatment Requirements Based on Safety 
Significance

    Where regulatory requirements are to be relaxed for SSCs 
categorized as low safety significant or where regulatory 
requirements are increased for SSCs categorized as safety 
significant, the IDP must document the functional requirements for 
the SSCs and describe the process to assure that these requirements 
are preserved. Based on the revised requirements, the IDP must 
document and justify the target SSC reliability and availability.

Element (5)(b): Assessment of the Change in Risk

    The potential impact of relaxing treatment requirements on SSCs 
must be evaluated in an integrated manner. Changes in CDF and LERF 
must be estimated by calculations where the failure likelihood of 
SSCs is changed to the level corresponding to the failure likelihood 
for the revised treatment requirements.
    Changes to CDF and LERF must be small. Plants with total 
baseline CDFs of 10-4 per year or less will be permitted 
CDF increases of 10-5 per year, and plants with total 
baseline CDFs greater than 10-4 per year will be 
permitted CDF increases of 10-6 per year. Plants with 
total baseline LERFs of 10-5 per year or less will be 
permitted LERF increases of 10-6 per year, and plants 
with total baseline LERFs greater than 10-5 per year will 
be permitted LERF increases of 10-7 per year.
    If a PRA model is not available to evaluate the change in risk 
from an external initiating event or plant operating mode, the IDP 
must provide justification, on the basis of bounding analyses or 
qualitative considerations, that the risk will not be significantly 
impacted.
    Subsequent changes to the categorization of SSCs for the purpose 
of further modifying regulatory requirements must be performed in 
such a manner where plant performance and previous changes to the 
licensing basis are taken into account. There must not be a pattern 
of systematic increases in risk as a result of repeated applications 
of the SSC categorization process.

Element (6): Documentation of the Integrated Decision-Making Process 
and the Decision Criteria Used

Requirements of the Integrated Decision-Making Panel

    Plant procedure: The IDP shall be described in a formal plant 
procedure which includes:

    (i) The designated chairman, panel members, and panel 
alternates;
    (ii) Required training and qualifications for the chairman, 
members and alternates;
    (iii) Requirements for a quorum, attendance records, agendas, 
and meeting minutes;
    (iv) The decision-making process;
    (v) Documentation and resolution of differing opinions; and
    (vi) Implementation of feedback/corrective actions.

    Membership: There shall be at least five experts designated as 
members of the IDP. Expertise in the following fields shall be 
represented on the IDP: plant operations, design engineering, 
systems engineering, safety analysis engineering, quality assurance, 
plant licensing, and probabilistic risk assessment. Members may be 
experts in more than one field, however excessive reliance on any 
one member's judgement should be avoided.
    Expertise: The licensee shall establish and document specific 
requirements for ensuing adequate expertise levels of IDP members, 
and shall ensure that expertise levels are maintained. There shall 
be at least three members of the IDP with a minimum of five years 
experience at the plant, and there shall be at least one member of 
the IDP who has worked on the modeling and updating of the plant-
specific PRA for a minimum of five years.
    Training: The IDP shall be trained in the specific technical 
aspects and requirements related to the categorization process. 
Training shall address, at a minimum--

    (i) The purpose of the categorization;
    (ii) Present treatment requirements for SSCs including 
requirements for design basis events;
    (iii) PRA fundamentals;
    (iv) Details of the plant-specific PRA including the modeling 
scope and assumptions;
    (v) The role risk importance measures including the use of 
sensitivity studies;
    (vi) The assessment of SSC failure modes and effects;
    (vii) The role of and the use of risk thresholds; and
    (viii) The defense-in-depth philosophy and requirements to 
maintain this philosophy. Each of these topics must be covered to 
the extent necessary to provide the IDP with a level of knowledge 
sufficient to evaluate and approve SSC categorization using both 
probabilistic and deterministic information.

    Decision-making: IDP decision criteria for categorizing SSCs as 
safety significant or low safety significant shall be documented. 
Decisions of the IDP shall be arrived at by consensus. Differing 
opinions shall be documented and resolved, if possible. If a 
resolution cannot be achieved concerning the safety significance of 
an SSC, then the SSC shall be classified as safety significant.
    Feedback and corrective actions: SSC categorization shall be 
revisited by the IDP when the PRA is updated or when the other 
criteria used by the IDP are affected by changes in plant 
operational data or changes in plant design or plant procedures.

