ML20214E164
| ML20214E164 | |
| Person / Time | |
|---|---|
| Issue date: | 05/31/1986 |
| From: | Connelly S NRC OFFICE OF INSPECTOR & AUDITOR (OIA) |
| To: | |
| Shared Package | |
| ML20210N704 | List: |
| References | |
| FOIA-86-651 NUDOCS 8611240361 | |
| Download: ML20214E164 (22) | |
Text
__ x _
g f- s, ,
, umem . : s Nuclear Regulatory Commission :" ~
...../
Report to the Commission
' ' ' ' n'-
y e g Review of Technical Specifications forNear-Term
.: Operating Licenses
?
E, h -
- - .i
- i 1
. Office of Inspector and Auditor May 1986 I
NOTICE This is An internal Management Document Prepared For The Commission Not to Be Released Outside Of NRC e'
. 8611240361 861110 1 l PDR FOIA
',o PLETTIN86-651 PDR
.- ~ .... _ - . - ..- .
i
. q
/ o,, UNITED STATES
! o NUCLEAR REGULATORY COMMISSION j
$ 'E WASHINGTON, D. C. 20555
' )
\ *****
] May 14,1986 MEMORANDUM FOR: Chairman Palladino Comissioner Roberts Cmmissioner Asselstine Comissioner Bernthal Cmunissioner Zech FROM: Sharon R. Connelly, Director t4h j Office of Inspector and Auditor
SUBJECT:
OIA REVIEW OF THE TECHNICAL SPECIFICATION REVIEW PROCESS FOR NEAR-TERM OPERATING LICENSES Attached is our report on the results of our audit of the technical specifications (tech specs) review process for near-term operating licenses.
We initiated our review to assess the effectiveness of improvements instituted,
. as a result of the experiences with the Grand Gulf technical specifications.
'Ihe objectives of our audit were to determine if tech specs are reviewed and approved in accordance with established NRC policies and procedures and if such policies and procedures provide for adequate internal control.
We conducted our review in accordance with generally accepted Government auditing standards. We performed the audit at NRC Headquarters and Region I intermittently frm May through August 1985 and full-time from September through November 1985.
Conclusions and Recomendations Our review disclosed that the Office of Nuclear Reactor Regulation (NRR) has made significant improvements to the tech spec review process since the Grand Gulf situation. However, we found that one improvement, a milestone schedule designed to provide adequate review time, is not being implemented as intended. Specifically, the staff compresses the tech spec review time to meet the scheduled full-load date, without due consideration of the impact on the review. We also found that other improvements implemented are less effective than they should be and do not provide for adequate accountability.
Specifically, NRR needs to establish formal policies and procedures to standardize the conduct of each tech spec review. l l
Our report makes three recommendations to effect corrective actions on the specific problems addressed in the report and to enhance the overall 1 effectiveness of tech spec reviews for near-term operating licenses.
Attachment:
As stated l m
_ _ , _ ~ . . _ . _ _
j l
. l
\
4 D cc: V. Stello, EDO S. Chilk, SECY j e M. Malsch, Acting OGC J. Zerbe, PE C. Kamerer, CA J. Fouchard, PA H. Denton, NRR J. Davis, NMSS J. Taylor, IE T. Murley, RI ~
J. N. Grace, RII J. Keppler, RIII R. Martin, RIV J. Martin, RV J. Roe, EDO T. Rehm, EDO L. Hiller, ICC
. THIS REPORT IS BEING PROVIDED FOR THE COMMISSION'S INFORMATION. NO ACTION BY 'IffE COMMISSION IS REQUIRED AT 'ITIIS TIME.
e i
I e
l
\
4
- - - - ~ ~ . - . - -
. ~ ~~ ' l i
.)
, 1 REVIEW OF W E TECHNICAL SPECIFICATIONS REVIEW PROCESS FOR NEAR-TERM OPERATING LICENSES e
EXECUTIVE
SUMMARY
During 1984, the Nuclear Regulatory Comission (NRC) experienced a major i problem relating to the develognent, review, and approval of technical
, specifications (tech specs) for the Grand Gulf nuclear power plant. A l subsequent Office of Inspector and Auditor (OIA) investigation of one aspect 4
of the Grand Gulf problem revealed that the entire tech spec process was i informal there were no formal records to substantiate the review work l conducted in the Office of Nuclear Reactor Regulation (NRR). As a result,
) accountability for the review work was limited.
l We initiated our review to assess the effectiveness of improvements instituted i as a result of the exparience with the Grand Gulf tech specs. S e objectives j of our audit were to determine if tech specs are reviewed and approved in accordance with established NRC policies and procedures and if such policies and procedures provide for adequate internal control.
FINDINGS Overall, we found that the process for developing, reviewing, and approving tech specs is much improved over the process as it existed at the time of the Grand Gulf situation. Although we conducted our evaluation of the technical specification process under a previous NRR organizational structure, we believe that NRR's improvenents, coupled with the recomendations contained in this report, will permit the improvements to be carried over to the new
. organization.
While the actions taken by NRR represent significant improvements to the process, our review showed that additional improvements are still needed in scheduling reviews.- Furthermore, we found that because there are no formal procedures to implement the corrective actions developed, the process is less
- effective and provides less accountability than it should. Specifically, we found that
In some cases tech spec review schedules were accelerated so that the reviews could be completed by the scheduled fuel load dates, thereby causing review schedules to be compressed. This was done despite the fact that the milestone schedule developed by the Tech Spec Review Group (TSRG), NRR, was designed to provide adequate time to produce accurate and reliable tech specs.
