10 CFR 50.

69
“RISK-INFORMED CATEGORIZATION AND TREATMENT
OF STRUCTURES, SYSTEMS AND COMPONENTS FOR NUCLEAR
POWER REACTORS”

REGULATORY ANALYSIS
 Table of Contents

I. Statement of Problem and NRC Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -1-

(a) History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -1-
(b) Objective for Rulemaking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -3-

II. Analysis Of Alternative Regulatory Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -3-

(a) No Action or Exemption Alternative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -4-
(b) New 10 CFR 50.2 Definition Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -4-
(c) Expand a § 50.2 Definition or Define a Currently Used Part 50 Term . . . . . . -5-
(d) New Section in Part 50 Approach (10 CFR 50.69) . . . . . . . . . . . . . . . . . . . . . -5-
(e) Categorization Requirements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -6-
(f) Conclusion Regarding Alternative Strategies . . . . . . . . . . . . . . . . . . . . . . . . . -7-

III. Estimate and Evaluation of Values and Impacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . -8-

(a) Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -8-
(b) Impacts to Licensees . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -9-
(c) Impacts to the NRC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -11-
(d) Impacts to Other Stakeholders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -11-
(e) Values of the Rulemaking for NRC, Industry, and Other Stakeholders . . . . -11-
(f) Decision Rationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -14-

IV. Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -15-

V. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -15-

ii
 Regulatory Analysis for §50.69

I. Statement of Problem and NRC Objectives

(a) History

The Nuclear Regulatory Commission (NRC) has established a set of regulatory requirements
for commercial nuclear reactors to ensure that a reactor facility does not impose an undue risk
to the health and safety of the public. The current body of NRC regulations and their
implementation are largely based on a “deterministic” approach. Requirements were devised
on the basis of a defined and analyzed set of events as “design basis events.” This approach
has employed the use of safety margins, operating experience, accident analysis, and
qualitative assessments of risk, as defense-in-depth philosophy. One element of this defense-
in-depth approach is the imposition of special treatment requirements on structures, systems,
and components (SSCs) that are important to safety to provide a reasonable assurance that
such SSCs will continue to function during the postulated design basis conditions. Special
treatment requirements are imposed on nuclear reactor applicants and licensees through a
number of regulations that have been promulgated since the 1960's. These requirements
specify different levels of special treatment requirements for equipment depending on the
specific regulatory concern.

As part of moving the Agency toward a more risk-informed regulatory body, in 1995, the
Commission published a Policy Statement on the Use of Probabilistic Risk Assessment (PRA).
To implement this Commission policy, the staff has developed guidance (Regulatory Guide
(RG) 1.174 “An Approach for using Probabilistic Risk Assessment in Risk-Informed Decisions
on Plant-Specific Changes to the Licensing Basis,” RG 1.175 “Risk-informed Inservice Testing,”
RG 1.176 “Graded Quality Assurance, RG 1.177 “Risk-informed Technical Specifications,” and
RG 1.178 “Risk-informed Inservice Inspection”) on the use of risk information for reactor license
amendments. In this respect, the Commission has been successful in developing and
implementing a regulatory means for considering risk insights into the current regulatory
framework. One such risk-informed application, the South Texas Project (STP) submittal on
graded quality assurance, is particularly noteworthy.

In March 1996, STP Nuclear Operating Company (STPNOC) requested that the NRC approve
a revised Operations Quality Assurance Program (OQAP) that incorporated the methodology
for grading quality assurance (QA) based on PRA insights. The STP graded QA proposal was
an extension of the existing regulatory framework. Specifically, the STP approach continued
to use the traditional safety-related categorization, but allowed for gradation of safety
significance within the “safety-related” categorization (consistent with 10 CFR Part 50
Appendix B) through use of a risk-informed process. Following extensive discussions with the
licensee and substantial review, the staff approved the proposed revision to the OQAP on
November 6, 1997. Subsequent to NRC’s approval, STPNOC identified implementation
difficulties associated with the graded QA program. Despite the reduced QA requirement
applied for a large number of SSCs in which the licensee judged to be of low safety
significance, other regulatory requirements such as environmental qualification, the American
Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel Code, or seismic
continue to impose substantial burdens. As a result, the replacement of such a low safety-
significant component needs to also satisfy other special requirements during a procurement
process. These requirements prevented STPNOC from realizing the full potential reduction in
unnecessary regulatory burden for SSCs judged to have little or no safety importance. In an
 -2-

effort to achieve the full benefit of the graded QA program (and in fact go beyond the staff’s
previous approval of graded QA), STPNOC submitted a request, dated July 13, 1999, asking
for an exemption from the scope of numerous special treatment regulations (including 10 CFR
50 Appendix B) for SSCs categorized as low safety-significant or as non-risk-significant.
STPNOC’s exemption was ultimately approved by the staff in August 2001.

Under Option 2 of 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 recommended that risk-informed approaches to the application of special treatment
requirements be developed as one application of risk-informed regulatory changes. Option 2
(also referred to as RIP50 Option 2) addresses the implementation of changes to the scope of
SSCs needing special treatment while still providing assurance that the SSCs will perform their
design functions. Changes to the requirements pertaining to the design of the plant or the
design basis accidents are not included in Option 2. These technical risk-informed changes
are addressed under Option 3 of SECY-98-300. The Commission approved proceeding with
Option 2 in a staff requirements memorandum (SRM) dated June 8, 1999.

