ML20024J453
| ML20024J453 | |
| Person / Time | |
|---|---|
| Issue date: | 10/14/1994 |
| From: | Milhoan J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | Tipton T NUCLEAR ENERGY INSTITUTE (FORMERLY NUCLEAR MGMT & |
| References | |
| NUDOCS 9410180219 | |
| Download: ML20024J453 (15) | |
Text
-___-____-_____-_
n na
[o s*g t
NUCLEAR REGULATCP A MTMSiON bj
}
/
- ..+
October 14, 1994
{
Mr. Thomas Tipton
{
Nuclear Energy Institute 4
1776 Eye Street, N. W.
Washington, D.C.
20006
Dear Mr. Tipton:
l l
1 am writing you this letter to address some of the issues that emerged during the last meeting of the NRC and NEl Graded Quality Assurance (QA) Working Groups on September 14, 1994.
Both the NRC and industry have made a substantial resource investment in trying to develop a workable approach to implement graded QA.
After our Steering Group meeting of July 21, 1994, it appeared that the major obstacles had been overcome for moving ahead with the generation of a sufficiently detailed guidance document for use in conducting the pilot project.
- However, your revised guidance document, " Draft Pilot Project Guideline for Implementation of a Graded, Performance-Based Approach to Quality" dated September 1, 1994 was still lacking adequate detail in several areas.
During the meeting of the NRC and NEI Working Groups on September 14, our major concerns with the document were articulated.
The staff also committed to provide you with supplemental guidance for implementing the graded QA pilot initiative. identifies certain portions of the NEI guide that are not appropriate for inclusion in the guide.
For example, the topic of regulatory assessment (i.e., inspection and enforcement) is an NRC function that is not subject to restrictions by the NEI guide.
Though we intend to perform inspections focused on the relatively risk significant issues, the NRC would not be precluded, for example, from examining the determination of component safety significance as low-risk significant and evaluating the graded QA treatment that the low-risk significant plant equipment receives.
Because some uncertainties exist about the appropriateness of some potential violations under graded QA, the staff will establish a peer panel to review proposed Notices of Violations (NOVs) during the pilot effort. provides supplemental information to the NEI guidance document in the areas of the Expert Panel function, the appropriateness of performance monitoring in-lieu of product quality monitoring, activities that should be accomplished as part of the plant corrective action program, and QA commitments.
This information is intended to augment that contained in the NEl guidance document and supersede the NEI information if they are dissimilar.
Additionally, the staff will be providing you with a comprehensive set of comments on your guide in the near future.
These comments are of less significance and chould not detract from the use of the guide during the pilot phase.
However, we will hold further discussions with you regarding these comments as the pilot project progresses.
01% h h RC 9
T. Tipton In addition to the level of detail provided by the NEI guidance document, there are several other issues related to implementing the pilot program that are wd?th noting.
The NEl* guidance document attempts to generically replace existing QA plan commitments without having each licensee properly evaluating whether the individual pilot plant QA plan changes would constitute a reduction in commitment as required by 10CFR50.54(a).
Each pilot licensee will be responsible to assess the intended QA program changes and arrive at i
their own determination in this regard.
The staff review of the revised NEI guidance document also noted that the terminology "non-risk significant" i
continues to be used when referring to safety-related low-risk significant components.
As we stated in the July 21, 1994 meeting, the term non-risk significant is inappropriate.
Prior to receipt of the September 1,1994, version of the NEI guidance j
document, the staff had understood that for high-risk significant plant 4
equipment the current quality program provisions would continue to be applied.
Your most current version of the graded QA guidance implies that the graded approach could be readily expanded at the option of the pilot plant licensees to include the high-risk-significant equipment and functional areas beyond those examined during the pilot phase.
The purpose of performing the pilot project is to establish the adequacy of the graded QA methodology.
Until the lessons learned from the pilot phase are analyzed and necessary revisions are made to the guidance, it would be premature for pilot licensees to extend the 4
application beyond that agreed upon for the pilot project.