2e absence of formal procedures has resulted in inconsistent distribution of the tech specs for review by the NRR staff and inadequate tracking and controlling of the technical responses to the tech specs. Specifically, we found that inadequate tracking and i
controlling techniques led to one branch's technical review not
, being documented for the River Bend tech specs. We also found that j URC's official docket records were incomplete as to technical review work done on the River Bend tech specs.
k
, . _ , , _ _ . _ . . _ _ - . - _ , _ . - _ _ _ . . _ . ,_ . . . , . , _,,_,, ~~ _ , _ - , _ . - ,,.__. .
- - - . a : = .-. =
z . ._ _ - ----
2 l
\ .
j -
Despite' the fact that changes to the tech specs are comonplace after staff review, the staff does not routinely review the final
- draft of tech specs. Tech specs could, therefore, contain changes and items not previously approved by the respective technical branches.
t 1 CONCIUSIONS We believe that the action taken by NRR in establishing a milestone schedule for tech spec reviews was a significant attenpt to improve the process and assure the reliability of the tech specs. However, our review has shown that
- three recent tech spec schedules were accelerated, m ile our review has not l established any adverse safety effects from the use of accelerated schedules,
. we question the wisdom of continually overriding a corrective action specifically designed to avoid the recurrence of significant tech spec j problems,.
i j Certainly, there may be situations where it is appropriate to accelerate a i
review schedule. However, such action should only be taken when there are i clear indications that the licensee will meet the projected fuel load date and j . the review will not be compromised by use of an accelerated schedule.
1 With regard to the absence of tech spec procedures, our review did not find
. that this situation had a major impact on the adequacy or accuracy of tech specs. However, we believe that the lack of procedures presents the potential for a technical review branch not getting the tech specs for review, for review coments not to be received by 'IERG, for documentation not to reach the Central Files, and for tech specs to be changed without proper approval. We
- believe that such procedures will provide controls needed to assure the l
integrity of the tech spec review process.
We recognize that under NRR's recent reorganization each division will coordinate and manage the technical specification process for each plant under its jurisdiction. We believe that the reorganization increases the need for tech spec review procedures to ensure that the quality of reviews remains consistent among the divisions.
RECOMMENDATIONS h e report contains 3 recmnendations to the Director, NRR to effect improvements in the tech spec review procesc for near-term operati'g n licenses.
Se recomendations are intended to address the need to 1) clarify NRR policy a with respect to tech spec review schedules and 2) establish uniform procedures to enhance the overall effectiveness of and accountability for tech spec
! reviews. .
AGENCY COMME 1ES On April 28, 1986, the EDO provided coments on a draft of this report (See l Appendix I to the report). In his coments, the EDO disagreed with-1 Recomendation 1 regarding the need to " Clarify for the NRR staff the policy 3
concerning licen.see review schedules, particularly as related to tech specs."
- R e EDO stated that staff scheduling is presently in accordance with agency
, - , ,--n g a - ----,-ae .,n-n,w,-s,,,
- e. g- -
, , w- , , , - -- , - - - - - r- - ,, ,,,,ng -,--,.--y -.-m- , -- - ~----e - -n e. ,--
._..--..m._______.._._____
3 policies, and that the staff does attempt to revise schedules when appropriate.
Despite the ED0's connents, we continue to believe that the EDO or the Director, NRR should clarify NRC's scheduling policy with regard to tech specs. Our report notes that the staff feels constrained to schedule strictly in accordance with NRC's official policy and that the Deputy Director, NRR, agreed that the scheduling policy with regard to tech specs may need to be clarified.
We have raised the issue of NRC's scheduling policies previously in our report entitled "OIA Review of the Operating License Review Process for Power Reactors." Since the Comnission at that time expressed the belief that the staff is in the best position to evaluate the adequacy of review schedules, we will defer to the EDO's judgement on this reccanendation and not elevate the matter for Commission consideration. We consider this reccanendation closed.
The EDO generally agreed with reconnendations 2 and 3 and advised that he
"...will establish written procedures for the Tech Spec review process, and will issue them formally to the staff." OIA will review and evaluate the effectiveness of these procedures at the time of our follow-up review.
1 O
r ,
. . , _ _ _ _ _a ----
TABLE OF CONTENTS INIRODUCTION 1 BACKGROUND 1 SCOPE 2 ,
FINDINGS 3 ACCELERATED TECHNICAL SPECIFICATION 4
- REVIEW SCHEDULES
! River Bend Technical Specifications 4 Millstone 3 and Perry 1 Technical 7 Specifications
- Conclusion 7 Recomendation 8 NEED FOR FORMAL TECHNICAL SPECIFICATION 8 REVIEW PROCEDURES Distribution of Technical Specifications t 8 Tracking and Controlling Responses 9 Fract Technical Branches Review Documentation in Central Files 9 Staff Review of Final Technical Specifications 10 Conclusion 10 Recommendations 11 i
AGENCY COMMENTS 11 Appendix I - Memorandum to Sharon R. Connelly from
- Victor Stello, Jr., dated Apr'il 28, 1986 l
t e e
4
, . , - I - - - - , y. .- 4.-. . . - - ~ . - . . - - .,
., .m . _
mm.,_ _ _ _ . _ _
i i, .
^
j .
j REVIEW OF THE TECHNICAL SPECIFICATION REVIEW PROCESS
- FOR NEAR-TERM OPERATING LICENSES i
! I!ERODUCTION
- t j
) During 1984, the Nuclear Regulatory Comission (NRC) experienced a major i problem relating to the review of the technical specifications (tech specs)
I for the Grand Gulf nuclear power plant. A subsequent Office of Inspector and '
Auditor (OIA) investigation of one aspect of the Grand Gulf problem revealed
- that the whole tech spec process was informal; there were no formal records to substantiate the review work Conducted in the Office of Nuclear Reactor
- Regulation (tRR). As a result, accountability for the review work was
! limited. For example, an IRC official comented to OIA that the draft tech f specs were so marked up that it could not be determined whether the changes i
were made by tEC or the licensee. S e then Director, Division of Licensing, j IRR, remarked to OIA that because of the informality of the process it was difficult to determine who within NRR conducted each review; there were no sign-off sheets indicating the responsible personnel at the various levels in
)l the organization. ,
l Subsequent to the Grand Gulf situation, NRR instituted changes to the process to ensure better accountability and to prevent a recurrence of the Grand Gulf
- situation.