Documentation of the IDP Process

    The following shall be documented and available for NRC review:
     Results of the relative risk importance of SSCs modeled 
in the PRA including the results of sensitivity analyses. This 
should include separate SSC importances for the external events 
initiators and for low power and shutdown operations when these 
events are modeled in the PRA.
     Results of the final SSC categorization including a 
summary of IDP deliberations for each SSC classified as low safety 
significant and each non-safety-related SSC classified as safety 
significant. Decision criteria in terms of qualitative assessments, 
assessments for initiating events and plant operating modes not 
modeled in the PRA, defense-in-depth, and safety margins must be 
included. Technical basis documents used to support the 
categorization shall also be available.
     Functional requirements for each SSC receiving revised 
treatment, the original treatment requirements for these SSCs, the 
revised requirements for these SSCs, target values for SSC 
reliability and availability, and the process that will be used to 
assure these functional requirements and target values will be 
preserved/met.
     The overall change in plant risk as a result of changes 
in treatment requirements, including the baseline CDF and LERF and 
the change in this CDF and LERF. Changes to plant risk from all 
previous changes to treatment requirements shall also be included.
     Requirements for the IDP including, the plant 
procedure, expertise, membership, training, and decision-making 
guidelines. Meeting minutes should also be included.

     The PRA used and the supporting analyses, together with 
a description of conformance of this PRA to the PRA Standards 
documents.

Element (7): Monitoring of the Impact of the Change in Requirements

    A performance monitoring and corrective action program must be 
implemented so that

[[Page 11496]]

early indication of SSC degradation can be obtained, and corrective 
actions can be implemented. This program shall include safety 
significant SSCs and safety-related SSCs classified as low safety-
significant. A mechanism for changing SSC categorization based on 
operating experience must be included in the program. SSC 
performance must be consistent with the level of performance 
allocated in the risk analysis or credited in the integrated 
decision-making process. Monitoring of the safety-significant SSCs 
is expected to be addressed by the Maintenance Rule as described in 
10 CFR 50.65.
    Results of the monitoring program must be documented and 
available for NRC review. Results of the monitoring program must 
also be incorporated into the PRA update process described in 
Element (2).

IV. Issues

A. Selective Implementation

    ``Selective implementation'' is defined as implementing the changes 
resulting from this effort for a subset of the affected special 
treatment requirements or implementing the changes for a subset of SSCs 
at a facility, or both. The NRC is considering the argument that 
selective implementation would tend only to reduce unnecessary 
regulatory burden and would not yield safety benefits where the risk 
importance of SSCs had not been recognized by the current regulatory 
framework. However, selective implementation may be possible and even 
necessary to some degree.
    The South Texas Project experience with the Graded Quality 
Assurance program has demonstrated that implementation of the resulting 
changes for only 10 CFR part 50, Appendix B, is not beneficial from a 
burden reduction perspective without exemptions from other regulations. 
The South Texas Project experience has further shown that 
implementation for a minimum set of rules, in combination with 10 CFR 
part 50, Appendix B, must occur before sufficient benefits are 
realized. The NRC believes that this feedback applies to most of the 
current set of regulations. However, even with the experience that 
South Texas Project had with 10 CFR part 50, Appendix B, the licensee 
did not request exemption from the full set of regulations identified 
as candidates for this effort. In addition, none of the potential pilot 
plant program participants have expressed interest in implementing the 
full set of rules being considered. As a result, the NRC currently 
believes that a sufficient amount of burden reduction can be achieved 
with selective implementation.
    The NRC intends to make rule changes so that exemptions will not be 
required for licensees wishing to implement the risk-informed 
regulatory regime that would result from this effort. Therefore, the 
NRC currently believes that it should not issue exemptions to allow for 
selective implementation after final rulemaking.
    With regard to safety, the NRC believes that, if the exemption 
request submitted by South Texas Project can be found acceptable, the 
NRC would have, in effect, determined that an adequate level of safety 
could be preserved without having to adopt all changes resulting from 
this effort. Therefore, the NRC will depend, in part, on the results of 
the South Texas exemption effort to decide this issue.
    Selective implementation of alternative regulatory treatment 
requirements would introduce additional complexity into the regulatory 
process and the NRC will need to assess the practicality of the 
approach. In addressing this issue, the NRC will need to establish an 
implementation approach which recognizes all of the NRC's outcome 
oriented goals, not just reducing unnecessary regulatory burden. The 
NRC is continuing to evaluate this issue and is seeking stakeholder 
feedback in Section V.F. of this advance notice of proposed rulemaking.
    Another selective implementation issue is whether licensees should 
be allowed to implement the alternative for certain systems and not 
others. The NRC expects that licensees would look at a comprehensive 
set of systems and components as it applies any individual risk-
informed regulation. If a comprehensive scope of equipment is not 
considered, the NRC does not believe that licensees can develop an 
appropriate risk-ranking process or identify risk-significant 
characteristics of equipment which may warrant additional control. For 
example, licensees would be expected to review systems and components 
outside current safety-related boundaries to identify the need for 
additional equipment qualification for risk-significant SSCs at the 
same time that it reviews the current equipment qualification scope for 
relaxation opportunities. The NRC does recognize, however, that 
implementation would take place through a phased approach by licensees.
    The NRC recognizes that licensees may elect to exclude certain 
systems from the detailed risk-ranking process based on their prior 
understanding of the importance of those systems to overall safety. 
Some systems, such as the reactor protection system, can be shown to be 
very important without an extensive risk evaluation. Other systems may 
not be relevant to facility safety at all. Licensees may determine that 
there is little benefit from a detailed risk categorization process for 
such systems. However, to ensure that this effort is implemented 
correctly, such systems may still need evaluation to assess the risk-
significant attributes from a risk-informed perspective.
    The Commission is continuing to evaluate this issue and is seeking 
stakeholder feedback on this issue in Section V.F. of this advance 
notice of proposed rulemaking.