The stated purpose of the “Option 2" rulemaking was to develop a voluntary, alternative
regulatory framework that enables licensees, using a risk-informed process for categorizing
SSCs according to their safety significance (i.e., a decision that considers both traditional
deterministic insights and risk insights), to reduce unnecessary regulatory burden for SSCs of
low safety significance by removing these SSCs from the scope of special treatment
requirements. As part of this process, those SSCs found to be of risk-significance would be
brought under a greater degree of regulatory control through the requirements being added to
the rule designed to maintain consistency between actual performance and the performance
considered in the assessment process that determines their significance. As a result, both the
NRC staff and industry should be able to better focus their resources on regulatory issues of
greater safety significance.

The Commission directed the staff to evaluate strategies to make the scope of the
nuclear power reactor regulations that impose special treatment risk-informed. SECY-99-256,
“Rulemaking Plan for Risk-Informing Special Treatment Requirements,” dated
October 29, 1999, was sent to the Commission to obtain approval for a rulemaking plan and
issuance of an Advance Notice of Proposed Rulemaking (ANPR). By SRM dated January 31,
2000, the Commission approved publication of the ANPR and approved the rulemaking plan.
The ANPR was published in the Federal Register on March 3, 2000 (65 FR 11488) for a 75-
day comment period, which ended on May 17, 2000. In the rulemaking plan, the NRC
proposed to create a new section within Part 50, referred to as § 50.69, to contain these
alternative requirements.

The Commission received more than 200 comments in response to the ANPR. The
staff sent the Commission SECY-00-194 “Risk-Informing Special Treatment Requirements,”
dated September 7, 2000, which provided the staff’s preliminary views on the ANPR
comments and additional thoughts on the preliminary regulatory framework for implementing a
rule to revise the scope of special treatment requirements for SSCs. The comments from the
ANPR are further discussed in Section IV.1.0 of SECY-02-0176 “Proposed Rulemaking to Add
New Section 10 CFR 50.69, ‘Risk-Informed Categorization and Treatment of Structures,
Systems, and Components’ ” dated September 30, 2002 (ADAMS accession number
ML022630007).
 -3-

The staff prepared a proposed rule package and provided it to the Commission in SECY-02-
176 dated September 30, 2002. The Commission approved issuance of proposed 10 CFR
50.69 for public comment in an SRM dated March 28, 2003. Consistent with Commission
direction, the staff subsequently published proposed 10 CFR 50.69 for public comment in the
Federal Register on May 16, 2003 (68 FR 26511). The Commission received 26 sets of
comments comprising more than 250 individual comments in response to the proposed rule.
The comments are discussed in section II of the final rule Federal Register notice.

(b) Objective for Rulemaking

As discussed above, the current scope of SSCs covered by the special treatment
requirements governing commercial nuclear reactors is deterministically based and stems
primarily from the evaluation of design basis events, as described in updated final safety
analysis reports (UFSARs). This regulatory framework provides reasonable assurance of
adequate protection (no undue risk) to the health and safety of the public. However, advances
in technology, coupled with operating reactor experience, have suggested that an alternative
approach, one that continues to maintain reasonable assurance of public health and safety
and common defense and security, while reducing unnecessary regulatory burden, is possible
and the utilization of such an approach could increase regulatory effectiveness. The new
approach embodied in the rule uses a risk-informed process to evaluate the safety significance
of SSCs and establish the appropriate level of special treatment requirements for SSCs. It is
important to note that this rule is a voluntary rule (it is not being imposed on licensees) that is
intended only to ensure that the scope of special treatment requirements imposed on SSCs is
risk-informed. The rule, however, does not allow SSC functional requirements to be
eliminated, or to allow equipment that is required by the deterministic design basis to be
removed from the facility. Instead, by restructuring the regulations to allow an alternative risk-
informed approach to special treatment, this rule enables licensees who elect to implement its
provisions, and the NRC, to focus its resources on SSCs with significant contributions to plant
safety. Conversely, for SSCs that do not significantly contribute to plant safety, this approach
maintains SSC functionality, albeit at a reduced level of assurance.

II. Analysis Of Alternative Regulatory Strategies

A number of rulemaking strategies were considered for risk-informing the special treatment
requirements. Those strategies considered most viable were evaluated in the rulemaking plan
attached to SECY-99-256. The evaluation of those strategies has been updated based on
additional information obtained since the issuance of SECY-99-256. The updated discussion
is provided below. The NRC continues to conclude that adding a new section to
10 CFR Part 50 is the appropriate approach for risk-informing the special treatment
requirements and hence this is the approach taken for the rule. However, a significant change
regarding the regulatory approach is being taken for the rule in lieu of what was concluded in
SECY-99-256. As discussed below (in section II.f) and as a result of additional interactions
with stakeholders, the NRC no longer concludes that the best regulatory approach is to include
an appendix to 10 CFR Part 50 that provides categorization requirements as part of the
regulatory approach.