1 Also in 1984, OIA cepleted a review of the operating license review process I
] for power reactors. During that review, NRC officials expressed the opinion J
- that tech specs are developed too late in the licensing process. Because of l these coments and the problems experienced with the Grand Gulf tech specs,
) we initiated a review of the developnent, review, and approval process for
{ tech specs for near term operating license (NIOL) plants, f
j BACKGROUND i
One of the major tasks in performing a license review is the developnent of '
, suitable tech specs. Se tech specs identify and define all the limits and
- requirements that the licensee must abide by without change unless specific approval (i.e. a license amendment) is obtained from the Imc. 10 CFR Part
- 50.36 requires the issuance of tech specs for nuclear reactors. Se tech' specs bec me part of the operating license and govern the subsequent j operation of the facility relative to the health and safety of the public and l the protection of the environment. E en the tech specs are cogleted, they
- are issued as Appendix A to the operating license.
I 2e tech rpecs must be substantially complete before the plant operating I procedures can be written and the plant operators can be trained. Se tech l specs are developed in much the same way as the Safety Analysis Report and j the Environmental Report. S e licensee provides input information and the t
! NRC staff reviews and evaluates the information. Se schedule objective is l for the Tech Spec Review Group to issue a " proof and review" draft of the j tech specs to the technical staff no later than 4 months before the scheduled fuel load date. !
i i
t_ _ , _ ._ _ . . - . _ _ _ , _ _ _ _ _ . _ , _ _ _ _ _ . _ _ - . . _ . - . _ _ . _ - . _ - . - _
2 Tech specs are derived from the analyses and evaluations include 0 in the safety analysis report, and amendments thereto. As required by 10 CFR 50.36, plant tech specs for power reactors are to include (1) safety limits and limiting safety system settings, (2) limiting conditions for operation, (3) surveillance requirements, (4) design features, and (5) administrative controls.
At one time, tech specs were developed individually for each plant. Sey were similar for plants of like design, but quite a bit of tailoring and scrne inconsistencies always existed. In the early 1970s, the IRC developed and required on a forward-fit basis, the use of standard tech specs (STS) to make the process more efficient and effective. Standard tech specs have been developed for each Nuclear Steam Supply System (NSSS) vendor design.
Operating licenses are now being issued with standard tech specs.
NRC policy dictates that NRR establish license review schedules based on applicants' projected fuel load dates, as estimated by the utility and confirmed by IRC's Caseload Forecast Panel (CFP) . Se CFP's function is to provide Imc management with an independent estimate of the applicants' fuel '
load dates.
Direction for establishing review schedules was provided to IGR in the Ccmission's Policy and Planning Guidance (PPG) for 1982 through 1985, which states that staff reviews and public hearings should be ccznpleted on a schedule that assures that the licensing process will not unnecessarily be a critical path item which would delay reactor start-up. It ha.s been and currently remains tac policy that the review schedule will parallel the fuel load date unless the applicant's estimated fuel load date differs by more than six months from the CFP estimated fuel-load date.
Prior to !ER's reorganization on November 24, 1985, the Technical Specification Review Group (TSRG) in the Division of Safety Assessment, was responsible for the overall management and integration of tech specs. Under the new IER organization, this responsibility is carried out by reviewers in each of three new divisions.
SCOPE We conducted our audit intermittently frcra May through August 1985 and full-time fran September through November 1985. S e objectives of our audit were to determine if tech specs were reviewed and approved in accordance with established IEC policies and procedures and if such policies and procedures provided for adequate internal control. Our audit was conducted in accordance with generally accepted Government auditing standar,ds.
We performed our rev'iew at NRC Headquarters and Region I. Our review included interviews with reviewers in TSRG, project managers, and management officials in IER and Region I. We also interviewed Region IV officials during a visit to Headquarters. We reviewed documents and correspondence related to tech specs in general, and in particular for the River Bend, Perry 1, and Millstone 3 nuclear power plants. We chose those plants because their tech specs were currently being reviewed and all were ccheduled to be licensed in 1985.
I
._ _ A
3 I
^
) .
FINDINGS j i
Overall, we found that the process for developing, reviewing, and approving tech specs is much improved over the process as it existed at the time of the Grand Gulf situation. Although we conducted our evaluation of the technical specification process under a previous NRR organizational structure,.we see no reason why the improvements NRR has already made will not be carried over to the new organization.
t
} Among the improvements NRR made to the process was the establishment of the
- TSRG under the Division of Safety Assessment. TSRG assumed overall
- coordination and management responsibility for tech specs and required all 1 '
technical reviewers to subnit the results of their reviews to TSRG in writing, through their respective Assistant Directors. %e staff reviews constitute formal confirmation that the tech specs are accurate and properly reflect the data contained in the Final Safety Analysis Report and the Safety Evaluation Report and its supplements. In addition, NRR required that the utility certify under oath that the tech specs are accurate and properly reflect the as-built plant.
To further improve the reliability of the tech specs, the Office of
,' Inspection and Enforcement issued inspection procedure 71301 on March 22, ,
1985 W e objective of this procedure is to ascertain whether the proposed s tech specs are clear, enforceable and address the systes installed at the j plant.
