ML993620415

Letter Transmitting Summary of Two Meetings Held on 991006-07 with Risk-Informed Technical Specifications Task Force

ML993620415
Person / Time
Issue date:	12/14/1999
From:	Beckner W Technical Specifications Branch
To:	Brooks A Nuclear Energy Institute
References
Download: ML993620415 (54)

Text

December 14, 1999 Mr. Anthony Brooks Nuclear Energy Institute Suite 400 1776 I Street, NW Washington, DC 20006-3708 Mr. Brooks:

The purpose of this letter is to transmit the summary of two meetings with the Risk-informed Technical Specifications Task Force. The first meeting was held at the San Onofre Nuclear Generating Station on October 6-7, 1999. The second meeting was held at the U.S. Nuclear Regulatory Commission (NRC) Headquarters offices in Rockville, Maryland, on November 10, 1999.

Sincerely, Original signed by:

William D. Beckner, Chief Technical Specifications Branch Division of Regulatory Improvement Programs Office of Nuclear Reactor Regulation

Enclosures:

1. Meeting Summary

2. Attendance List

3. October Meeting Presentations

4. November Meeting Presentations cc: See attached list DISTRIBUTION: See attached.

DOCUMENT NAME: G:\RTSB\GILLES\MTGSUMRITSTFI 11 099.WPD OFFICE INRR/DRIP/RTSB NRR/ICTSB NRCRDRIP/RTSB I Vilsr j/ RL-Den@ - WDBpeckner tL, Af.j DAE 12/t /U/99 12/ /f/99 12.iU99 1 OOFICIAL RECORD C' PY-2

>QL ao)L (Dt 396ý (

Multiple Addressees DISTRIBUTION:

E-Mail w/o Enclosures 3 & 4 SCollins/RPZimmerman JJohnson BWSheron DBMatthews SFNewberry GMHolahan WDBeckner RJ Barrett RLDennig FMReinhart MLWWohl NTSaltos TSB Staff JAZwolinski JFWilliams AWMarkley MACunningham, RES MMarkley, ACRS Staff HARD COPY FILE CENTER

"\ PUBLIC TSB R/F NVGilles

UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555-0001 December 14., 1999 Mr. Anthony Brooks Nuclear Energy Institute Suite 400 1776 1Street, NW Washington, DC 20006-3708 Mr. Brooks:

The purpose of this letter is to transmit the summary of two meetings with the Risk-Informed Technical Specifications Task Force. The first meeting was held at the San Onofre Nuclear Generating Station on October 6-7, 1999. The second meeting was held at the U.S. Nuclear Regulatory Commission (NRC) Headquarters offices in Rockville, Maryland, on November 10, 1999.

Sincerely, William D. Beckner, Chief Technical Specifications Branch Division of Regulatory Improvement Programs Office of Nuclear Reactor Regulation

Enclosures:

1. Meeting Summary

2. Attendance List

3. October Meeting Presentations 4' November Meeting Presentations cc: See attached list

Multiple Addressees cc:

Mr. Alan Hackerott Mr. Dennis Henneke Omaha Public Power District San Onofre Nuclear Generating Station Ft. Calhoun Nuclear Station Southern California Edison P.O. Box 399 5000 Pacific Coast Highway Ft. Calhoun, NE 68023-0399 San Clemente, California 92674-0128 Mr. Noel Clarkson Ms. Sharon Mahler Duke Energy/Oconee Cooper Nuclear Station Mail Code: ON03RC Nebraska Pulic Power District Highways 130 & 183 (29678) P.O. Box 98 P.O. Box 14393652 Brownville, NE 68321-0098 Seneca, SC 29679-1439 Mr. Frank Rahn Mr. Greg Krueger Electric Power Research Institute PECO Energy Company P. 0. Box 10412 Mail Code 63A-3 Palo Alto, CA 94303 965 Chesterbrook Boulevard Wayne, PA 19087 Mr. Donald Hoffman EXCEL Services Corporation Mr. Wayne Harrison 11921 Rockville Pike, Suite 100 South Texas Project Electric Generating Rockville, MD 20852 Station STP Nuclear Operating Company Mr. Jack Stringfellow P. O. Box 289 Southern Nuclear Operating Company Wadsworth, TX 77483 P.O. Box 1295 Birmingham, AL 35201-1295

NRC/INDUSTRY MEETING OF THE RISK-INFORMED TECHNICAL SPECIFICATION TASK FORCE Meeting Summaries October 6-7 and November 10, 1999 Two meetings between the NRC staff and industry representatives comprising the Risk Informed Technical Specifications Task Force (RITSTF) were held on October 6-7 and

,November 10, 1999. The attendees are listed in Enclosure 2. The meetings were a continuation of earlier meetings where the NRC staff and the industry discussed ongoing risk informed technical specification initiatives and the creation of a fully risk-informed set of standard technical specifications (STS).