Alternative regulatory approaches for risk-informing special treatment requirements were

discussed in the ANPR (attached to SECY-99-256). For example, the NRC discussed use of
exemptions if only a limited number of plants were interested in this approach, as well as
several variations for proceeding with rulemaking (e.g., including within each special treatment
requirement any alternative requirements). The NRC did not receive ANPR comments that
 -4-

disagreed with the NRC’s suggested approach to add a new section to Part 50. However,
negative comments from stakeholders were received with regards to the use of a detailed
appendix (i.e., Appendix T) to support a proposed new CFR section (refer to section II.f for a
summary of these ANPR comments).

(a) No Action or Exemption Alternative

This alternative describes what would occur without the rule. One way to risk-inform special
treatment requirements is to do it without a rulemaking, and instead to process exemptions per
10 CFR 50.12. Such an approach was followed for STPNOC who filed an exemption request
(from numerous special treatment requirements) which was then subsequently reviewed and
approved by the NRC. This exemption review for STPNOC was a “proof-of-concept” of the
categorization and adjustment in special treatment concept. While other plant-specific
exemptions could be processed, when there is sufficient industry interest, rulemaking is the
most efficient means for implementing the type of generic changes encompassed by this
effort. Rulemaking, when compared to the exemption process, also provides an opportunity
for input from all stakeholders about the requirements that the NRC is considering to
promulgate for the contemplated risk-informed process. If only a small number of facilities are
interested in risk-informing special treatment requirements, then review and approval of a
limited number of exemptions under 10 CFR 50.12 would probably be more efficient. Based
on the industry’s response to the ANPR, and subsequent industry participation in this
rulemaking effort to date, the NRC continues to conclude that there is sufficient industry
interest in this initiative to warrant the NRC continuing to expend its resources to develop the
rule. For these reasons, the NRC did not choose this alternative.

(b) New 10 CFR 50.2 Definition Approach

This alternative rulemaking approach would entail the development and incorporation of a new
definition into 10 CFR 50.2. This new definition (e.g., define a new term such as “safety-
significant”) would describe, for the purposes of the special treatment requirements within
Part 50, which SSCs are safety-significant and, therefore, need to be within the scope of the
special treatment requirements. To implement this approach, this new term would need to be
incorporated into each special treatment rule, thereby enabling the scope of these special
treatment rules to be revised per the new definition such that SSCs that are not “safety-
significant” would no longer be subject to the special treatment provisions of the applicable
rules. Licensees could voluntarily revise the scope of SSCs that are subject to special
treatment requirements by implementing a risk-informed categorization process that
determines which SSCs are safety-significant. To determine which SSCs are safety-
significant, the Commission could issue a new Part 50 appendix or new section that contains
the requirements governing the categorization of SSCs, or alternatively, a regulatory guide
could be issued that contains the SSC categorization guidance.

A significant problem with this approach is that unless new requirements are placed into
Part 50 to address the low safety significant SSCs (no longer subject to special treatment
requirements) and ensure that their design basis functions are maintained (once special
treatment is removed from these SSCs), the design basis functional capability could be lost.
This is not consistent with the ground rules for this rulemaking. For this approach to work and
meet the rulemaking objective of preserving the design basis, it would appear that these
additional requirements (to maintain the design basis functions for low safety significant SSCs)
would either need to be incorporated into each and every special treatment requirements
 -5-

section, or be incorporated into a separate section. In this later case this approach becomes
very similar to the approach selected for the rule. This rulemaking alternative appears to
require duplicate changes to multiple rules, and it is less coherent when compared to an
approach that combines all the relevant requirements into one section. Relative to the
preferred rule change, this alternative is disadvantaged because the changes to NRC
regulations are more extensive, and thus the NRC rulemaking effort should be more time
consuming and costly than the recommended alternative. In addition, as a vehicle to risk-
inform the regulations, it is likely to be less effective than the recommended rule change
because it would be more confusing and less coherent to licensees. For these reasons, the
NRC did not choose this alternative.

(c) Expand a § 50.2 Definition or Define a Currently Used Part 50 Term

This alternative is a variation of the approach just described above, but instead of using new
language (define a new term in § 50.2), it would expand the definition of a currently defined
term such as “safety-related,” or it could define another term currently used (but not defined) in
Part 50 such as “important to safety.” This approach has one advantage over the “new
definition” approach discussed above such that this approach uses the same terminology as
already exists in each of the special treatment requirements. Therefore, it would not be
necessary to change the language in any of the special treatment rules. However, a significant
effort would be required to review all the regulations to ensure that inadvertent revisions to any
non-special treatment rules will not occur and to make appropriate changes to preclude such
occurrences. In a similar fashion to the “new term” approach, this consideration would also
need to be supplemented with a new Part 50 appendix or section that contains the
requirements governing the risk-informed categorization of SSCs. This approach has the
problem of the previously described approach (new definition approach) in that a separate
section would be required to contain the requirements needed to maintain the design basis of
SSCs removed from the scope of special treatment requirements (in which case this approach
becomes very similar to the approach selected for the rule) or the requirements would need to
be incorporated into each and every special treatment requirements section. 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. For these reasons, the
NRC did not choose this alternative.

(d) New Section in Part 50 Approach (10 CFR 50.69)

This alternative rulemaking approach is the approach taken for the rule, and entails the
development of a new rule that would be added to Part 50 that licensees could voluntarily
adopt. The rule contains the categorization requirements (supported by a regulatory guide).
Additionally, the rule contains the new “treatment” requirements that apply to SSCs based on
their associated risk-informed safety class (RISC) categorization.