B. Impact on Other Regulations

    The NRC has determined that implementation of risk-informed 
alternatives in Part 50 may affect implementation of other regulations. 
For example, the NRC has determined that changes to Part 54 may be 
required to accommodate license renewal for a facility that had 
implemented risk-informed changes encompassed by this effort. The scope 
of Part 54 is explicitly defined using the traditional deterministic 
approach. Therefore, Part 54 does not, without change, accommodate the 
alternative the risk-informed scope that would result from this effort. 
The goal of the license renewal program is to establish a stable, 
predictable, and efficient license renewal process. The NRC believes 
that a revision to Part 54 at this time would have a significant effect 
on the stability and consistency of the processes being established for 
preparation of license renewal applications and for NRC review. 
Allowing a voluntary alternate scoping criteria would necessitate the 
development of an alternate license renewal process. Guidance would 
need to be developed regarding format and content of a renewal 
application, NRC review criteria, and inspection guidance for 
conducting onsite scoping inspections.
    In other cases, such as operator licensing (Part 55), rule changes 
may not be necessary. Nevertheless, licensees may need to make changes 
to programs implementing these regulations in order to ensure 
compliance.
    The Commission would like to identify all such impacts early in 
this effort and is, therefore, seeking stakeholder input on this issue 
in Section V.G. of this advance notice of proposed rulemaking.

C. Need For Prior NRC Review

    The preferred approach for this effort is to avoid the need for 
prior NRC review and approval of either the licensee's categorization 
process or the results of that process. The Commission intends on 
achieving this by issuing a

[[Page 11497]]

detailed and enforceable appendix which would yield consistent, 
objective, and inspectable results. This appendix is being developed, 
in part, from existing guidance such as RG 1.174 and from experience 
gained by review of the South Texas Graded Quality Assurance 
methodology. Several significant aspects of the proposed categorization 
technique rely upon subjective and qualitative judgement. For example, 
it is expected that an expert panel will consider defense-in-depth and 
margin of safety as part of the assessment of the significance of SSCs. 
However, these terms are often defined only in a qualitative, not 
quantitative, sense. These terms are difficult to translate into 
enforceable regulations yielding consistent, objective, and inspectable 
results. Therefore, use of these concepts within an appendix creates a 
significant challenge to the NRC. If the NRC cannot develop criteria 
which result in consistent, objective, and enforceable results, some 
level of NRC review and approval will be necessary.
    No prior NRC review of a licensee's categorization process may 
affect the public participation process concerning the implementation. 
With no prior NRC review, public participation would be limited to the 
rulemaking process. For example, the public could participate by 
providing input on this advanced notice of proposed rulemaking, on the 
notice of proposed rulemaking, in public meetings, etc. However, public 
participation allowed by the licensing amendment process (i.e., for 
implementation), including hearing rights on the licensing action, 
would not be part of the implementation of this effort because no 
licensing action would need to take place.
    The Commission is seeking comment on this issue in Section V.H. of 
this advance notice of proposed rulemaking.