The main purpose of the October 6-7 meeting was to have a more detailed discussion of the probabilistic risk analysis (PRA) work done to support the current technical specification (TS) initiatives being prepared for submittal to the staff. There was limited RITSTF representation at this meeting. Southern California Edison has volunteered San Onofre Nuclear Generating Station (SONGS) to be the industry's lead plant for the majority of the seven initiatives currently being pursued by the RITSTF. These include:

1. Define preferred end states for TS actions (e.g., hot shutdown vs. cold shutdown)

2. Increase the time allowed to delay entering required actions when a surveillance is missed

3. Modify existing mode restraint logic to allow use risk assessments for entry into limiting conditions for operation (LCOs) with inoperable equipment based on low risk

4. Develop a risk-informed extension of current allowed outage times based on a configuration risk management program (CRMP)

5. Optimize surveillance requirements (SRs)

6. Modify LCO 3.0.3 actions and timing by extending minimum time to begin LCO 3.0.3 shutdown from 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and allowing for a risk-informed evaluation to determine whether it is better to shut down or continue to operate

7. Define actions to be taken when equipment is not operable but is still functional The staff and the industry discussed the meaning of the term "risk-informed" as it relates to regulatory applications. The industry stated that it was their general philosophy to use qualitative risk assessments where they believed the benefits of a proposed change were obvious, and to use quantitative assessments where the outcome was not as obvious. The industry also stated that they take into-account other aspects such as defense in depth and safety margins when considering a proposed change. The industry believed such an approach was consistent with the guidance in Regulatory Guide 1.174, "An Approach for Using Probabilistic Risk Assessment in Risk-Informed Decisions on Plant-Specific Changes to the Licensing Basis," to use PRA to improve decision making and regulatory effectiveness. The staff reserved judgement of such an approach noting that benefits that are obvious to one person or group might not be obvious to another.

SONGS representatives then presented some details regarding their PRA work in support of some of the RITSTF initiatives. The San Onofre presentations are contained in Enclosure 3.

The SONGS representatives presented information regarding their living PRA and Safety Monitor, including the full power, transition, and low power and shutdown risk models. The 1 Enclosure 1

SONGS representatives stated that they had their own transition model development document that would likely be referenced or included in the SONGS plant-specific submittal for Initiative #1.

The staff and the RITSTF discussed some specific submittal and implementation and technical issues related to a few of the initiatives. With regard to Initiative #1 related to safe end states, the RITSTF indicated that the Combustion Engineering Owners Group (CEOG) was preparing a draft report to support this initiative and that the report would accompany an industry request for a generic change to the STS. The RITSTF indicated that changes to the TS end states would be proposed for the vast majority of LCOs in the STS. The RITSTF indicated that uncertainties and sensitivities of PRA results will be investigated. The group discussed that additional known shutdown issues which may impact the results (e.g., external events and boron dilution) should be addressed.

Representatives from the Boiling Water Reactor Owners Group (BWROG) stated that they are still considering this initiative and that the risk of operating in hot shutdown and in cold shutdown is relatively the same, whereas forthe pressurized water reactors there appears to be a clear risk benefit to operating in hot shutdown in most cases. They indicated that they would be more interested in changing the current TS to allow them to use PRA in deciding whether to stay at power orshut the reactor down. They also indicated that they may be interested in extending the time requirements for going to cold shutdown. This would allow them to operate longer in hot shutdown conditions. The BWROG is considering a pilot plant for this issue.

The RITSTF and the staff discussed how the SONGS assessment results for Initiative #1 could be applied to other plants. The Westinghouse Owners Group (WOG) indicated that they would attempt to show how the SONGS results apply using qualitative assessments and comparisons, and that they did not plan to perform any further plant-specific analyses. The Babcock and Wilcox Owners Group indicated that they intended to pursue a similar approach.

The group indicated that they could use the SONGS model to help develop sensitivity studies to address some of the design differences between the plants. The staff stated that they could not make any definitive decisions on what type of submittals they would find acceptable and that the burden of proof of similarity to the CEOG work would rest with the other owners groups.

With regard to Initiative #2 related to missed SRs, the group discussed the PRA aspects of that initiative. In order to assess the increase in risk it is necessary to make an assumption about the frequency of the various expected missed surveillances. This requires an understanding of the reasons, the nature and circumstances under which surveillances are missed. The RITSTF pointed out that even if the failure rate of a component is doubled, due to the missed surveillance on that component, the plant risk would not be affected significantly.

The staff agreed that most likely the risk increase would not be significant unless some licensees abuse the proposed flexibility. The development and implementation of an appropriate regulatory oversight process could address this issue. The industry and the staff agreed that the staff's review of this issue would involve more of a policy decision than a technical decision.

With regard to Initiative #3 related to mode restraints, SONGS has been studying this issue by comparing the relative importance of functions and associated systems at various modes of 2 Enclosure 1

operation. The proposed change would allow a licensee to use a CRMP to decide whether to enter into a mode or other specified condition within the applicability with inoperable equipment. The staff agreed that the study of conditions and risks associated with various likely transitions could help focus the issue and identify an appropriate regulatory oversight mechanism.