The staff recognizes that licensees have flexibility to implement a graded approach in i
an alternative manner that would necessitate additional interaction with the staff.
The staff agrees that the NUMARC 93-01 methodology provides a starting point for guidance in the graded QA area.
However, we see that given the broader scope and complexity of activities covered under graded QA, NUMARC 93-01 would be built upon and supplemcated accordingly.
We feel that Probabilistic Risk Assessment (PRA) insights can play an important role and that they should complement, and not replace, deterministic evaluations to identify the risk significance of structures, systems, and components (SSCs).
In graded QA, the Expert Panel will play a critical role in determining SSC risk significance based on PRA insights and will compensate for PRA limitations.
Originally, we had envisioned that several licensees could initiate the pilot project simultaneously.
Given the level of staff resources available to support this task, we propose that two lead pilot plants use the supplemented methodology to implement the project initially.
If the lead pilot plants desire to implement an alternative approach from the j
supplemented guidance, the staff and pilot plant licensee would need to discuss those alternative approaches to assure a common understanding and acceptability of the planned licensee project.
The lead pilot plants would then proceed with their project.
After the staff has gained confidence that the supplemented guidance is acceptable, and after some revisions to the supplemented guidance that may be necessary based on insights from the initial implementation phases of the lead pilot plants, the remainder of the pilot plants could proceed at that time.
In this manner, the supplemented guidance will be verified and refined in stages to ensure that it contains the requisite level of detail.
At the conclusion of the pilot phase,
i T. Tipton October 14, 1994 0
lessons learned from the seven pilot plant applications would be incorporated into the guidance.
The staff would develop a draft regulatory guide endorsing the guidance for public comment.
The correct determination of high and low-risk significant SSCs is vital to the effective implementation of the graded QA program.
For example, at one plant we understand that the reactor protection system (RPS) is considered to be low risk significant.
This causes us great concern, because the PRA results are predicated on a highly reliable RPS and that high reliability is, in large part, the result of the QA program.
Therefore, in the near-term, if requested by any pilot plant, the staff would agree to review the process for determining risk significance.
The staff is committed to the graded QA project.
We hope that the pilot plant licensees continue to see the mutual benefits that can be derived from implementing the concept.
We remain optimistic that discussions can be initiated in the near future with the lead pilot plants to start the practical application of the supplemented guidance for graded QA.
Please contact Suzanne Black, the NRC Graded QA Working Group Chairwoman at (301) 504-1017 if you have any questions and to arrange subsequent meetings on the topic of graded QA.
James L. Milhoan Deputy Executive Director for Nuclear Reactor Regulation, Regional Operations and Research
Enclosures:
As Stated cc:
Jack Skolds, South Carolina Electric and Gas, Inc.
William Bohlke, Florida Power and Light Daryl Prisby, Commonwealth Edison - Byron Roger 0. Anderson, Northern States Power Mike Meisner, Entergy Operations - Grand Gulf Edwin Froats, Florida Power Corporation Carter Rogers, Arizona Public Service Company Stephen Eisenhart, Virginia Power Ed Rogers, Entergy Operations - ANO DLSTRIBUTION: See next page DOCUMENT NAME:g: FINAL \\NEILTR9.27
- SEE PREVIOUS CONCURRENCE Concurred in draft To receive a copy of this document, indicate in the box: "C" = Copy without enclosures *E" = Copy with enclosures "N"
= No copy 0FFICE
- RPEB:DRIL l
- RPEB:DRILl C:RPEB l
- SPSB:NRR l
- D:0E l
NAME RLatta RGramm SBlack*
WSwenson JLieberman DATE 09/28/94 09/28/94 10/7/94 09/28/94 09/29/94 0FFICE
+0GC l
D:DRIL l
ADT:NRR l
- RES l
DEDR l
NAME JCutchin RLSpessard*
ACThadani*
JCraig JMilhoan DATE 09/28/94 9/28/94 10/08/94 09/30/94 10/n /94 W w cenmwa j
l
T. Tipton lessons learned from the seven pilot plant applications would be incorporated into the guidance.
The staff would develop a draft regulatory guide endors'ing the guidance for public comment.