The “new rule section” approach embodied in 10 CFR 50.69 has the benefit of grouping and
integrating all the risk-informed requirements into one rule. This contributes to regulatory clarity
and makes it easier for both licensees and the staff to implement the regulation (as opposed to
having risk-informed requirements incorporated into each regulation). Additionally, the “new
section” rule approach enables the NRC to identify in one place what the regulatory treatment
 -6-

requirements will be for each risk-informed safety class (RISC)1. RISC-1 and RISC-2 SSCs will
continue to meet applicable special treatment requirements and will also have requirements
that ensure that key categorization assumptions that relate to credited performance in beyond
design basis scenarios are technically valid, and updated consistent with the process feedback
requirements in the rule. RISC-3 SSCs will have requirements that maintain with reasonable
confidence the capability of performing their safety-related functions under design basis
conditions. RISC-4 SSCs will be removed from any applicable special treatment requirements
and have no additional requirements imposed by § 50.69 (recognizing that any
technical/functional requirements continue to apply unless they are changed via the normal
design change process including § 50.59). This approach of utilizing a separate section in
Part 50 to contain the revised special treatment requirements is a significantly more coherent
regulatory approach than any of the previously described approaches. Revising each specific
special treatment rule, as suggested by the other alternatives, would be more difficult and
confusing because it involves changing each specific regulation, and in addition each of these
specific special treatment requirements would need to be modified to address RISC-2 and
RISC-3 SSCs. Given that these requirements were structured for "design basis" events, this
would be a difficult task. In the case of RISC-2 SSCs, this would mean revising the current
Part 50 regulations which have a design basis focus to address SSCs that are important for
beyond design basis events. In the case of RISC-3 SSCs, this would mean revising the current
Part 50 regulations with respect to the special treatment requirements, and replacing these
requirements with similar, but less extensive treatment requirements. The potential for
increased confusion is significant for such an approach. Further, since 10 CFR 50.69 is a
voluntary alternative to existing requirements, changing the individual sections could potentially
be confusing for those licensees who elect not to implement the new alternative requirements.
These considerations led to the decision to develop a separate section to contain the new
requirements. As already noted, the stakeholder comments agreed with this portion of the
suggested regulatory approach.

(e) Categorization Requirements

The NRC considered two alternative approaches for incorporating the categorization
requirements into the new regulatory framework: 1) adopting a rule (i.e., a new appendix) that
sets forth in significant detail, objective, nondiscretionary criteria governing the categorization
that licensees could implement without prior NRC review and approval or 2) placing higher-
level, less-detailed categorization requirements in the rule with the need for NRC to review and
approve a submittal prior to implementation of § 50.69.

Incorporating the categorization requirements into an appendix, such that a no prior review
approach could be pursued, would require the appendix to contain a sufficient level of detailed
requirements such that the NRC would be able to determine, in an objective, non-discretionary
manner involving no substantial professional judgement on the part of NRC staff reviewers, that
a § 50.69 licensee complies with the appendix categorization requirements. This “appendix”

1
Safety-related SSCs that a risk-informed categorization process determines are significant contributors to
plant safety are termed RISC-1 SSCs. Nonsafety-related SSCs that the risk-informed categorization determines to
be significant contributors to plant safety are termed RISC-2 SSCs. Safety-related SSCs that a risk-informed
categorization process determines are not significant contributors are termed RISC-3 SSCs. Finally, nonsafety-
related SSCs that a risk-informed categorization process determines are not significant contributors to plant safety
are termed RISC-4 SSCs.
 -7-

regulatory approach was the approach the NRC originally concluded was the best approach
(see SECY-99-256 and the ANPR). This approach appears to have the following advantages:

 Provides a stable and predictable regulatory framework.

 Reduces and potentially eliminates NRC and industry resources that would be expended
on a submittal and associated review.

 Simplifies inspection and enforcement.

The disadvantages of the appendix approach were pointed out in the ANPR comments as
follow:

 Incorporating detailed requirements into the regulations can, and has in the past
(e.g., Appendix R), resulted in numerous exemption requests from licensees who wish
to pursue alternative approaches. The review and approval of these exemption
requests is very resource intensive.

 Incorporating detailed requirements into the regulations stifle new creative approaches
(i.e., forces licensees to pursue exemption requests for alternatives which can be costly)
and ultimately can cause licensees to not pursue these new creative approaches, which
may be technically superior.

 It appears that there would be a need for the NRC to review some aspects of the PRA to
determine its acceptability for application to § 50.69 under any circumstance. As such,
a true “no-prior-review” type of approach simply does not appear to be technically
feasible at this time. As a result, some level of prior review and approval appears to be
needed, and this in turn removes much of the attractiveness of contemplated Appendix
T “no prior review” approach.