In order to provide an additional check on tech spec reliability, tRR has an
, independent contractor review the completed tech specs and compare them to the Final Safety Analysis Report, the Safety Evaluation Report, and changes thereto. % e contractor reports its findings directly to NRR and in, particular notes the consistency or inconsistency of the tech specs Vith the above mentioned documents.
To provide the level and depth of review needed to produce an adequate and 1
accurate set of tech specs, TSRG developed a milestone schedule for the
, develognent, review and approval of tech specs. With this schedule, TSRG established a 56 week time frame for developing and issuing the tech specs prior to NRR's issuance of the low-power license. We schedule is extended to 72 weeks when the time line is extended to the full power license.
l While the actions taken by NRR and IE represent significant improvements to the process, our review showed that additional improvements are still needed in scheduling reviews. Furthermore, we found that because there are no j formal procedures to implement the corrective actions developed, the process is less effective and provides less accountability than it should.
Specifically, we found that:
l In some cases tech spec review schedules were accelorated so that the a
reviews could be completed by the fuel load dates, thereby causing .
, review schedules to be compressed. 2 1s was done despite the 'act that the milestone schedule developed by TSRG was expressly designed to provide adequate time to produce accurate and reliable tech specs.
t e absence of formal procedures has resulted in inconsistent distribution of the tech specs for review by the NRR staff and inadequate tracking and controlling of the technical responses to the tech specs. Specifically, we.found that inadequate tracking and
_ _ _ _ _ _ . _ . , - _ . - _ _ - _ _ . - - ~ _ _ , _ _ . __. _._._ ._ __ - -_ __
-4
+
controlling techniques led to one branch's technical review not being documented for the River Bend tech specs. We also found that the NRC's official docket records were incomplete as to technical review work done on the River Bend technical specifications.
Despite the fact that changes to the tech specs are conunonplace after their initial review by the technical staff, current procedures do not provide for staff review of the final tech spec package. Instead the ,
staff reviews the entire package only at the " proof and review" stage, and there are no procedures to ensure that changes to the tech specs after that review are approved by the technical staff. Tech specs could, therefore, contain changes and items not previously approved by the respective technical branches.
Se details of our findings are presented in the following sections.
ACCELERATED TECHNICAL SPECIFICATION REVIEW SCHEDULES As we stated earlier, TSRG developed a milestone schedule to provide for !
adecuate review of tech specs. According to TSRG's milestone schedule, the i Proof and Review copy of the tech specs is to be issued by TSRG approximately 120 days prior to the fuel load date. Se Proof and Review stage is the first formal opportunity that the technical staff has to review the tech specs, he schedule allows the staff four weeks to review the draft and report the results of their reviews to TSRG. S e final draft of the tech specs is supposed to be completed six weeks prior to fuel load and the independent contractor review is supposed to be cmpleted four weeks prior to fuel-load.
As part of our review we examined the review schedules used by NRR for the River Bend, Perry 1, and Millstone 3 nuclear pnwer plants. Our review showed that all three review schedules were accelerated (compressed). Additionally, '
we found that the River Bend review was accelerated and remained accelerated despitetheconcerngoftheTSRGreviewerresponsibleforitandonehigh level NRR official. ,
River Bend Technical Specifications
- As of April 1985, River Bend was scheduled to load fuel in June 1985. At that same time, however, NRR determined that the review of the River Bend tech specs was behind schedule and could not be completed by the anticipated fuel load date. As a result, NRR decided to accelerate the schedule to accomodate the June fuel-load date. Because of FSAR amendments and other review work which needed to be completed, however, River Bend did not receive its low-power license until late August 1985.
1 All three tech spec reviews were in process at the time TSRG instituted its milestone schedule. Berefore, we did not evaluate the adequacy of the total review time (56 weeks) allocated for tech specs.
9 1ww-w y -32,-ne, same-- -w.r,-i ,--yyw-e *-w rv r- ve -w w w~rr ww w w-w w=-v-- -in--- "
- w v--,-m-* -v,- ---=- y -+' +yvv ~
TM----m vt-~~ * * - - = ~ * , "--f -
+-vem- * - "--~w-
5 ll .
-1 l
l
- . l
- .l l In revising and adhering to the accelerated schedule, however, NRR made
, several changes which we consider to be contrary to their objective in
! establishing the milestone schedule, i.e., to provide for adequate review of -
J the tech specs and to prevent a recurrence of,the Grand Gulf situation, f i
l
- In accelerating the review of the River Bend tech specs, NRR decided to by-pass the Proof and Review stage of the review, and instead opted to send a
, final draft to the staff for review. At that point, it was not anticipated that the staff would get an opportunity to review the tech specs again before l they were issued. In fact, however, the Division of Licensing (DL) did j request the staff to review the tech specs again in late July because of the ntnerous changes made from the time the staff originally reviewed the tech i specs in April and May.
! In transmitting the tech specs to the staff for review on April 19, 1985, the i Assistant Director for Safety Assessment and the Assistant Director for i Licensing, DL, stated:
By issuing these technical specifications in the final draft form ,
at this time without benefit of a formal proof and review period 4 . between the second draft and the final draft, we are making a j . significant deviation from our normal technical specification ,
preparation process. R is deviation is being made in the .
. interest of expediting the review process so as not to unnecessarily delay the planned start up date of some time in
] June 1985. It is important to note, however, that even though we
{ are skipping one of the steps in the preparation process we are .
l not skipping any of the steps in our independent review and l certification process to assure the safety adequacy of the technical specifications.
Additionally, the Assistant Director requested that the staff respond by May 13, 1985, thus allowing the staff three weeks to review the tech specs instead of the established review time of four weeks. Furthermore, if River Bend had loaded fuel as scheduled, the tech specs would have been issued for review just 70 days prior to fuel load rather than 120 days, as prescribed by the milestone schedule (assuming a fuel-load date of June 30, 7 1985).