The correct determination of high and low-risk significant SSCs is vital to the effective implementation of the graded QA program.
For example, at one plant we understand that the reactor protection system (RPS) is considered to be low risk significant.
This causes us great concern, because the PRA results are predicated on a highly reliable RPS and that high reliability is, in large part, the result of the QA program. Therefore, in the near-term, if requested by any pilot plant, the staff would agree to review the process for determining risk significance.
The staff is committed to the graded QA project. We hope that the pilot plant licensees continue to see the mutual benefits that can be derived from implementing the concept.
We remain optimistic that discussions can be initiated in the near future with the lead pilot plants to start the practical application of the supplemented guidance for graded QA.
Please contact Suzanne Black, the NRC Graded QA Working Group Chairwoman at (301) 504-1017 if you have any questions and to arrange subsequent meetings on the topic of graded QA.
James L. Milhoan Deputy Executive Director for Nuclear Reactor Regulation, Regional Operations and Research Enclosures As Stated cc:
Jack Skolds, South Carolina Electric and Gas, Inc.
William Bohlke, Florida Power and Light Daryl Prisby, Commonwealth Edison - Byron Roger 0. Anderson, Northern States Power Mike Meisner, Entergy Operations - Grand Gulf Edwin Froats, Florida Power Corporation Carter Rogers, Arizona Public Service Company Stephen Eisenhart, Virginia Power Ed Rogers, Entergy Operations - AN0
I. Reaulatory Assessment - NRC enforcement policy and the inspection program jurisdiction discussed in the 5th bullet item on page 6 are not suitable topics for inclusion in the graded QA guidance document.
For low-risk significant SSCs, inspectors could legitimately pursue.a notice of violation (NOV) if the circumstances warrant.
In recognition that some uncertainties may exist about the appropriateness of some potential violations under the graded QA project, the_ staff will form a panel comprised of representatives from the Office of Enforcement, Office of Nuclear Reactor Regulation, and regional staff to convene on a periodic basis to review proposed NOVs-issued under the graded QA effort.
This review will be done prior to issuance of the NOV to ensure proper consideration has been given to: safety significance, degree to which the licensee has implemented the graded approach in a technically sound manner, the degree to which existing regulations and license requirements are 1
fulfilled, and a consideration of ramifications on plant performance.
The panel review may also be re-initiated should a. licensee contest an issued NOV.
The scope of the panel review will be limited to functional areas / systems
)
addressed under the graded QA pilot.
At the conclusion of the pilot project, more specific guidance will be developed for inclusion in the NRC inspection i
Reference to regulatory assessment should be deleted from NEI's draft graded.
QA guidance document.
1 II. 10 CFR 50.54(a) - The assertion on page 23 that the graded QA effort for the pilot plant implementation will not constitute a reduction in commitments is not appropriate for NEI to make.
Each licensee must make that i
determination dependent upon their unique QA plan provisions.
The regulatory requirements associated with QA plan changes are contained in 10 CFR 50.54(a).
A pilot plant licensee may currently have language embodied in its QA plan that allows for a graded application and no QA plan changes may be needed to implement its pilot effort.
However, we request pilot plant licensees that do not envision the need to amend their QA program, to inform the NRC in writing of their intention to employ the supplemented NEI guidance, or alternate approach, to implement their graded QA project, and of the scope of application, either on a functional area (e.g., procurement) or system basis.
Licensees that need to revise their current QA program must evaluate the extent.f the changes to ascertain whether they constitute a reduction in commitment.
If they do not, then 10 CFR 50.54(a) does not require an advance NRC review of the changes.
However, as indicated above, we request that the licensees participating in the pilot project inform the NRC in writing of their intention to employ the supplemented NEI guidance, or alternate approach, to implement their graded QA project and describe.the scope of application.