As evidenced in the ANPR comments, stakeholders generally did not support the detailed
appendix approach. In response to the proposed rule, two comments were provided on this
issue. Both comments were from industry owners groups, and both did not support putting
more detailed categorization requirements back into the regulatory framework. Since this is a
voluntary rulemaking initiative, and since it was clear that industry would not utilize the appendix
approach, it was not appropriate, nor an efficient use of NRC resources, to continue to develop
the appendix approach. Accordingly, the NRC elected to incorporate less detailed
categorization requirements into the rule, and to require licensees to provide a license
amendment submittal for NRC review and approve prior to implementation of § 50.69. This
approach (regarding the incorporation of more high level categorization requirements into the
rule versus a detailed appendix) is supported by industry based on the comments on the
proposed rule.

(f) Conclusion Regarding Alternative Strategies

The NRC concludes that:

1. Rulemaking is the most effective tool for implementing the type of generic changes
encompassed by this effort. This conclusion is based on the industry’s continued support for
the rulemaking as evidenced with their ongoing development of implementation guidance, their
 -8-

continued participation in pilot activities, and the fact that no major impediments to wide industry
use were identified in the proposed rule comments.

2. Adding a new section to Part 50 that contains the necessary requirements, but without a
supporting appendix as initially suggested in the ANPR, is the best approach for rulemaking.
The final rule reflects this decision.

III. Estimate and Evaluation of Values and Impacts

(a) Overview

The NRC’s chief concern in moving forward with this regulatory approach is ensuring that
sufficient requirements have been incorporated into the new regulation to maintain adequate
protection of public health and safety (please refer to section III of the statement of
considerations supporting the rule for a discussion of the technical basis for § 50.69). Once the
NRC has satisfied itself that the new regulation will maintain adequate protection, then the
NRC’s next concern is whether the regulatory approach is cost-beneficial. Since
implementation of this rulemaking is voluntary, it is not in the NRC’s interest to continue
developing a regulatory approach that would not be adopted by industry. Hence from this
perspective, the NRC’s interest in estimating the values and impacts of the regulatory approach
is to determine whether the approach is likely to prove cost-beneficial. If the approach should
prove not to be cost-beneficial, then the NRC will not expend additional resources on
development of the rulemaking since it would not be utilized by industry.

Available cost information has been utilized in this regulatory analysis. However, some of this
analysis is qualitative with regard to the potential values and impacts of the rulemaking. It is not
possible to develop a more quantitative regulatory analysis that has a reasonable level of
certainty for this rulemaking. The NRC requested cost and benefit information as part of the
ANPR, but did not receive the requested information. However, the nuclear power industry,
through the efforts of the Westinghouse Owners Group (WOG), was able to generate some
cost and benefit information as a result of a detailed examination of the costs and benefits for
implementing § 50.69 based on its understanding of § 50.69 (then in draft form). This
information has been incorporated into this regulatory analysis. No additional information
involving costs and benefits for implementing § 50.69 was received in response to the proposed
rule request for comment, and as a consequence, there has been no adjustments made to the
previous estimates for costs and benefits that were provided in the draft regulatory analysis that
supported proposed § 50.69.

It should be recognized that the costs and benefits of implementing § 50.69 will vary widely for
licensees dependent on facility design, vintage, and licensing history. A further complicating
factor is that § 50.69 is really a “process approval.” Licensees will not know the actual cost
savings until they begin implementing the new process (categorizing SSCs, revising treatment,
replacing SSCs) at their facilities. As a result, the only facility that has developed real cost
information is South Texas (whose exemption request was approved in August 2001). South
Texas represents the bounding, cost benefit situation since the facility has the greatest potential
to realize the greatest cost savings from risk-informing special treatment requirements. South
Texas is a more recent facility, with a complex design (three train), large safety-related
equipment list (i.e., list of equipment which receives special treatment), and a large number of
applicable regulations. However, some cost benefit information was provided by Dominion from
their Surry pilot activities. This information is incorporated into this regulatory analysis.
 -9-

Additionally, based on the § 50.69 categorization pilot efforts, the staff developed rough
estimates of the costs (in terms of days and number of people involved) associated with
categorizing SSCs on a system basis.

In addition to facility design, vintage, and licensing history, the specific issues addressed below
(as impacts) will also influence whether § 50.69 is a cost beneficial endeavor for licensees.

(b) Impacts to Licensees

Licensees that wish to implement § 50.69 will, at a minimum, incur the following impacts:

 PRA: The licensee will need to address PRA quality issues. At a minimum
licensees will need to have a PRA that reflects the current plant configuration, is
sufficiently complete for the intended application, meets a quality standard
(RG 1.200), and is up- to-date. Depending on the state of the licensee’s PRA,
this activity could involve a significant commitment in resources. NRC notes that
many licensees have already made investments in development of a PRA and
having the PRA peer- reviewed for use in various applications, such as
implementation of section 50.65(a)(4). Those licensees who choose to
implement this risk-informed alternative would be likely to already have incurred
many of these costs, and would be interested in additional opportunities for using
the PRA. Another key factor is the NRC’s requirements for submittal of PRA
information and the resultant level of resources that § 50.69 licensees need to
expend to provide the requested information (i.e., the effort to address the NRC’s
issues associated with NEI 00-02).