With regard to Initiative #6 related to changes to LCO 3.0.3, the RITSTF stated that this initiative is very closely tied to the maintenance rule as it addresses what actions to take for a loss of safety function. The industry indicated that the frequency of plant conditions for which this TS change is applicable is low. The staff mentioned that the identification and study of examples of plant specific LCO 3.0.3 entries and conditions that would drive the plant to shutdown, including associated risks, could help focus the issue.

The group discussed the process for plants to adopt approved changes to the STS. The staff briefly discussed processes being considered to make adoption of such approved changes more efficient.

Finally, the group discussed submittal schedules for some of the initiatives. The RITSTF stated that they expected to submit Initiatives #2 and #3 to the staff around October 30, 1999, and that they might be able to submit some of their PRA assessment for Initiative #1 by December 31, 1999. The RITSTF stated that Initiatives #4 and #6 would come sometime in 2000.

The November 10, 1999 meeting was a meeting of the full RITSTF to discuss high-level objectives and the status of the current initiatives. All presentations for the meeting are contained in Enclosure 4. The staff opened the meeting with a general discussion of the status of risk-informed regulatory activities at the NRC. The staff pointed out that there was a high level of interest in the RITSTF activities and that we needed to understand how broad the interest in the industry was. The staff presented several thoughts on a long-term vision for risk-informed TS and acknowledged that structure and resource issues will need to be worked out. The staff also pointed out that it will need to address the four strategic performance goals identified in the Commission's Nuclear Reactor Safety Strategic Plan. Those performance goals are: (1) Maintain safety; (2) Increase public confidence; (3) Reduce unnecessary regulatory burden; and (4) Make NRC activities and decisions more effective, efficient, and realistic.

The RITSTF presented some of its thoughts on a long-term vision for risk-informed TS. Much of the discussion focused on a slide which presented the Westinghouse Owners Group (WOG) five year risk-informed TS strategy. The group discussed at what point rulemaking might be needed to accommodate all of the envisioned changes to TS. The group also discussed the issue of PRA quality and the fact that the ASME PRA standard currently being developed did not address configuration risk management tools like those envisioned as necessary to implement many of the TS changes under development.

Representatives from the South Texas Project presented a concept of a fully risk-informed set of TS that essentially relies ona CRMP as the backbone of the TS. They likened the proposed risk limits in their concept to radiation protection limits (i.e., ALARA limits). The NRC regulates ALARA limits at a high level and licensees control these limits at a lower level administratively. The group also discussed whether there was a need for an instantaneous 3 Enclosure 1

risk cap for TS. The RITSTF stated that the major question is the cost benefit of going to this extreme. For example, if plant PRAs essentially become the TS, then licensees would have to control changes to the PRA model to the same degree as TS changes are currently controlled.

The RITSTF presented a status of the seven initiatives currently under development. The group discussed Initiative #1 related to safe end states. The RITSTF stated that the expected results were confirmed by the PRA work done by SONGS. The group again discussed what work the other owners groups planned to do to justify the changes for their plant types by taking advantage of the SONGS and CEOG work.

Updated schedules for the various initiatives were discussed. The RITSTF stated that they expected to submit Initiative #1 in February or March 2000. Initiatives #2 and #3 were expected to be submitted in the very near future. Initiative #4 was planned for submittal in late 2000. Initiatives 5, 6, and 7 were also expected in mid to late 2000.

The group briefly discussed the staff's planned process for reviewing and adopting these initiatives as changes to the STS. The group also discussed support for the December 16, 1999 meeting with the Advisory Committee for Reactor Safeguards Reliability and PRA Subcommittee and agreed to a possible future meeting in late February 2000.

4 Enclosure 1

Meeting Attendees October 6-7, 1999 Name Affiliation Dennis Henneke Southern California Edison Sharon Mahler Southern California Edison Gary Chung Southern California Edison Brian Woods Southern California Edison Thomas Hook Southern California Edison Ed Scherer Southern California Edison Don McCamy TennesseeValley Authority Kent Sulton Nebraska Public Power District S.. ViswesWaran General Electric Thomas Sihko Vermont Yankee Jerry Andr6 Westinghouse Mike Kitlan Duke Power Rick Wachowiak Nebraska Public Power District Frank Rahn Electric Power Research Institute Nicholas Saltos NRC/NRRJSPSB Millard Wohl NRC/NRR/SPSB Nanette Gilles NRC/NRR/RTSB Enclosure 2