If a licensee concludes that the' change is a reduction in conoitment, there appear to be two options available.
The first option would be for the licensees to submit a normal QA program change in accordance with 10 CFR 50.54(a) that delineates what elements of the program will be revised ENCLOSURE 1
and a justification for concluding the program will continue to meet the requirements of Appendix B, As outlined above, the licensee should also describe the scope of application for the pilot effort.
This submittal would receive an expedited NRC review to ensure that the start date of the pilot effort is not impacted.
If the changes are found acceptable, the NRC would approve the changes for the pilot effort.
Following completion of the pilot project,'the NRC would expect the licensee to eventually conform with the revised NEI guidance which will incorporate lessons learned from the pilot project.
Licensee programs may need to be revised to account for changes in the NEI guidance.
The second option would be for a pilot project licensee to request a temporary exemption, in accordance with 10 CFR 50.12, to the reporting requirements of 50.54(a). As a basis for requesting the exemption, the licensee would indiciete that the supplemented NEI guidance, or alternate approach, would be utilized in-lieu of the current QA commitments for low-risk significant SSCs.
The licenste should define the scope of application, either on a functional area (e.g., procurement) or system basis.
If the NRC can make a finding that the supplemented NEI guidance, or alternate approach provides an acceptable way to meet the requirements of 10 CFR 50, Appendix B, the NRC would grant a temporary ewmption fr:m the requirements of 50.54(a) for a time period limited to the pilot project on the ground that compliance with the explicit requirements of 10 CFR 50.54(a) is not necessary to achieve the underlying purpose of the rule.
At the conclusion of the pilot project, the NRC would expect the licensee to conform with the revised NEI guidance which incorporates lessons learned from the pilot project phase.
Licensee programs may need to be revised to account for changes in the NEI guidance.
Reference to an advance determination that a generic substitution of existing commitments would not constitute a reduction in commitment should be deleted from NEI's draft graded QA guidance document.
III. Non-Risk Sianificant - The use of the term "non-risk significant" to describe safety related SSCs is misleading.
Referencrs to risk-siguificant and non-risk significant (safety-related) SSCs in NEI's draft graded QA guideline should be-replaced with the terminology "high-risk significant" or
" low-risk significant" respectively. ENCLOSURE 1
1 l
SUPPLEMENTAL GUIDANCE FOR NEl GRADED QA DOCUMENT
\\
l I. Expert Panel Evaluation The nature and magnitude of an.SSC's contribution to plant risk should i
determine the type and amount of QA controls and practices applied to that i
SSC.
An essential element of the graded QA process involves the thorough
'l evaluation of SSC safety functions and risk significance to discern the aspects that should be taken into account when determining what quality elements in 10 CFR 50, Appendix B, to apply in a graded manner.
The Expert Panel described in NUMARC 93-01, and further expanded upon in NUMARC 93-02, "A Report on the Verification and Validation of NUMARC 93-01,"'would have a central role in this process. The panel's role would be two-fold: first, to establish the high-risk significant and low-risk significant SSC categories';
and second, to determine the appropriate application of QA controls for the-SSC or activity under consideration.
i A. Expert Panel Composition The utilization of the Expert Panel, as described in the previously referenced NUMARC documents, provides a basic framework and valuable insights that can be applied to the process of evaluating SSC risk significance for the graded QA process.
Based on the lessons learned documented in NUMARC 93-02, the panel would nominally include experienced representatives from each of the following disciplines:
Operations Maintenance Engineering PRA I
Additionally, given the emphasis on quality considerations in the graded QA process, the following two disciplines should also be included on the panel:
Quality Assurance Procurement B.
Risk Significance Determination The initial function of the Expert Panel would be to evaluate both
. probabilistic and deterministic information available regarding each SSC (or broad classes of functionally similar SSCs) within the defined scope to reach a conclusion as to the risk significance of SSCs.