 Infrastructure for Categorization: The licensee will need to develop the

infrastructure to support the risk-informed categorization of SSCs to determine
safety significance. At a minimum, this involves the development of procedures
governing the risk-informed SSC categorization process (e.g., for Palo Verde’s
pilot activities, procedure 70DP-0RA04 “Component Risk Significance
Determination” was developed based on the NEI 00-04 guidance), establishment
of the integrated decision-making panel (IDP), training of the IDP, and
establishment of a supporting working group that provides the IDP with the
relevant information to enable the IDP to make the categorization decisions.
Some of this infrastructure may already exist from previous categorization efforts
to meet maintenance rule monitoring and for other purposes (e.g., risk-informed
Inservice Inspection (ISI) applications may have categorized the passive
components in the system). Training, based on the pilot experience, is
estimated to take at least one day for the IDP members. This training would be
to familiarize the IDP with the PRA and the IDP decision-making process.

 Performing the Categorization: The licensee will need to expend significant

resources in evaluating the SSCs to determine safety significance, both for the
working group to complete the initial work of developing and gathering the
relevant information on SSC/function significance and for the IDP to convene
and make the decision regarding SSC categorization. This will be an ongoing
cost and it is a function of the number of systems the licensee decides to
 -10-

categorize. Based on the pilot experience, it is estimated that the working group
(estimated to be three people at a minimum) would need to spend about two
weeks developing and preparing the information for presentation to the IDP. It is
estimated that the IDP (estimated to be 5 members plus the 3 working group
presenters) would need to spend an average of 3 days per system reviewing the
information and making the categorization decisions. For less-complicated
systems, these numbers would be much less, while for more involved systems,
the estimates increase. Also, it is expected that over time, the process would
become much more efficient, and these costs probably can be reduced,
particularly if efficiencies are identified for categorizing groups of components.

 Implementation of § 50.69 Revised Treatment: Following categorization, the

licensee will incur impacts that result from revised treatment. These include
changes to 1) plant procedures to implement the revised approach
(e.g., changes to procedures governing procurement, receipt inspection, testing),
2) equipment specifications, 3) plant data bases, and 4) training of plant
personnel to implement the revised approach.

 Monitoring: To implement § 50.69, licensees will incur impacts that result from
additional monitoring activities. It is expected that current maintenance rule
monitoring efforts will address a significant portion of the § 50.69 monitoring
requirements for RISC-1 and RISC-2 SSCs. However, monitoring activities must
be expanded to consider all SSC performance issues (in addition to maintenance
related issues) and include RISC-2 SSCs that may fall outside the scope of the
maintenance rule in order to meet § 50.69 monitoring requirements. From a
practical standpoint, licensees typically evaluate all failures for maintenance rule
impact, therefore expanding the monitoring scope to consider failures other than
those that are maintenance related can be readily addressed by current
programs. Additionally, a level of monitoring is needed for RISC-3 SSCs to
ensure that the condition and performance of SSCs is consistent with
categorization sensitivity studies, and that design basis functions are being
maintained per § 50.69(d)(2).

 Updating: To implement § 50.69, licensees will incur impacts that result from the
need to periodically (every other refueling outage) update the PRA and
categorization process to reflect the data collected from plant monitoring, or from
industry, and to reflect any changes to plant configuration that impact
categorization. Licensees have already developed much of this infrastructure in
order to comply with the PRA quality guidance being implemented in support of
the maintenance rule.

 Submittal Review and Approval: Licensees will incur an impact resulting from the
need for the NRC staff to review and approve a submittal prior to implementing
§ 50.69. This impact includes the licensee’s effort to develop a § 50.69
submittal, and the impact from the staff’s review of the submittal including the
need to support any requests for additional information from the staff.

The Westinghouse Owners Group (WOG) estimates that the total cost for implementation of
§ 50.69 at a single unit site is $2,400,000. For a dual-unit site, with identical plants, the costs
 -11-

are estimated at $3,300,000. These are the total costs for program development,
implementation and maintenance, and these costs include both utility and contractor support.
All of the above costs are included within this estimate. Additionally, these costs were
estimated for the categorization of 12 systems, and were assumed to occur over a three year
period.

(c) Impacts to the NRC

 NRC would expend resources to review and approve § 50.69 submittals. If

licensees adopt the NEI 00-04 guidance as endorsed by the RG 1.201, then
review costs will be minimized (and this is the objective of this effort concerning
the development of implementation guidance). This review effort will focus on
the results of the PRA peer review, and the licensee’s disposition of peer review
findings. This impact is therefore a function of the number of licensees who
choose to voluntarily implement § 50.69, the degree to which licensees adopt the
RG (i.e., exceptions will require NRC review), and the number of key peer review
findings (i.e., the size of the submittal). An estimate of this impact is that the
staff will receive four (one per year) § 50.69 submittals and expend 400 staff-
hours on each submittal for a total of 1600 hours of staff review time. This
estimate could vary substantially is significantly more licensees implement
§ 50.69 that the estimate.

 There would also be additional resource impacts for adjusting inspection

guidance or processes to take into account the existence of alternative
requirements, and to perform an audit or inspection at some point in the future
for some licensees following adoption of § 50.69 requirements. The initial effort
to develop the inspection guidance is estimated to take 4-5 person weeks.

(d) Impacts to Other Stakeholders

 The NRC has not identified any impacts upon other stakeholders. Any costs of
implementation will be borne by the licensees. The NRC does not expect
licensees to implement § 50.69 unless they conclude it is cost-beneficial for their
facility.

