ML20210J591
| ML20210J591 | |
| Person / Time | |
|---|---|
| Issue date: | 07/29/1999 |
| From: | Bill Dean NRC (Affiliation Not Assigned) |
| To: | NRC |
| References | |
| NUDOCS 9908050055 | |
| Download: ML20210J591 (68) | |
Text
g>n Clivg g?1
$11 UNITED STATES g
j NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 2066H001
+9
..,+
July 29, 1999 MEMORANDUM TO:
Public Docum oom FROM:
William M. Dean, Chief M.
l y
inspection Program Branch Division of Inspection Program Manageme Office of Nuclear Reactor Regulation
SUBJECT:
RESPONSE TO QUESTIONS ASKED DURING REGULATORY OVERSIGHT PROCESS PILOT WORKSHOP The Nuclear Regulatory Commission (NRC) held a public workshop from May 17 through May 20,1999, at Philadelphia, Pennsylvania to provide information to the NRC, industry, and public representatives of the participating pilot sites regarding the revised oversight processes.
Among the topics discussed were the performance indicator reporting, inspection program, significance determination process, assessment, communication, enforcement process, and pilot and fullimplementation of the program. During the workshop the participants asked numerous questions regarding the revised oversight processes. The questions and answers discussed during the workshop are attached to this memo. This information reflects the response to the questions as of July 1,1999.
Please place this document in the NRC's Public Document Room.
Attachment:
As stated l
r m e ec 050010
~
9900050055 990729 PDR ORG NRRA d t i
Que: tion 3 cnd An:wera Regulatory Oversight Process Pilot Workshop May 17 - 20,1999 Philadelphia, PA Topics Contained in this Document:
Page Communication 1
Pilot Program /FullImplementation 2
Performance Indicator (PI) Data Submittal 6
Reactor Safety Strategic Performance Area 7
General 7
- Initiating Events Pls 8
- Mitigating Systems Pls 10
- Barrier integrity Pls 13
- Emergency Preparedness Pls 14
- Reactor Safety inspections 15 Radiation Safety Strategic Performance Area 16
- Radiation Safety Pls 16
- Radiation Safety inspections 19 Safeguards Strategic Performance Area 21
- Physical Protection Pls 21
- Physical Protection Inspections 29 Baseline inspection Program 30
- General 30
- Corrective Action (Pl&R) Inspections 34
- Event Followup Inspections 35
- Inspection Documentation 36
-Inspection Planning 37 Supplementalinspection 38 Significance Determination Process (SDP) 39
- General 39
- Reactor Safety SDP 46
- Emergency Preparedness SDP 54
- Radiation Safety SDP 55
- Physical Protection SDP 59 j
Enforcement 59 Assessment Process and Thresholds 61
- General 61
- Action Matrix 61 Attachment
i:
Questions on Communication O - The NEl Director of Regulatory Reform indicated that an NEl Website would be available for NEl users with passwords to give answers to questions in this process. He also indicated that NRC would be tied in closely to provide timely answers. Will there be a Website where NRC employees can view these answers to questions on the program? So far questions from the Illinois and Atlanta workshops and the Change Coalition questions have not been shared so all can see the answers. In many cases the questions have not even been answered to the individuals who asked them. Please help get this information out to the multitudes.
A - NEl and NRC are planning to coordinate issuance of O&As. The NRC will have two websites. One will be for internal use by NRC employees. This website is in operation. O&A's have been submitted to the Chief Information Officer for inclusion on this website. The initial Q&A's (from previous conferences) were expected to be included on the internal website by the end of the week of May 17.
The NRC is also planning an external Web page which will be divided into two sections.
Section 1 willinclude a basic description of the program and Q&As. Section 2 willinclude feedback on Pl data provided by the licensees. The external Web page is currently under review by various NRC offices. While some information is currently available, Pi information is expected to be available following the August information Submittal.
O - Who is the contact for training and communications plans from NEl? Do you have examples or handouts available for review at this conference? Is it available on the NEl web?
A - Training and communications materials are being developed by Bob Evans at NEl. The initial set of materials were sent out to licensees during the week of May 24. The website for the J
oversight process was planned be up and running the first week in June.
O - Will there be additional opportunities for shared industry /NRC training?
A - Additional workshops are planned for feedback on the pilot. Regional training workshops t
will be held in early 2000. The NRC can also support limited ad hoc training (i.e., at ANS). The f
Sequence Determination Process (SDP) is a target area for such training.
1 O - What mechanisms / forums will be used by NRC to communicate the new assessment process to the public? What will be done for the pilot plants? Specifically, the NRC web page, by itself, may not be enough. Different stake holders may need different levels of detail.
A - There were public meetings in each local area during the June / July time period. Also, the NRC has revised and issued its plain language description of the new process (NUREG 1649).
Also, see previous O/A.
O - Are there plans to issue a "line in the sand" type document to recognize that future inspection reports and PIMs will only contain negative findings? (Lack of negative indicates areas are acceptable to NRC). Public perception / anti-nuclear perception may be affected.
A - Our trarssition into the new process has been documented in SECY papers as we progress through the pilot into full implementation. The new process is also described in NUREG 1649. It
is believed that the new process, due to the frequent and more easily understood processes for determining NRC involvement beyond the baseline inspection program and the quarterly updates of licensee pedormance, will provide an appropriately balanced view of overall licensee performance.
O - The NRC model uses terminology for the white and yellow bands that will confuse the public, namely, " minimal" or "significant reduction in safety margin." NEl commented on this and suggested different language (i.e., " challenges to cornerstones [may) or (require) increase (d)
NRC response"). It will be difficult to explain to the public that a plant may continue to operate when cornerstone objectives are met with significant reduction in safety margin. Why not use the alternate language?
A - The intent of the language is to reflect that there is margin to actual safety performance and that performance degradation, as reflected in PI thresholds being crossed, are still reflective of safe performance, but that margins to safety are being affected. Experience may show that alternate language may be appropriate.
Questions on the Pilot Proaram/ Full implementation O - How will NRC regions and headquarters get feedback from licensees during the pilot?
A - Feedback will be obtained through existing regulatory feedback processes, through periodic 1
visits by the staff, through NEl on an ongoing basis, through a solicitation of public comments in
{
a forthcoming Federal Register Notice, and through a public workshop near the emi e' the pilot.
l O - On page 5 of the conference materials, the second slide,5th bullet says " Pre-established 4
criteria will measure success." What are the pre-established criteria?
Related O - On page 7 of the conference materials, the second slide,2nd bullet says
" Qualitative and quantitative measures established." What are these measures? Where are they written down and how we can get a copy of them?
A - The pre-established criteria refer to the pilot program success criteria that were established for utilization at the start of the pilot program. Qualitative and quantitative criteria will be used to measure the effectiveness and efficiency of the new oversight process. See SECY 99-007A, which is available on the NRC Website, for more details.
O - With respect to the Pilot Program Evaluation Panel (PPEP): Can anyone be a member?
Are the regions represented? How is this consistent with the Federal Advisory Committee Act (FACA)? How often will they meet and where will the meetings be held (in DC or near the pilot plants)? What product will result and how will the public get it?
A - The PPEP members will be selected to have broad, high-level representation of pertinent stakeholders. Regions will be represented at the division director or deputy regional administrator level.
The PPEP meeting process (frequency, location, etc.) is being developed to be consistent with FACA. PPEP meetings will be public meetings, and results will be provided in meeting l
2 1
summaries. They are also expected to provide a final report to the Commission, noting their observations on the implementation of the pilot program. The first PPEP meeting is scheduled for July 28,1999.
O - The NRC should consider including a FEMA representative on the PPEP to coordinate onsite and offsite EP reasonable assurance findings. The present Pl and inspection findings program proposals would not provide licensees or the NRC with an adequate measure of EP performance and advance indication of possible offsite exercise " deficiency" findings by FEMA.
Public confidence in the regulatory program is also a FEMA concern.
A - We agree that FEMA concerns regarding a licensee's state of emergency preparedness are appropriately integrated into the NRC's oversight process. While we have not included a FEMA representative to the PPEP, the staff has conducted specific interface meetings with appropriate FEMA representatives.
O - Who is the primary contact for the PPEP program? How do we get a hold of them?
A - Frank Gillespie, Deputy Director of NRC's Division of Inspection Program Management, is the panel chair and can be reached by phone, e-mail, or regular mail.
O - Is it anticipated by both NEl and NRC that inspection procedures and/or Pts may fundamentally change during this pilot process?
A - We do not expect " fundamental" changes, but there will be ongoing refinements as lessons are learned. We have to be careful to assure that we get adequate experience with all aspects / elements of the pilot program, if a substantive change occurs, it will be in reaction to a substantial issue or problem that was not foreseen.
O - Will the NRC develop a group to change, modify, or evaluate Pis during the pilot program?
Related O - Are there additional Pls being developed?
A - We will continue current Pl development efforts, working with NEl's working group on Pls.
Additional near-term Pls are being considered, such as a reliability indicator, Also, the NRC's Office of Research (RES) is currently developing risk-informed PI's that will be considered for inclusion in the program, possibly in 2001.
1 O - Are under-development areas going to be phased in during the pilot?
l A - They will be issued as soon as processes are developed and piloted, so they can be used l
during the pilot.
O - Have NRC points of contact been established to provide guidance that may be necessary during initial development of the PI program? If yes, who are the contacts? If no, which NRC office should be contacts?
A - Yes, Don Hickman of the Inspector Program branch in NRR is the POC for Pts for the pilot i
program. Additionallong-term development of Pls is under the auspices of the NRC's Office of Research, of which Pat Baranowsky and Steve Mays are key contacts.
3 l
3 l
1 O - The pilot is for six months, but we are counting the same indicators over 12 quarters. Will you see anything?
A - We have asked for data for 2 years prior to the pilot.
O - What reports go to the public during the pilot?
A - We will do periodic inspection reports as well as, quarterly and mi6-cycle reviews.
Significant inspection findings and Pls will also be available on the NRC website.
O - Will each of the Pts be verified at EACH pilot plant or will each of the Pls be verified at one pilot plant in each region?
A - The pilot program's goal is to exercise all the procedures. The Pts should be verified at j
each pilot plant. However, in some rare cases (i.e., if there is no emergency preparedness drill during the pilot) ihere will not be any Pl data to verify.
O - What is the status of the inspection module preparation? We have seen drafts of the modules about a month ago. Have they changed since then? If so, how? Will we see them again soon? When?
Related O - When and where can we get a copy of the inspection procedures?
A - The procedures have been changed in response to comments received. Draft documents intended to be used for the pilot were issued during the week of May 24. They are available j
through the Public Document Room (PDR). When issued for full implementation next year, the i
procedures will be available on NRC's website.
1 O - For the pilot plants, will previous (i.e., prior to 6/1/99) inspection findings go through the SDP and then be used for assessment or will all plants start the pilot on 6/1/99 with no previous Inspection findings?
A - Issues associated with plant performance at pilot plants were captured during the recent plant performance review and where appropriate, were incorporated into the inspection plan for each pilot site. There may be some additional inspection effort and use of the SDP to complete review of these issues.
O - Will SDP charts be finalized by implementation of the pilot program?
A - The SDPs that will be applied during the pilot process are finalized for purposes of i' ' al implementation. However, ongoing refinements may occur during pilot implementation.
One exception is the fire protection SDP, which is still under review. Additionally, the staff is exploring the application of similar processes to other aspects of plant operation (e.g.,
shutdown operations).
O - How close are inspectors to being ready to use the SDP process?
4
A - Pilot plant inspectors were trained at a workshop in Atlanta in May. As second training workshop was conducted in June in the Dallas area. It is expected, participating early in the pilot program that they will need to work closely with senior reactor analysts to properly characterize inspection findings during the process.
O - it was stated that the pilot plants would be provided with a manual / document of scenarios and systems we need to use/ consider when using the SDP and Phase 2 risk estimates for
. multiple initiating events. What is the name of this document or when will it be available?
Related O - The SDP and Table 2 have been revised from the SECY 99-007A documents. How will non-pilot plants be made aware of these changes?
Related O - Within the SDP there are a number of areas with recently developed " rules" as well as areas still being discussed. How will the recent " rules" and the " rules" to be developed be documented and given to the pilot plants? It was stated that the " rules" will be given to the inspectors. The pilot plant licensees also need the rules.
A - The latest SDP was issued in draft form during the week of May 24. The NRC will include this draft in a federal register notice request for comment in the ve;y near future. The NRC will continue to refine the SDP process during the pilot. As appropriate, we will issue changes to tha guidance and it will go to the pilot plants as well. We will issue the final guidance as a manual chapter prior to full implementation in early 2000.
O - Will the NRC, NEl, and industry reconvene prior to January 2000 to go over any changes brought about by the pilot program?
A - We expect to have a workshop near the time that the pilot program ends (early January
{
2000) to discuss lessons learned and potential changes. We also intend to have industry l
workshops in each region, similar to this one, to prepare licenstes for full implementation (Feb/ Mar 2000).
O - It was mentioned that routine 6 week inspections will continue during the pilot. How will the nature of these inspections change? Will these inspections continue after the pilot?
A - it is the periodicity of the resident inspection cycle that dictates when reports will be issued.
This will be continued during the pilot, but this is an area NRC is considering adjusting. This will be evaluated during the. pilot program.
O -Plants that are not in the pilot program will be assessed differently in terms of performance and enforcement than the pilot plants. How will the NRC accommodate this difference in PPR reviews?
A - Plants in the pilot process will be assessed as outlined in SECY 99-007/007A and will not be under the PPR process that exists today. For non-pilot plants, the mid-cycle review will be in Aug/ Sept 99, for pilot plants Nov/Dec 99.
O - Will there still be enforcement boards and pre-decisional enforcement conferences during the pilot?
i 5
A-For a violation evaluated under the SDP with no actual safety consequences, there may be a j
need to have a conference to discuss the significance of the violation. Violations for which the SDP does not apply, or those with actual safety consequences, we would follow the current process, which includes provisions for an enforcement conference.
O - How and when will the decision as to whether to delay formal implementation of the revised oversight process from January 2000 to April 2000 be made, and how will this be communicated to licensees?
A - The Commission provided the staff this airection in its staff requirements memorandum (SRM) of June 18,1999. This document has been made public.
O - What should non-pilot plants be doing now? What must they have done by March 20007 1
A - Non-pilot plants should continue to learn about the process. They should also be collecting
)
Pl data for calendar year 1999, to be reported as historical PI information in January 2000.
O - What will happen to the pilot plants / program in the interim period between December 31, 1999 and March 31,2000, if full implementation is delayed?
A - The NRC will continue to implement the pilot program at those sites.
O - How will the baseline inspection program be incorporated into the training / education of NRC j
inspectors?
)
A - The plan is to train all inspectors before full implementation. Lead inspectors who will be j
conducting inspections at pilot piants will be trained prior to the inspection, j
Questions on Pl Data Submittal O-- What requirement (e.g.,10 CFR 50.xx) should be cited when submitting the Pi data?
A - This is a voluntary program, so is not required by regulations. However, completeness and accuracy of information submitted to the NRC does fall under 50.9 requirements. If a licensee opts not to report the PI information, it would result in additional inspection to cover that area.
O - Reconstructing data for notification / classification for past drills may not always be possible.
Will it be necessary to go back 2 years to have 2 years data at the start of the PI period?
Related O-For the non-pilot plants, how far back should we go to search on historical data? Is January 1,1998 acceptable, even for 12 quarter data?
Related O - How much past data must we collect and enter for the various performance indicators? For safety unavailabilities, are we to have 3 years worth of data in place at the time the program goes into place? This question applies to all indicators.
A - Non-pilot plant licensees should report in January 2000 data from January 1,1999, through December 31,1999. This should be obtainable for the alert and notification system (ANS) and 6
drill and exercise performance (DEP) Pls. For DEP, if more data is available (i.e.,1998) it would be useful. A review of critiques and IP scenarios should identify the opportunities and successes. ERO should be easy to accomplish, as it is just the percentage of key ERO members that have participated in the past 8 quarters as of December 31,1999.