The_ Expert Panel would need to carefully weigh the PRA insights, including the associated limitations of PRAs, as part of its process to categorize each SSC by risk significance.
)
PRA results should be viewed as a screening mechanism to provide insights about the relative risk significance of SSCs.
The output from the PRA and/or IPE should be augmented with other critical information, such as SSC outage 2
configuration experience, for the panel to reach its conclusions regarding risk significance.
Other sources of input information that the Expert Panel ENCLOSURE 2
may feel appropriate to consider include: design bases documents, design specifications, failure modes and effects analyses, plant operating procedures, normal and abnormal plant configurations and alignments, and plant licensing bases documents.
Risk significance should be determined using a methodology similar to that described in NUMARC 93-01, section 9.3.1, revised to account for non-maintenance related failure modes that will be of importance when applying the PRA approach in the QA area.
NUMARC 93-01 defines risk significance in terms of prevention or mitigation of core damage, containment incegrity, or a reduction in the release probability or consequence to the public, as determined by the PRA/IPE and Expert Panel process.
NUMARC 9L-01 also provides quantitative guidance, including the application of lisk Reduction Worth, to determine which SSCs should be considered risk significant.
Factors such as potential common-mode failures, human errors of omissions and commissions, defense-in-depth, and the maintenance of safety margins would also be considered.
Both high-risk significant and low-risk significant categories could include safety-related and non-safety related SSCs.
In some cases, a deterministically important SSC., one whose failure is important from a standpoint of design, operations, and maintenance considerations might not make a significant contribution to plant risk as modeled by the PRA.
In addition, some of those SSCs may not be explicitly modeled in the PRA, or the PRA boundary assumptions may be constructed to assume that they do not fail.
Thus, a change in QA practices could have significant impact on the continued validity of the boundary assumption.
The risk significance of plant features such as containment and the containment isolation function should be thoroughly evaluated because a strict PRA assessment with respect to core damage frequency would not be representative.
The Expert Panel should assess these factors during its deliberations and document the results.
Some important limitations of PRA that would have to be considered during the Expert Panel deliberations are outlined below:
Plant systems are modeled to varying degrees (explicitly, implicitly, or not at all) in the PRA.
For example, systems such as the nuclear boiler system and remote shutdown may not be explicitly evaluated in typical PRAs.
The fact that an SSC is not modeled does not provide sufficient basis to amend the QA treatment of that SSC.
Further Expert Panel evaluation wtuld be warranted.
The scope of plant PRAs and IPEs may be at the Level 1 analysis that provides estimates of core damage frequency and identifies core damage contributors.
Consideration of external events such as seismic, fire, smoke and soot effects, toxic gas dispersion, high winds and to' nadoes, flooding, and core damage mitigation (e.g., recovery factors, etc.) should also be addressed, i
i ENCLOSURE 2
The level of PRA performance is a determining factor with respect to the degree the results can be utilized.
If the PRA was performed at the system level, then graded QA insights would be derived at the system level.
The failure modes modeled by the PRA may not be all inclusive.
The type of data for equipment failure rates, unavailabilities, and initiating event frequencies may be either plant specific or generic.
If generic data is utilized, an evaluation is warranted to assure the appropriateness of extrapolating the data to the plant specific QA realm.
The modeling of human activities and human errors, including cognitive and comprehension errors, is subject to considerable uncertainties in the PRA.
These can affect dominant accident sequences and dominant contributors to core damage.
Truncated low probability events (below approximately 95 percent of core damage frequency) may affect res lts from the PRA.
For example, the u
reactor pressure vessel is not normally included after truncation yet the reactor pressure vessel obviously plays an important role in assuring plant safety.
Plant specific PRA modeling practices (in contrast to actual design differences) that could skew the plant specific PRA results when compared to the generic population of similar plant PRA results.
Some baseline validation to assure consistency of insights would be warranted.
l The treatment of software driven solid-state control and protection devices are not readily amenable to being analyzed by
)
PRA.