Meeting Attendees November 10, 1999 Name Affiliation Ray Schneider ABB-Combustion Engineering Nuclear Fuel Company Alan Hackerott Omaha Public Power District Dennis Henneke Southern California Edison Sharon Mahler Southern California Edison Biff Bradley Nuclear Energy Institute Noel Clarkson Duke Power Wayne Harrison South Texas Project Rick Grantom South Texas Project Donald Hoffman EXCEL Services Jerry Andr6 Westinghouse Jim Andrachek Westinghouse Jack Stringfellow Southern Nuclear Don McCamy Tennessee Valley Authority E. D. Ingram Southern Nuclear Glenn Warren BWR Owners Group David Stellfox McGraw Hill John Fehringer INEEL J. E. Rhoads Energy Northwest Richard Harris Entergy Mike Kitlan Duke Power Rodney Johnson Detroit Edison Bert Morris Tennessee Valley Authority Gregory Norris Entergy Rick Wachowiak Nebraska Public Power District Scott Newberry NRC/NRR/DRIP Rich Barrett NRC/NRR/SPSB Mark Reinhart NRC/NRR/SPSB Mark Rubin NRC/NRR/SPSB Millard Wohl NRC/NRR/SPSB Nick Saltos NRC/NRR/SPSB William Beckner NRC/NRR/RTSB Bob Dennig NRC/NRR/RTSB Jack Foster NRC/NRR/RTSB Nanette Gilles NRC/NRR/RTSB Enclosure 2

ENCLOSURE 3 OCTOBER 6-7, 1999 MEETING PRESENTATIONS Enclosure 2

Risk-Informed Technical Specifications Low Power Shutdown Risk I

'At Agenda

+ Background

+ Low Power Shutdown (LPSD) Risk "AssessmentMethodology

+ Major Assumptions

+,: Low Power vs. Full Power Success Criteria

+ LPSD Risk Sensitivities

+. Summary I

4q

Background

• . LPSD Risk Performed Since 1990

+ Started LPSD PRA Models in 1993 l .+ LPSD Risk Used For Outage Risk Planning and Monitoring, RI-IST, RI-TS, and Outage Safety Significance Determinations "Low Power and Shutdown Risk Assessment Methodology

'*  :.:*WINNUPRA/Safety Monitor

• Full Event Tree/Fault Tree Model

+. Only Internal Events Modeled

+. Complete System Models

• Full Power Models used with LPSD enhancements

.4 2

4 LPSD Methodology (con't) 44

+*41 Low Power/ Shutdown Plant Operating States (POS)

• Vent size and availability, RCS level, time since shutdown, RCS draining, equipment availability

-4

• 3 - 5 POS dominate risk profile A

5 Risk Profile 6

3

UNIT 3 CYCLE 10 REFUELING OUTAGE Safety Monitor V2.0a : Safety Monitor V2.0a 1.1OOE-02

'I- Mode 5 Einry. 5- Draining to mid-loop.

6- Mid-loop. 7- I FI Below RVF. 8- I Fi Below RVF (Ilead Off/Swyd Maini),

9- 13 FI Above RVF. 10- Fuel Olfloading, I I- Fuel in the SFP, 12- Fuel Reloading.

13- I FR Below RVF. 14- Draining to mid-loop. 15t RCS at mid-loop. 16- Mode 5 Pzr Normal Vented. 17- Mode 5 Pzr Normal/Solid No Vie,"; 18- Mode 4 I lot Shutdown.

1.1OOE-03 aI 2' 1.00E-04 Ial, 6 1 q iu*%) 12-3/4/1999 05:48am 62/1999 6 :48am 1. 00E- 05 INSTANTANEOUS CORE DAMAGE RISK PROFILE

LPSD Methodology (con't)

+ Human Reliability Analysis (HRA)

Methods Are The Same As Full Power (i.e. Dr. Swain's THERP method)

• Mission times based on time to core boiling "andcore uncovery

• HRA probabilities are conservative

+:: System Models And Support System Dependencies Are Essentially The Same As Full Power 7 t* Initiating Events

+..Grid-Related Loss of Offsite Power

+:Plant-Centered Loss of Offsite Power

+ Loss of Shutdown Cooling

• modeled using fault tree (vs point estimate) 4 includes loss of support system such as CCW, HVAC 4

Initiating Events (con't)

+ Loss of Inventory (LOI) event frequency based on NSAC data

-LOIfrequency reduced by factor of 10 (judgement) when not in draining or filling operation SSystem Alignments i-...;

The SONGS LIPSID includes multiple

'",*alignments S*-Componentfor the following Cooling Water systems:

° Salt Water Cooling

- Shutdown Cooling 0 Containment Spray

- High Pressure Safety Injection

- CVCS 5

11 Major Assumptions i ..+ Core Damage is Defined To Be Core Uncovery Large Early Release Not Possible At LPSD 4+* Gravity Feed Is Not a Success Path (Surge Line Flooding) Unless RPV Head Is Removed

"+Containment Spray Pumps Backup The LPSI Pumps For SDC Full Power vs. LPSD Success Criteria FulH Pover LPSD

• ]']PSI 2 of 4 injection Eims I of 4 injection lines LPS1 (SDC) 2 of 4 injection lines 1 of 4 injection lines

_I of2 pumps I of4 pumps (incl CS pp)

___________I__ of 2trains I of2 pumlps (SDC backup)

UCWIof2 trains same of3 pumps SSWC __________________I I of 4 pumps same I of 2 trains AFW I of3 pumps I of2 pumps (turbine drive purrmunavailable)

Electrical I of2 trains same Core Damage Core Uncovery same 12 6

LPSD Risk Sensitivities tit! Conservative HRA Leads To Conservative Results

+ Loss of Inventory Event Frequency

• Dominant During Draindown POS

+..Initiating Events Contribute Rather Equally For POSs Other Than Draining 13 Significant Operator Actions