(e) Values of the Rulemaking for NRC, Industry, and Other Stakeholders

 The NRC concludes that this regulatory approach can be accomplished while
maintaining public health and safety. This rulemaking will allow licensees to
remove RISC-3 and RISC-4 SSCs from the scope of special treatment
requirements. This rulemaking will not allow SSCs to be removed from the
facility, or for the design basis functional requirements of RISC-3 or RISC-4
SSCs to be changed or eliminated (i.e., for RISC-3 SSCs, design basis
functional requirements are to be maintained, albeit at a reduced level of
assurance, and in all cases, licensees must follow existing design change
control requirements if they desire to change an SSC’s design basis). Some
SSCs are expected to be "scoped" into regulatory treatment (i.e., RISC-2
SSCs), and it is possible that these SSCs will receive enhanced attention
thereby increasing the level of assurance that such previous "nonsafety-related"
 -12-

SSCs will perform as expected (i.e., as required by § 50.69(d)(1)). This element

of the rulemaking may contribute to enhancing safety. Importantly, the
regulatory approach will include a "performance-monitoring" element, such that
if the performance of equipment degrades substantially (to the extent that it is
not reasonable to expect that the SSCs can meet functional requirements, or
that the assumptions that supported the SSC categorization are no longer
valid), or if operational experience indicates that an SSC may be more important
to plant safety than previously thought, consideration can be given to revising
the SSCs categorization and associated treatment (as required by § 50.69(e)).

 As an indication of the potential savings that could be achieved through a risk-

informed special treatment approach, the following information was provided by
the licensee for the South Texas Project (STP) during a presentation to the
Advisory Committee on Reactor Safeguards in July 1999. The STP licensee
estimated that full implementation of its exemption request (which addresses
the same set of special treatment requirements specifically involving relief from
§ 50.49; § 50.34 and 10 CFR Part 100; § 50.65; 10 CFR Part 50 Appendix B;
10 CFR Part 50 Appendix J; and 10 CFR Part 21) would result in several million
dollars in savings a year at STP Units 1 and 2. This estimate is judged to be an
upper bound on the potential savings that can be realized by a given licensee
based on STP's unique three-train design, which results in a larger number of
SSCs whose special treatment requirements can be relaxed and based on a
comparison with WOG estimates provided below. Part of the cost savings
would arise if replacement components could be procured with less-prescriptive
(and thus less expensive) quality and administrative impacts.

 Table 1 has some examples of procurement savings for STP that have resulted
from approval of their exemption request (this information comes from a
presentation at the Tenth Annual International Conference on Nuclear
Engineering in Arlington Virginia, from April 14-18, 2002). As of April 2002,
STP had saved an estimated $300,000 in labor and $60,000 in parts as a result
of being able to modify the scopes and frequencies of preventative maintenance
for SSCs categorized as low safety-significant or nonrisk-significant (i.e., the
equivalent of RISC-3 for § 50.69). In addition, STP noted that there are other
less quantifiable benefits, such as reduced outage time (arising from not having
to test certain isolation valves), and greater flexibility in maintenance
(procedures and scheduling). In fact STP is modifying the scope and focus of
post-maintenance testing to streamline the testing for low safety-significant
SSCs while maintaining an adequate level of assurance.
 -13-

Table 1: Some Examples of Procurement Savings for STP

Item Safety-Related Nonsafety-related

Spent Fuel Pool Heat Quoted, safety- Identical commercial guides
Exchanger Outlet Valve flow related/qualified price = = $842 (for two)
guide $34000 (for two)
Generic Purchase of 1" vent $2400/valve $500/valve –total savings for
and drain valves for lot of 100 valves =$190,000
100 valves
Flow switches used in 45 To buy all 45 switches Nonsafety-related cost
applications (18 safety- safety-related costs $1200/switch –changed out
related and 27 nonsafety- $9000/switch every 5 years – by
related. purchasing all commercial
and evaluating life savings
on these switches =
$900,000

 The WOG estimated that the total cost savings for adopting § 50.69 on a per
unit basis per year is approximately $1,100,000. Based on the single unit costs
($2,400,000 incurred over three years) and dual-unit costs ($3,300,000 incurred
over three years) the corresponding payback periods are approximately 2.2
year and 1.5 years respectively. Extending these savings to the entire fleet of
Westinghouse plants (and assuming that all plants implement §50.69 and have
an average licensed-life to 2020 and extended life to 2040), and calculating a
net present value results in the cost savings shown in Table 2. These savings
are significant, and when considered for the entire fleet of 48 Westinghouse
plants could potentially exceed 500 million dollars.

Table 2: WOG Estimate of Cost Savings

Average WOG Plant Single Unit Site Dual Unit Site
Net Present Value Net Present Value
Licensed Life (2020) $6,800,000 $14,800,000
License Renewed (2040) $11,200,000 $23,400,000

 Additional information was provided by Dominion (shown in Table 3) during a public

meeting held on Feb 21, 2002. See the notes for the table for an explanation of the
information provided.
 -14-

Table 3: Procurement Cost Comparison: Safety-Related vs Dedicated vs Nonsafety-related

SSCs For Surry
Item Safety-Related Dedicated2 Nonsafety-related
Relief Valve 1 ½" X2" $11,000 $4400 $3600
Operator (valve) $30,000 $15,000 $9900
Gate Valve 3" SS $7000 $800 $130
Butterfly Valve 36" $36,000 $13000 $9500
Operator (large bore) $70,000 $23,000 $18,000
Check valve $3200 $1000 $320
Ball Valve 2" $3500 $1000 $560
Gate Valve 6" $15,000 $2600 $600
Butterfly valve 20" $30,000 $7000 $5000

Notes:
1. These are estimated procurement savings from actual SSCs (taken from purchase
orders) procured at Surry, an older, Westinghouse designed, 3-loop plant.
2. The information is meant to estimate the potential savings for procuring a similar
component as either safety-related, dedicated (for safety-related application), or
nonsafety-related.