D. Identification and Control of Attributes Requiring Special Treatment

    The NRC anticipates some SSCs that are not presently subject to 
special treatment requirements to be identified as significant to plant 
safety (i.e, RISC-2 SSCs). The NRC further anticipates to find that the 
existing special treatment requirements do not fully address some risk-
significant characteristics of SSCs that are significant to plant 
safety (RISC-1 and RISC-2 SSCs). This is anticipated to occur because 
the risk-informed categorization processes will address some severe 
accident concerns that are not currently addressed by the special 
treatment requirements. The Commission expects to develop regulatory 
controls for RISC-1 and RISC-2 SSCs to ensure risk-significant 
characteristics of these SSCs are adequately preserved.
    The Commission expects some SSCs that are presently subject to 
special treatment requirements to be identified as being of low 
significance to plant safety (i.e, RISC-3 SSCs). However, it is not the 
intent of this effort to redefine the design basis events that a plant 
must analyze to demonstrate compliance with the regulations. Therefore, 
this effort will not allow for elimination of these components from the 
plant. In addition, these components must remain functional to meet the 
design basis. Accordingly, the Commission expects to develop regulatory 
controls for RISC-3 to ensure that they would be maintained functional.
    The Commission is considering how to identify the risk-significant 
attributes for RISC-1 and RISC-2 SSCs and what regulatory controls to 
establish for them to ensure that they are adequately preserved. The 
Commission is also considering what regulatory controls to establish 
for RISC-3 SSCs to ensure that they would be maintained functional. The 
Commission is seeking comment on this issue in Section V.E. of this 
advance notice of proposed rulemaking.

V. Specific Questions

    Comments, advice, and recommendations on a proposed rule reflecting 
the features presented above and any other pertinent points are invited 
from all interested persons. Particularly, comments and supporting 
reasons are requested on the following questions arranged by topic:

A. Approach

    A.1. If the NRC elects to pursue a phased rulemaking approach, how 
should the rules identified be prioritized/phased?
    A.2. Proceeding with changes to special treatment requirements 
before establishing a risk-informed design basis (establishment of a 
risk-informed design basis is being addressed by a separate task) may 
create inconsistencies between the treatment of SSCs and the functions 
they serve for the deterministic design basis. Are there any 
detrimental effects (licensing or otherwise) associated with changing 
the special treatment requirements before changing the design basis? 
Please provide a discussion of the detrimental effects that you believe 
would result.
    A.3. (a) What should the proposed rule state in order to clearly 
identify the scope of SSCs in each special treatment requirement for 
which the rule provides a regulatory alternative? (b) If the Commission 
should decide to impose alternative requirements to the special 
treatment requirements and/or if the Commission should decide to impose 
risk requirements on RISC-1, RISC-2, and/or RISC-3 SSCs, how should the 
proposed rule be constructed in order to clearly identify the scope of 
SSCs for which the alternative requirements apply?
    A.4. If the Commission should decide to impose alternative 
requirements to the special treatment requirements and/or if the 
Commission should decide to impose risk requirements on RISC-1, RISC-2, 
and/or RISC-3 SSCs, how should the alternative requirements be 
expressed to ensure clarity (please provide examples of how the 
requirements should be phrased)? Should the alternative requirements be 
expressed prescriptively or in a performance-based approach? Should the 
alternative requirements be placed in each specific special treatment 
regulation for which an alternative is being provided, or should the 
alternative requirements be included in the proposed new rule?
    A.5. Please provide an estimate of the expected costs and benefits 
of implementing risk-informed special treatment requirements.
    A.6. Please comment on the benefits of risk-informing 10 CFR 50.36?

B. Screening

    B.1. Are the screening criteria reasonable and have the rules that 
have been evaluated (see the attached Table) been screened correctly 
against the screening criteria? Please provide rule-specific comments 
on reduction of unnecessary burden and the need to modify a rule in 
order to maintain safety (Criterion III).
    B.2. Are there any other rules, in addition to those that have been 
evaluated, that should be considered as part of this effort? Please 
provide specific comments identifying any rules that you belief should 
be considered and the reasons for recommending their inclusion.
    B.3. Are there any rules that have been identified for inclusion 
that should not be included? Please provide specific comments 
identifying those rules and the reasons for recommending their 
exclusion.

C. Categorization Methodology

    C.1. Are the elements identified for the appendix appropriate and 
adequate for establishing a risk-informed process to categorize SSCs 
with respect to their significance to safety?
    C.2. Is the appendix written at a level sufficient to support a no 
prior NRC review approach? Are there specific

[[Page 11498]]

areas that warrant additional requirements?
    C.3. The approach described in this ANPR would define two levels of 
safety significance. Would it be better to define more than two levels? 
For example, South Texas uses a four level approach where they 
categorize equipment as having high safety significance, medium safety 
significance, low safety significance, and no safety significance. 
(Note however, that South Texas is not proposing to apply four 
different types of treatment for the four levels of significance.) What 
are the benefits of using an approach where more than two levels of 
safety significance are defined? Would it be better to define more than 
two levels in this rulemaking?
    C.4. Importance measures are strongly affected by the scope and 
quality of the PRA. For example, incomplete assessments of risk 
contributions from low-power and shutdown operations, fires, and human 
performance will distort the importance rankings. What should be the 
requirements for assuring PRA quality? What should the scope of the PRA 
be in terms of initiating events and plant operating modes? If modeled 
in a PRA, how should the contributions from external event initiators 
and low power and shutdown operating modes be factored into the results 
(taking into account that modeling for these events is usually not as 
complete as that for the internal events)?
    C.5. Even with a full-scope, high quality PRA, the importance 
measures have limitations. How should these limitations be addressed in 
Appendix T? What is the role of sensitivity and uncertainty analyses? 
What is the role of delta risk measures and absolute risk measures?
    C.6. It is essential that the implementation of 10 CFR 50.69 and 
Appendix T be scrutable and auditable. What requirements are needed to 
ensure that this is the case? What documents should be available for 
NRC inspection (e.g., the risk assessment, technical bases documents, 
inputs to and deliberations of the expert panel)? Please provide a 
discussion to support your comments.
    C.7 Does the proposal provide adequate guidance on the use of 
expert judgement in the form of the integrated decision-making panel to 
ensure consistent categorization of SSCs across the industry?

D. Pilot Plant Program

    D.1. How should the pilot plant program be constructed and 
implemented in order to adequately pilot the elements in the appendix?
    D.2. Please comment on the need or lack of need to pilot each of 
the rules affected by this effort.

E. Identification and Control of Special Treatment Attributes

    E.1. How should the special treatment requirements for SSCs that 
are currently safety-related for one reason but found to be safety 
significant for a different reason be modified? Should special 
treatment of safety-related SSCs be modified to address risk-
significant attributes that are identified as a result of a risk-
informed categorization process? If so, how should treatment be 
identified and controlled?
    E.2. What regulatory treatment should be applied to safety-
significant SSCs which are not currently safety-related?
    E.3. Explain whether the design control and procurement 
requirements in Appendices A and B of 10 CFR part 50 should apply to 
safety-significant SSCs which are not currently safety-related (i.e., 
RISC-2 SSCs).
    E.4. (a) Should 10 CFR part 21 requirements be imposed upon vendors 
who supplied safety-related components to licensees who subsequently 
select the new regulatory approach? If not, what regulatory basis would 
there be for not imposing such requirements on those vendors? Would the 
failure to impose Part 21 requirements on such vendors be inconsistent 
with the underlying statutory basis for Part 21, viz., Section 206 of 
the Energy Reorganization Act of 1974, as amended? What regulatory 
provisions are necessary to assure that the underlying purpose of 
Section 206 and 10 CFR part 21 are fulfilled under the alternative 
regulatory approach?
    (b) If such requirements are imposed, what difficulties would such 
vendors experience in fulfilling their Part 21 responsibilities and how 
could these difficulties be addressed in this rulemaking? What specific 
rule provisions are necessary in order to fairly impose Part 21 vendors 
who supply basic components to licensees who at some point decide to 
adopt the alternative approach?
    (c) Discuss whether the alternative regulatory approach, with 
respect to the new categories, is inconsistent with the definition of 
basic component in Section 223.b of the Atomic Energy Act (which 
imposes criminal liabilities for knowing and willful violations of NRC 
rules, regulations orders and license conditions that result, or if 
undetected could have resulted in significant impairment of a ``basic 
component''). If there is an inconsistency, does it have any adverse 
effects on licensees? What rulemaking provisions could eliminate or 
minimize such adverse effects?
    E.5. What regulatory treatment requirements are necessary to ensure 
the functional capabilities of SSCs that are safety-related because of 
the plant's deterministic licensing basis but found to be of low safety 
significance are maintained?
    E.6. To what degree should severe accidents be incorporated into 
the licensing basis under the regulatory effort to risk-inform special 
treatment requirements?