I Our review has raised questions regarding the need to accelerate the i River Bend review schedule. Specifically, we believe that NRR had sufficient information available to it before the schedule was accelerated to indicate
, that the June fuel load date would not be met. For example, as early as
! March 25, 1985, the staff had indications that the schedule was slipping due j to delays in receiving submittals from the applicant. On that date, the TSRG j group leader sent a memo to a DL licensing branch chief noting the concerns i of the TSRG reviewer responsible for River Bend tech specs. S e TSRG reviewer
- was concerned that the applicant was already late in submitting data to the staff. Furthermore, he indicated that the issues resolved by the applicant
! up to that point were relatively easy ones, and implied that even more time i would be lost when the applicant addressed the difficult issues.
Additionally, in March 1985 the number of River Bend SER outstanding issues requiring resolution began to increase, thus adding more work to that l required of the licensee. In our opinion, this was an indication that the l .
- - - _ _ _ _ _ _ _ - - -._.. - _ _ - - - - ~ _ _ , _ - - . - _ _ _ _ , _ - _ _ _ . . _ . - _ _ _ _ .-
. . _ - - - _ _ _ . a . e.e,._ m -- - -
6 -
scheduled fuel load date would slip even further. (In fact, the number of outstanding issues rose frm about 18 in March, to about 25 in June; and the fuel load date did slip. Ultimately there were almost 30 issues that needed resolution by the time the low-power license was finally issued in late August. All outstanding issues were resolved at the time NRC issued that-license.)
On May 1, 1985, the TSRG reviewer sent a memo to his group leader expressing several concerns about the quality of the tech specs, the pace of the review, and his belief that the tech spec process for River Bend was paralleling that '
of Grand Gulf. 21s particular reviewer had also served as project manager for Grand Gulf. In that memo, he expressed his belief that the plant would not be ready for licensing before September 1985 and further stated:
It would appear that our accelerated review schedule is inappropriate and our " Final Draft" questionable. Derefore, I feel that the future developnent of RBTS (River Bend Tech Specs] should be reassessed upon the conclusion of the current review, on or about May 15, 1985, and see if a normal schedule (enclosed) would not be the better way to proceed.
. Se licensing process for River Bend follows that for Grand Gulf during the last year of review. Grand Gulf was always ready for
. licensing in the next month or two from November 1981 to June 1982. Due to this anticipated action, technical reviews and Tech Spec developnent could not be performed under ideal conditions.
We are all aware of the problems that surfaced afterwards on GGTS (Grand Gulf Tech Specs] but yet it appears to me that we have ignored some of the lessons learned from the GG case by accelerating the RBTS development, thus giving reviewers less time to perform their review and eliminating some logical steps in our established procedure.
Due to the uncertainties associated with the cur' rent safety review, I would recomend that the TS situation be reassessed in the near future and a realistic schedule be set. If many problems exist or are still to be resolved, a normal schedule ending in August should be adopted.
We spoke with staff and officials throughout NRR in order to assess the impact of an accelerated review schedule on the conduct of a tech spec review. Almost everyone with whom we spoke believed that an accelerated ;
schedule did not necessarily have a negative impact on the conduct of the i review. S ey believed instead that an accelerated schedule causes them to !
change their priorities to accomodate the schedule. No officials at the l Assistant Director level and one reviewer, however, expressed the belief that an accelerated schedule has the potential for a negative impact by causing the review to be done in a " hurry-up" atmosphere.
tRR officials advised us that it was a conscious decision to accelerate the review of the River Bend tech specs. Furthermore, they said they were aware of the concerns expressed by the TSRG reviewer, and determined that the review should remain accelerated despite those concerns. One official directly responsible for the review said that he was uncomfortable with the
. review schedule, and would not conduct another review in the same fashion.
He said that the decision to proceed with the accelerated review came from
. . ; - =; .
'7 l
- 5. .l i .' ,
l above his level. Another official advised us that it is NRC policy not to change a schedule date unless the utility and NRC estimated fuel load dates j differ by more than six months. This official also expressed his belief that i i the River Bend tech spec review provided a reasonable level of assurance of '
! the tech specs' reliability, but believed that a higher level could have been
!- achieved under a normal review schedule.
j We spoke to the NRR Deputy Director to get his perspective on IttC's policy
- for scheduling tech spec reviews in particular and licensing reviews in i
- general. Since he was a previous Director, Division of Licensing, we l believed that he could provide insight fra the point of view of having l
- directly supervised the process and fr a a management perspective. He
- confirmed that NRC's policy is to work to the fuel load date unless NRC and
- the utility disagree on that date by more than six months. He believes,
- however, that whenever there are indications that working toward an
} unrealistic date may compromise or have the potential to compromise a review, i management should be notified and the dates should be changed. We advised him that this was precisely the case with River Bend but that the schedule was not changed. He expressed his belief that the current scheduling policy j -may need to be clarified for the staff.
1 Millstone 3 and Perry 1 Technical Specifications ,
! . Se review schedule for Millstone 3 was accelerated up to the point where the l proof and review version of the tech specs were finalized. This was partially due to the fact that, although the utility submitted their tech-specs to the NRC in December 1984, inadequate staff resources prevented review of the package until about March 1985.- 2 en the Division of Safety Assessment sent the proof and review copy to the staff for review, the staff
! was requested to respond in two weeks instead of the established four weeks.
l Again the available review time was cut short.
Se project manager and TSRG reviewer for Perry 1 advised us that the review of this plant was accelerated initially; however, the utility subsequently
! modified the fuel load date and that resulted in more review time for the j staff.
i j conclusion l We believe that the action taken by NRR in establishing a milestone schedule . -i j for tech spec reviews was a significant attempt to improve the process and i assure the reliability of the tech specs. However, our review has shown that i i three recent tech spec review schedules were accelerated. m ile our review l i has not established any adverse safety effects from the use of accelerated schedules, we question the wisdom of continually overriding a corrective action specifically designed to avoid the recurrence of significant tech spec
. problems.