Quantitative results should be treated with considerable l
caution in this area.
l PRAs normally address only 100% power operation.
The effects of partial I
or low power, shutdown, and refueling modes, on plant safety will also need to be considered by the Expert Panel, i
Generally, fault trees are not developed and generic event data is used for modeling the switchyard and emergency diesel generator.
The Expert Panel should exercise caution when extrapolating PRA insights in this area.
Containment performance provisions including containment isolation functions may not be fully modeled.
The Expert Panel should consider this limitation during their deliberations.
Potential influences of aging on component reliability are not evaluated by the PRA.
This may warrant additional evaluation by the Expert Panel. ENCLOSURE 2 wa.
,.4 4
ev.
4e-
,4
..s
-~,
Low-risk significant components not required to support safety function but whose failure could adversely impact safety function performance (i.e., seismic II/I system interaction) may not apppropriately evaluate The Expert Panel evaluation will have to make appropriate by the PRA.
compensations in this area.
Before an SSC would be categorized as low-risk significant, the following screening criteria needs to be reviewed by the Expert Panel:
If a failure of the SSC would not be detectable through performance monitoring methods, it would not be categorized as low-risk significant.
If the SSC is used in multiple applications in the plant and is susceptible to generic or common mode f ailure that could impair redundant or diverse safety systems / trains, it should be carefully In considered before being categorized as low-risk significant.
addition, if a f ailure should occur, the root cause analysis must In the determine whether the failure was random or common mode.
case of a generic or common mcde failure, the loca failure mode must be evaluated, and appropriate corrective action must be taken.
Other regulatory requirements, such as seismic qualification, environmental qualification, applicable codes, e.g., ASME, etc.,
would not be affected by the implementation of the graded QA program.
C.
Quality Element Assessment The following information supplements section 4.1.2 of the NEI guide.
Subsequent to the determination of SSC risk significance using a metho similar to that described in NUMARC 93-01, the selection of appropriate is a quality assurance (QA) requirements for the low-risk s d
safety margins.
The appropriate identification of QA requirements is based on plant sp considerations, in particular, the safety analysis which identifies design basis safety functions and operational parameters.
1.
High-Risk Significant SSCs In a two-tier categorization system (i.e., high-risk significant and low-ris for significant), the current QA practices would be retained by licensees safety-related high-risk significant SSCs.
It is likely that a certain number of SSCs currently classified as non-safe For these SSCs, related will f all into the high-risk significant category.
ENCLOSURE 2 4' 4
4
.3 4.,
4<
4 4.a
the licensee would institute more rigorous quality practices than would normally be done for those risk-significant non-safety-r elated SSCs.
This would entail the application of augmented quality practices for those SSCs in recognition of their relative importance to plant safety.
2.
Low-Risk Significant SSCs The pilot plant licensee would determine the specific nature and extent of the QA controls and practices to be applied to the SSC to, among other things, support an effective root cause analysis and corrective action program.
This determination would include consideration of the safety function of the SSC and non-maintenance related factors such as design, procurement, fabrication, construction, installation, testing, and human f actors issues. As previously noted, all failures would have to be considered for their potential generic or common mode effect on other plant equipment.
The licensee must be able to locate similar equipment that is susceptible to the failure mechanism and consider the effect of these failures on redundant trains or systems.
Adequate corrective actions must be taken.
TV milot.44tJicog.mesvould ensure that regulatory requirements beyond Appendix B germane to the SSC would not be revised by the graded QA program.
Technical provisions of the regulations (e.g., seismic qualification, code requirements) would be maintained while the quality assurance implementation practices would be adjusted.
3.
Consideration of Additional Quality Assurance Requirements for Non-Safety Related SSCs in the past, the NRC has determined that certain categories of non-safety related SSCs warrant some QA program treatment.
This has included station blackout equipment, Anticipated Transient Without Scram (ATWS) equipment, and fire protection equipment.
These decisions did not result in re-classifying these SSCs as safety-related; however, they recognized the increased importance of these SSCs to plant safety.