+ Operator Isolates Coolant Diversion Prior To Loss Of SDC

• Operator Initiates Backup SDC Prior To

• Boiling

+ Operator Initiates RCS Make-Up Prior to Core Uncovery

• .Operator Initiates DG Cross-tie To Other Unit

,* Recovery of Offsite Power 1 7

SSummary

"+.LPSD Risk Analysis Can Give Meaningful,

• f :Quality Results Comparable to Full Power

• . Analyses Given:

1

• System Success Criteria Are Accurately

& Captured

• Operator Recoveries Are Understood

. Sensitivity to Operator Action Probabilities Are Appreciated

• Level of Detail of the System Models Is 4# Equivalent To Full Power Models 8

Risk Informed Technical Specification Task Force SONGS PRA Presentation Dennis W. Henneke SONGS Nuclear Safety Group 10/6/99

SCE Living PRA/Safety Monitor if The SONGS 2/3 PRA model is developed and maintained on WINNUPRA, and can be solved on either WINNUPRA or Safety Monitor.

if Scope of the PRA includes:

- All modes 1 to 6, refueling and offloaded.

- External Events for modes 1-4 (Fire/Seismic)

- Fault Tree Initiating Events for Support systems (e.g., CCW or SDC) 2

SCE Living PRA/Safety Monitor

/Differences between WINNUPRA and Safety Monitor include:

- WINNUPRA:

• Solve individual event trees, sequences etc., or the Safety Monitor top logic model.

o Software helps in troubleshooting results, viewing solution steps, performing sensitivity, etc.

- Safety Monitor:

. Solves whole model (top logic model) each time.

• Can easily run selected configurations (3-5 min). 3

SCE Living PRA/Safety Monitor (Living PRA:

-PRA is constantly being updated, as new PRA information becomes available or modeling enhancements are performed.

-,Failure Data for major equipment is updated each plant cycle.

- PRA modeling basis and changes are tracked electronically 4

Full Power and Transition Mo-dels JThe

' following categories are used for the SONGS 2/3 PRA:

- Full power: Mode 1

.....- Transition: Modes 2, 3, and 4 on AFW

- Shutdown: Mode 4 on SDC, Modes 5,6 and

.. offloaded.

lFull Power and Transition models include both internal and external events (EEs).

.- Es are not used for comparison to shutdown.

5

Full Power and Transition Models - Continued f Transition Models are similar to Full Power, with some changes:

- Pressurizer Safety Lift less likely in modes 2,3 and not possible in mode 4.

- ATWS Less likely in mode 2 and not possible in modes 3,4.

- Loss of AFW used instead of Loss of MFW for modes 3, 4 (MFW Not Available for SONGS).

AFW TD Pump not available in mode 4 for SONGS I- nit-LOP increased for modes 3-6, offloaded. 6

Full Power and Transition Models - Continue.d

. TT, Rx Trip, etc. set to zero in modes 3/4.

- More time available ( 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> versus 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />) for recovery of offsite power and MFW/Condensate.

- LOCA Initiating Events reduced by a factor of 20 for mode 4.

- Loss of MFW increased by 4 in mode 2.

- Other Model adjustments needed for conditional events, such as operator responses, fast bus transfer, or conditional loss of offsite power.

7

Full Power and Transition Models Continued

• /Humanactions for modes 1-4 are mostly the same except time related actions.

. Important IEs change from a typical PRA result in mode 1 to dominated by loss of

.. AFW in mode 3-4, or loss of offsite power.

8

. I '.

I-Total CDF, all Initiating Events DRAFT - SONGS 10/4/99 SONGS Transition Risk Model Draft, Rev. 2 1.OOE-04 9.OOE-05 8.00E-05 7.00E-05 6 .00E-05 CL 1..

5.00E-05 IL.

C.) 4.00E-05 3.00E-05 2.00E-05 1.00E-05 0.00E+00 Mode 1 Mode 2, Initial Mode 3, Initial Mode 4, on AFW Mode 4, on SDC Mode 5, vented Shutdown Shutdown Plant Operational State

Full Power and Transition

. odess- Continued f .Model Sensitivity:

- MFW assumed available had little affect on the result, since condensate pumps are already assumed available on all non-LOP sequences.

- Feed& Bleed/PORV availability will lower results in mode 1-4, but only slightly.

- SONGS 2/3 Emergency DG Crosstie removal would raise the PRA results for all modes, but the relative risk would remain similar.

iiii550 0

Full Power and Transition Models Continued IVModel Sensitivity:

- TD AFW Pump being available in mode 4 would lower mode 4 results, with a greater reduction in mode 4 AFW.

-Containment Spray is assumed available for SDC backup. Removal of this results in a factor of 1.5 to 2 for credited modes.

II

Full Power and Transition Models - Continued:

"lModelSensitivity Conclusions.

- Major sensitivities looked at above do not change the general results that mode 4 on AFW has the lowest shutdown risk, and most defense in depth.