1 i
Based on our review it appears that the use of accelerated review schedules l, is directly tied to fuel load dates. It should be noted that the final stage
, of the tech specs review occurs at the end of the licensing process and the
! review remains geared to the fuel load date despite the fact that there may be problems with finalizing the tech specs.
l -
I 1
. , - - , - . _ .--.,-,,---_,-.--.----.-,-.-_.-.~..~~-i . - - - - - - - - - - - - - - - - -
8 Certainly, there may be situations where it is appropriate to accelerate a review schedule. However, such action should only be taken when there are clear indications that the licensee will meet the projected date, and the review will not be compromised by use of an accelerated schedule.
Reconnendation Since we found that the staff feels the need to accelerate tech spec reviews to meet fuel load dates, we re w m A 4 that the Executive Director for Operatiors and the Director, NRR:
, 1. Clarify for the'NRR staff the policy concerning license review schedules, particularly as related to tech specs.
l NEED FOR FORMAL TECmTICAL SPECIFICATION REVIEW PROCEDURES i Our review disclosed that despite NRR's attempts to improve the process, the absence of formal tech spec review procedures causes the process to be less i effective and provide less accountability than is desirable. Furthermore, 1
under NRR's 1985 reorganization, the tech spec process is now being carried out in three divisions instead of TSRG alone.' Therefore, we believe that the need for formal procedures to standardize the conduct of each review becomes even more important. .
i We believe that formal procedures are required in four areas:
(1) distributing the tech specs to the technical review branches, i (2) tracking and controlling the responses frczn the technical branches, (3) ensuring that the central files get copies of all review related material, and (4) ensuring staff review of the final set of tech specs.
Distribution of Technical Specifications We reviewed the distribution of the River Bend, Perry 1, and Millstone 3 tech
{ specs to the technical staff. In that review, we noted that the Perry 1 tech
' specs were not given the same distribution within NRR as the other two.
Specifically, the Mechanical Engineering and Chemical Engineering Branches were left off the distribution, and the Equipnent Qualifications Branch was listed twice. While these branches ultimately provided input to the tech
' specs, we believe that the distribution list should be standardized so that chance does not play a role in a branch receiving tech specs for review.
The Group Imader of the TSRG advised that he was aware of the discrepancies
- in the distribution of the Perry 1 tech specs and action was taken to get the
- t'ech specs to those branches. He advised that TSRG did use a standardized l distribution list and that this case represented an isolated error. While it may have been TSRG's policy to use a standardized distribution list, one TSRG l
, reviewer we spoke to did not know of its existence. When asked for a copy of the distribution list that was used for his plant, he advised that he would have to use an organizational chart to determine which branches should get
! the tech specs for review. .
I
+
- , , , . ~ e _ -w, ,,- n
.,,-e, -ne,----
--7, ,, ,, --v---+.,,-- , , - ------.-,,e---,,-.p--,--- , + m-,--e,y - - g-
9 r
_Trycking and Controlling Responses From Technical Branches As we noted earlier, the former Director, Division of Licensing, remarked that.the tech spec process was so informal that it was difficult to determine who conducted a review because there were no sign-off sheets indicating the responsible personnel at the various levels of the organization. Although NRR later required written responses to document each technical review, NRR did not develop procedures to ensure that responses would be adequately tracked and controlled. m ile their process appears to have generally been reliable, we believe that the absence of standard procedures resulted in one branch's review not being documented and that situation going unnoticed by TSRG.
Our review disclosed that TSRG reviewers did not use a formal checklist or tracking sheet, rather they made notations on the distribution list as to tM status of the responses. Furthermore, this annotated distribution list is not made a part of the permanent docket records to support a completed review. W e reviewers with whom we spoke indicated that the tracking list is their personal record and does not get ingorporated into the official docket files.
In order to assess the effectivenes.s of the current process, we audited the receipt of branch responses to the tech specs for River Bend, Millstone 3,
. and Perry 1. We reviewed TSRG documentation and spoke to TSRG reviewers to attempt to account for responses from all technical branches. At the con-clusion of this review, we were unable to account for a response from the Materials Engineering Branch (MrEB), Division of Engineering on the River. Bend tech specs. TSRG was not aware that they lacked documentation for this review. Subsequent OIA review and discussion with NRR officials and technical reviewers disclosed that MrEB had done its review but had not documented it. S e review, which was done in April or May 1985, was documented on November 7, 1985, in a memorandum from the Division of Engineering to the Division of Licensing.
Review Documentation in Central Files During our examination of documentation to support tech spec reviews, we noted that NRR's official docket files (central files) do not contain complete records of technical reviews. We found this situation for two of the three technical specification reviews we examined. Our revie,w also indicated that this problem may not be confined to tech spec reviews only.
I We discussed this situation with officials throughout NRR and the Chief, Records Services Branch, Office of Administration. 21s branch has the responsibility for maintaining and updating the central files. Our l discussions revealed that there is no consistent understanding concerning who l is responsible for ensuring that central files is on distribution for review related material. Some officials in review organizations (e.g. the Division of Engineering and the Division of Systems Integration) believe it is their responsibility. Others in those organizations believe that it is DL's responsibility.
2e Chief, Records Services Branch, told us that on occasion NRR offices have advised her that certain material does not need to be sent to central files.
It is her belief that each NRR office decides what to send to central files and she knows that all necessary material does not get sent to central files.
Our review of tech spec documentation confirms her statement. We found
10 1
- \
s numerous documents in NRR files that were not contained in the central files.