These additional QA measures should be considered, as appropriate, during the Expert Panel determination of graded QA treatment.
D.
Documentation The licensee should establish adaquate documentation with respect to the assignment of risk significance that integrates with existing QA program documentation and includes:
the qualifications and composition of the Expert Panel performing the risk significance determinations; the basis for these determinations; adequate explanation regarding the application of graded elements such that failures of SSCs can be evaluated to determine if a lessening of administrative controls for QA activities was the cause or contributing factor; and ENCLOSURE 2
4 provisions for revising these determinations based on feedback from performance monitoring and corrective action programs.
II. Performance Monitorina vs. Product Ouality The following information supplements section 4.4.2 of the NEI guide.
The guidelines in NUMARC 93-01 describe a performance monitoring program.
The practices identified in this document could be extended into the graded QA efforts to validate, with respect to some observable parameters, that plant SSCs continue to be able to perform their safety function.
Benefits should be derived from providing operational feedback from a performance monitoring program to ensure the effective implementation of the quality program provisions for safety-significant SSCs.
For the low-risk significant SSCs, a performance monitoring system similar to that described in NUMARC 93-01 would yield beneficial feedback by providing a measure of confidence of the effectiveness of the graded QA provisions.
With respect to graded QA, the performance criteria would take into account all failure modes, not just those associated with maintenance preventable failures.
It would be appropriate for an Expert Panel to delineate the performance criteria based on the SSC (or broad classes of functionally similar SSCs) risk significance and safety function..
For aspects not amenable to verification by testing, performance criteria could be met by condition monitoring (e.g., nondestructive examination, visual inspection, vibration, deflection, thickness, corrosion, or other monitoring methods as appropriate).
In other cases component, train, system, er plant level criteria could be considered by the panel depending upon the particular SSC under evaluation.
The QA program is designed to assure that SSCs will function when called upon to do so.
There will be human issues, such as design and operating plant staff activities, and process controls that would warrant other types of criteria to judge the success of the program rather than relying on plant performance level criteria.
These types of criteria would be selected more in the vein of assuring product quality (e.g., sampling technical integrity of design output documents).
III. Corrective Action Proaram Activities The following information supplements section 4.6 of the NEI guide.
With respect to corrective action, the Expert Panel would identify the information and records necessary to conduct effective root cause determinations and corrective action activities for low-risk significant SSCs.
These records should include applicable design, procurement, installation, maintenance, and other information germane to the identification of material characteristics related to potential SSC failure modes.
The records should be sufficient to evaluate the locations of similar SSCs, or SSCs that perform corresponding functions, that may be included in the scope of necessary corrective actions.
The records would support the ability to perform an in-depth root cause analysis of low-risk significant SSCs and to institute the ENCLOSURE 2
requisite corrective actions to minimize the potential for repetitive failures.
It would not be appropriate to limit corrective actions for low-risk significant items to only those cases where there is a failure to meet a performance criteria or where there has been a safety system functional failure.
Oversight of plant activities could identify a design related issue that has not yet manifested itself in a functional failure.
In this type of situation the expectation would be that appropriate corrective actions be initiated as determined necessary prior to the occurrence of a failure.
Nor will it be appropriate to have repetitive failures prior to performing a root cause analysis and corrective action implementation.
Operational feedback may well demonstrate that performance criteria have been inappropriately set.
In that case, the corrective action would be to adjust the performance criteria to be consistent with the operational feedback insights.
In the event that the specified performance criteria are not met, then the appropriate corrective actions would be implemented, including a sufficient root cause analysis.
The analysis should consider whether the specified quality treatment (either graded or full quality program implementation) is adequate. The licensee would need to maintain the capability to trend adverse equipment conditions to provide a mechanism for elevating SSCs from low-risk significant to the high-risk significant classification based on repetitive failures (see NUMARC 93-01, Section 9.4.4).