12

ENCLOSURE 4 NOVEMBER 10, 1999 MEETING PRESENTATIONS Enclosure 2

Risk-Informed Technical Specification Strategic Vision November 10, 1999

. How do RI-TS fit in with RIP50?

How do RI-TS mesh with the Maintenance Rule 50.65(aX4):

"The second NEI concern addressed the apparent overlapping regulatory requirements ortechnical speclifcatlons and the proposed maintenance rule." [SECY-99*-133]

"* Results of GAO Study: "NRC has not developed a corn prehensive strategy that would move Its regulaton of the safety of nuclear power planta from its traditional approach to an approach that considers risk."

" Concerns about incrm ental changes without overall viion 1inow I

Risk-Informed Alternative Technical Specifications Issues

"* Bring maintenance rule ad technical specificadior into congruence - address fumndaental problem of potentially getting differtni auwers from two major regulations

" Band-aidd rMltionMhp ofsuppor/suppoft*d systems in STS

"* Growing use offunctiosality in TS vice 'trains"

"* STS structure dt has ýsilod" LCO aligned by design revicew and responsible design review organization

" Consider what can be accomplishd without changing 50.36 It/ I 0*9 EXAMPLE Reconstruct Current Technical Specifications Category I Cateory 11 TSCriteria &2 TSCriteri3 &4 Regiie RIP50 Opt4ion 3Modeled in PRA Safety Limits ' Arge LCOs & SRs by Mode Limiting Safety System Soft shutdown requirements Settings Containment Risk Mauamst*,

Etc. (things tot we not modeled, initial Conditions, Threaholb & Aeteas core P0r10erm, etc.)

One or mom' LOOs not met l-A.ss plant Configuration Armnge by Mode??

2-Rmtm fwxtionality 3-Retore rdZImdncy (Le., exit HaMdshutdown requirentmv LCO)

Exoeed safety limit OR

[50.36MeXXiXA)] Use SFDP and quantify i1/10, 2

WOG Five Year Risk Informed Tech Spec Strategy 1999 ... .. - =.. --- 2001 -- M -- -- m- --2005 Infrastructure Major Format Risk Informed Changes Evolutions Je)(JOG) Integrated Safety Specifications (RUSS) STP Lead

-Determine appropriate end states -Delete SRs not related Issue 1 (Task 1, MUHP-3015) to safety functions Rule Changes (Issue 5a) 2002- 2003

• Relax requirement for missed SR *Risk-inform 50.36 Issue 2 (Task 2 of IfMUP-3015)

-- Relocate STIs to format Licensee Controlled 50.36 Format/

-Relax mode change requirements Program (Issue 5b) Content "-Risk-inform Issue 3 (Task 3 of MUHP-3015) Criteria 1, 2, 3

-Risk Informed AOTs Issue 4b (MUHP-3010) a r

-Extend Time in LCO 3.0.3 I Consistent with I *"Floating" AOTs with a "backstop" based on Issue 6 (Task 4 of MUHP-3015) IMaintenance Rule CRMP (Issue 4a)

• RTS/ESFAS AOT/STI Extensions -.. *SSCs that are Inoperable, (MUIP-3045) but Functional (Issue 7)

Current WOG Programs New WOG Programs 99S9PPT - Rough Draf-99G13005 103/ 5799 -

5

Ia INITIATIVE 4a

-AOT EXTENSIONS

• Seen as two part Initiative.

4a Generic risk informed AOT's with a backstop

, FOR EXAMPLE: If component X is inoperable, restore to operable status within 7 days, or operation may continue for up to [30] days if the configuration is acceptable in accordance with the Configuration Risk Management Program (CRMP).

0 ,,

INITIATIVE 4a AOT EXTENSIONS LiENERIC RISK INFORMED AOTS' WITH A BACKSTOP:

EXAMPLE FORMAT Condition Action Completion Time A. One required A. 1 Restore required [subsystem] to Operable status

[subsystem] [ 7 days]

Inoperable or A.2.1 Determine if the configuration is acceptable for an

[ 7days ]

Action A. 1 Completion Time extension not to exceed [30 days] in accordance with the CRMP.

And A.2.2 Restore required [subsystem] to Operable status Action A.2 Limit B. Required Action A and associated Completion Time not B. I Be in Mode 3 met 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br />

2.0 Configuration Risk Management Program 2.1 Purpose - The Configuration Risk Management Program (CRMP) is used to monitor and assess the risk impact of equipment out-of-service and to maintain station risk at desired levels. The CRMP is used to assess risk impacts for planned and unplanned equipment outages that are modeled in the STP Probabilistic Risk Assessment (PRA). The CRMP is applicable to systems, structures, and components (SSCs) within the scope of the station's PRA as reflected in the Risk Assessment Calculator (RAsCal) for plant Mode I and 2 operation and the Shutdown Risk Assessment for plant Mode 3,4, 5, and 6 operation.