Data for Pls that are calculated over more than 4 quarters should be reported if relatively easy to obtain. We are asking for a licencee's best effort to obtain this data. No verification of the accuracy of past data will be performed by the NRC.
O - Currently, plants provide the NRC a monthly operating report that contains much of the I
same information that will be reported quarterly on the Pls. Will the monthly report still be
{
required?
A - Pl data was developed without regard to other reporting requirements. Once we have this process in place, we will review the interface with some of our other processes (such as rnonthly reperting) and make changes as necessary.
O - What are the data quality and retention requirements for availability data? For example, if a system engineer reports that a train was unavailable for 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> during maintenance (based on observation, interviews, etc.), is this acceptable, or must the exact maintenance times be j
formally logged separate from the Pl data gathering process?
A - The data requirements are as described in NEl's current draft guidance on PI's, NEl 99-02 Draft B.
O - is there a process in place to resubmit / correct data which was previously submitted and found to be in error / incorrect?
A - Yes. Licensees will enter the corrected Pl data for the appropriate quarter, and annotate in the comment block for the current quarter to indicate what PI was changed and in which quarter.
Q - The 14-day reporting requirement for Pl submitted is not reasonable. 30 days would be more appropriate. What is the big rush to get the quarterly data?
A - The 14-day reporting requirement was developed through discussion with the NEl working group. The only Pl that may need to be delayed is associated with RETS/ODCM. If the pilot program indicates that the 14-day period is unreasonable, if may be adjusted.
Questions on the Reactor Safety Stratealc Performance Area General C - In general, please remove ambiguity in some of the Pls. It is understood through your discussions so far that the Pls are supposed to meet certain intents. However, these Pts such as leakage or those for radiation protection need to have clarifying statements to remove any disparity between the site residents and the utikty. For example, unintended exposure can be 7
interpreted by the resident as that which is potentially identified by alarming dosimetry, whereas utility would identify unintended exposure as that exceeding the work package planned dosages.
A - Issues associated v ith implementation of tho program, such as that described above, will be collected during implementation of the pilot program for possible program refinements.
Initiatina Events Pls O - Why are the time frames over which the initiating event Pls are measured different? (4 quarters for scrams,12 for risk significant scrams,4 quarters for unplanned power changes ability)
A - Because scrams with loss of normal heat removal occur less frequently than the other two Pls, a longer time frame is appropriate.
O - The " purpose" statement for the scram with loss of heat removal Pl conflicts with the
" indicator definition." The " purpose" references use of alternate means and the " indicator definition" references loss of normal heat removal. Many low power scrams can have a " loss" of normal heat removal without the "use" of alternate means.
A - Low-power scrams within the capability of the PORVs are not counted if the main condenser is not in service at the time. We may make this guidance more general to state that any time the main condenser is not in service, scrams with loss of normal heat removal will not be counted.
O - For unplanned power changes, how is 20% calculated? Is it 20% of current steady-state level or 20% of full power no matter what the current level is? Is this thermal power? What if you're not steady-state (increasing power after RFO)?
A - 20% power change means 20% of full power as measured by nuclear instrumentation. It counts swings in power greater than 20% starting from any instantaneous power level (steady-state, increasing, or decreasing power).
O - Power change initiated less than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> following discovery may be at the discretion of load dispatcher and not the plant. As an example, take the case of a problem requiring power reduction >20% discovered on Friday and which the plant feels they could live with for weeks, but the load dispatcher prefers the power reduction this weekend; this would be counted. You could have the same problem, but the load dispatcher prefers the following weekend, and it would not count.
Related O - If a problem is identified and the repair is planned in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, and the weekend comes when it is economically best to make the repair within the 72-hour window, won't the Pl drive people to wait another week to fix it? Is this really what we want to measure? How about i
a failure that must be repaired or causes a power reduction in 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> to prevent an even worse event?
A - We will review this situation during pilot program implementation and change the guidance if appropriate.
I 8
1
O - With most plants being able to plan down power transients in less than 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> (canned work orders, forced outage plans), are there any plans to reduce the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> criteria?
A - We will revisit thic issue as lessons are learned from the pilot program. The intent of this Pl was to measure those incidences which required a plant to reduce power in a fairly short time period to address an emergent issue.
O - Why are equipment failures counted in the unplanned power changes but instrument
{
problems excluded? Instruments are necessary to monitor many plant conditions and if they do not perform properly, they can contribute to unavailability.
A - The unplanned power changes indicator counts power changes greater than 20%; it does not consider equipment failures or instrument problems. The safety system unavailability indicator counts unavailable hours for any reason-equipment failures or instrument problems.
The guidance for this Pi indicates it will not count an apparent (not actual) power change due to instrument failures.
O - Why do pu believe the indicator for unplanned power changes doesn't inhibit " conservative decision making" with respect to bringing a unit down into a conservative operating region to effect certain activities? For example, I find a problem I want to fix right now. I can do the fix at 100% but I limit my potential for a significant transient or trip if I do the activity at lower power.
A - The green-white threshold for unplanned power changes is greater than 8 transients per 4 quarters, which should allow for licensees to take the kind of actions to address situations described in this question. In addition, if the threshold is crossed, the NRC will assess the reason for the change in the PI before it takes any further action. If the licensee has taken appropriate precautions, effectively assessed the root cause and acted to maintain plant safety, no further NRC action should be necessary.
O - Is entry into technical specification 3.0.3 considered an unplanned power change per 7000 critical hours, even if no actual power change occurs, since this entry reauires a change of power greater than 20% (to be at hot shutdown in 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />).
A - This would count only if an actual power change occurs. If the problem is corrected within the technical specification LCO time limits, this would not be considered a transient for PI purposes.
]
O - Where does 2400 hours0.0278 days <br />0.667 hours <br />0.00397 weeks <br />9.132e-4 months <br /> and its basis come from? (It's the lower limit of validity for the scram /7000 critical hours indicator.)
A - This will ensure that when a plant has few critical hours in a 4 quarter period, a single scram won't cause it to cross the green white threshold.
O - Why does the red threshold go down for scrams with loss of normal heat removal vs for unplanned scrams?
A - The thresholds are based on risk information and the relation of these events to a change in core damage frequency (CDF).
9
Mitiaatina Systems Pls O - For planned unavailable hours, what does " automatically overridden" mean with respect to testing? " Automatic" implies no human action. Is this correct? Or do you mean "promptly overridden"?
A - During surveiliance testing a train is available if an automatic signa! or operator action can initiate the train's safety function. A more definitive description of acceptable operational action is currently under discussion.
Q - Why are manual actions limited to " surveillance testing"? How about troubleshooting? How about routine testing not associated with a technical specification surveillance requirements?
A - We will revisit this issue as experience is gained with the pilot program.
Q - Is the " planned" definition for safety systems the same 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> definition as for transients?
A - That is one option. Another is to define preventive maintenance as planned unavailable 1
hours. We will revisit this issue as experience is gained with the pilot program.
Q - Pleaso explain safety system unavailability with regard to 12 or 18 month survey intervals i
and how time is calculated. Provide an example.
A - Fault exposure (FE) hours are computed based upon the time of the last surveillance test that would have exposed the discovered fault. See guidance in NEl 99-02, Draft B, p. 33, line 18-23.
Q - NUREG-1022 does not require plants to report inoperability that may have occurred due to surveillance testing failure unless there is clear evidence that indicates when the failure occurred. The use of the T/2 between surveillance tests does not comply with this requirement.
This may lead to contradictory information (equipment out of service for a long period of time with no LER submitted). What is the benefit of this policy? Does breaking a Pl barrier on one instance warrant the increased cost in manpower for the licensee and NRC7 A - The safety system unavailability indicator was chosen because it uses the same data licensees are supposed to be reporting to WANO, and therefore imposes minimal additional j
burden on licensees. There is little connection between NUREG-1022 and the new Pls. Only 3 of the 11 indicators in the initiating events, mitigating systems, and barriers cornerstones are required to be reported per 10 CFR 50.73 -- scrams, scrams with loss of normal heat removal, and safety system functional failures.
Q - S;nce people are reporting safety system functional failures, doesn't that provide " reliability" data that is needed better than fault exposure hours?
A - No, safety system functional failures counts conditions as well as actual failures.
Q - What governs the maintenance of spare equipment (whether installed, portable, or governed by system alignments)? How is the use of spare equipment tracked under this indicator or how should it be tracked?
10
A - Installed spare equipment is defined as equipment not needed to satisfy the plant's design basis. It is included in the safety system unavailability indicator only when it has been placed in service to replace normally in-service components. In those cases, any unavailable hours are counted against the train that it is serving. At all other times (when it is not in service),
maintenance of spare equipment is at the licensee's discretion and is not included in any Pl.
O - Our current plans will have HHSI unavailability above the green / white threshold. Our PRA 4
shows this to be below 1x10 for a 3 safety train plant. How can we get " official" recognition of this so that we will not be in the white band?
' A - Plants that by design are regularly borderline white, in particular, should make an annotation in the comment field to this effect. NRC's Pl verification inspections will provide NRC insights into the cause for their condition.
O - Is it recognized and accepted that the use of " fault exposure unavailable hours" for safety systems promotes shorter surveillance intervals? For example, a HPCI system is tested every 90 days. It has not failed for 3 years, and its (NRC-approved and PSA validated) maintenance rule performance criteria allows 2 failures every 3 years. Now, the system fails Q surveillance due to control instrument drift (time of onset unknown). HPCI will be a white for the next 3 years due to this single failure. (45x24/7000x3 = 5.1%, white > 4%). A good way to avoid this issue is to reduce the surveillance interval. I would suggest limiting fault exposure time to no more than 15 days as an alternative.
A - Yes. We are working on an unreliabihty indicator so that fault exposure unavailability can be removed from the unavailability indicator.
O - Doesn't the safety system unavailability (SSU) Pl penalize BWRs who have diverse single train systems? Shouldn't HPCI and ADS count as two trains since they have the same purpose
{
rather than HPCI as one train and ADS as other train. In PWRs HPSI is two trains but not diverse.
A - ADS is not included in the safety system unavailability indicator. A two-train system has a slight advantage in that a good performing train can mask a poorer performing train. However, counting HPCI in SSU is not necessarily a penalty as long as its availability is maintained.
O - Why was service water not included in the systems considered in the SSU Pl? It can be a significant contributor to CDF.
A - To mir imize the impact on licensees, we used the systems that are being reported to INPO/WANO.
O - Why was the SSU indicator not tied to the maintenance rule indicators? (i.e., How much of this indicator is site specific or should be considered site specific?)
A - The maintenance rule requires each plant to establish its own goals. We intended to establish plant-generic, not site-specific indicators. The maintenance rule will, as assessed through the baseline inspection process, provide added assurance of system availability and reliability.
11
C - More clarification and examples dealing with system unavailability and updating past reports are needed.
A - The unavailability Pl is intended to replicate the information provided to WANO in the calculation of the unavailability and reliability WANO indicators. It is currently being assessed by the NRC, INPO, NEl and industry.
a Licensees update " data" for errors or new information that would impact the current indicator report. " Data" submitted to NRC for the current indicator would update trend charts. Licensees do not submit " corrected" indicator reports for previous periods. The program or licensee only holds data to support the current indicator. (e.g. the SSU data is held for 12 quarters; the barrier indicator for 12 months.)
O - Since your definition of unavailability by train is different than the WANO definitions, how can you use WANO normal data to set the green-white threshelds?
Related O - What value does fault exposure hours provide? It's the most cefusing area, most subject to potential violation and least useful from a risk perspective. If PSA, have reiiability and unavailability from discovery, I think they have what they need to understand the risk.
Related O - Why use fault exposure hours when mere is no threshold associated with it? Is it simply added to the unavailability data? Were the thresholds set based on this criteria?
A - If the NRC is successful in developing a reliability indicator, the need to include fault exposure lessons will diminish. Staff will assess these issues during implementation of the pilot program.
O - Are the safety system functional failures (SSFFs) e camples listed in the NEl guidance under SSFF purpose all consistent with 50.73 (a)(2)(5)?
Related O - The Safety System Functional Failures indicator is predicated on LERs submitted per 10CFR 50.73 (a)(2)(v). The LER iule 10CFR 50.73 is currently being evaluated for change.
If 10CFR 50.73(a)(2)(v) is changed and the change has an impact to the guidance for this cornerstone, will the cornerstone be changed accordingly?
A - The SSFFs indicator is consistent with the current version of 50.73, however,50.72/73 are currently being revised, which will require the staff to revisit this issue in the future.
I O - Do you realize that the SSFF rules are inconsistent with maintenance rule requirements, especially for risk significant systems?
A - Yes, SSFFs are identified from licensee event reports, which are only required when an event or condition could prevent the fulfillment of a safety function. Train failures are not reported in LERS.
O - Since failures are reported under 10CFR 50.73, tvill the SSFF PI be reported by the licensee or will the NRC review LERs? I believe the licensee should report, retain rationale, and prevent interpretation confusion caused by reading LERs. Keep LERs and Pl separate.
12
p A - The licensees are to report the Pls. The NRC will inspect to validate the information being reported.
Q - The SSFF Pl definition uses the phrase " number of events or conditions..." If the licensee issues a single LER that discusses more tha.n one potential SSFF, such as for a generic breaker, motor operated valve, Appendix R, etc. issue, does it count as one event or more than l
one condition?. (Asume the finding involves the same condition in multiple components.) -
4 A - If there's one cause, there's one count. See p. 68 of the NEl draft guidance NEl 99-02, line 40.
Q - If the yellow and red thresholds are risk informed (^CDF of10 and 10 respectively) how 4
d can BWR thresholds for RHR be lower than for HPCI since BWR accidents are dominated by high pressure events?
A'- These thresholds were calculated to correlate to the aforementioned ^CDF figures. If there are issues related to the PI's thresholds, NEl or NRC staff responsible for Pl development should be contacted with the appropriate risk-informed information.
Barrier Intearity Pls Q - Identified leakage can be replaced by total leakage 2nJIy if specified in the guidelines!!
There should be no ambiguity!
' A - For plants using standard technical specifications, identified leakage can be obtained by subtracting unidentified leakage from total leakage.
Q - What does maximum value mean? Is it the maximum value of the minimum pathway or the maximum value of the maximum pathway?
Related Q - Explain the rationale for excluding containment leaks that are found and repaired in an outage. Isn't it likely that the leaks would have existed dering part of the previous cycle?
Could a T/2 (fault exposure) concept be adapted to better reflect actual containment performance?
A - It's the maximum value of whatever method a licensee uses to calculate containment leakage. The validity and utility of this Pl.will be assessed through the pilot program.
C - For containment leakage, is the Pl " maximum" value for the penetrations (so you take the best valve in each) or is it maximum value for the valves (so you take the worst value at each 1
penetration). If using worst values, a single failure to close can put you over 2L,.
)
O'- Explain the reason for recording unmeasurable leaks as 1.0 L,7 A - 1.0 L, is the default value when leak rate is unmeasurable.
' O - The definition of containment (Type B and C) leakage includes the term " monthly." We do not perform type B or C leak tests monthly; therefore, how can we report the " monthly" maximum total type B and C leakage (as a percentage of L.) for each month of the quarter?
13 2
1 Related Q - How do you get L, on a monthly basis?
A - You are required to know L,at all ?mes using the method you currently use. Report the latest (last) value you have.
I
- Q - Why doWt you account for fault exposure time in the type B/C containment leakages Pls?
Failed LLRTs are generally identified during outages. The Pl could routinely go white during outages, but be green throughout the operating cycle.
A - Fault exposure time applies to random failures rather than to degrading equipment. Also, isolation valve leakage is unpredictable and can drift up or down with time. We do not 'ntend, as part of this new process, to specify a new method for calculating containment leakage.
- O.- The PI on containment leakage appears to be BWR/PWR dependent, (i.e., smaller BWR containments push BWRs closer to 0.6 L, due to containment size). Should we have a separate BWR and PWR containment leakage Pls?
A - We will look at this carefully during the pilot program to see if separate thresholds for PWRs and BWRs are warranted.