3. This information does not contain the increased cost due to § 50.69 regulations. But
this is estimated to be approximately $50–100 per component.

4. For valves procured as “ASME Section III” valves, it is estimated that the column 1
numbers would be a factor of 1.5 higher.

5. At Surry, the general practice is to “dedicate” safety-related equipment (this should be

obvious from the substantial cost savings that are achieved)

6. Presumably, § 50.69 would enable cost savings for procurement to be similar to

column 3 (close to nonsafety-related SSCs) with some additional costs associated with
application of § 50.69 requirements

(f) Decision Rationale

This regulatory analysis is largely a qualitative analysis of the potential costs and benefits
associated with § 50.69. This is due to the uncertainties that currently exist regarding
implementation, as well as the major factors that can affect the costs and benefits associated
with implementation of the rule (facility design, vintage, and licensing history). However, the
NRC utilized all available cost information to inform the regulatory analysis where the

2
This refers to licensee’s effort to qualify commercial equipment for safety-related applications. Refer to 10
CFR Part 21 for more information.
 -15-

information was available. Because of the voluntary nature of this rule, the NRC is not
attempting to justify implementation on the basis of cost information. With respect to values
and impacts, the decision rationale that the NRC chose is whether there is reasonable
expectation of a favorable value/impact from developing and implementing this rulemaking.
Therefore, with respect to costs, efficiencies will be realized by incurring a one-time rulemaking
cost in lieu of expending many of the repetitive costs of individual exemption requests.
Relative to benefits, rulemaking is also preferable because it will add greater clarity and
certainty to risk-informing the SSCs which in turn is likely to encourage more licensees to
participate than would be the case if they had to rely on the vagaries of successfully receiving
an exemption. Based on the available information, and noting the industry’s continued interest
in pursuing this rulemaking effort, it is the NRC’s judgement that the values (including the cost
savings and other benefits) described above outweigh the identified impacts. It was expected
that better estimates of costs of implementation could be identified by the industry when they
have had a chance to review the proposed rule, supporting SOC, and associated guidance in
detail. However, no comments were provided on the regulatory analysis through the public
input on the proposed rulemaking.

IV. Implementation

NRC is issuing a new rule section that defines the requirements and the process for
transitioning from existing requirements to the new requirements. Implementation guidance
will also be provided that discusses the categorization process requirements. The NRC is
currently reviewing an industry-developed guidance document for categorization. The NRC
plans to endorse the industry guidance document through a regulatory guide.

Section 50.69 requires licensees or applicants who wish to implement the requirements of
§ 50.69 to make a submittal to the NRC for approval of the categorization process prior to
implementation. NRC plans a focused review of the PRA that undergirds the significance
determination as well as of the integrated decision-making process. NRC has prepared review
guidance to assist the staff in reviewing this submittal to determine whether the PRA is
adequate for this application. Under the rulemaking approach, a licensee who implements the
alternative rule requirements would not provide to NRC the actual list of specific SSCs and
their new category per § 50.69 (i.e., RISC-1, RISC-2, RISC-3, RISC-4), nor would the licensee
provide NRC with a description of the revised treatment applied to RISC-3 SSCs. Rather,
NRC will review the categorization process before implementation begins (i.e., process
approval), and following this approach, the licensee would proceed to categorize SSCs and to
implement treatment processes that satisfy the rule requirements over time. Until SSCs are
categorized per § 50.69 (i.e., categorized as RISC-1, RISC-2, RISC-3, or RISC-4 such that the
treatment requirements associated with each category in § 50.69(d) can be applied), existing
requirements remain in effect. NRC oversight of implementation would be through the routine
inspection process.

Given the NRC’s expectations that implementation guidance will be issued in conjunction with
the final rule or shortly thereafter, the NRC expects that the final rule can be made effective
immediately upon publication (or within a reasonably short period of time such as 30 days) in
the Federal Register.

V. Conclusion

The risk-informed approach embodied in this rule for establishing an alternative scope of SSCs
subject to special treatment requirements is a regulatory approach that maintains safety and is
 -16-

consistent with the NRC’s efforts to risk-inform its regulatory activities. The risk-informed
approach 1) is consistent with the defense-in-depth philosophy, 2) provides reasonable
assurance that necessary safety functions will be performed, 3) provides reasonable
confidence that any increases in core damage frequency or large early release frequency (and
therefore risk) are small, 4) is consistent with the safety goal policy statement, and 5) utilizes a
performance measurement strategy. The overall value/impact of the rulemaking has been
examined from a qualitative standpoint, and NRC concludes that the expected benefits
outweigh the expected costs.