2.2 Description - Licensees shall be capable of determining the risk in terms of core damage frequency associated with all historical and planned plan configurations of defined plant critical safety functions. Plant risk levels shall be managed via the most restrictive concurrently applicable risk limits (i.e., lowest quantitative limits and most stringent associated action levels) prescribed in this section. In Sections 2.3.1 and 2.3.2. the plant configuration risk must meet both weekly cumulative and incremental risk limit criteria to justify operation under the green action level. In all cases, incremental risk limits, though calculated over a one-hour time period, apply to any CDP values calculated over time periods of one hour or less (i.e., to meet the risk limit criteria, no instantaneous risk levels greater than the incremental risk limit values are permitted).

2.3 Requirements -The licensee shal be capable of determining the risk associated with plant configurations and shah operate the plant In accordance with tMe allowable risk limits identified in this section.

2.3.1 Allowable risk limits for Modes I and 2:

Risk Significance Region Allowable Weekly Core Damage Allowable Incremental Core RequIred Action Non-Risk Significant Region Probability, CDP__-u-il. Damage Probability, CDP*,,-

CDP,,* < 15.00E-07] CDPkII.--=, < [5.00E-o8j Level GREEN Potentially Risk Significant Region [5.00E-07J < CDPmy,< [1.00E-06] [5.OOE-08] < CDPw.m. < [1.00E-07J YELLOW Risk Significant Region [1.OOE-06] 5 CDPs,... < [2.ooE4)6] 11.OOE-07] 5 CDP.... < [2.OOE-07] ORANGE Highly Risk Significant Region [2.OOE-06) < CDP-..* [12nOOE-07) < CDP.m. RED 2.3.2 Allowable risk limits for Modes 3. 4.5, and 6:

Risk Significance Region Allowable Weekly Core Damage Allowable Incremental Core Required Action CDPwwft Damage Probability,SProbability, CDP;,._,.. Level 0uu.-r-,* ognmca n Region CDP, myw < [5.00E-061 CI*P..... < f* t'V'*,:,.rJH*'l

-D . *Diiy . ,, - .U. j *6-Potentially Risk Significant Region l5.0OE-061 <5 CDP,,f < 11.00E.051 VI=I I t'ttM Risk Significant Region S DP~i~ 1.OE05 Ib.OE06 I12.OOE.O061 < CflP6--. < rR nnpFniVLva ^

[1.00OE-051 < CDP,,"u.< 14.00E-051 t'tO A t, ir*t*

Highly Risk,Significant Region -_~~~ 1`5 tI 001=-m -r~

IR" 14.OOE-Os1c L .OFJI 14.00E-05] <5 CDP,,"* I*l::rt I 4OE-5 C P.1 i.OE05174S CDP&-..I....

I!

RliSMaBW4lpt Nd 14

m iB 2.3.3 Allowable annual risk limits for all Modes:

Risk Significance Region Allowable Outage Core Damage Required Action Probability, CDP..-_ Level Non-Risk Significant Region CDPwwuw < [2.00E.05]

GREEN Potentially Risk Significant Region [2.OOE-05J < CDP <[5.00E-05] YELLOW Risk Significant Region [5.00E-051 < CDP._,- < [1,00E-04]

Highly Risk Significant Region ORANGE

[1.00E-04] < CDP., REDn Plant-spec values (inbrackets,[ D)to be approved by the NRC (Note: Mode transitions risk Is subsumed vwtin these limits.).

A W-4 Actions

2.4.1 Green

Follow normal operating and business practices 2.4.2 Yellow:. The Control Room Staff SHALL takethe following 2.4.2.1 actions:

Notify the Duty Operations and Duty Plant Manager that the Potentially Risk Significant Region has been 2.4.2.2 Identiry and Implement compensatory measures as entered.

approved by the Duty Plant Manager. Compensatoy Include but are NOT limited to the following: measures may 2.42.2.1 Reduce the duration of risk sensitive activities.

2.4.22.2 Remove risk sensitive activities from the planned work 2.4.2.2.3 scope.

Reschedule work activities to avoid high risk sensitive equipment outages. or maintenance states.

2.4.2.3 Ensure any measures taken to reduce risk are recorded In the Contrl Room Logbook.

2.4.3 Orange

Perform Action B and Immediately make notication to NRC.

2.4.4 Red: Perform Action C and transition toa Plant Mode that reduces the overall risk A!

RIllS MScp.ppt Side 18

Initiative 1- Endstate Level of Justification Determination of Appropriate Endstates Identified as the First "Short Term Success" Initiative at the Initial Meeting in December 1998 It was Generally Agreed that it was Intuitively Obvious that Hot Shutdown was a Safer Endstate than Cold Shutdown SONGS Evaluations Confirmed the Intuitively Obvious Endstate Conclusion

"* TSTF Justification will Summarize the Results of these Evaluations to Provide a Basis for the Change

"* Other OGs Intend to Compare their Plants to the SONGS Evaluations and Discuss Significant Differences

° Do not Intend to Perform Similar SONGS Quantitative Evaluations Determination of Appropriate Endstates Should not Require Quantitative Risk Analyses

Initiative 3 LCO 3.0.4 Mode Restraint Flexibility

~ ~ ~

V;~~~ 11= ,J.1ýjiJS Example CE Plant

nmary of Technical Specifications Current TS Allows e Itrv into the Mode of Applicability only if associ Rted Actions permit continued 4 ~operatiol is for an unlimited period of time.

"Proposedt

• in additic]

Ohange into current provision, allows entry into the Ihlode of Applicability based on appropri He Management Review and Approval land relyiing on the associated Actions].

Example

>+ Cycle 8 Containment Spray Pump

• Containment Spray LCO Applicability- Modes 1, 2 and 3.

• Required Completion Time is 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> for 1 Pump.

• Repair was for seal leakage.

tR***i::*outage.

• Repair was begun Mode 4 coming out of the

Exampe [continuedi

-Ccle 8 Containment Spray Pump U* Difficulties with

¶0?

the repair resulted in an 07 hour8.101852e-5 days <br />0.00194 hours <br />1.157407e-5 weeks <br />2.6635e-6 months <br /> critical path extension,

, Major difficulties were encountered in the beginning of the repair.

4:

til 1

Example (continued)

Risk Impact

...a..

• Delta CDF for 1 spray pump out in Mode3 is lE-06.

• Delta LERF for 1 spray pump out in Mode 3 is lE-0O.

" Compared to Regulatory Guide 1.177 or 1.174 Criterion change would be acceptable.

I Enample (continued]

S.Potential Cost Savings Ii't "While the enti re critical path extension could not have been averted by-the proposed TS at change, itislI elieved that once there was a high confiden ce that the repair-would be completed we 11 within the 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> completion time [e.g., 48 hoUrsi, the proposed TS could

> have saved cr1itical path.

p,

S Example

• Potential Cost Savings (continued]

• Using $1millionlday for lost revenues.

• 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> of critical path savings.

• $2 million dollars.

STATI ISRITSTF INITIATIVES 4, 5,v 6and7 p r~

ITIATIVE 4 INITIATIVE 4A

- RISK INFORMED FLOATING AOTS WITH ABACKSTOP

- OGS ACTIVELY DEVELOPING

- TSTF INLATE 2000 N!IN INITIATIVE 4B

' pI* - RISK INFORMED AOTS TSTF PURSUING SELECTED AOTS INPARALLEL INDIVIDUAL OGs HAVE PROGRAMS ADDRESSING SELECTED AOTS

- RITSTF WILL COORDINATE TO SUPPORT GENERIC APPLICATION

- ONGOING THROUGH 2000

L STATUS RITSTF INITIATIVES 4, 5, 6 and 7[CONTINUED)

¶0? INITIATIVE 5

- INITIATIVE 5A RELOCATE SURVEILLANCE REQUIREMENTS (SRs) THAT DO NOT DEMONSTRATE OPERABILITY OF SAFETY FUNCTIONS

- TSTF/RITSTF DEVELOPING

- TSTF INLATE 2000

• INITIATIVE 50,

- RELOCATE SURVEILLANCE TEST INTERVALS (STIs) TO LICENSEE CONTROLLED PROGRAM

- RITSTF DEVELOPING

- TSTFIN 2001

STI lTUS RITSTF INITIATIVES 4,5,6and7 INITIATIVE 6 EXTEND TIME UPON ENTRY INTO LCO 3.0.3 TO INITIATE AND DEVELOP RISK INFORMED COURSE OF ACTION INITIALLY SCOPE WAS TO CHANGE CURRENT LCO 3.0.3 ONE HOUR TO 24 HOURS tp INITIAL EVALUATIONS DID NOT SUPPORT CHANGING ALL SPECIFICATION ENTRIES INTO LCO 3.0.3 TO 24 HOURS

SSTATUS RITSTF IINITIATIVES id 7 i

S-INITIATIVE 6 [CONTINUED ]

• t Im" EXPANDED SCOPE OF INITII ITIVE 6 HAS 3 EFFORTS

- LCO 3.0.3 ONE HOUR TO 241lOUR CHANGES

- ADDRESSING CONDITIONS !]iFINDIVIDUAL "SPECIFICATIONSWHICH DII

]ECT ENTRY INTO LCO 3.0.3 TO PROVIDE ACOMPLETION' TIME INTHE INDIVIDUAL SPECIFICATION

STATUS RITSTF INITIATIVES 4,5,6and7 NITIGATIVE 6 (CONTINUIE D)l

- PROVIDING CONDITIONS, IREQUIRED ACTIONS AND COMPLETION TIMES FOR THOSE INDIVIDUAL SPECIFICATIONS WHERE RlONE EXIST REQUIRING ENTRY INTO ICO 3.0,3

-PRA EFFORT ONGOING

~jm~ TSTF IN2000

STi iTUS RITSTF INITIATIVES 4,5,6and7 IfE7 1E 7 DEVELI PACTIONS. FOR EQUIPMENT THAT IS Ilk 4>

• INOPEfi ABLE BUT FUNCTIONAL

..... ~ ... V AL DATING MAINTENANCE RULE aA AND AVAILABLE TSTF INLATE 2000 Ui-.