O - Was there any consideration for making the containment leakage Pl risk informed? That is, a level 111 PRA with a realistic source term will generally show little or no dose impacts from leakage less than or equal to Technical Specifications (TS).
A - This is a good idea, but is not practical at this stage. The efficacy of this PI will be evaluated 4
during the pilot program.
i Emeroency Preoaredness Pls Q - When a key Emergency Response Organization (ERO) member is added to the organization or changes from one ERO position to a different ERO position between drills, is there a grace period for having these people participate in drills?
A - If the new position is related to the ir.dividual's old position, the previous participation would be applicable (i.e., was the previous participation a proficiency enhancing experience for the
~ ' new position?) If the new position is unrelated and the previous participation is not applicable, then the next time the Pi is calculated (last day of the quarter) the individual will not have participated in the previous 8 quarters for reporting purposes.
Q - How is the length of a delay factored into performance on rfrill/ exercise performance notifications? If a protective action recommendation (PAR) is incorrect, can you factor in the significance or magnitude of the incorrect PAR 7 Is there a method to weigh the significance of J
missing personnel responsibilities in ERO participation? A reduced head count will not reflect a missing EPO function, for example missing PAR capability or offsite notification.
A - A Pl aggregates data in a manner that provides summary information useful to management (or oversight) of a process. There is no intent to ascertain the significance of an 14
l individual failure. That is done through the significance determination process, if the event is a l
violation of NRC regulations. However, poor performance in a drill / exercise is not generally a j
violation, nor is reduction of the number of key ERO merhbers, as long as the function remains viable.
Q - Could it be implied that for each classification opportunity, there may be several associated notification opportunities due to the need to notify several different state / local authorities? A!so,
)
If a classification opportunity is missed completely, are the associated notification opportunities also failed, or should they be considered not applicable?
A - No, initial notification is counted as 1 opportunity, regardless of how may offsite agencies are notified.
Q - On page 83/84 of NEl 99-02 (draft Rev. B) dated May 7,1999, the following given under
" definition of terms" " Shift communicator - provides initial offsite (state / local) notification." We understand fliat the shift communicator is not the phone talker but is the decision maker who fills out the initial notification form. Is the intent of this PI to include the person who declares the i
emergency, not the person who makes the phone call?
A - The intent is to track the key members responsible for the initial notification, not those responsible for just phone dialing and message reading. There is no intent to track a large number of such phone talkers. The licensee should identify the key ERO members thought to best represent the function and for whom proficiency enhancing experiences are necessary to maintain capability, i.e., drills are probably not necessary to maintain phone dialing and reading skills.
.Beactor Safety Insoections
'O - For post event human error measurement, why did the new program designers choose to k ok at operators in training and requalifications, and not to observe operator performance in re e life plant situations (except under very limited event response or surveillance test '
. siteations)?-
A - la addition to training and requalification, human performance is observed under the nonroutine plant evolutions inspectable area. The program has de-emphasized NRC
~ observation of operator performance during normal routine activities it was determined that the Pls and inspectable areas provide substantial insights into human performance on an indicative basis and as such, prevalent human performance issues will be effectively identified by the total process.
Q - Is the maintenance rule monthly inspection limited to those high safety significant SSC's that are in goal setting? (For example, suppose only 10 HSS SSC's are in goal-retting during j
the year.)
A - The maintenance rule monthly inspection will focus on performance issues for all high safety significant SSC's regardless if the SSC is in an (A)(1) status. The main objective of the inspection is to determine if the performance problem is appropriately resolved through j
maintenance rule processes.
15 l
L
O - Why does maintenance rule (MR) implementation require status as an inspectable area when the baseline inspection for MR implementation have already been completed?
A - Maintenance Rule (MR) implementation will include an annual review of the program, even though MR baseline inspections have been completed. Annual inspections will review changes since the baseline and other aspects including the balancing of reliability and availability, and i
the periodic evaluation performed by licensees. These three aspects of the MR change over I
time, hence, the annual inspection. Thn annual inspection is not intended to duplicate the MR baseline inspection.
O - Will MR inspections be risk informed? (i.e., review HHSI vs communication).
A - Yes, the MR inspections will focus on risk-significant SSCs, but some review of low risk significant SSCs may be done. (Example: the basis for reclassification of a system from high to low risk significance.)
Q - Now that MR baseline inspections are behind us, why are there over 250 hours0.00289 days <br />0.0694 hours <br />4.133598e-4 weeks <br />9.5125e-5 months <br /> per year looking at compliance? The emphasis should be event based to follow performance aspects and effectiveness at finding and resolving issues. Also, it should review the work planning to assure risk is assessed and managed. Compliance based inspections are the opposite of the philosophy on which this new assessment process is based, and will lead to both NRC and utilit:es spending resources in very low value activities. The MR inspectable area reviews actual failure for proper classification and handling under the MR.
A - The MR implementation IA does emphasize performance based issues. The intent is to l
look at SSCs that exhibit failures or performance issues. Work planning / risk assessment is inspected under several different IAs. We do not believe this is a " compliance-based" inspection. However, compliance with the MR should result in effectively addressing many (i.e.,
maintenance related) equipment performance issues.
O - Walking down verification of alignment of an in service train while the redundant train is out of service will cause 3.0.3 entry and unnecessary cycling of control room staff and focus where the level of safety significance is normally very low. It may be more appropriate to focus on line-ups, efforts to place equipment back in service, and configuration control.
i A - The walkdown of the redundant train should not cause the system to be made inoperable.
We do not require plants to cycle equipment to verify operability; therefore, there's no TS 3.0.3 entry unless problems involving functional inoperability of equipment are found. We agree that returning the inoperable system to service should remain the focus of the plant staff.
Questions on the Radiation Safety (RS) Strntealc Performance Area RS Pls O - In the occupational exposure control effectiveness slide on page 72 and 73, the 1st bullet refers to a "significant" high radiation area (HRA) occurrence, yet the 2nd bullet talks about just a very HRA occurrence. Both of these have the same criteria. Why is the term "significant" used?
16
A "Significant HRA"is used to. distinguish between a HRA of > 100 mrem / hour but < 1000 mrem / hour and a HRA of > 1000 mrem / hour. Both significant HRA and very high radiation
. areas are referred to by the term " radiologically significant areas."
Q - The locked high radiation door program requires an HP technician to verify doors are locked
~ fellowing retum of the key by the work group. The HP technician checks the door and finds it unlocked. Is that a PI hit?
A - The answer hinges on the timeliness of the check. If the check is part of a control prograr-and is performed in response to key return, it is not counted under the Pl. If the check is a routine check sometime later to verify technical specification compliance, it would be a Pl hit.
' O - Is the following scenario a " hit?" The wrong key to a high radiation area is inadvertently provided to a worker. Barriers such as REP, HP coverage, and HP control point still existed to prevent the individual actually using the key and no entry was made.
A - This would not be a PI hit as it's t ot a failure to meet technical specifications.
1 Q - Would the following be counted in the indicator for the occupation exposure cornerstone:
. individual working in a locked high rad area receives a dose alarm but chooses not to exit the area until the work is done. He receives 18 mr additional dose.
Related Q - Would the following be counted in the indicator for occupation exposure cornerstone: an individual working in locked high rad area doesn't hear the dosimeter alarm (due to high noise). He realizes the alarm after a rad tech informs him and he leaves the area, receiving 2 mr additional dose.
A - No, since in these examples there is only 18 or 2 mr unintended dose (not >100 mrem). But the first instance should be counted as a violation of station procedures.
Q - On page 73, the top slide says "a single nonconformance with a requirement for access control and work within..." What does access control jitn.d work mean? Does this mean entry n
into the area must be made in order for this to be a hit? Is not the intent to have a hit anytime access control methods fail (i.e., no posting, no barrier, unlocked) even if no entry is made?
- A - Yes, any failure to comply with technical specificatic:,s, either access control (i.e., failure to lock) gr work controls (i.e., issue Radiological Work Permit [RWP), issue a Digital Alarming Dosimeter (DAD), etc.), is a hit, j
Q - If the NRC inspector identifies the nonconformance does it get counted as a Pl? Would it also go through the SDP?
A - The answer to both questions is "yes."
Q - Exposure within regulatory limits is well defined. How do you define exposure in ALARA?
A - ALARA deals with collective dose.
i 17
Q - As part of a routine survey program an HP technician enters an area maintained locked and normal dose rates < 1000 mrem per hour. The technician identifies increased dose rates >
1000 mrem per hour. Does this count on the Pl?
A - No, this is not a failure.
Q - For multi-unit sites, how are PI's handled for things like occupational exposure? Are the occurrences counted against each unit, counted against the site, totaled and divided by the number of units?
A - Pls are counted for the site.
Q - Does unintended exposure occurrence refer to the individual's entry and to the total job estimate?
A -it refers to the individual's exposure event. It is not necessarily associated with an entry into a radiation significant area. The total job estimate is more associated with ALARA and collective dose, which is outside the scope of the Pl.
O - The occupation exposure PI that defines exceeding a digital alarming dosimeter (DAD) i setting as an unintended exposure renders the DAD an ineffective tool in maintaining dose ALARA. This encourages licensees to " pad" the setting to the upper limits established on an REP. Why do we use the " DAD" setting as the trip instead of the REP limits?
Related Q - Routine work planning involves establishment of an ALARA budget or dose budget for the job. Efforts are made to be very conservative in establishing the budget. When this budget is exceeded, we will take a hit on the Pl. Won't this PI encourage more liberal dose budgets for work in high dose jobs in order to avoid negative reporting on the Pl?'
A - if there is no other information we will use the DAD set point. If intended dose is formally documented elsewhere (i.e., ALARA planning package or RWP) licensees can provide a lower DAD setting for ALARA purposes.
Q - The regulations allow the licensee nine months to assess final transuranic exposure. How will the licensee accommodate this in the SDP process and action matrix?
A - Where the evaluation of the event requires time to fully assess, the process will allow j
licensees to go back in time and revise the color of the finding,if appropriate.
Q - How often will X be recalculated for the occupational radiation safety Pl?
A - Once.
O - It was mentioned that the NRC is looking at decreasing the rolling average time period from
" previous 12" to " previous 8" quarters for the radiation safety Pl. We feel this is a good way to go because:
1.
Rolling averages reflect prior performance rather than current performance.
i 18 l
i 2.
Each quarter, the "old" data that gets dropped out of the average is as important as the new data added. As a result of this the graphed change from one quarter to the next reflects the delta between what dropped off and what was added.
3.
Rolling averages could result in a long time lag before poor performance is dropped, especially with using a 3-year window on the mitigating systems Pl. For example, a system has average performance for 12 quarters, then has 2 quarters of below average performance before returning to 12 quarters of average performance. If the two bad quarters raise the rolling average above the poor performance threshold, then that system will remain there for 3 years until the bad 2 quarters drop out.
Is the staff (based on discussion above) considering reduction of all 3-year averages to 2-year rolling averages (12 quarters vs 8 quarters)?
A - There is a difference between rolling averages and rolling sums. The problems identified in this question are of most concern with rolling sums, such as the occupational radiation safety indicator. Only the safety system unavailability and the EP indicators use a rolling average, and this is necessary to smooth the data to avoid large spikes in the indicator value.
Radiation Safetv Insoection O -it seems that more has been added to this part of the inspection program than subtracted.
Will the overall hours go up? How many hours are expected in the old versus the new program?
Related O - The number of inspection hours has increased. Isn't the purpose of the new program to decrease the number of hours?
l A - Inspection hours will be determined after completion of the pilot. Hours were estimated before procedures were written. Part of the pilot program assessment process is to ascertain if the appropriate scope and level of effort have been identified.
Q - The intent for inspection is to look at a few key areas, not all areas not covered by Pls. I do not believe the RS team has been challenged in this concept.
A - The process identified all the areas that were thought necessary to verify to determine if j
cornerstone objectives have been met. The program will be critically evaluated on a regular basis to improve the program, removing areas or inspections that are unproductive and emphasizing areas that provide better inspection indications.
j Q - There seems to be lot of programmatic review. Isn't the emphasis supposed to be l
performance based, independent of a program?
I A - The new procedures are performance based, not focused on procedures, training, or
)
programs.
O - It seems like you're conducting enforcement by inspections. The procedure claims many 10CFR Part 20 requirements as a basis when, in fact, there are not requirements or the interpretation is in dispute.
i 19 1
a l
l l
I
r A4 Inspection procedures are not designed to fit the regulations but to be risk informed or dose based. Many regulations are not inspected by the baseline inspection program. Inspections are designed in combination with Pls, to effectively assess the objectives of the applicable cornerstones.
C - Industry and NRC RS experts did not write the RS module and are not in agreement with it.
A - Inspection procedures were written by NRC staff who had the appropriate expertise. They were also provided for comment.to all pertinent NRC organizations and comments were appropriately resolved. The need for refinement of the procedures will be evaluated during the pilot program.
Q - Wouldn't the Pl on occupational exposure control effectiveness apply to the inspection area "ALARA planning and controls"?
Related O -Why are the occupational radiation safety findings duplicating Pis?
A - While the Pls provide ongoing trending and indication of performance in a particular area, valuable insights are frequently obtained in inspecting issues associated with specific events.
The SDP is the process for screening out significant individual events that in themselves provide a separate input to assessment.
Q - The ALARA inspection module appears to be "out of syne" with the other inspectable areas and Pls. The inspection seems greatly expanded over what is required by 10 CFR Part 20 and what is currently in place, even though industry performance in this area has steadily improved (as evidenced by WANO indicators).
.The framework philosophy behind the ALARA area seems to be more INPO-like (in search of continued improved performance) than the stated objective of this NRC initiative (to optimize health and safety objective while minimizing regulatory burden).
A - The purpose for modifying our inspection approach was to make it risk-informed. In the 2 radiation safety cornerstones, this has resulted in a focus on dose. Collective exposure as well as individual exposures are risk-significant. In recognition of this, the ALARA inspection area has been more broadly developed. In light of overall industry improvement in this area, the SDP recognizes those nuclear power plants that achieve overall good ALARA performance. The past 3 years of industry dose experience was used to derive average BWR and PWR annual collective exposure constant values. Any nuclear power plant with overall ALARA performance better than those_ values will result in " green" assessments for all ALARA findings. With respect to 10 CFR Part 20 requirements, the statements of consideration indicates that the NRC agreed
' with a commentor that ALAHA should include a lower bound (which the SDP now reflects). The rule also indicates that licensees' level of effort should reflect the magnitude of potential exposures, both individual and collective doses. The ALARA inspection procedure also provides a terraced approach of inspecting licensees according to overall collective dose performance and with respect to source term. This approach is consistent with the ALARA rule.
Q - The ALARA draft inspection procedure included criteria associated with exceeding an ALARA goal by a given percentage. Is this still included? It appears that including such a criterion will effectively eliminate challenging ALARA goals from station outage planning.
20
A - Yes, the current draft ALARA inspection procedure provides quantifiable criterion with respect to ALARA job performance. This is based on exceeding the ALARA estimate by greater than 50% for high collective dose jobs. The estimate is not the same thing as an ALARA challenge goal.
Q - Give examples of ALARA findings.
. Related O - What is an ALARA finding?
A - Occurrences where the licensee has failed to implement engineering or other controls to maintain the total collective dose for a job package constitute an ALARA finding. Individual doses are not ALARA findings.
Q - Please describe what a source control finding is as referenced in the ALARA flowchart on p.
284, and provide a couple of examples. Why was example 5 on p. 286 of the conference I
materials a source control finding?
Related O - What is a source control finding? I'm used to sources being controlled, inventoried sources.
A - A source control finding occurs when there is a loss of control of the radiation source--l.e.,
j that which caused increased dose rates at a job site--is the cause of the ALARA finding, j
Q - Doesn't the SDP result in counting those events already captured under the Pis? If so, doesn't this drive more NRC response than intended due to multiple white findings being
)
evaluated under the action matrix?
A - There may be an element of double counting, but the SDP is intended to focus on those
. events / issues that are significant enough to have an independent input to assessment.