IV. 0A Commitments The following information clarifies the third bullet on page 6 of the NEI guide.
For the purposes of the pilot project, existing QA commitments for low-risk significant SSCs could be replaced by the graded QA methodology that relies upon an Expert Panel to determine the appropriate quality elements that will be applied to the SSCs.
The provisions of 10 CFR 50.54(a) related to QA plan changes will continue to be applicable unless a temporary exemption is granted by the NRC.
Adjustments under the graded QA project would be limited to the following Regulatory Guides:
REGULATORY GUIDE TITLE 1.28 Quality Assurance Program Requirements (Design and Construction) 1.30 Quality Assurance Requirements for the I
Installation,. Inspections, and Testing of Instruments and Electrical Equipment 1.33 Quality Assurance Program Requirements (Operational) ENCLOSURE 2 i
1.37 Quality Assurance Requirements for Cleaning of Fluid Systems and Associated Components of Water-Cooled Nuclear Power Plants NOTE:
The technical provisions for Reg Guide 1.37, Section 3 - which establish water quality requirements remain in effect 1.38 Quality Assurance Requirements for Packaging, Shipping, Receiving, Storage, and Handling of Items for Nuclear Power Plants NOTE:
The technical provisions of Reg Guide 1.38 Section 3.2 which define levels of packaging remain in effect.
1.39 Housekeeping Requirements for Water Cooled Nuclear Power Plants 1.58 Qualification of Nuclear Power Plant Inspection.
Examination, and Testing Personnel 1.64 Quality Assurance Requirements for the Design of Nuclear Power Plants 1.74 Quality Assurance Terms and Definitions 1.88 Collection, Storage, and Maintenance of Nuclear Power Plant Quality Assurance Records 1.94 Quality Assurance Requirements for Installation, inspection and Testing of Structural Concrete and Structural Steel During the Construction Phase of Nuclear Power Plants 1.116 Quality Assurance Requirements for Installation, inspection, and Testing of Mechanical Equipment and Systems 1.123 Quality Assurance Requirements for Control of Procurement of Items and Services for Nuclear Power Plants 1.144 Auditing of Quality Assurance Programs for Nuclear Power Plants 1.146 Qualification of Quality Assurance Program Audit Personnel for Nuclear Power Plants ENCLOSURE 2 e
V. Definitions The following information supplements the definitions in Appendix A of the NEl guide.
Assessments:
A wide spectrum of methods for plant personnel to gain input relative to plant activities, processes, and equipment.
Assessments include design reviews, monitoring, surveillances, inspections, audits, and testing.
Expert Panel: A group of experienced and knowledgeable facility personnel that determine the safety significance of SSCs based on PRA and deterministic evaluations involving operations, maintenance, engineering, and quality assurance insights.
They, or a technically equivalent group, are also responsible for establishing the appropriate graded QA provisions which apply i
to SSCs and for determining the necessary performance monitoring criteria.
i Graded Quality Assurance:
The proportionate application of quality
]
verification to SSCs and/or activities, commensurate with their risk significance.
Low Risk Significant SSCs:
The set of SSCs (safety related and non-safety j
related) that is determined by an Expert Panel through evaluation of both PRA l
and deterministic approaches to have relatively low risk significance.
Quality Element:
Quality attributes, controls, criteria, processes, or practices necessary to provide reasonable assurance that an SSC would be able to perform its intended safety function.
Safety-Related SSC:
From 10 CFR Part 100 and 50.49/50.65: A structure, system, component, or part necessary to assure:
(1) The integrity of the reactor coolant pressure boundary; or (2) The capability to shut down the reactor and maintain it in a safe shutdown condition; or (3) The capability to prevent or mitigate the consequences of accidents which could result in potential offsite radiation exposures comparable to the 10 CFR part 100 i
guidelines.
High-Risk Significant SSC:
The set of plant equipment that is determined by an Expert Panel, through evaluation of both PRA and deterministic approaches, to have a relatively high risk significance. ENCLOSURE 2
-