F. Selective Implementation

    F.1. What are the potential advantages and disadvantages of 
selective implementation with regard to selection of rules and 
selection of systems?
    F.2. What bounds should be set on the scope of SSCs evaluated under 
a risk-informed regulatory framework? Should all systems be evaluated, 
or can some subset be considered?
    F.3. What limits should be placed on the set of rules for 
implementation? Should licensees be required to implement all risk-
informed rules? If not, what limitations are appropriate?
    F.4. How can the NRC ensure that additional attention is given to 
risk significant components if selective implementation is allowed?

G. Impact on Other Regulations

    G.1. What regulations may be affected by risk-informed changes to 
special treatment requirements in Part 50 and how are these regulations 
affected?
    G.2. For those licensees implementing the new regulatory approach: 
(a) What, if any, GDC will require exemptions? (b) If exemptions would 
otherwise be necessary, is there a way and/a regulatory basis for the 
rulemaking to exempt, in whole or part, compliance with those GDCs for 
those licensees choosing the alternative regulatory approach?
    G.3. Part 19 currently requires all licensees to post NRC Form 3. 
Would it be more or less confusing if all licensees posted a single, 
NRC-developed Form 3 that covered both licensees who remain with the 
existing regulatory regime as well as licensees that choose the 
alternative regulatory approach; or should an alternative Form 3 be 
developed, with the licensee required to post the applicable Form 
depending upon whether it chose to implement the alternative regulatory 
approach.
    G.4. If a licensee were to adopt the alternative regulatory 
approach, would there be any inconsistency or discrepancy created 
between the term ``operability'' as currently used in technical 
specifications'' limiting conditions for operations (LCOs) and the 
concept of ``functionality'' as

[[Page 11499]]

proposed for SSCs in RISC-3? Please describe any adverse effects in 
detail, and discuss the manner in which these adverse effects can be 
avoided or minimized.
    G.5. What changes should be considered to provide consistency 
between affected regulations and risk-informed scope of special 
treatment?
    G.6. Please comment on the need and appropriateness of applying a 
risk-informed scope to license renewal (i.e., Part 54)?

H. Need for Prior NRC Review

    H.1. Given that the means for public participation for this effort 
is through comment in response to this advanced notice for proposed 
rulemaking and in response to a proposed rulemaking, is there a need to 
have an NRC review process such that there will be additional public 
participation as part of the licensing amendment process?
    H.2. What level of NRC review is appropriate for a facility making 
the transition to a risk-informed regulatory regime?
    H.3. What regulatory controls need to be placed on licensees to 
implement risk-informed changes to special treatment without prior NRC 
approval?
    H.4. Please comment on the need for revising 10 CFR 50.59 to 
facilitate the risk-informed approach?
    The preliminary views expressed in this document may change in 
light of comments received. In any case, there will be another 
opportunity for additional public comment in connection with any 
proposed rule that may be developed by the Commission.
    The authority citation for this document is: 42 U.S.C. 2201; 42 
U.S.C. 5841.

    Dated at Rockville, Maryland, this 25th day of February, 2000.

    For the Nuclear Regulatory Commission.
Annette L. Vietti-Cook,
Secretary of the Commission.
BILLING CODE 7590-01-P

[[Page 11500]]

[GRAPHIC] [TIFF OMITTED] TP03MR00.003


[[Page 11501]]


[GRAPHIC] [TIFF OMITTED] TP03MR00.004


[[Page 11502]]


[GRAPHIC] [TIFF OMITTED] TP03MR00.005


[[Page 11503]]


[GRAPHIC] [TIFF OMITTED] TP03MR00.006


[[Page 11504]]


[GRAPHIC] [TIFF OMITTED] TP03MR00.007


[[Page 11505]]


[FR Doc. 00-5016 Filed 3-2-00; 8:45 am]
BILLING CODE 7590-01-C