Questions on the Safeauards Stratealc Performance Area Physical Protection Pts
)
I O - The security intrusion detection system (IDS) indicator uses a " normalization factor." If there is 1 IDS segment, the metric requires equipment to perform at 95% reliability / availability.
If there are 2 IDS segments, the metric requires equipment to perform at a 97.5%
reliability / availability, if there are 39 IDS segments, the metric requires equipment to perform at 99.975% reliability / availability!
Is there an opportunity during the pilot program to revise the normalization factor to reflect a 97% availability?
Related O - Your slide shows (p. 6 89) 95% reliability. The normalization factor requirer a 99.975% reliability for a 39 zone system! Is the normalization factor going to be re-done?
21
A - See the calculation factor in NEl 99-02 (draft B) on pages 97 & 98 and note that normalization factors are general approximations and may be modified as experience dictates (see note on page 98 of NEl 99-002).
O - What is reported quarterly for physical protection? Is it hours out of service divided by hours required? Is any failure of a camera or perimeter detection, regardless of size or area reviewed, counted?
A - Each zone of detection and each camera is counted. However, if one officer can cover two zones then only one compensatory hour is counted.
O - If two people complete one zone, is that two completed hours or is it just counted as one?
A - That is two completed hours.
O - At a utility with multiple nuclear sites, if a fitness for duty (FFD) or access authorization (AA) failure is attributed to an individual with access to both sites, which site should report this with their Pls?
Related O -Is a failure in AA/FFD assigned to a plant or a site etc.?
l A - The site the individual normally works at would report this, unless at a plant site such as Salem / Hope Creek. Then the security organization reports it for both plants. Failures are generally assigned to a site security organization.
Physical Protection Insoection O - 10 CFR 50.13 states licensees DO NOT have to design or provide measures to protect against attack by enemies of the United States. So what is the legal basis for OSRE7 50.13 appears to excuse plants from having to respond to OSRE.
A - 10 CFR 50.13 refers to an enemy in a military operation.
O - Page 106 of the conference materials says " interdict and defeat" the design basis adversary force. The regulation says " protect against." What is the expectation difference between your words and the regulation words?
A -it will depend on what each security plan requires, but if the security organization cannot
' protect against" a threat then they have failed to effectively interdict that threat.
O - In the currently existing program,93 hours0.00108 days <br />0.0258 hours <br />1.537698e-4 weeks <br />3.53865e-5 months <br /> are allocated for a dual unit site. How many i
hours are expected from the new piogram?
A - Access Control: Estimated 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> - Annually Access Authorization:
Estimated 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> - Annually Response to Continge ncy Events: Estimated 104 hours0.0012 days <br />0.0289 hours <br />1.719577e-4 weeks <br />3.9572e-5 months <br /> - Biennially (2 years) 29
O - Inspections are supposed to be risk-informed and performance based. Conducting interviews to determine if a manager is aware of behavior observation requirements is NOT performance based.. Failure to detect fitness for duty problems is performance based and is all that should be inspected.
A - For this area, it was determined that in order to assess performance, interviewing has a direct correlation to the supervisor's understanding of the problem.
O - Is the personnel screening Pl counting positive;?
A - No, it is only counting failures of the program.
Questions on the Baseline inspection Proaram General Q - There are three cross cutting issues. Only one, problem identification and resolution (Pl&R),
4 has an identified inspection procedure, Why are there not inspection procedures identified for safety conscious work environment and human performance as there are in Europe?
Related Q - How will human performance, safety conscious work environment, and problem identification and resolution be evaluated / assessed and what measurable criteria will be used?
Related O - The 2515 document states the program "is intended to place inspection emphasis on elements of licensee activities most important to reactor safety....etc." The document later states "the baseline program does not include routine oversight of operations." Does the NRC not consider operator performance, or inspections thereof, as activities important to reactor safety? Would not reviews of communications, command and control, and procedure adherence during routine operations give important indications (as opposed to diagnostics) of how operators might perform in situations important to safety? Non-routine events, event followup and requalification are too limited and infrequent to be an indicator.
A - The three cross cutting issues are imbedded within all aspects of the inspection program and it is a basic premise of the new oversight process that problems in these areas will manifest themselves in degraded performance indicators or significant inspection findings. Among these cross-cutting issues. Pl&R is treated differently because of its impact on all aspects of plant evaluation, and the emphasis the new process places on licensees having an effective and i
comprehensive corrective action program.
Human performance is specifically inspected in the inspectable areas of operational
- requalification and non-routine events. Performance in this area is also measured on an indicative basis through the Pls and other inspectable areas and observations made by the I
resident inspection staff while performing the plant status portion of the inspection program as well as other inspectable areas will continue to provide appropriate case results.
We are pursuing with'the Commission the directica on safety conscious work environment.
Currently, it is incorporated into Pl&R to ensure inspectors maintain an awareness of indications that a problem in this area may exist. The agency's allegation program continues to serve as a key indicator of potential problems in this area.
i 30
- Q - What's to prevent double-dipping, that is, having a PI that the inspector also writes up?
A - This might happen for some Pts (e.g., scrams). If the finding is risk significant, the NRC needs to take appropriate action so it is ok to double count in these instances.
Q - Are findings pegged to one comerstone? (Some findings may apply to multiple comerstones.)
A - Yes, the inspector needs to associate a finding to the one cornerstone most closely related to the issue.
Q - If an inspector follows up on a white finding and finds it's more significant, how is the finding handled?
LA - The inspector should recharacterize the initial finding, but do not record a second finding.
O - Many of the "inspectable areas" apply to multiple cornerstones. Should we expect multiple inspections or a single inspection that covers all cornerstones. For example, problem identification and resolution applies to all 7 cornerstones. Will we see 7 inspections or something in between?
A - For Pl&R there will be one annual inspection that covers all cornerstones. Inspectors will also include some inspection of Pl&R in each of the baseline inspections, inspections will be scheduled and conducted similarly to the current program, scheduling according to inspectable areas, not procedures. Inspectable areas that cross cornerstones will not be scheduled separately.
Q - When you say that initiative inspections will not be done for performance in the green zone, against what do you measure the " performance" - a finding, a cornerstone, a strategic performance area or at a plant overall?
A - Agency action beyond baseline is determined by the number, of affected cornerstones and significance (i.e., white, yellow, or red) of individual Pls and inspection findings as described in the action matrix.
Q - Will regional inspection be only in the non-green areas or in all areas?
A - The baseline inspection program is the minimum inspection for all plants. Regional specialist inspections are included as part of the baseline inspector program. Inspections over and above the baseline will be based on "non-green" findings or Pls or significant events.
Q - You said there were 37 inspectable areas. Page 93 shows 2 slides that are very similar-
- however, the second one adds " drill and exercise inspection." Between pages 93 and 94 I add up 37 or 38, depending which chart on page 93 is read. Also, on page 100 the second chart I
identifies 36 inspectable areas. Please comment.
\\
A - There are 37 inspectable areas (IAs). Errors were introduced into the table when it was converted for the book. It should be noted that the number of lAs may also change in the future based on lessons learned during the pilot program.
31 h
Q - Currently there are " core inspections." How do the new " baseline inspections" compare?
Specifically, is there any area which will have more inspection than under the old system? [ Note
- the new draft radiation protection baseline hours did exceed the old module).
A - Many areas being inspected are the same, although the focus or emphasis may change.
For example, approximately 65% of the core was focused on operations, while about 35% of the baseline is focused on operations. The hours provided right now are estimates and we expect them to be modified to reflect experience gained during the pilot program.
- Q " Supplemental" inspection is being used to mean two things: Baseline inspection that supplements a Pl and initiative that supplements the baseline when the green / white threshold is crossed. Please use more distinct terms.
A - The inspection program has used " supplemental" relative to inspections to cover areas not covered by a Pl. We will strive to keep these concepts clear. We are not trying to
(
" institutionalize" the terms " complementary" and " supplementary" within the baseline program other than to indicate that inspections serve a 2-fold purpose with respect to covering inspectable areas not touched on by Pls. We are still looking at the terminology to describe inspection effort beyond the baseline.
O - The different terms used for inspection types is a good idea. It helps capture the type of inspection and what is being done. Please provide a short definition for each of the types, the hierarchy between them, and which is being replaced.
A - BASELINE: In the new inspection program, this is the minimum inspection activity at each reactor site.
SUPPLEMENTARY: A term used to describe baseline inspections in an inspectable area not fully covered by a Pl.
COMPLEMENTARY: A term used to describe baseline inspections in an inspectable area where Pls provide no insights.
REGIONAL INITIATIVES: A term to describe current (not the new oversight program)
)
inspection effort instituted at the discretion of regional offices to inspect areas or issues of l
concern beyond the core inspection program.
SUPPLEMENTAL: In the new inspection program, this is the current term used to classify inspection used to provide additional oversight in areas in which a PI threshold has been crossed or a significant finding has been identified.
O - Please review the following activities to the baseline. Some annual, monthly, and 4
semiannual activities were mentioned. Are these additional activities, mini-audits or are they done off-site?
A - Each inspectable area has an established frequency. Hence, the various monthly, semi-annual, etc. requirements. Some event followup is contained within the baseline program.
Additional event response will be performed at the site or in the regional office, as appropriate.
32
~ Q - it seems like there will continue to be substantial inspections required under the new oversight process. Is there a plan to reduce the number of inspectors in the next two years assuming continuing improved performance by the industry?
- A - One of the purposes of the pilot is to determine how much effort is needed to implement the
. baseline inspection program. This will continue to be assessed through fullimplementation of the program. Experience will help dictate the overall scope and breadth of the baseline inspection program, which is expected to be reviewed on an annual basis.
Q - Will the NRC still have 3 residents at dual unit sites?
A - No staffing changes are required during the pilot program, but these sites may need to engage the 3rd resident in other activities besides the baseline inspection program (e.g.,
training, inspections at other sites, program development).
Q - What roles do the regional SRAs play in this new process? Please give some specifics.
A - SRAs will provide support in reviewing issues that are evaluated using the significance determination process (SDP) as well as participate in inspection planning for selected inspections to assure appropriate risk insights are incorporated.
Q - Regarding the issue as to what to call " findings," we need something more tentative so that the item can be further explored as to validity and significance. Two possibilities: 1)-Questions (already used a lot too), 2) areas of inquiry (more elegant and new).
Related O - What are the two types of findings discussed on page 143 of the conference materials?
Related Q " Finding" has two meanings. Please review / explain the two meanings of
- finding."
The explanation during the presentation was unclear.
Related Q Please reiterate the "two meanings" of a " finding." The explanation was unclear. Do you mean to say that it has one meaning before it is subjected to the SDP process (i.e., a
" preliminary finding") and another (risk-informed) meaning after it is subjected to the SDP process?
Related O - We need a clear definition that a finding is a non-compliance, not anything that an inspector wants to question.
A - Findings are those issues that have more than negligible impact on meeting a cornerstone
. objective. The two meanings of findings are: 1) a finding which merits risk significance assessment in the SDP but is not yet assessed, and 2) an outcome finding as assessed by the SDP. We need to re evaluate our use of term " finding" and its definition in IMC 0610. The program already defines inspection " observations" and that may be sufficient.
. Q - Under " example of summary of findings"it is not clear to me why the item is shown as a
" finding" since there appears to be r.o violation.
33
A - Findings, as they apply to the baseline inspection program, are based on risk-significance as well as compliance with the regulations. Enforcement is not a driver of the process, but an output of assessment. The example is a finding because it has risk significance.
O - Is the 40501 inspection module being resurrected and how prevalent will it be?
A - Procedure 40501 is not appropriate for the baseline. The reference to this procedure was in the context of a regional initiative to address a particular concern. Procedure 40501 will still be used in this way when appropriate.
O - Has the NRC completed a risk assessment to ensure that the current level of effort in the areas of steam generator (SG) replacement and spent fuel storage installation is appropriate to this new process?
A - Some licensee activities, such as SG replacements or independent spent fuel storage installation, are of such a magnitude in terms of scope and level of effort that for purposes of public confidence alone, absent any risk evaluation, they warrant NRC attention and oversight.
Part of the development process for inspections beyond the baseline will assess the appropriate level of effort for activities such as these.
Corrective Action (Pl&R)Insoections O - Effective corrective action has been a subjective " fall back" for the NRC in the past. Will the new process change or better define what is effective corrective action or place some risk-significance on specific issues?
A - While the majority of the baseline inspection program is indicative in nature, the efforts to assess the effectiveness of a licensee's Pl&R are spread throughout the entire inspection program. It is expected that the NRC will be able to make qualitative assessments of the licensee's corrective action program (CAP) and that these will be documented in inspection reports. It is presumed that the licensee's CAP is effective and does not warrant additional inspection effort if PI's are in the licensee response band and there are no other than green j
inspection findings.
O - Does the 71152 procedure include specifics on effectiveness of licensee Pl&R7 A - It includes guidance on assessing the effectiveness of corrective action programs. There 1
are no specific criteria. The NRC and NEl are discussing the possibility of an industry standard i
for CAPS. See previous question.
O - What initiatives are ongoing to allow flexibility to OA programs to focus on risk-significant areas / activities?
A - The baseline inspection program does not limit licensees' OA efforts. There is an initiative to risk-inform the 10 CFR Part 50 regulations that will include Appendix B.
O - If minor violations are not worth the NRC inspector's time to document and follow up, is it wo-th the utility's time and effort to correct the minor violation?
34
A - All violations should be placed into the CAP for appropriate prioritization and disposition.
Minor violations should not normally be documented in inspection reports. If the violation has no value, perhaps action should be initiated to modify or eliminate the requirement. However, compliance with the regulations is an important basis for assuring safe plant operation and the NRC will continue to identify such issues.
O - it was mentioned on the annual inspection that NCV's would be reviewed for corrective actions to prevent recurrence. Per 10CFR 50 Appendix B, prevention of reoccurrence only applies to significant conditions adverse to quality. Is the application of prevention of reoccurrence being changed? Currently, this is where the trending aspect of the CAP is typically applied for identification of issues to address.
- A - No, corrective action to prevent reoccurrence is still only required for SCAQ. The annual inspection of Pl&R will review a sample of significant issues and NCVs to ensure 1) compliance is restored and,2) corrective actions required to prevent recurrence have been taken.
Event Followuo Inspections O - What triggers an event followup inspection? It would seem that a clear threshold is needed.
What criteria will be used to determine if additional regional initiative inspections will be conducted? Is it the significance determination process thresholds?
l A - Event followup is normally triggered by 50.72 events, but may be triggered by significant events that aren't 50,72 issues. The NRC is currently developing guidance that will be more risk informed in determining the levels of NRC response to events.
O - If an event relating to EP, RS or safeguards occurs will those specialists use the event followup procedure? If not, what will they use to assess the significance of an event?
A - We will develop an " event followup procedure" that can be used by specialists. The existing event followup procedure was written for all areas, but it is anticipated that residents would determine, with appropriate regional guidance, if followup outside the baseline is necessary.
Efforts are in process to develop a process similar to the significance determination process as
)
a guide for determining if an event needs followup. Regional specialists could use the current i
procedure for initial review of LERs and events in their areas.
O - Does the NRC wait until a LER is issued to perform the SDP on an event, or do they do it after they have gathered enough information shortly after the event?
A - First, the SDP is not used on events which involve initiating events (scrams, LOSP). Risk assessment in these situations is performed by SRAs. For IE-type events, inspectors will not wait for LER issuance to assess if the event is risk significant. For events which are not really events but require 50.72 and 50.73 event reports, residents will use SDP screening criteria and, if the issue is a SDP candidate, will continue to gather information. If the issue screens out, they will wait for the LER.
i 35 i
Inspection Documentation 0 - Please explain the six week inspection process once again. Are we going to have a six week exit and a six week inspection report? I was of the understanding that the six week process was being canceled.
A - For the pilot program, the six week resident inspector (RI) inspection period will continue.
Inspection reports and RI exit meetings will continue on a six week cycle.
Q - If performance indicators are provided on a quarterly basis, why should resident inspections be issued every 6 weeks? Why not time them to coincide with the PI report?
A - This is under consideration. It may be a lesson learned from the pilot program.
O - Page 151 of the conference materials says that there is no entry in the PIM for minor violations. Aren't green results " minor" violations? For example, the examples on p.151 (top slide) are green noncited violations. Isn't this minor? Why is it in the PIM?
A - First, noncited violations are not minor violations in the new enforcement process for the revised oversight program, minor violations are less than " green" findings. Green findings will result usually in a noncited violation. This is, by definition, more significant than a " minor" violation.
Q - On page 143 of the conference materials, the second slide says " positives are NOT documented." What ' oes the NRC inspector document if all observations are positive? (For d
example, how would EP inspections with no negative issues, including no green findings, be documer.ted?)
A - The inspector would document what inspectable areas were looked at, and to what extent they were inspected and note that there were no findings of significance.
O - If an inspection report has a finding with no color, how is this tracked if there are no j
inspection follow-up items?
A - Until the SDP process is complete, an ongoing issue would likely be carried as an apparent violation if a requirement has potentially been violated. The NRC is looking at the best way to characterize such issues.
O - What is the disposition of an inspection observation?
A - It is not documented, just turned over to the licensee for inclusion, as appropriate, in the corrective action program.
O - Must each resident inspection report contain findings for every comerstone?
4 A - No. However, every finding will need to be aligned with the most appropriate cornerstone to avoid double counting.
O - If only negative comments / findings are going to be documented in inspection reports, the public perception of nuclear power is going to be very skewed. P! ease comment. The NRC will 36
need to articulate that this program enhances our safety role by allowing us to better focus on those things would challenge safety evaluations.
A - Our efforts to include positive findings have had mixed results in the past due to a lack of criteria and resultant inconsistency. In the revised oversight process, the public will have more information than what is just contained in the inspection reports (e.g., Pis). Also, evaluation that takes place as part of the SDP will result in positive observations of plant conditions to mitigate potential accidents.
Q - When will the public be notified of findings? When are they docketed?
A - NRC findings will be documented in inspection reports (issued every six weeks), as they are today. The final decision is also docketed in an inspection report.
Q - Will the NRC inspector be given a master computer disk for preparing their reports. This would help make standard a format for reporting. Will the findings and report be placed on the NRC Web page?
A - There are no current plans for HQ to provide a master disk, although guidance on inspection report format has been provided and the program office will review reports during the pilot program to assure consistency. PIM entries are intended to be available on the NRC's I
Web page.
O - Will there be an effort to eliminate " narrative" " diary" inspection report write ups?
A -The new guidance on inspection report writing is intended to minimize the level of narrative.
inspection Plannina O - Are the plant specific risk information matrix (RIM) documents available for plant.
use/ review? Or when will the RIM documents be made available?
A - Plant specific RIM 2s are draft documents for use by inspectors during the pilot. They were provided to the pilot residents at a workshop in April. They will be made available to public and the pilot plants in the near future.
O - A plant specife risk significant systems matrix will be used for baseline related activities.
What is the source of this matrix (i.e., licensee's MR determination of IPE information, which is potentially outdated, etc.)? Who will determine the content of this matrix?
Related O - Who develops the site specific risk table (RIM 2) and how? How are the results j
validated?
A - Plant specific RIM 2s were developed by the Office of Research (RES). They are drafted and are still being reviewed. The plant specific risk information matrices will be developed by RES's contractor from plant PRAs and IPE/IPEEEs.
O - RIMS are developed from IPE insights. Some plants' PRAs show differences in risk from the IPEs. How is this reflected?
37
l A - The new oversight processes are risk-informed, not risk-based. Plant-specific RIM 2s are being developed. Plant specific information will be captured to support inspection planning using RIM 2. Where differences exist, it is expected that licensees will share this information with residents.
Questions on Supplemental Inspection f
Q - It was noted that a single failure could move a safety system out of the green band if it has a long surveillance window and " fault exposure" hours are applied. Plants were told to "make a i
notation" to this effect when submitting data. How will this case be handled with respect to sucolemental inspections? What about the case where unavailability moves a system out of the green but the site PRA says the unavailability is still acceptable? How will supplemental inspections be handled? (Will the situation be resolved promptly, before upper management contact results in new, unnecessary commitments on system performance)?
A - Inspections will focus on assessing the root cause of failure, extent of condition, and adequacy of corrective actions. We will need to consider revision of the threshold if the situation described here is the case. Performance may still be questionable and we need to understand why the threshold was exceeded.
O - Will supplemental inspections be used "in lieu of" augmented inspections or are they similar in scope and intent?
A - AIT's will be incorporated as part of supplemental response to determine root cause, extent of condition and adequacy of corrective actions when the situation warrants. This guidance is currently being reviewed for applicability in the revised oversight process.
O - How are the three main objectives / attributes of the supplemental inspection program different from the problem identification and resolution annual and cornerstone inspections?
A - Except for plants with multiple degraded cornerstones, supplemental inspection is focused on specific problems within the affected cornerstone of safety and the root cause that led to them. Pl&R inspection attempts to provide a general indicator of performance in Pl&R areas across all cornerstones.
O - Is follow-up inspection looking at the CAP?
A - Yes, the follow-up would look at the CAP in the area of the original finding.
O - Supplemental inspections seem to be based on a " white" indicator due to a problem. What about a " white" that is due to a well planned and thought-out plan, i.e, safety system unavailability (SSU) that is planned and below the PRA value?
A - This case would indicate a need to re-evaluate the appropriateness of the PI threshold. If such a situation were to occur, the supplementalinspection effort should come to the same conclusion and use a minimal effort by the NRC to ascertain.
38 J
l i
Questions on the Sianificance Determination Process General O - How does the new risk-informed process factor in the NRC (regional and NRR) responses to " resident concerns", part;cularly those with a short turn-around time? For example, equipment is taken out of service for preventative maintenance and is pre-analyzed by the site PRA as acceptable. The resident still has " concerns" and contacts the region. Will NRC be able to provide quick risk-informed feedback? (In a recent event, feedback arrived from region and NRR six months later, after utility had addressed NRC " concerns" by committing not to do the work in the future. Feedback indicated doing work was acceptable.) Feedback was way too late.
A - The SDP will provide the inspector with a tool to more quickly determine the level of risk
. significance at a rough approximation. The complexity of the issues will r*fect the timeliness of
. responses from regional and HO specialists.
O - Is the feasibility review (DC Cook, Millstone, St. Lucie, and Waterford) available for review?
How can it be obtained?
A - It was induded in SECY 99-007A as Attachment 3. It is on the NRC External Web page.
O - In Morris Branch's presentation on the results of the feasibility study for the new significance determination process, he stated that the feasibility study results indicated that
" actions proposed by the new assessment process were similar to past actions, but the new process is more scrutable." In the case of DC Cook, application of the SDP process did not result in identifying the need to conduct a detailed design look via an AE type inspection. The
. AE inspection resulted in the identification of a number of significant design issues. The NRC would not have initiated an AE inspection based on the SDP results for findings existing before
, the AE inspection was initiated, hence, the focus being placed on design insp'ection activities. I suggest you address the DC Cook situation and put it in proper context. I have probably answered my own question / point, but I think this aspect of the feasibility study results should be mentioned so that licensees understand why the NRC is focusing on design type inspections to
- the degree it is.
A - The feasibility review for DC Cook did identify that the NRC assessment process would not have predicted the problems. Because of this effort, the feasibility review made recommendations to add more inspection effort to the baseline inspection program in the area of engineering and design verification.
i O - It was stated that the NEl table top exercise resulted in a " red" for complete loss of ECCS recirculation for a long term for the medium break LOCA. Please explain how this was determined since it looks like the significance determination process matrix (p.199) never gets beyond " yellow" for any medium break LOCA or large break LOCA condition?
t A - The frequency of medium LOCA was reduced by about one order of magnitude between the draft AEOD study (NUREG-5499) used for the feasibility study and the final version of NUREG-5499. Additionally, the small break LOCA which has a greater likelihood of occurrence may also drive the risk of the degraded condition even to " red."
39
r Q - Since the ASP process used LERs as its initial input for determining risk significance of events, shouldn't the LER process also be risk informed? What efforts are being taken to risk inform the LER process?
A - Work is being done on revicing 10CFR50.72/73 guidance and should be consistent with other agency initiatives. Industry needs to be aware that LERs currently apply a significance categorization to the issue and any information that can be used to properly communicate this to the NRC and public should be considered.
Q - is the SDP process only for NRC findings? If it is, how did the BWR4 example with the problem condensate pumps end up going through the SDP7 Related O - Does it make any difference who identifies the findings that are evaluated by the various SDPs?
A - The SDP is intended to be applied to issues identified by either the licensee or the NRC.
Additionally, the NRC may also review issues involving failures of non-safety related equipment t
as well as safety-related equipment if it represents an issue of potential risk significance.
O - What issues get you into the SDP? LFRs? DRs?
A - Any finding that deals with plant risk should be screened. This includes LERs and DRs if they are handled through inspection.
Q -is the N3C or NE1 going to create a computer-generated SDP system similar to Turbo Tax
(" Turbo Finding")?
i A - While the SDP could lend itself to a computerizC process, and may be considered in the future as more experience is gained with the process, the assumption-driven aspects of the Reactor Safety SDP necessitates, for the near-term at least, that the process be utilized manually.
O - Will the screening work sheets be " codified"? If so, based on 50.59 experience, we should all anticipate a significant high volume of documentation to manage. Since a large fraction of findings will screen to green, how will this be controlled? Or, what does NRC expect to require?
Related O - Will the SDP worksheets be published? If yes, where? Will licensees be required to document conclusions using them? Will the NRC be required to document conclusions using them?
Related O - Will the licensee have the opportunity to review significance determination process (SDP) worksheets that are completed by resident and regional inspectors for findings to ensure that assumptions listed in the worksheets are accurate? For regional inspections, will SDP worksheets be shared with the licensee prior to the exit meeting so that any potential misunderstandings can be resolved before the exit meeting?
A - No, the licensee will not be provided " draft" copies of the worksheet. The worksheets are not mandatory. The information, including all assumptions that an inspector uses to conduct their review, will be documented in an inspection report. The worksheets are only intended to be a tool available to inspectors to assist them in executing the SDP.
40
n O - Is it expected that the licensee's Pl&R process go through ito own SDP for all non-conformances?
Related O - Does significance determination imply anything about whether the issue is a condition adverse to qua",ty (CAQ) or a significant condition adverse to quality (SCAO)? For example, is a yellow finding always a SCAO and a white finding never a SCAO or some other such correlation?
A - Licensees currently control resolution of issues in their Pl&R pocess.10CFR, Part 50, Appendix B, as well as, Generic Letter 91-18 revision 1, provide additional guidance as to what is necessary for effective problem resolution. The significance of the hardware, human i
performance or procedure issue will be evaluated and this will be used to characterize the significance of it e PIR identified problem. While licensees need to comply with their Appendix j
B OA program policies and procedures, they should not completely rely on the SDP to Jetermine significance as it applies in this context.
O - Why does the SDP process use old terminology--i.e., risk significant, non-risk significant--
when Regulatory Guide 1.160 Revision 2 specifies "high safety significant" and " low safety significant"? Can we assume they are the same?
A - Safety significance for the maintenance rule was determined by the licensees using an expert panel with certain PRA insights as an input. The primary PRA input to the process was PRA importance measures such as RAW and RRW. The SDP process estimates a change in core damage due to a performance problem at a plant. They have similar meaning but different metrics.
O - Non-PSA/PRA people will be using Tables 1 and 2 in the Phase 2 risk estimation process.
These tables do not appear to have any margin for error, and may even be an order of magnitude on the non-conservative side (e.g., SGTR, MFLB, stuck open PORV) or quite complex (LOCA determinations). What conservatism protects us against non-PSAS/PRA people in Phase 27 A - Initiating event frequencies in Table 1 are NRC's current best estimates. Conservatism is built into the Table 1 columns (higher exposure time value used in each column) and in Table 2 (generally conservative failure probability estimates for trains, systems, and operators), in addition, the pilot program will utilize at least 3 levels of review for SDP determinations:
Regional SRAs, NRC HQ risk analyst review, and SDP oversight panel review.
O - I've heard that the resident inspectors have had only two weeks of PRA training. Is that enough to use this process? Is further training planned?
A - The two week course should provide sufficient " background" in PRA to use the SDP.
However, training on SDP specifics will also have to be accomplished. The pilot program and the efforts to prepare inspectors for the pilot plants will help in determining how much training will be needed. The intent is to provide enough training for inspectors to use the SDP.
O - What is Phase 3 evaluation and where is it used?
i 41
En L
A - SECY-007A notes that Phase 3 will be performed by a risk analyst and wi'I utilize whatever methods / tools / approach are most appropriate for the specific issue. This was described on general terms to allow maximum flexibility in gaining the best risk analysis.
, O - Are there predetermined CDF or delta CDF thresholds for the colors for Phase 3 of the
- significance determination process?
A - Yes. As described in SECY 99-007A.
' O - Is Phase 3 optional or can a licensee request it so that a quantitative assessment can be applied?
A - The SDP is the NRC's process and the NRC has final responsibility for making a significance determination. If a licensee performs a more detailed PRA evaluation and th; result differs from the SDP, then we expect to dialogue with the licensee to understand why tnis difference exists. If reasonable assumptions are made by the licensee analysis, then, in view of i
the fact that SDP is a very approximate " order of magnitude" tool, we would likely accept the results of the more detailed analysis.
Q - Will licensees have access to or be involved in the SDP process? Will there be a process to I
resolve disagreements?
Related Q - At what stage of the SDP does the NRC interact with licensees to ensure
)
assumptions and conclusions are based on facts and current, up-to-date information?
Related O - How does the licensee get involved in determining significance of findings? When do findings get evaluated? When do site risk analysts get involved in enforcement?
Related Q - What is the process for a licensee not agreeing with the significance of a finding?
Related O - What is the licensee role in this process? Is the scope " limited" to inspection findings and LERs?
Related Q.What is meant by " assumption" in the SDP process? Do you revisit assumntions in Phase 37
- A - An example of an assumption is whether the equipment was operable. Assumptions are reviewed for applicability during the Phase 3 analyses.
Licensees need to understand how the NRC plans to use the SDP on a day to-day basis.
We expect that inspectors will interface with the licensee throughout the process. The SDP L
procedure requires interface after a Phase 2 review determines an issue is other than green.
Phase 3 of the SDP will involve discussions between NRC and licensees on risk cheracterization. Phase 3 SDP results will be documented on the docket. No final color is assigned to a finding until after the Phase 3 review. The NRC bears final responsibility for all significance determinations.
42 l
I
During the pilot and the first year of implementation, the staff will have a SDP review panel consisting of HQ and regional personnel to review the results of SDP evaluations to assure consistency and identify needed refinements.
i
. Q - Why are subjective, vague words still used in this process (most aspects are specific)?
Examples from the mitigating systems SDP questions include:
No expected impact" -
" Substantial potential"
" Compromised ability".
" Low risk" "Some risk" Affects" equipment in maintenance rule (Inoperable? Degraded? Inoperable but still functional?
A - This step was not intended to be vague. If the item affects reliability or availability of
^'
equipment that is significant under the maintenance rule, it needs to be evaluated.
Q - Licensees update their risk models periodically, but do not report changes to NRC. It-appears there may be a lot of time wasted dialoguing over issues using inconsistent models.
Will this result in licensees having to maintain NRC with a living PSA? Also, how can licensees be sure that the NRC model (in the RIM) is even valid since we did not review it?
A-There is no requirement for licensees to submit changes to their PRA models. While there may be some inefficiencies in the process because of using different data the NRC will use the
' best data available to make its final decisions. The risk-lnformed tools have been developed to allow inspectors to make a general rule assessment. It is recognized more detailed information is available in site-specific PRAs. Licensee feedback on the RIMS will be accepted for
. consideration.
Q - Are we free to use licensee success criteria in the SDP if analysis exists to support it?
A - Yes, the NRC and licensee may negotiate which success criteria to use, but the NRC retains responsibility to decide what assumptions are reasonable.
Q - It is recommended that " utilities translate PSA information into a form compatible with the process"(see p.~ 161) of the conference materials. Will Tables 1 and 2 be plant-specific? (It is noted that PSA results don't seem to be used in completing the Phase 1 worksheet and Phase 2 estimation.) What guidance is planned to best assure plant-specific activities are handled g
consistently across industry?
Related Q a It was recommended that licensees convert PRA information into a form that is compatible with the SDP process. Should this transformed information be provided to NRC? If so, is it provided to headquarters or the residents? -
43 l
l i.
A - There is no plan to make Tables 1 and 2 plant specific. The slide on page 161 was associated with an industry presentation and should not be interpreted as the "NRC-recommendation."
i
' O - A Phase 3 detailed risk significance evaluation implies the need to use the latest utility PSA.
)
The NRC does not have access to the models but is ultimately responsible. How will this be
)
resolved? Will the utt'4 provide the evaluation with NRC review?
A - During the revicw of an issue, the inspector and SRA will have access to data available to the licensee. The Phase 3 process may be done by the licensee with the NRC reviewing and evaluating the results or the NRC may perform their own Phase 3 refined review.
2 0 - The NRC is amending the maintenance rule to require an assessment of the impact of removing risk significant equipment from service for maintenance. Can the SDP be used to perform this assessment? It would appear that the SDP offers a relatively simple, yet consistent, way of evaluating the significance of removing equipment from service for maintenance.
A - The SDP requires equipment that is used for mitigation to be within the scope of the maintenance rule. The NRC is using the SDP to evaluate the risk significance of inspection findings. The NRC has not assessed the feasibility of using SDP for evaluating there risks associated with removing plant equipment for maintenance activities.
O - In cases where the SDP yields multiple findings, would there just be a single (worst-case) window finding?
A - Yes. Each event analyzed through the SDP will result in one disposition, even though the event may have multiple aspects reviewed through the SDP.
Related O - If you have two findings, one green and one white, would you use the color to prioritize what to fix or whether to look at technical specification changes?
A - The SDP was developed only for processing inspection findings. While it provides insights into the risk significance of a specific inspection finding, it is not intended to replace existing work prioritization schemes.10CFR, Part 50, Appendix B and GL 91-18 revised also provide guidance in this area.
O - Do all green findings become NOVs?
A - No, green findings will for the most part be NCVs.
O - Would repeat greens generally be expected to be observed in routine inspection reports?
If so, will low risk significance be acknowledged in the report?
A - No. It the repeat green is a failure of Pl&R, it should go through the SDP and be dispositioned accordingly.
C - White indicators should all have the same risk to health and safety.
44
A - While the NRC agrees with this comment, it should be noted that the SDP is not a tool designed to measure accurately the impact of an issue on risk, but to provide a rough approximation and to help define the appropriate regulatory response.
Q - it appears there are different entry conditions for moving into the SDP charts. For example, security uses " identification of non-conformance issues" vs a finding or violation identified as a Pl&R problem for EP. Are these just terminology or real differences in entry conditions?
A - This is just an artifact of the ongoing development process and efforts to achieve appropriate level of consistency are still in process.
_Q - The four reactor safety SDPs all provide for a pre-SDP " finding" to be evaluated through the SDP as a non-finding (i.e., no color assigned, therefore only an inspection observation if it doesn't meet a certain threshold). However, the two radiation safety SDPs and the physical protection SDP do not provide for this conclusion, i.e., the best any pre-SDP " finding can do is result in a green finding (per all the flowcharts). Was this intentional, or is there some pre-analysis of an observation supposed to be made before entering a radiation safety or safeguards SDP?
. A - There is some pre-analysis of radiation safety and safeguards findings similar to reactor -
issues.
Q - What kind of SDP is there for (1) human performance issues that may indicate operators or maintenance workers won't take actions required by procedures; (2) common cause failure issues; (3) safety culture issues; and (4) shutdown risk issues? If there is no SDP for them, what is the timetable to be able to address these issues through SDP process? Where do these issues fit in enforcement space, since they don't seem to be within the SDP process but also aren't willfulissues?
A - If there is not a clear tie between the identified concern and an increase in initiating event frequency other than a reactor trip or a reduction in mitigation equipment availability or reliability, the issue would be screened as a green finding having little or no negative impact on the cornerstone, if there are human errors or procedure inadequacies that impact mitigation, then the item may affect the cornerstone. The agency is still working on methods to evaluate the above'noted items contribution to risk.
Q - Why is there an SDP for EP human performance (planning standard during exercise performance and problem identification) but none for operator performance of the controls of the reactor (or simulator) or problem identification?
A - There is a requirement in 10CFR 50, Appendix E that requires a licensee to learn from drills and exercises. Therefore, items that continue to reoccur are not being identified and corrected.
Also, the reactor safety SDP does evaluate human performance and procedure quality as it affects initiation events and mitigation systems and barriers.
. Q - Is the new significance determination process overly weighted toward EP at the expense of l~
reactor safety?. For example, failing 3 or more RSPSs in a drill / exercise (with no actual event) results in a " red" whereas all mitigating systems actually inoperable for a large LOCA or ATWS for 29 days is still "greon." is this appropriate?
45
n s
A - The SDP for the at-powe'r reactor safety cornerstone was based on PRA insights. The EP area is more closely based on alignment with current regulations and policy. Failures of 3 or more RSPS is equivalent to complete failure of the EP program. We will monitor the thresholds in all areas during the pilot program and may have to adjust our values if necessary.
Q - In the reactor safety discussion, in the BWR4 example of both trains of standby liquid control inoperable for 18 days, the most significance attached by the GDP for a total inability to combat an ATWS was " white." However, in the physical protection d;scussion, one individual successfully making an unauthorized entry and damaging some equipment would be a " red" and the plant would be shut down.. Does this seem to be an appror iate actions for both of
- these occurrences?
A - The reactor safety cornerstone area uses risk insights to a much greater degree than the others and as experience is gained during the pilot program adjustments to the various SDPs may be warranted. The assertion that a single " red" inspection finding results in the plant being shutdown is not correct.
)
C - With respect to notification and emergency information, N1, N2, N3, and N4 all involve administrative violations of CFR requirements with no safety implications or consequences, yet they are considered " white." This is not consistent with othei parts of the new assessment process where similar violations are " green."
Related O - What is the basis for including failure to notify in the transportation area a white finding? There are many reparting requirements in the other cornerstones, but failure to notify is not tied to white findings.
A " Failure to notify"in the transportation area reflects the potentially close proximity of the general public to a source of radioactive material involved in an accident (on a freeway, for
. example). The first responder could be a local law enforcement official. This individual needs the emergency response information prescribed by 49 CFR 172.602 to effectively mitigate.an -
incident involving radioactive material. Thus, this can lead to a white finding based on safety significance.
Reactor Safety SDP Q - While performing an SDP for an actual event (say a relief valve stuck open with
. complications),'do we assign a risk frequency value of 1 in Table 17 How do we enter this in Table 27 Or, do we assign a frequency postulated in the table?
l A - SDP is not currently designed to assess actual events, only degraded conditions. Efforts are l
ongoing to design an SDP-like process for assessing the risk significance of events.
O - How does the screening process handle the difference between an actual system failure vs 1
a potential system failure (i.e., a ringle active failure potential which could fail the system)?
. A - Initially, the SDP can be used assuming the affected system is unavailable (failed). If the
' SDP renuits in a " green" significance, it may not be worth spending more regulatory effort.
However, if the finding is white or greater, then the issue should be reviewed in more detail by
' the Phase 3 process (accounting for the estimated increase in failure probability).
46
i C - The significance determination process appears to be primarily based on only one component of risk (i.e., the likelihood of occurrence of a specific event). It is hard to tell where other components of risk, such as consequences of the event are taken into account in particular, high consequence but low likelihood events (such as large LOCA, ISLOCA, ATWS, and MCP LOCA) are almost always green. Please explain where the SDP includes consequences.
A - The SDP is looking at only one consequence (core damage). There is nothing in the SDP process that would allow licensee noncompliance to the rules. TS and design bases which provide the design margin or defense-in-depth for design basis events are still required to be followed and complied with.
O - Where are the consequences of events taken into account in the significance determination process? The table seems to be totally based on a belief that the event won't happen (i.e.,
likelihood of occurrence).
A - The SDP process is used to estimate the risk associated with inspection findings relating to the degradation of the reactor safety cornerstones. The SDP process does not estimate the risks associated with actual plant events though afforts are in progress to develop an SDP-like
. process for this purpose. The NRC will assess the risk associated with actual events though other processes, such as the accident sequence precursor program or event analysis by NRC risk experts.
{
Q - is there an order of magnitude difference in risk in the SDP tables in both the vertical and horizontal directions?
A - Yes, but this is not true in the diagonal direction.
O - During the SDP presentation, the speaker stated that if no cornerstone is affected, you should not be in the SDP process. However, page 2 of the Phase 1 screening form clearly states "lF NO cornerstone is degraded then the condition semens OUT as green and is not assessed further by this process." Please explain the conflict. I believe that inspector judgment decides the items that go into the SDP regardless of cornerstone affiliation and the SDP properly assesses them as " green." In other words, any finding that is included in a report must go through the SDP. See page 7 of MC610* for a good listing of issues that would be noncornerstone degraded items. For example," insights of human performance" could be from just about any activity a licensee does.
A - The statement that if "there is no impact you are not in the process" was meant to imply that simple screening items may not even need to be documented. *MC0610,* along with the documentation guidance in MC06xx, should be applied.
Q - Do we expect the NRC to interact with the licensee on every LCO out of service time to calculate a risk impact? If so, it is a waste of utility and NRC time. If not, it was a bad example.
A - No, this would screen out in step 1.2 of the SDP.
Q - Please provide the AEOD NUREG number that formed the basis for the initiating event frequencies in SDP Phase 2.
47
A - It is NUREG-5499.
O -If a condition is found which raises the estimated frequency of a large LOCA from <110 yr 5
to 1 per 5 x 10' yr, do we pick the likelihood rating letter for Row VI or Row IV of Table 17 A - The user can pick whatever Table 1 row is appropriate for the condition being evaluated, but must justify any choices if they differ from the examples shown in Table 1. It may be useful to do a " sensitivity" evaluation to see if changing the IE f requency really will cause a color change in Table 2.
O - How do you relate the failure to the scenario?
A - The NRC will provide inspector guidance documents relating failure type to a specific accident scenario.
O - What is the source document for determining which IEs to consider? Where do you find out that condensate is a miti ating system?
0 A - Use the plant PRA; you will need to look at many of the event initiators.
O - Why don't we intend to use design basis documents to determine what to credit?
i A - Use whatever provides the best estimation of risk, including design basis documents.
C - Many systems are credited in the PRA but not in the design basis document.
A - Correct. Licensees must still meet their design and licensing bases and technical j
specifications requirements. The SDP goes beyond design bases in some respects because of its focus on risk significance.
O - For each plant, what PRA source will the NRC use to determine the SSC/IE and mitigating system relationships?
A - The best available information will be used.
O - In the SDP process, are safety system functional failures going to be limited to the 15 systems / functions covered by the performance indicator vs the 26 systems / functions AEOD uses to inonitor vs any reported under 50.73(a)(2)(v)??
A - The SDP process evaluates inspection findings, which are not limited to the systems i
i monitored in the safety system functional failures indicator, O - How would going " white" or " yellow" for a mitigating system PI be factored in to the SDP? Is
" yellow" for a PI the same as " yellow" in SDP?
A - Pl data is used by the plant assessment process and not the SDP. SDP and Pts use essentially the same scale for significance determination.
48
I O - Does the phrase " realistic accident anelysis - not Chapter 15" used in the reactor safety SDP mean that non-safety related equipment can be credited for mitigation?
A - in general, yes, however SECY 99-007A currently limits credit only to equipment scoped within the maintenance rule.
Q - Is it correct that all instances of a degraded containment barrier, no matter how large, would
- screen out of the SDP as not significant? For example, an equipment latch is left wide open, it only affects the containment barrier cornerstone, and does not contribute to a LOCA or a trip.
A - No, this needs to be determined. Such issues will be sent to a risk analyst for significance determination until further guidance is formulated.
- Q - What does one use to analyze events affecting containment?
Related O - Hon do containment integrity issues run through the SDP?
A - Currently, the guidance to an inspector will be to refer the issue to a risk analyst for help.
We will also have an operations support group in the PRA Branch in NRR to help. Additionally, there will be a risk oversight panel to provide guidance and direction. The current guidance -
does require any impact on containment heat removal to be considered when evaluating its impact on the ECCS systems that use the containment sump for long-term recirculation.
j j
Q - The step 1.2 SDP screen is entered only with findings that have more than negligible impact on meeting cornerstone objectives. What constitutes "more than negligible impact"? This is an important criterion since it determines what merits inclusion in the inspection report summary of
. findings and therefore the PIM.
A - The SDP is intended to be used to review an increase in initiating event likelihood or a reduction in the reliability or availability of a mitigation system. Therefore, if there is no impact on the above conditions, the issue would be considered to have little or no impact on the cornerstone.
i l
Q - What if an event is a loss of a single train < technical specifications AOT but is also risk l
significant according to the maintenance rule? The first criteria would " screen out," the second criteria would " screen in"? Essentially every system or train with a technical specification AOT is in maintenance rule. So the new maintenance rule screening question appears to negate the single train < > technical specification AOT screening question?
A - The Phase 1 screening has been revised. If the component is covered by a technical specification and the duration of the degraded condition is less than the technical specification allowed outage time than the finding Phase 1 screen pertains only to plant equipment. This is maintenance rule high safety significant, but not governed by a technical specification. Findings
' for these non-technical specification equipment will be screened out if the duration of the degraded condition existed for less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
'49
1
. Q - One screening criterion is that the loss of a single train is less than the technical specification AOT. Have you considered using the maintenance rule unavailability performance criterion? This parameter is a better indication of the cumulative time that a component can be
? out of service.-
- A - A selected group of SSCs were selected for PI monitoring similar to how risk significant
' equipment is monitored under the maintenance rule. The SDP tries to measure the risk impact of a single degraded condition time period, not the cumulative impact.
Q - Since the tech specs are not risk-informed, I could probably find a scenario whereby there is an actual loss of safety function of a system less than the tech spec allowed outage time that could be a white (since we have custom TSS). How would this be treated? I think Question 2 on the SDP screening table should be revised to be >TS AOT similar to Question 3 on screening.
A - This is precisely the question that prompted us to incorporate Question 4 under Mitigation Cornerstone Phase 1 screening. Question 4 should catch the type of issues inferred here.
Q - During a simulator scenario, a crew takes an inappropriate (not allowed by procedure)
-. action which renders all remaining safety systems for mitigating a large break LOCA inoperable (during a large break LOCA scenario). What would be the inspector's characterization of this finding? (i.e., through SDP). What will be the NRC response? If needed, assume a BWR, and that an error made the remaining electrical bus inoperable. Will this be considered a good finding if the training department finds it? Or will it be considered a significant finding, indicative of increased risk of operator error?
A - This would be considered a good finding with respect to the training program if identified and corrected by the licensee. However, if the inspector finds that this is a widespread problem, as yet uncorrected, then the NRC's actions may be different. However, as previously answered the impact of the concern has to have an impact on the cornerstone to be considered as an issue of high risk significance. The NRC's C'fice of Research is also working en a method to better risk inform human error concerns.
1 Q - Can a utility use their PRA to formally submit and have approved plant specific event types which correspond to the values listed in the Approx. Freq column of Table P 1
A - Table 1 frequency values are bassd on a recent industry-wide study. The intent at this time -
was not to create site-specific frequency tables. However, site specific information may be considered during the Phase 3 SDP review.