NRR officials agreed with our assessment that the central files should contain such documents. The Director, Division of Licensing, advised us that he would take action to ensure that all necessary material gets sent to the central files.
We note that the issue of an incanplete docket file 'is not a new one. On March 3, 1978, the then Acting Director, NRR, sent a memo to the NRR staff stating that "Recent events involving several EOIA requests and Congressional Inquiries have demonstrated a need within NRR to establish general guidelines for maintaining office files. Currently, no consistent procedures exist and manpower is being wasted searching for internal documents which should have been placed in Central Files initially." He further noted that the failure to provide copies to central files results in an incomplete agency record. l He therefore mandated that one copy of all applicable correspondence be transmitted to central files to becane the permanent agency record.
Staff Review of Final Technical Specifications As we noted previously, the proof and review stage provides the staff with
. their first formal opportunity to review the tech specs. D e tech specs are supposed to be in final form at that stage. W ere are, however, several reasons for changing the tech specs after the proof and review stage. For example, the tech specs may be refined or enhanced by the utility as a result of preoperational testing, or they may be changed as a result of questions raised by the NRC staff. Regardless of the reasons for the changes, the staff is not routinely afforded the opportunity to review the entire tech spec package after proof and review. Several officials believed, however, that changes to tech specs are sent back to the appropriate technical branch or resident inspector for review. We found, however, that there is no procedure to ensure that changes will be sent to the appropriate technical branch. .
A former Assistant Director in the Division of Systems Integration noted that his technical reviewers felt that they were ultimately responsible for- the accuracy of their tech specs; however, since the reviewers do not see the final tech specs, they have to assume that the final tech specs read the same as when they gave their approval. Region I officials said that they too have to make the same assumption. We do not believe that reviewers should have to make this assumption; they should know this through a review of the final package.
We believe that a procedure for staff review of all tech spec changes and review of the final tech specs would provide further assurance that the tech specs properly reflect the staff review.
Conclusion Our review did not find that the absence of procedures had a major impact on the adequacy or accuracy of tech specs. However, we believe that the lack of procedures presents the potential for a technical review branch not getting the tech specs for review, for review catinents not to be received by TSRG, for documentation not to reach the central files, and for tech specs to be changed without proper approval. We believe that such procedures will
y 11 provide controls needed to assure the integrity of the tech spec review process.
We recognize that under NRR's recent reorganization each division will coordinate and manage the technical specification process for each plant under its jurisdiction. We believe that the reorganization increases the need for such procedures to ensure that the quality of reviews and the documentation of reviews remains consistent among the divisions.
Reccanendations We recomend that the Director, NRR:
- 2. Take action to develop and establish uniform procedures for technical )
specification reviews within NRR, consistent with the particular issues discussed in this report.
- 3. Reiterate and enforce NRR's policy with respect to ensuring that copies of all pertinent documents are sent to central files.
MENCY N
. On April 28, 1986, the EDO provided coments on a draft of this report (See Appendix I). In his coments, the EDO disagreed with Recomendation 1 regarding the need to " Clarify for the NRR staff the policy concerning licensee review schedules, particularly as related to tech specs." The EDO stated that staff scheduling is presently in accordance with agency policies, and that the staff does attempt to revise schedules when appropriate.
Despite the EDO's coments, we continue to believe that the EDO or the Director, NRR, should clarify NRC's scheduling policy with regard to tech specs. Our report notes that the staff feels constrained to schedule strictly in accordance with NRC's official policy and that the Deputy Director, NRR, agreed that the scheduling policy with regard to tech specs may need to be clarified.
We have raised the issue of NRC's scheduling policies previously in our report entitled "OIA Review of the Operating License Review Process for Power Reactors." Since the Ccmnission at that time expressed the belief that the staff is in the best position to evaluate the adequacy of review schedules, we will defer to the EDO's judgement on this reccanendation and not elevate the matter for Comission consideration. We consider this recomendation closed.
The EDO generally agreed with recomendations 2 and 3 and advised that he
, ...will establish written procedures for the Tech Spec review process, and will issue them formally to the staff." OIA will review and evaluate the effectiveness of these procedures at the time of our follow-up review.
J 6
c -
APPENDIX I y
ff $
g UNITED STATES NUCLEAR REGULATORY COMMISSION
. j WASHING TON, D. C. 20555
%, ., J APP 2 8 G86 MEMORANDUM FOR: Sharon Connelly, Director Office of Inspector and Auditor .
FROM: Victor Stello, Jr. -
Executive Director for Operations
SUBJECT:
COMMENTS ON DRAFT OIA REPORT ON TECHNICAL SPECIFICATIONS REVIEW PROCESS By memorandum dated March 5,1986, you forwarded to me for comment a draft report on the results of OIA's review of the Technical Specification review process for near-term operating licenses. The draft report states that although significant improvements have been made to the tech spec review process, that one improvement, a milestone schedule, is not being implemented as intended. The report makes three recomendations to correct specific problems identified during the review and to enhance the overall effectiveness of tech spec reviews. Our coments on each of the recommendations are enclosed.
In sumary, I believe that the current policy on scheduling is understood by the staff and adhered to without resulting in inadequate review. I will have the staff issue procedures for tech spec review. I agree that the role of central files is very important and will ensure that the policy is understood and followed by the staff.
. w-
, .- V
,< _,-s,. .d(ss~~
Victor Stello, Jr.
Executive Director for Operations
Enclosure:
As stated 8
- ~ - -.-
.. . . - . _ . . _ - _ - ~ _ . g_ _--- . _ _
4 Enclosure -
COMMENTS ON OIA RECOPMENDATIONS '
l OIA Recomendation #1 l
Clarify for the NRR staff the policy concerning license review schedules, I
} particularly as related to tech specs.