C - Question 4 in the Phase 1 screening process asks if the condition affects risk-significant MR equipment not subject to a TS AOT. Taken to its extreme, any out of service time could be questioned and put into the Phase 2 evaluation. At the same time, the maintenance rule
. requires on line maintenance to have a risk review to support the work. We could be doing j
^
Phase 2 evaluation for things already reviewed and authorized in accordance with the MR. This seems duplicative. I suggest that a screeriing time limit--like 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />s-be added to Question 4.
Related Q - Does step 4 on the Phase 1 screening process worksheet under mitigation systems imply that all risk significant equipment including non-tech spec equipment would be subject to 50
enforcement?' For example, some plants might have portions of instrument air or the condensate system as MR risk significant. These systems are non safety related. Would they now be subject to violation space? Their risk significant element would be in mitigation, but no event may have occurred.'
Related O - Are only maintenance rule systems considered as mitigators in the reactor safety SDP?.
Related O - Does the maintenance rule screening question apply only to non-TS equipment? If it applies to all equipment, TS or not, then nothing in the maintenance rule scope can ever be screened out, even if < TS AOT.
A - Mitigators are safety systems in the plant PRA, plus non-safety systems if they are scoped j
into the maintenance rule. The impact of non-safety equipment will also be considered when
' evaluating the risk significance of failures. We will change our guidance to clarify what is meant by maintenance rule equipment impact.
Enforcement will be based on violations of NRC requirements and the maintenance rule may apply with some enforcement impact on non-safety equipment.
Q - On the Phase 1 screening process, the check boxes for the " barrier integrity" cornerstone
)
are listed separately as (1) RCS barrier, (2) fuel barrier, and (3) containment barrier. The form j
states "if more than one cornerstone is affected, then go to Phase 2." Does an event under RCS barrier and containment barriers count as two affected cornerstones or one (i.e., barrier integrity)?
(
A - The guidance presently says to go to Phase 2 if rnore than two cornerstones are affected, meaning any of the three barriers in the barrier cornerstone counts only as one cornerstone.
However, if the RCS barrier is affected such that it has a higher likelihood of leaks or ruptures I
then the " mitigating event" comerstone is also affected.
l Q - In the Phase 2 risk evaluation worksheet, item 3 and note 1, who determines that sufficient time is available, that environmental conditions are acceptable, that adequate procedures exist, and that adequate training has been conducted?
' A - The answers will be situation specific dependent on the scenario and deficiencies assumed.
The value of the SDP is that it forces us to acknowledge that we are giving credit to certain
' equipment and operator actions. In doing so, we can either satisfy ourselves that such credit is reasonable or we can challenge these assumptions.
. Q - It looks like the Large LOCA (BWR) and ATWS-BWR frequency ranges are high. First,
)
large LOCA in a BWR can range from 6" to 36" with a great range of frequencies and short-1 4
4 term mitigation requirements. Should there be two ranges: Large LOCA (10 to 10 ) and DBA 4
LOCA (<10 )? Second, INEEL-recently released a report demonstrating that BWR RSs 4
reliability is 3.8 x 10. Will these frequency ranges be revised? Will we allow plant specific binning of initiators?
51
A -The idtiating frequencies used in Table 1 of the SDP Phase 2 process were obtained from a recent NRC study document: NUREG 5499. New information regarding initiating frequencies will be considered and incorporated into the process if appropriate.
Q - The design basis for BWRs typically indicates that low pressure injection systems are important to mitigate large and medium break LOCAs. The SDP process indicates those
- systems are not important, in that when they are not available for LOCA mitigation we have a green finding and the NRC should not engage. Can you reconcile these two positions?
A~- For the SDP process, a risk informed approach was used to assess the significance of inspection findings incorporating event likelihood and available mitigation systems. The design basis, while very i.nportant in assuring defense-in-depth and appropriate design margins, does not always utilize risk insights and is used for different purposes than the SDP process.
- Q - Are the Phase 3 color band conclusions based on a CDF/LERF that is consistent with the Pl's? 1.e., White is equivalent to a a CDF E-5/LERF E-6, Yellow is equivalent to a a CDF E-4/LERF E-5.
A - Yes.
Q - Why do you not consider driving control rods in the BWR4 Case 1 example?
A - We eliminated this aspect for simplicity, but we would consider them in a real plant.
Q - In the BWR4 Case 1 example, what would happen if you could credit recovery?
A - If you could manually operate the standby liquid control (SLC), you would use column 6 and the result would be green.
Q - In the BWR4 Case 2 example, why do we enter the SDP on the condensate pump? I thought we couldn't get to the SDP without a violation.
i A - The condensate pump is scoped in under the maintenance rule. We assess inspection findings with risk significance, not only violations of NRC regulations.
Q - In the example of the BWR loss of CD PP with ATWS as an initiating event, if the system is capable of immediate use, you would not include PCS. Since the CD PP is part of PCS, why is the event included as an ATWS initiating event?
A - The loss of PCS may not have been the result of a loss of a condensate pump. The condensate pumps may still be available even if PCS is not (i.e., MSIV closure). During an ATWS, PCS is assumed lost due to MSIV closure due to oscillating reactor vessel level.
Q - In the BWR4 Case 2 example, on ATWS, why not credit reactor core isolation cooling (RCIC)?
A - RCIC is not an adequate make-up source.
52 s
O - In the BWR4 Case 2 example, why doesn't the logic that says RCIC is not an adequate make-up source so is not credited, apply to SRV?
A - RCIC does have adequate make-up in that case.
O - In the BWR4 Case 2 example, why is the condensate pump in the maintenance rule?
A - The condensate pump is included because it can cause a plant trip.
O - In the SDP example involving the condensate pump failure for the BWR, the failure is in-scope for maintenance rule purposes, but is classified as non-risk significant. Why would this be processed any further once it has been determined to be a non-risk significant failure?
A-This finding should not be further processed once it is determined to be green in the SDP process.
O - Is the pump in the BWR4 Case 2 example assumed to be in stand-by or operating? If in stand-by, would it change the column entry?
A - It is assumed to be operating. Yes, if the pump was on stand-by, the column entry would change.
O - In the PWR Case 1 example, there's a different perspective in the PRA vs Chapter 15 regarding one train of ECCS equipment out of service. From the risk perspective, we don't need ECCS. Why don't we credit this in the SDP?
A - We are looking at the change in CDF, so certain things cancel out from the nominal case to the case under consideration. Here, there's a change in function due to the accumulators.
O - Why is the PWR Case 1 example different from the BWR 4 Case 2 example? Do you still have remaining mitigation?
A - The success criteria require one train of ECCS and injection capabilities; this requires all accumulators -- not having one accumulator says mitigation capability isn't functional.
C - What's missing in the PWR Case 1 example is the information that the first 10 minutes is where immediate flow isn't avai!able if all four accumulators aren't functional.
O - If the failure documented in the PWR Case 1 example also affected other IE's, would you also lock at these?
A - Yes.
C - The PWR Case 1 example assumes that all accumulators are needed to mitigate medium and large LOCAs. You would only get a yellow if the accumulators were out for a year. I think it should be a red.
A - For a medium LOCA, you couldn't get to a red given the assumed probability of failure of mitigation. This wouldn't generate a change in CDF 1 x E-4 because the frequency isn't high enough.
53
O -If the finding in the PWR Case m aple is a! ready green, would anyone do the engineering calculat!on look into whether the accumulator would have perbrmed its function, or would we just handle it through CAP?
A - If the situation existed longe,.han 30 days, it wouldn't be green so the licensee would have to fix it or it would go white. The issue is that the licensee must fix the problem.
O - Where do the success criteria for the PWR Case 1 example come from? Success criteria do not usually use SITS as mitigator.
A - The success criteria come from the plant PRA, but remember that this is just an example, so these success criteria may or may not apply to other facilities.
O - In the PWR Case 1 example, what would enforcement be if all four accumulators were empty for two and a half days?
A - There would be a loss of function, so enforcement would be a Severity Level ll maybe with a fine in the old process, and a NCV in the new process.
O - With respect to the PWR Case 3 example, most PWRs are designed for an alternate suction source.
A - This may be true, and the licensee could take credit if they meet the criteria for recovery actions.
Emeraency Preoaredness SDP Q - On the EP SDP example we had 1 or 2 yellows and a couple of whites. Would the program get a red with agency action required--that is, would you aggregate the findings?
A - The agency engagement for 2 yellow inputs is substantial, but there is no intent to aggregate the items up to a higher level--l.e., red.
O - What are the criteria for the "3 or more failures to meet RSPS" on the EP SDP7 A - The findings must be contemporaneous.
O-FEMA bases its offsite " reasonable assurance" finding on exercise performance evaluation.
An onsite EP failure may result in an exercise " deficiency." How will NRC reconcile its "No i
violation" decision for EP requirements with a FEMA deficiency which could be serious enough to withdraw its " reasonable assurance" iinding and recommend shut down?
Related 0 - How do we reconcile findings of FEMA vs. NRC7 A - it there is no violation of NRC requirements, then NRC has no reason to engage the licensee. If FEMA determines that a deficiency exists, they must pursue resolution through its established mechanisms. The two processes are independent.
54
O - Exercises are designed to determine how the liensee will respond in an actual event. Why does exercise performance carry less weight than an actual event?
A - The regulatory basis for response to emergencies differs substantially from the regulatory basis for drill and exercise conduct.
O - What is considered to be a failure of Pl&R with respect to EP?
A - A failure to identify a problem with EP requirement compliance (either to meet or to implement) or a failure to resolve in a timely manner problems identified.
O - Are we to assume that if it's "No" to all questions on sheet 4 of the EP SDP, then it is not a failure of Pl&R7 A - In many cases this may be the result, but it could also be a failure of minor significance.
k O - What is the difference between the EP SDP and the EP Pis?
A - There is an overlap here and in radiation safety. We will be cognizant of the potential to double count and make adjustments, as appropriate, based on lessons learned from the pilot program.
O - Is the difference that the EP SDP counts number of misclassifications to get to a white finding and the PI looks at number of opportunities?
f 1
A - Inspection is looking at the significance of the issue, while the Pi is looking at indications of program effectiveness.
Radiation Safety SDP Q - Will the 3 year rolling average dose be a fixed number perhaps based on '98 values, or will the rolling average and, hence the criteria, be revised annually, thereby increasing the requirement?
A - It will be with respect to a fixed number, based on '96-98 data.
O - For ALARA findings: (1) I thought the new process was to evaluate against thresholds, not other plants' performance and (2) using a 3 year average will almost guarantee being below the median when 2 refueling outages are included vs a different 3 year period with only one RO.
A - The threshold standards (median & 4* quartile) will be set consistent with current industry exper:ence/ performance.
O - The flow chart on p.170 of the conference materials indicates that 50% of the plants cannot have an ALARA finding worse than green. Is this correct?
A - Yes.
55
/
l-Q - Why wouldn't the green finding reported on p. 285 of the occupational radiation safety discussion be a NCV?
A -it would be;
- Q - Help me understand the substantial potential decision gate.
A - It has the same consideration and definition as in the current enforcement policy as outlined
'in NUREG-1600.
- Q - it looks like we only consider actual exposures or releases in the public radiation safety SDP. Shouldn't a finding involving a large potential release or exposure be considered for significance?
Related _Q - Are the exposure decisions for radioactive material released to unrestricted areas based on actuel exposures to the public or potential exposures (what someone could have j
received)?
A - We consider actual exposures only, due to the scope of 10 CFR 50 Appendix l, which applies to normal operation and anticipated occupational occurrences. For the actual exposure i
situation, it is not necessary to determine whether an individual was actually present at the j
location of interest (e.g., nearest residence) during the passage of a plume containing airborne radioactive material.
Large potential releases would be assessed under the reactor safety cornerstone.
O - Where do HAZMATs affecting workers and the public get processed in the SDP?
A - They are not part of the SDP.
Q - The ALARA flowchart discriminates between colors of finding, using plant performance
' quartile as a determinant.- This is inconsistent with the balance of the new assessment process.
- The difference between a poor performer (bottom quartile) and top performer is expected to continue to decrease and, in fact, may not be significant.
A - Once a standard is set, it is possible and desirable that NO plants wil: be within the 4*
quartile.
Q - Since regulations cover limits for TEDE, CDE, SDE. LDE, why isn't substantial potential for determination of risk tied to the likelihood of exceeding some multiple of applicable dose limit? It seems that including only dose rates is missing some important risk perspectives.
A'- it is difficult to get to an overexposure from an intake so it is very unlikely that an event poses a substantial potential for a more significent exposure.
- Q - On the occupational radiation safety SDP chart question, "Is this a source control problem?", "yes" takes you to a less significant white finding while "no" takes you to a more signi, ant yellow finding.
i 56
A - Correct. This means you have programmatic problems, but the finding is not related to
" program failure."
C - The' SDP process includes decision making diagram logic blocks. In some instances, the number of previous events is used in a block. Presumably, the number in the block does not include the event / problem that is being evaluated for significance determination. In other rtords, if the number in the block is > 5 (for example) and the number of previous events is 5 (from Pl data), the decision would be "no" (not greater than 5) when evaluating the significance of an event / problem that just occurred (that's not yet in Pl data). Please confirm / clarify.
A - The current event being evaluated is not included in the decision box. Thus, if 5 events had already occurred and a new one was being evaluated, the answer to the question "are the >5 events?" would be "no."
Q - In the radioactive material control diagram <.005 rem goes to < 5 events. Over what time period are the 5 events counted? Is it per year or per 12 quarters?
Related O - Over what period of time are the 5 events in the public exposure SDP7 A - The time period is the inspection frequency for the public exposure area, which is nominally two years.
Q - What limits pertain to inadvertent release of solid contaminated materials and the offsite transport of radioactive materials and wastes?
A - There are no " release limits" given in 10 CFR Part 20 that pertain to the release of contaminated materials into unrestricted areas (public areas). Material that may be contaminated must be surveyed for the presence of licensed material using survey techniques appropriate for the material. The material can be released for unrestricted use if no licensed radioactive materialis detected.
Radioactive materials which have been packaged for transport are subject to the surface contamination limits (for the package itself, as opposed to the contents) specified in the DOT regulations (19 CFR).
- Q - The transportation and Part 61 assessment process appears to be much more complex than risk significance would justify. In fact, the determination process appears to be more associated with compliance with regulations than with actual risk. Why is this considered risk-informed, since the' complexity of this determination is driven by the complexity of the regulation, as opposed to risk?
A - The NRC staff agrees that this SDP is complex, consistent with the broad scope and
. complexity of the DOT /NRC regulations in place to protect the public. However, given the potential, direct impact (risk) on the public from the radiation levels, package breaches, etc., the staff does'not agree that the SDP is not risk informed. For example, if the licensee is unable to L provide timely information to the state /localinitial on-scene responders during an emergency,-
inadequate mitigation may result (see N2, N3 on " Failure to make notifications...information") in undue risk to the public.
57
L Q - The high level decision points on the transportation and Part 61 SDP are shown in a
' decision sequence that implies that they (all 5) are inclusive rather than exclusive of each other.
However, a single event could include all 5 problems. Why not take out the filter chain and go directly from the finding to each of the 5 decision points?
A-The inspectors have been instructed when using the existing flow diagram to go down all the vertical gates to gauge the significance of the individual finding. If multiple " hits" are recorded (and only one hit per finding is used for assessment purposes), the highest risk
' significance finding is selected.
However, this comment will be considered further.- It has the benefit of possibly clarifying the process.
C - The SDP for public radiation safety indicates that a radiation material control finding resulting in a public exposure of >.1 rem is a white finding and >.5 rem is a red finding. If the
- problem is related to transportation and a breach of the package during transit, a public exposure of > 25 mrem is white,25 - 100 mrem is yellow, and > 100 mrem is red. This appears inconsistent. Please explain.
A - The NRC staff is in the process of reviewing this issue and the benefits of better aligning the public dose criteria.
Q - How would failures in the radioactive waste process control program for 10 CFR 61 classifications be reviewed under the SDP? Undetected, underclassified waste could be inadver1ently accepted by the burial site.
A - The NRC inspection program will not significantly change. There will be the same focus on the licensee's waste characteristics and classification program.
Q - What color is it if the licensee is denied access to a low-level waste site?
A - Yellow.
Q - is the logic backwards in the ALARA SDP flow chart?
A - No, the logic is correct.
Q - Please categorize the Haddam Neck radioactive blocks which got offsite in terms of public radiation safety.
A - The situation at Haddam Neck involved numerous events which resulted in radioactive material being released to an unrestricted area. A " white" finding would have beon the likely outcome based on the number of events, not on the calculated dose from the material.
Q - What is the most significant event which has ever occurred with respect to public radiation safety, and how would you categorize it?
' A - The most significant event that has ever occurred at a nuclear power plant with respect to public radiation safety was the accident at Three Mile Island. Based only on the estimate of the 58
maximum dose received by a member of the public, the release of radioactive material would have been classified as a " white" since the Part 20 public dose limit was not exceeded.
However, there are other cornerstones that would have to be evaluated (reactor safety and emergency preparedness), and those would likely have greater significance.
Ehysical Protection SDP Q - The entry of an individual into a protected area doing actual damage to safety equipment is red. However, if all safety equipment needed to mitigate a medium or large break LOCA is broken or damaged (not by an intruder) the issue is green. Why is this inconsistency allowed?
A - This is a red issue for the physical protection cornerstone, as a breakdown in the multiple barriers has allowed damage to equipment important to safety. While the probability of a medium /large break LOCA in and of itself is of a low enough frequency that its risk significance due to inoperable equipment would not result in a red characterization, within the physical protection cornerstone, such an event would warrant such a characterization based on the impact on public confidence alone. Also, PRA does not account for malicious intent.
O - When the TMI intrusion event was run through the chart on p.175 of the conference materials, it ended up with 3 different colors depending on assumptions. Instructions to security were to remove the number of subjective judgments from the chart so this would not happen.
J Has the chart been changed since that meeting per NRC instructions? It does not appear to
]
have been.
A - It has not been changed yet. Work continues on refining this SDP.
Q - How is the risk significance determination for physical protection to be performed?
Shouldn't PRA information be a significant factor in this?
A - Fundamentally, security falls outside the PRA process but is tentatively linked at the level of operational solutions.
Questions on Enforcement O - Since the SDP does not consider design basis requirements (such as loss of offsite power or single failure criteria), how will non-compliances with design bases or technical specifications
. be dispositioned in enforcement space? Will there be a resolution of discrepancies between the i
non-compliance and the risk determination if the SDP screens to green.
A - Enforcement will use the results of the assessment process (including SDP) for all issues except 1) willfulness / integrity issues and, 2) issues impeding NRC's ability to do its job.
Therefore, if the SDP result is green the enforcement result will be a NCV or a cited violation if one of the 3 exemptions exist.
O - The description of violations not covered by the SDP only lists those that are (1) willful and
- (2) impact NRC's ability for oversight. However, there appear to be several areas that the SDP doesn't yet fully address such as: containment barrier issues, cross cutting issues such as CAP breakdown, human performance failures, etc., any issue requiring a Phase 2 evaluation. How are these issues going to be treated in enforcement space?
59
i A - Containment barriers violations will be reviewed by a risk analyst for significance. Because this will be analyzed and placed into the action matrix, the enforcement approach will be that determined by the SDPs (NOV or NCV). For cross cutting, human performance and corrective action issues, if plant equipment is affected, such that a white, yellow, or red violation results, an NOV will be issued. If the issue is green, a NCV will result with licensees placing the issue into the CAP.
O - Would deliberateness be associated with 50.9?
A - It could be. The Office of Enforcement will rely on the Office of Investigations to make the first assessment on deliberateness.
O - Will enforcement be issued for violations of procedure steps which are elective (licensee put a step in for non-regulatory reason)?
A - All violations should be put into the CAP to correct. An elective step with no impact on a regulation may be de%rmined to be minor and would therefore not be normally documented.
However, all violations (of greater than minor concern) of green significance should be documented as either a NCV or NOV as appropriate. Work is expected to continue on the subject of procedure compliance through the pilot program.
O - As used relative to enforcement, does white, yellow, and red refer to SDP Phase 2?
A - The assignment of colors by the SDP process may be accomplished with a Phase 2 or Phase 3 analysis of a violation's significance. Enforcement will use the output of the SDP process to determine whether to use a cited or non cited violation.
O - Is the difference in enforcement that the previous policy did not consider event frequency, only consequence?
A - The current policy is more deterministic, while the new policy is more risk informed. The old policy wasn't wrong; it just had a different basis. Although in the new policy, risk is considered in making assessments of significance to determine severity level and penalty amounts. Deterministic compliance with requirements is also considered. The new policy relays more heavily on the violations effect on changing core damage frequencies.
O - Would an unsubmitted or inadequate 50.59 be viewed as impeding the process in enforcement space?
A - An unsubmitted 50.59 would be considered as impeding; an inadequate one would not.
O - Will the timeliness of enforcement change?
A - The timeliness on enforcement action will be determined by SDP assessment of the significance of the finding. As a result, it is not known at this time what the timeliness will be for enforcement actions.
i 60
Questions on the Assessment Process and Thresholds General Questions O -is there any plan for the NRC to rank plants for assessment purposes based on Pis? For i
instance, if most plants have all or mostly " green" Pls, would the NRC try to differentiate those plants that are "more green" vs those that are "less green"?
f i
A - There is no intent to rank plants. One of the advantages of this approach is that performance, as measured by Pls and inspection findings, determines the actions taken by the NRC as outlined in the action matrix, j
O - In the September / October '98 workshop we discussed the concept of " rebuttable presumption" of performance indicators. At that time, it was supposed to take a
" preponderance of inspection findings" to contradict a Pl. It looks like now a sinale white inspection finding carries the same weight as a Pi in that area. This doesn't look like the same l
deal. A single finding is a " preponderance"?
l A - The thinking has evolved based on the development of the baseline inspection program.
Because the inspection program looks at areas not covered by performance indicators, the rebuttable presumption concept does not effectively apply to the Pls. It takes both the Pls and the inspection findings as two distinctly separate inputs to determine whether the objectives of a cornerstone are met.
Ouestions on the Action Matrix O - When do we start the clock on an issue?
A - The clock starts with the discovery of an issue.
O - Are the dates of findings counted from the date of the inspection report? If so,if the j
inspection report is signed-out the last day of the quarter, the NRC actions might be different than if it got signed out several days later (i.e. first of the next quarter).
Related 0 - Will the beginning and ending dates of NRC inspections (baseline and supplemental) be consistent with the beginning/ending dates of the quarters for which the coding of findings, if any, will be applied? The reason for the question is that an inspection with a finding (s) should begin/end within the quarter for which the coding will apply. An inspection with a beginning or ending date that overlaps the quarter's beginning or ending date that has a finding (s) willintroduce suspense / confusion regarding the assignment of finding (s) and coding relative to the applicable quarter.
A-The PIM will be issued quarterly just as the Pts are issued quarterly. The resident inspector reports will also be timed to be completed at the end of a quarter. Inspection reports that overlap two qur.rters will be coded based on the close of the report period.
O - Can we use the PIM to provide updated status of a finding?
61 L
- A - No, only findings are shown in the PIM, not resolution of problems.
Q - How will regional management consider the schedule of inspections based upon the potential for findings and subsequent actions from the action matrix?
Related O - Does the potential exist for a licensee to be notified of a reduction in a current 6-month inspection plan?
A - When inspection plans change, we will inform and coordinate with licensees.
C - The " color schemes" for Pls and inspection findings should be consistent. A " green" inspection finding should denote that an inspection was conducted in this area and the results are ok. This resolves the means of designating whether an inspection was conducted or not--
currently there is no mechanism for designating this. Inspection findings should also be
. consistent with the action matrix, i.e. no action is taken. [Why bother coloring an inspection
. finding " green" as is presently done (which means that everything is NOT OK, but conveys to the public that it is okay), if there is no action ever based on it?] The action matrix states "All
- Assessment inputs (performance indicators (Pis) g_nd inspection findings) green, cornerstone n
objectives fu!!y met." This conveys, by definition: Green Pl = Green inspection finding = OK.
, Related C - A Pi indicator of green means good! However, in the inspection findings, a green box, means a finding has occurred. Please consider human factoring this so that outside agencies and others don't confuse them.
C - Consider not carrying inspection findings for four quarters if the follow-up inspection shows effective corrective action, or shows that corrective action was satisfactory.
Related C - Consider leaving inspection findings active until follow-up identifies that appropriate and effective corrective actions have been taken. If the problem is corrected sooner (than in 4 quarters), it would be indicative of an effective corrective action program, if the problem isn't corrected in 4 quarters, it might be indicative of an ineffective program.
)
A - We will continue to assess this part of the program through the pilot process. The above recommendations may be appropriate for consideration.
Related O - What is the rationale for keeping a " white" finding on for the entire 12 months of the assessment cycle although it was corrected in second quarter? Isn't this a double hit?
A'- This provides equivalent treatment with Pis. However it has been suggested that an inspection finding should be characterized once corrective actions have been effectively implemented. This issue will be assessed during the pilot program.
Q - How are bi-annual inspections treated with respect to carrying findings forward?
A - Bi-annual inspections are treated the same way as more frequent inspections. There is no provision to carry their findings longer.
O - If you have a scram in quarter 1, none in quarter 2, does the quarter 1 scram stay with you?
A - No. Pls toggle on and off and can change each quarter.
62 L
e O'- Are actions based on inspection findings and Pls or the reviews?
A - The Agency's regulatory actions are the outputs of the reviews that are conducted, based on inspection findings and Pls.
O - The persistence of inspection findings gives them 4 times the weight of Pls.
A - Pls get continuous review, while inspection findings are periodic. Significant inspection findings may not generate additional action, even if they're carried forward. This aspect of the program will be assessed during the pilot program.
Q - Follow-up findings constitute multiple findings from one issue. Why not toggle the original finding off?
A - It does not toggle off because the original finding is still indicative of performance.
(Remember that it takes a loss of function to go white on an inspection finding.)
O - The public will be interested to know that a problem has been corrected. Can we add a field to show that follow-up has been conducted and the problem has been corrected?
, A - The NRC will assess this issue as part of the pilot program.
O - What color is the inspection area if no inspection occurred or if there were no findings?
There should be differentiation between "no data" and "no findings."
A Currently, it is expected that the data display will be crossed out. Carrying data when no new inspection had occurred would give the wrong view.
Related C - The two cases for "no findings" data need to be dietinguished on the Web display, where Case 1 is no findings because of no inspections on that cornerstone area for the quarter, and Case 2 is no findings from inspections in that cornerstone area for the quarter.
i If green is a non risk significant finding, it should not also represent "no findings" (we looked) or l
"no data"(we didn't look). One option is X - no inspections findings - green white yellow red based on risk-significance NONE - no findings.
I O - Since one " yellow" gets a team inspection as a response, which can have a significant impact on the NRC and station resources, how can we continue to use fault exposure hours (FEHs) with unavailability knowing that FEHs can lead to a " yellow" indicator with one failure?
The team inspection isn't warranted in this case.
A - The team inspection may not be warranted in this case. Keep in mind that 1 yellow does not reauire a team inspection response. This action is one of the potential tools available to the region to address the issue.
O - For performance indicators that have rates (i.e., scrams /7000 hrs, unplanned power changes /7000 hrs, etc.) the indicators can exceed thresholds for many months just due to going 63 L
into a refueling outage if you have several months with zero houm critical replacing months at full power. What happens when you have several white ir. puts.nM were actually caused by something a year ago?
A - We are aware that this situation can occur. We will, during the pilot program, develop methods to handle it appropriately.
O - Does it take 2 white findings in the 1st quarter and 2 in the 5th quarter to trigger a repetitive inspection finding (4 findings total)'or, does it take 2 white findings in each of 5 consecutive quarters (10 findings total)?
A - An inspection finding characterized at a risk-significance of white or higher will be retained in the system for 4 quarters. If other inspection findings of risk significance occur on the same cornerstone of safety such that 5 consecutive quarters exist with two white findings, a repetitive degraded cornerstone would exist.
O - Is assessment done on a per reactor or per site basis? Pls appear to be tracked on a per reactor basis. Do inspection findings hit all reactors?
Related O - How will some inspection findings be allocated to multiple units at a site? For example, would an EP drill finding be charged against the unit simulated in the drill (unitized) or against both units (because it is a common program)? Similarly, will an ALARA finding be against the unit where it occurred or against the program (both units)?
Related Q - is a radiation protection finding assessed to one unit or the site at a multi-unit site?
A - The answer depends on the finding. Some Pts are site-specific, some are reactor specific.
The same situation will exist for inspection findings.
Assessment will be done on a reactor basis. However, as noted above, some Pls and inspection findings will be impacting all units on a multi-unit site. If the finding is that a programmatic breakdown has occurred that affects the entire site (such as the example of an EP drill failure or a security system breakdown), these findings will be counted against both (all) units. On the other hand, if the finding is that a single failure occurred in one unit, this will only be counted against the affected unit.
During the pilot program an oversight panel will be ef.ablished to review this and other similar issues. This panel will provide program guidance and ensure consistent application of the program. This panel will also provide recommendations for improvements, as appropriate.
O - Would a yellow finding in the first quarter result in the scorecard being yellow at the annual review?
A - As noted earlier, current guidance is that inspection findings will be retained in the system for 4 quarters. Therefore, a yellow finding that emerged during the first quarter would be in the system at the time of the annual assessment.
O - If more Pls are added, will the action levels change?
64 i
A - No, actions are based on level of significance of the thresholds, not the number of indicators.
O - Why is an inspection finding of a historical issue just found treated as poor performance when it's really indicative of good performance?
A - The finding is handled through a graded response. If it has risk significance such that it is characterized as other than green, part of the regions determination as to what action it should take will be based on the circumstances surrounding identification of the icsue.
O - How do Pl&R team inspection findings fit into the overall assessment?
A - Individual inspection findings associated with previous or newly identified risk significant issues may emerge. Also, a qualitative assessment of program effectiveness should result. The Agency's response when a threshold is crossed would be affected by the assessment of the CAP.
O - What and when is the regulatory conference referred to in the action matrix?
A - The regulatory conference will most likely occur following the annual end of cycle review and assessment report. The focus is on the events and performance of the previous four quarters.
This doesn't preclude other meetings as conditions may warrant.
C " Engagement" takes place according to the action matrix. " Healthy interaction" can/should take place throughout the process to help move the plant towards strong safety performance.
O - Because security is both a cornerstone and a strategic performance area, does two whites equal a degraded cornerstone?
A - No, but this should drive close observation of the related processes and programs as part of the NRC's followup activities.
C - Application of the action matrix is not just a " bean count." Judgment rnust be used in its application depending on individual circumstances (e.g., Two whites occur in quarter 1, which are resolved. Then two new whites in the same cornerstone in quarter 4 would not necessarily lead to coming before the Commission.)
O - There are 19 Pls and 7 inspection inputs per cycle, is action reasonable for only 2 whites?
A - Thresholds have been set to make some follow-up reasonable in such a situation. The response will be graded according to the circumstances associated with the two white indicators.
O - Over what time period does the top row of the action matrix apply?
l A '- Action will be taken in the quarter when indicators / inspection findings cross a threshold. The NRC will not wait for the annual review to act if it appropriate to do so.
O - How do you determine the "most significant" inspection result if all findings are green?
65 I
i i
l
A - Findings are aligned to cornerstones. Findings are arranged by quarters, and while we may have multiple findings under the block, the block color is determined by the most significant finding color associated with that cornerstone.
O - The assessment input graph shows 4 Pis for physical protection, but aren't there now only 37 A - Correct.
O - In the planned website assessment display, can you transition from summary to raw data?
A - Yes, this feature is being incorporated into the associated software.
O - If you have a situation where a single Pi in a cornerstone is white in the first quarter and green in the second, then there is a white inspection finding in the third quarter, is this a degraded cornerstone?
A-No.
O - If a repetitive degraded cornerstone occurs over 5 quarters, why are we only displaying 4 quarters?
A - Previous data is intended to be available within the system for review, but this comment may warrant further review.
66