Staff Coments With regard to setting milestone schedules, the draft report criticizes the staff for compressing the tech spec review schedules for River Bend, 4
Millstone 3, and Perry to meet the applicant's fuel load date. It is
! important to note that the NRC policy is to use the applicant's fuel i
load date for scheduling licensing activities including Technical Specification reviews, unless there is a six month discrepancy between ,
- the staff's and applicant's estimates. The staff scheduling was in accordance with this policy. I believe that the staff is aware of the NRC policy concerning review schedules. The quality of staff reviews i : is not compromised by implementing this policy, although adjustment to work assignments may have to be made from time-to-time. Also, during j
the last several months prior to licensing, the staff frequently re-assesses the progress of the facility and if the applicant's fuel load i
appears unrealistic, the applicant is asked to justify or revise the schedule and we adjust our schedules accordingly.
1 OIA Recomendation #2 '
Take action to develop and establish uniform procedures for Technical Specification reviews within NRR, consistent with the particular issues j discussed in this report.
t
{ Staff Coments This recomendation stems from several findings, the first of which is that in one instance a branch was inadvertently left off a distribution list while another branch was listed twice. This was obviously a clerical error which
! would not necessarily be eliminated by establishing review procedures. The
- j report recommends that a standard distribution list be established, when at the time in question such a list did in fact exist. The report also stated that tech spec reviewers do not use a formal checklist or tracking sheet for tracking review branch responses, and the informal list is not part of the permanent record.
However, since each technical branch written response which is included in the docket files, provides it does not a formal seem-
!. necessary to produce a fomal checklist for inclusion in the docket file.
The one instance in which a technical branch did not femalize their review by memorandum was apparently an oversight that could have occurred even if 1
the tracking list had been incorporated into the official docket files, i
l i
-. . -. . n. - .~ _ -
Enclosure
- The draft report suggests that the staff establish a policy of providing the final version of the tech specs to the technical staff for their review. In fact, the final version of the tech specs was routinely sent to the technical
. Division Directors for their certification prior to issu'ance of the tech specs I with the license. As with any final licensing document (for example, an SER), I the Division Director is held accountable for those sections prepared by his !
staff and in generel these documents are returned to the reviewers for their i recommendation regarding this certification. :
We do not believe that the basic process needs to be changed. However, we l will establish written procedures for the Tech Spec review process, and will' issue them formally to the staff.
OIA Recommendation #3 Reiterate and enforce NRR's policy with respect to ensuring that copies of all pertinent documents are sent to Central Files.
Staff Comments
- I agree that this is important and will have the Director of NRR issue a memorandum on this subject.
. l e
0 l
I
.a_ # ,..,,-ade ,. ,e-.-mm-.. .-. - , ,
p * ' .'-' ,, , ..e '4=w el- Q f?^'h hW. ; A yt.' y [.] ,f . Q%t J
.n~--
s, ~~d d. y':N9 ,.
df/ "'[,.?/E S-F18 M.m+ En%D,1
- Oi [' '
,, . ._ T.ayN, - ,.T.M. , , _,<.,i]...-.
v ' s
- j. M Q h.a ...,-
Ti 's .
,W h. . .a.
G-sfy ,
, _ , : {k ;
- ' bky h?
n m i,+.-" -q -J -
q
- ' ',4 y , ,tggg,
-i
. ..y y (
, nnn s %-...:
- - - -i /gtl/ r
. @.. g., ., . g m
_ pFgg. g_ y 4
A 4,. ;&a$'
%. .W' ;'
e**. ' . . , .
%. .i "I. 4, $. :g_ .(;,f [Q'- .T- ' [ ,*L.p+rq e '*
'.' g n y,g;r
.p y .,
- u,..y- : -
"n- ,
+ <
. n. -a. + ms-
, m. e- + ---
+
- w ;.,.r.. ',', . ;. p[f . < e -IM, ;,s3>;h m' X 54Tfg*.Mq rg< , ;; _ '
.,.-4>* 1 z-
+..+,~ ,
v ,, .7,~ : ~'
- O 1 _ - - , s- [) ., . w _+ t *
. , - ~
_w;w# ,
, , _m g< I.,? m#uy ..
; h ' 3 ..'p -
4 5 .a 5
' 't l.h"is5 Thia'i, b _l[y) N y., m ?vk)h$4f. h., .
, % + 1
~
Y vj_ . y.3lh.I 4
.,..l
. ,h,,' h '
.. ;sq.v.? f -
.&,a: we
+ . e 4 h, #_-
li': ' -
,'.y -
5' ).20 ' "'
'p _y,,yf - ~' -
?./ t iN,,-
f.
3 ..t -
.-s,6 h fr ? w'
, . .. , - . s k-J.4A+=
m
,<y. . gm
+.c e, a. v. s .
r 'e ,.'
- t . d%7& ~lk.
. .-v., . ag . %.: r. ...
. L. ; .r,. -'y:;Rev ;,]u
. , , . , . -g,
+
l.*
- ..% t~ Y" 4: , .hk 3 +4+
. . , . . W-
?
,y
-* pch- .
.s,&..,
1
- - ga r,* --. . .
s.
v
+p
'**(i-0
,h*'.
^
' 1/
,i, 4. g. .
tb.
i e i
=(1'.
V I [% '5'
(
' [ s l iN'lf's
-adet & h, ,
4 -l 1
4 i
g .
m A Amey. e. .
r-- 9 e
s'
. w ., A / i-4 e
t
, sb N'S m,. . --
g . ieL-r k_____ _ _ _ _ _ . - _ . _ _ _ . _ . _ - _ _ _ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ - . _ _ . _ _ _ . . _ _ _ . _ . _ _ - _ . _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ .