ML20147G632

Forwards Matl Re NRC Statement on Risk Assessment & the Reactor Safety Study Rept(Wash 1400)in Light of Risk Assessment Review Grp Rept

ML20147G632
Person / Time
Issue date:	12/12/1978
From:	Gossick L NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
To:
References
RTR-WASH-1400 SECY-78-637, NUDOCS 7812260274
Download: ML20147G632 (53)

Text

,

he 4?

UNITED STATES December 12, 1978 NUCLEAR REGULATORY coMMISSloN SECY-78-637 INFORMATION REPORI Rdqtrc, pyy FOR: The Comissioners IcROM: Lee V. Gossick Executive Director for Operations

SUBJECT:

PROPOSED NRC STATEMENT ON RISK ASSESSMENT AND THE REACTOR SAFETY STUDY REPORT (WASH-1400) IN LIGHT OF THE RISK ASSESSMENT REVIEW GROUP REPORT DISCUSSION:

For your use in the upcoming Commission meeting on this subject, I am enclosing material forwarded to me from NRR. Enclosure 1 is a niemorandum containing a simple chronology of the history of the RSS and a revised draft of the statement on Commission views and actions in response to the Risk Assessment Review Group's report.

The redraft represents principally.the efforts of NRR who had discussions with SD, NMSS, IE, RES, as well as with the ED0 staff and OPE.

As pointed out in Mr. De'nton's memorandum, this version was prepared by the staff with the intent that NRC should (a) respond in a positive and constructive wa Assessment Review Group;avoid (b)y to the recommendations overreaction of the Risk to isolated statements in the Risk Assessment Review Group report; and (c) present a clear, balanced view. )

Mr. Denton also indicates that the following two points are not explicitly dealt with in the statement. First, there has been a range of views on the RSS among the offices. The proposed. statement .

deals with the issues raised in the Review Group report but does not attempt to characterize the range of individual views on the RSS and its use. Second, the previous draft called for a plan for j systematically guiding future applications of the RSS methodology. I While several groups have called for such a plan, the staff concluded I that this need not be part of a statement on the RSS.

SECY NOTE: This paper is currently scheduled for a Commission briefing on Thursday, December 14, 1978.

7812260M S

s 4' Enclosure 2 is a report, prepared by NRR, on the staff's survey of the use of.the RSS; this report will aid in your review of the

~

proposed statement.

A separate memorandum is being prepared on the FY 79 and FY 80 resource impTications for implementing the Risk Assessment Review Group report recommendations. That paper will complement this one and should be available to you shortly.

NRR, SD, RES and TA/EDO have concurred in the revised policy statement. The MPA staff and IE are preparing comments _to be forwarded prior to the meeting. NMSS notes that, because of limited impact on its projects, they have no objections to this policy statement.

'9)k y W/

v Lee V. Gossick Executive Director for Operations

Enclosures:

1)DraftStatementonRisk Assessment / Lewis Committee Report

2) Summary of Results of NRR Survey of the Use of WASH-1400 in the ~

Licensing Process DISTRIBUTION Commissioners Commission Staff Offices Exec Dir for Operations Regional Offices ACRS Secretariat

^t ( -

k. c-

. g ,, . ..,l"'

. m.. i r

.- i 1

, l '.

k I

n

\

s e l l

r 4

.1 1

1

. ENCLOSURE 1 .

P l

l i

I

.y

'..).

0 L

h*' _ . , . _ t

,y .

UNITED STATES 4,O*

[g 4 NUCLEAR REGULATORY COMMISSION jo.,,,!

I # j t

WASHINGTON, D. C. 20555 w

DEC08518 MEMORANDUM FOR: Lee V. Gossick Executive Director for Operations l FROM: Harold R. Denton, Director Office of Nuclear Reactor Regulation

SUBJECT:

NRC STATEMENT ON RISK ASSESSMENT AND THE REACTOR SAFETY STUDY REPORT (WASH-1400) IN LIGHT OF THE RISK ASSESSMENT REVIEW GROUP REPORT Enclosed is a revised draft of the statement on Commission views and actions in response to the Risk Assessment Review Group's report. NRR has had discussions with SD, NMSS, IE, RES, as well as with the ED0 staff and OPE.

This version was prepared with the intent that the NRC should respond in a positive and constructive way to the recommendations of the Risk Assessment Review Group. We have bten careful to avoid overreaction to isolated statements in the Risk Assessment Review Group report and have tried to present a clear, balanced view. The proposed statement is concurred in by NRR, SD, RES, and NMSS.

There are two points which are not explicitly. dealt with in the state-ment. First, you should realize that there has been a range of views on the RSS among the Offices. Some staff members have continued to be very critical of the RSS while others have found the benefits of the RSS to greatly outweigh its technical flaws. These views have no. doubt shaped the various uses (or lack of use) of the RSS. More importantly, they have shaped the various views on the significance of the Review Group report to the licensing process. The proposed statement deals with the issues raised in the Review Group report but does not attempt to characterize the range of individual views on the RSS and its use.

A draft report on the staff's survey of the use of the RSS is being transmitted separately and will aid in your review of the proposed sta tement. A simpie chronology of the history of the ~RSS is also en-closed to assist in that review.

Second, the previous draft called for a p1an for systematically guiding future applications of the RSS methodology. While several groups have called for such a plan, it was concluded that this need ENCLOSURE 1

.?

'f, Lee V. Gossick DE0 0 81978 not be part of a statement on the RSS. The revised draft encourages" the use of the RSS methodology and notes that additional guidel.ines will be developed but does not call for a specific action plan. -l While it was concluded that the above items should not be included in the NRC statement, we will be prepared to discuss them at the up- ,

coming Commission meeting. '

Harold R. Denton, Director ,

Office of Nuclear Reactor Regulation (

1

Enclosures:

1. Chronology on the RSS

2. Draft Statement on Risk Assessment /

Lewi s Committee Report l

1 l

l l

, iu,,

DEC 0 81978 Enclosure 1 CHRONOLOGY

1. December 4, 1970 Letter, Commissioner Larson to Senator Gravel, noted that AEC plans to hav,e a stiudy made and

- report prepared to cover area covered by ->

j WASH-740.

• 2. March 26, 1971 SECY R-199 provided possible approaches on

- study of nuclear-risks and benefits. .

3. June 2, 1971 Regulatory Information Meeting No. 482, proposed '

study of nuclear risks and benefits requested by <

Commission. Relationship of proposed study to Price-Anderson Act emphasized.

i d

4. April 5, 1972 SECY R-432 Policy Paper on MIT proposal for study  ;

.4 risks due to accidents in nuclear power i reactors (intended to be separate from but coordi- .;

nated with a companion study in RDT of all other risks / benefits of electrical energy process).

5. August 4, 1972 Reactor Safety Study (RSS) initiated.

6. August 20, 1974 DraftRSS(WASH-1400)distributedforcomments.

7. August 23, 1974 AEC Interim General Statement of Policy released, 39FR30964 (cf. SECY-R-75-62) l

8. September 1974 AEC Regulatory staff review group formed to review draft RSS. .

9. December 3, 1974 Initial report of Regulatory staff review of draft RSS (cf. SECY-R-75-133).

10. January 19, 1975 NRC formed, RSS study group assigned to RES.

Responsibility for RSS effort transferred from .

AEC to NRC.

11. March 4, 1975 Comission briefing on WASH-1400 draft and comments (cf. SECY 75-51)

12. April 28, 1975 Report to the American Physical Society by the Study Group on Light Water Reactor Safety.

13. May 30, 1975 Detailed Regulatory staff coments on draft RSS.

_m__ m ,_ _ _ w, - . . m..

14. October 30, 1975 Final RSS (WASH-1400) issued (Press release No.75-259 included Chairman Anders' statement -

on RSS).

15. November 15, 1975 Preliminary Review of WASH-1400 (Final) by NRC staff / consultant review group.

16. June l'1, 1976 Udall Subcommittee oversight hearing on -

continuing criticism of RSS.

17. February 7, 1977 Udall Subcommittee observations on the RSS. .

18. March 2, 1977 Memo Gossick to Commissioner Kennedy regarding '

- staff position on application of the RSS to the licensing process. .

19. March 14, 1977 Udall requests formation of a review group to prepare new Executive Summary for RSS.

20. April 4, 1977 Chairman Rowden commits to form review group with .

different charter.

21. May 13, 1977 Memo Chilk to Gossick on Cotmission's desires for use of risk assessment methods in licensing practices. y .

a

22. July 1, 1977 Risk Assessment Review Group (Lewis Committee) formed. <

t

23. September 7, 1978 Risk Assessment Review Group Report (NUREG/CR-0400) published.

G

-- ______-_a _ _ _ ___

i ,4

. DEC 0 81978 flRC STATElENT. ON RISK ASSESSi1ENT

~'

AND THE REACTOR SAFETY STUDY

~~

Protection of the public health and safety is a paramount objective of the Nuclear Regulatory Commission in regulating the design, construction -

and operation of nuclear power plants. The operation of nuclear power ,

plants can never be conDietely risk-free. The safety objective of the .

NRC has always been to assure that the risk from normal operation and ac-cidents is maintained at an acceptably low level and to assure that the likelihood of accidents more severe than those considered in the design is extremely small. '

).*

This safety objective has not been set forth in numerical terms in the Commission's regulations, largely because ij his not been possible to g

49' make quantitative estimates of the risks to the public from nuclear power plants with sufficient precision to be us,eful as licensing criteria.

The Reactor Safety Study (RSS) was an attempt to use the em.erging disci-plines of risk analysis and cecision analysis to develop quantitative estimates of the risks of nuclear reactor accidents. At the time the study was initiated, it was recognized that the study might not be suc-cessful in reaching this ggal and that further research and development might be necessary before quantitative estimates of risk could be de-

~

veloped with sufficient precision to be useful for licensing purposes.

I

, DEC08 E ,O,

-2 -

In August 1974, the report on the Reactor Safety Study was is. sued in draft fonn for public comment. In commenting on that draft, the Atomic Energy Commission stated that the study, when completed, would be the subject of -

~ ~

a thorough evaluation "...with respect to both t5e 5asic question whe'ther the risks portrayed by the study are acceptab'le from the standpoint of the ,,

Commission's statutory responsibility to protect the health and safety of ,

i the public, and the related question whether any changes in the Commission's ~ _

safety or environmental regulations are warranted."1I In thi s statement, the AEC also set forth an interim position that "...the contents of the draft study are not an appropriate basis for licensing decisions."  ?

Many individuals and groups reviewed the draft RSS arid provided comments. '-

In addition to the request for public corine,nts, the AEC requested members of the regulatory staff not previously assignep to the study to perfonn a f

review of the draft study. The comments of 'the ~AEC regulatory staff and others pointed out deficiencies in the RSS.

In October 1975, the final report of the Reactor Safety Study was issued (WASH-1400/NUREG 74-014 ). Although the final report included responses to comments received on the draft report, critics of the final report maintained that the study still contained serious flaws and pointed out that many adverse comments on the draft were not suitably taken into ac-count. In particular, comtents were made that the Executive Summary of -

1/ 39 FR 30964, " Interim General Scatement of Policy," August 23, 1974.

gjg78 3 -

the Reactor Safety Study was presented in a manner which created a mis-leading impression of the certainty and comprehensiveness of the study's conclusions.

These concerns were considered in connection with a Congressional review _

of the study. Following publication of the results of this review and an -

exchange of correspondence between the HRC and Congressman florris R. Udall, .

the i;RC formed an independent group to further consider this matter. The .

Commission established the following charter for the Review Group:

"The Review Group will provide advice and infomation to the Commission regarding the final report of the Reactor Safety Study, WASH-1400, and the peer comments on the .

' Study, advice and recommendations on developments in the field of risk assessment methoaology a'nd on future courses of action which should be taken to improve this methodology and its application. This advice ahd infonnation will as-sist the Commission in establishing pblicy regarding the use of risk cssessment in the regulatory proving the base for the use of such a,sse' process,Itin im-s sments.

will also clarify the achievements and' limitations of the Reac tor Safety Study."

The Review Group presented its findings and recommendations to the Commis-sion on September 7,1978 and its report was issued as an NRC document NUR EG/CR-0400. In general, the Commission agrees with the findings and recommendations of the Review Group, The Review Group endorsed the methodology of the Reactor Safety Study and greater use of risk assessnient in the regulatory process. However, they also agreed with the critical views of others that the final report on the .

DECTT Ur-

'. r,

- 4 -

Reactor Safety Study suffers from a number of technical deficiencies and that the final report did not respond adequately to ths peer comments on the draft. In view of the Review Group's comments, the Commission has re-cently requested a rev.iew of past Commission and staff uses of the RSS.'

The review has shown that there has been limited use of such assessments by the 14RC staff in the past. There is a lack of explicit requireme'nts or guidelines for the use of probabilistic assessments, and particularly ,

probabilistic risk assessments similar to those of the Reactor Safety Study. Howe /er, as analytical techniques for carrying out such assess-nents have become more v;cepted and as more operating experience, with ,,

its associated data, has become available, there has b'een increased use of probabilistic assessments. The Reacto Safety Study contributed much

~

to familiarizing the technical staff with the use of these assessments

\

and made considerable advances in risk assessment methodology.

N Probabilistic techniques were in use by the staff before publication of the RSS as support for some of the judgments reached in the reactor 11-censing process. The.ir use has ranged from the simple semi-quantitative arguments in support of a judgment that plant operation can continue for a limited time pending implementation of a needed modification, to quanti-tative fault tree / event tree analyses to help detennine the need for and nature of additional licensing requirements (e.g., as in consideration of The primary .

protection against anticipated transients without scram).

result of the staff's application of the RSS in 9e licensing process has m=c:= _ .

1-+ + w -== +is = e., s

s ' *. . .'

CEC ^ T 1376

~

-5 -

been the imposition of additional requirements for protection against l

those accidents identified in the RSS as major contributors to risk, rather than reductions i,n existing requirements.

Since the issuance of the final RSS report the Commission has also viewed application of the RSS methods as a valuable supplement to present li-censing practice, and has encouraged the staff to use those methods *in this fashion in making recommendations on important safety issues before the ,

Commi ssion. As an example, emphasis has been given to the use of risk as-sessment in providing a means for more rigorous characterization of cur-rent value-impact analyses.

The Commission believes that uses of the RSS ,such as described above are consistent with the statement of the Review Group that the fault tree / event tree methodology used in the RSS "should be gmo'ng 'the principal means used to deal with generic safety issues, to fomulate new regulatory require-ments, to assess and revalidate existing regulatory requirements, and to evaluate new designs." The Commission concludes that the staff should con-tinue to utilize fault tree and event tree analyses to aid in reviews of various safety issues. The Commission also will continue to support im-provements and extensions of risk assessment theory, methods, data devel-opment and statistical analyses to promote their prg:r and effective use by the NRC staff.

However, the record of past uses of the RSS by the staff and Commission is not entirely favorable. The Review Group noted that there have been

,,i, DEC 0 81978

- 6 -

instances when the RSS was misused as a vehicle to judge the acceptability of reactor risks. The Commission's review confirms this finding. Some Commission and staff statements have cited results from the RSS, particu-larly those in the Executive Summary, without identifying the uncertainties associated with the results as presented in the RSS itself. Moreover, as .

reflected in the report of the Review Group, it is now generally agreed that the range of uncertainty associated with the RSS risk estimates is far greater than that presented in the RSS.

The Commission believes that while the Reactor Safety Study concludes that reactor accident risks are very low, the range of uncertainty in these estimates does not permit an unqualified can'clusion that they are clearly lower than other natural or man-caused rdtk,s." The existing assessments of risk are sufficiently uncertain that it is bet}er to characterize the RSS

.s z

results in terms of a range of estimates and make any comparisons of risk based on that range. Such use would provide an added perspective on ac-cident risks not otherwise available.

In sum, the results of the Commission's review of past uses of the RSS indicate that there has been limited application of and no undue reliance on the RSS in licensing decisions. There have been a number of statements by the Commission and the staff whose language was not properly qualified.

To some extent such statements may have created a misleading imprestion of the comprehensiveness and certainty of the Reactor Safety Study results. .

The Commission has instructed the staff to develop whatever guidelines

c. ,

M;ee1978

. 7 .

may be determined necessary to ninimize the potential for uncri.tical use of the RSS in the future.

In addition to the general observations discussed above, the Review Group made a number of specific findings and recommendations dealing with var-ious parts of the Reactor Safety Study and the use of risk assessment methodol ogy. The Commission generally agrees with these findings 'and -

recommendations and will take them into account in its future actions. .

In addition, the Commission has determined that the following specific actions are appropriate: ,-

~

(1) Copies of the Risk Assessment Review, Group Report (NUREG/CR-0400) and of this statement will be sent to all known recipients of the

}

RSS and the Executive Summary. CopYes of the RSS, including the Executive Summary bound with the main r'eport, will be accompanied by copies of the Review Group's report and this statement.

(2) The Commission will review the staff's current practices and pro-cedures for peer review of significant staff reports to identify any needed improvements. ,

(3) Major NRC reports which nay have the potential to affect policy, such as the RSS, and which receive extensive criticism or require substantive changes when issued for comments, should be recircu-lated for a second round review and comment before being issued -

as a final report.

1

g;  ; 1978

- 8 -

(4) The Commission is giving support to ongoing work aimed' at up-grading the use of risk assessment methodology in the regulatory process. Follow-on studies to the RSS will address the many technical issues and deficiencies raised by the Review Group.

Reports on these studies will indicate the full range of un-certainties associated with the studies and acknowledge existing ,

criticism and these uncertainties and criticisms will also be dis- ,

cussed in any Executive Summaries.

\

(5) sNstaff's envirornental statements on LWR applications contain a disctission of accident risks and include reference to the RSS. _

All new or revised LWR enviromental statements will, in their

. y -

discus,sion of the RSS, also discuss"the 1 imitations or range of

/

uncer}tainty associated with the RSS res<ults.

1 The RSS should be seen as one step, already accomplished, in the contin-uing development of risk assessment metho'dology. While the RSS has de-veloped risk assessment methods of considerable utility to the regula-tory process, care will have to be taken in any use of these quantita-tive risk techniques. To keep the Congress and public infomed, the Commission will publish the results of the efforts described above, in-cluding the programs to prcmote and insure more effective and careful use of the RSS methodology.

m. -um um-s-

{

I, - e* 7 1

2

,f 1

+

m, e

S f

ENCLOSURE 2 m)'

1.

• J

'\_' 4

,# UNITED STATES 4r,& NUCLEAR REGULATORY COMMISSION 3i l 3 .,I y$ WASHINGTON, D, C. 20555 1

% . .w. ./ ./

• December 11, 1978 MEMORANDUM FOR: Lee V. Gossick Executive Director for Operations FROM: Harold R. Denton, Director Office of Nuclear Reactor Regulation

SUBJECT:

REVIEW 0F REGULATORY ACTIONS AND STAFF POSITIONS WHICH RELY ON WASH-1400 As you requested on October 27, 1978, we have surveyed the NRR staff to identify uses of WASH-1400 in the licensing process. We also re-ceived and categorized the responses of other Offices. The results of the survey are summarized in Enclosure 1. A synopsis of each of the issues identified by the survey, along with a recommendation for further action, is included in Enclosure 2. Copies of the documents identified by the staff are provided by Enclosure 3.*

To summarize, the staff identified many instances where the Reactor Safety Study was mentioned or discussed, but only a few where the

.RSS played a substantive role in the licensing process. The re-

~

sponses indicate that the use of the Reactor Safety Study has been increasing since the issuance of the final report on the RSS. This is consistent with the guidance from the Commission (cf., memorandum from S. J. Chilk to L. V. Gossick dated May 13, 1977).

While it is difficult to assure that the survey has identified all documents in which the RSS has been used, we believe it has re-vealed all substantive licensing actions where the RSS played a major role. We were also provided by Commission Offices copies of Congressional correspondence and prior Commission statements re-garding the RSS (memos C. C. Kammerer to S. J. Chilk of October 31, 1978 and K. S. Pedersen to Commissioners of October 11,1978). As discussed in Enclosure 1, these have also been considered in assessing the results of the staff's survey.

Of all of the material identified, only three were determined to re-quire reconsideration in view of the Risk Assessment Review Group recommendations. They are summarized below.

• Enclosure 3 available in the Office of the Secretary.

- - - - - - - - - -- ENCLOSURE 2

Lee V. Gossick -

2 -

December 11, 1978

1. In reviewing the Clinch River application, the staff used the RSS analyses of the time to containment failure for various core melt sequences as an aid in determining what licensing requirements would assure comparability of residual (Class 9) risks between the CRBR and LWRs generally. If the Clinch River review is reactivated or another similar review is re-quested, this licensing position should be reconsidered.

2. In the report on ATWS, the NRR staff used the RSS estimates of the overall probability of core melt as a benchmark in recommending a quantitative safety objective for ATWS. The staff is reconsidering the degree of reliance on the RSS in light of the Review Group report and expects that the forth-coming supplement to NUREG-0460 will take an approach which is consistent with the Review Group's recommendations.

3. In addressing the concerns of an ACRS consultant relating to d.c. power supply reliability, the staff utilized WASH-1400 to confirm the staff's conclusion that adequate protection of the public health and safety had been provided, and that the evaluation of this generic issue was proceeding at a reasonable pace. The use of WASH-1400 in the staff evalua-tion of this issue is being reconsidered as a part of the resolution of Task Action Plan A-30 dealing with the ade-quacy of d.c. power supplies.

The perception of the majority of the staff is that there has been lim-ited use of the RSS in the licensing and regulatory process. However, some of NRC's correspondence and analyses have not clearly set forth the degree of reliance on the results of WASH-1400 relative to a given topic, and most correspondence on the subject does not properly qualify the uncertainties associated with the RSS results. This raises a ques-tion of the extent to which the RSS results may have been used to im-properly allay concerns about a specific technical issue or otherwise contribute to an imperfect decision-making process. Some have argued (cf., memorandum from D. L. Basdekas to S. J. Chilk of November 28,1978) that staff reliance on the results of WASH-1400 has contributed to faultf regulatory decisions and faulty representations to the Congress regarding the significance of certain safety issues.

The extent to which the RSS has colored the staff's views on various safety issues is a matter of subjective judgment, which cannot be clearly

Lee V. Gossick -

3 -

December 11, 1978 determined from the record. However, we view the record as a whole as showing a cautious and prudent application of the RSS by the staff. Its principal application has been to supplement or con-firm the main stream of analyses and judgments reached by the staff.

4t h l

Harold R. Denton, Director Office of Nuclear Reactor Regulation

Enclosures:

As Stated cc: S. Levine, RES R. Minogue, SD W. Dircks, NMSS J. Davis, IE N. Haller, MPA K. Pedersen, PE

ESCLOSURE 1 SU!UMRY OF OVERALL RESULTS OF NRC SURVEYS REGARDING USE .

OF WASH-1400 IN THE LICENSING PROCESS On October 27, 1978, the Executive Director for Operations requested the _

major program offices to conduct a survey of their staff to identify uses of WASH-1400 in the licensing process. It was al.so requested that. .,"

- the type of use be described and five broad categories were set forth. ,

These categories ranged from use of numerical risk estimates as given b in the RSS to make a specific licensing decision (Category 1) to use of the RSS methodology without relying on the specific numerical esti-

~

mates in the RSS (Category 5). In addition to the five categories _.

defined there, a sixth category has been added to the list. It con-tains those actions which did not propehly' fall into Categories 1 -

through 5. NRR was requested to coordinate the.i responses of the survey z

conducted by the other NRC offices.

In addition to NRR, six other NRC offices (MPA, IE, NMSS, SP, RES, and SD) provided responses to the survey request. A matrix of the number and categorization of issues identified by the responding groups is at-tached. Of the actions identified by the staff, only two were cate-gorized as Category 1 with the rest falling into the remaining categories.

As discussed in the cover memorandum, only three issues were determined to require reconsideration. Thus, it is evident that there were only a few instances in which WASH-1400 was a principal basis relied upon to -

- 2 ..

make licensing'. decisions. The preponderance of the actions identified were those instances in which the staff use of WASH-1400 absolute ac-cident risks was restricted to relative comparisons of risks, or the estimates of WASH-1400 were used to illustrate or confirm staff con-clusions on the-disposition of an issue, or in which the' methodology or values of WASH-1400 were independently used or modified to reflect .

new information. .

To some extent, it can be argued that the RSS has shaped or influenced ,

the direction of licensing actions and any reference to the RSS by the m

NRC implies a use of the RSS. There is a considerable body of corres-

..M pondence and staff and Commission speeches regarding the RSS and its role j in the licensing process (which we woulp place in the "Other" or Category 6 grouping). However, in most instances, the use of the RSS has been to

' s buttress and add perspective to th'e normat staff review process.

Listed below are the descriptions of the various categories and the types of applications of the RSS that were identified in this survey.

4 l

i l

\ .

, N CATEGORY 1 De fi ni t ion includes those actions in .thich an absolute value of accident risk as set forth in WASH-1400 was relied upon the licensing process to make a speci fic licensing decision. Included in this category would be any .

reliance on an overall probability for core melting or on the proba-bility of a given event sequence leadi'ng to core melt. A p'ossible example is the use of the RSS to develop quantitative estimates of ,

health risk from the coal and nuclear fuel cycles.

Example .

The two items identified in this category include the example in the _

definition and the use of the numerical estimates of core melt prcbability f rom WASH-1400 to derive proposed safety objectives for ATWS. In both of these instances, either the final report or the planned supplement will include use of WASH-1400 in a manner consistent with the Review Group recommendations.

CATEGORY 2 Definition }

Includes those actions in which the absolute vqlues of accident risks of WASH-1400 were used in the licensing process,, but'when such use was re-stricted to relative comparisons of risks.

Included in this category would be any reliance on the overall probability of core melting of the RSS to draw comparjsons between two design concepts.

Possible examples are the use of the RSS to compare an FNP to a land-based plant and the use of the RSS to develop perspectives on overall ATWS risks.

Example There were 9 items in this category. Typically, items in this category utilize the numerical risk estimates of the RSS (such as a core melt probability of 5 x 10-6 per reactor year) but only in a relative sense.

These assessments did not require that the values used be precise since they were used to compare the relative differences between two or more alternatives or concepts.

m

CATEGORY 3 De fi ni t ion includes those actions in which the quantitative estimates of fault tree /

event tree analyses of WASH-1400 were used in the licensing process to f ilustrate or confi rm staf f conclusions on the disposi tion of a potential ~

safety issue or to aid in selecting the preferred of several alternate regulatory requi rements. One possible example is the NUREG-0138, " Treat-ment of Non-Safety Grade Equipment in Postulated Steam Line Break Eval- ~

uations." .

Example ,

i l Approximately 88 Identified issues fit into this category. For these ~

items, WASH-1400 was used to further support or buttress a staff conclusion.

WASH-1400 was not the principal basis for the staf f action. Rather, the quantitative estimates or the analytical techniques aided the staff in reaching a conclusion. Some of the items contained in NUREG-0138 and NUREG-0153 utilized information from WASH-1400 to help respond to the concerns raised by some individuals that the priority or progress of re- ~

solut f or, of certain issues was not proceedir}g satisfactorily.

Since the values or techniques were only used in a support Ive role or to help select a preferred of several althrpatives, all but two do not require any reconsideration. One of those two, CRBRP design criteria will be reconslJered if the review is reactivaged. The other (reliability of d.c. power suopliesT is being reconsidere,d as-b pa'rt of generic issue A-30.

CATEGORY 4 pe fi ni t ion includes those actions in which values of WASH-1400 were modi fled by the staff to reflect di fferent data base or experience and were then used in the licensing process.

A possible example is the adjustment of the RSS estimates of scram unre-1iabi1Ity in NUREG-0460.

_E_x amp l e There are 12 I tems include *d in this category. Typically, the issues identified used WASH-1400 data as modified or supplemented by the staff to reflect added experience or a different data base before using the more complete ,

information in the licensing process. For example, WASH-1400 data on pipe ruptures was considered along with data obtained by the staff during its review of water hammer events at operating plants.

While the additional failure rate information gathered from operations provided a more complete data base, the decision to proceed with water hammer as a generic issue was based principally on other considerations.

3 CATEGORY 5 Definition includes those actions in which the event t ree/ faul t tree methodology of

'! ASH-1400 were used in the licensing process, but no reliance was made on the specific numerical estimates of WASH-1400.

Exanole .

There were 47 items identified in this category. The item's in this '

category used the evaluation techniques of WASH-1400. An example of this ,

use is in the evaluation of vendor proposed ccmputer pro.tection systems. . ..

In thses reviews, the staff performed preliminary reliability assessments using WASH-1400 methodology. These results aided the staff in their .

de l i be ra t i ons . -

Category 6 .

Definition _

This category was added after the responses were received. Issues were placed in this category when they could not b,e considered to fit into any other categories, included here are (ngtances when the staff considered using WASH-1400 in the licensing process but dismissed it and staff reviews of WASH-1400 informatipn used by other agencies in their evaluations. Only 8 . i tems were 14clude'd in this category ~

and any use of WASH-1400 could not be considered to have either signi-ficant or direct impact on the licensing process.

O e

NUMBER OF ISSUES IDENTIFIED

• i i i i i GROUP CA1. 1 li CAT. 2 li CAT. 3 il CAT. 4 ,l CAT. 5 l ,

CAT. 6 I i i i 1 i i i i i, i , , i

. . . t i i

~

AIG 0 I, O l 3 l 0 l 0 l 0

, . . i t i f I i e DOR 0 l 0 l 9 l 1 l.

1 1 l i

1 - --

, , ,i , , .

DPM 1 l 0 l 4 l 1 l 6 l 2 -

, , t , , .

1 I t a i OSE 1 4 l 15 l 1 -l 2 l 1 -

i i

1 i

1 1

i 1

Dss 2 l 6 l 9 l 4 l 16 l 0 i i , , ,

i i t i 1 4 1 I I t

____________. .__________..________.;_________4.________4.________a_________

1 I i 1 I l l l l l -.=

NMss 0 l 0 l 1 ,l 0 _li 0 l 1 I 1 i 1 1 1 0 1 i IE O l 0 l 0  ! 0 l 0 l

i

. 3) a li i

i i

i RES 0 l 16 l 14 l 4 l 37 I O i , i . .

i I ii g 3 i so 0 l 1 l i

3 \li 1 l 0 l i

'O 4 1 1 1 1 MPA 0 l 0 l 1 l 0 l 1 l 3 i i i i i i 1 i t i SP 0 l t

0 I i

l- l i

0 l i

0 i

l 0 i 1 1 i 1 1 i i i 1 e i i i e

...___......,..______.J,._________'________..l________.'_..______

i . , i t i t I i TOTAL 4 l 27 l 63 l 12 l 63 l 8 i i e i i e i e i i I i I t 1

• Note that the total issues identified above is larger than the number 1

in which brief synopsds are provided. This occurred because more than one group reported the same issue and some issues were recategorized ,

to more accurately reflect the type of use of the information.

3 )

ENCLOSURE 2 SYN 0PSIS OF ISSUES

I ..

Synopsis of Category __1 Issues

1. Synopsis: Using the results of WASH-1400, regarding the probability of core melt, the staff reconnended in NUREG-0460, that the safety objective for ATWS events be changed froIn 10~ FRY to 10~0/RY. The staff further recommended that systems to be used to mitigate ATWS events be safety grade or that they could be shown to be reliable using RSS estimates or an updated data base. Other portions of the ATWS study i

where WASH-1400 is addressed fall into Categories 2, 3 and 4.

l l

~

We recommend that these actions be reconsidered and the staff is reconsidering the degree of reliance on the RSS in light of the Review Group report. The forthcoming supplement to NUREG-0460 will take an approach which is consistent with the Review Group's recommendations.

1-1

il

,\

2. Synposis: Health Effects Attributable to Coal and Nuclear Fuel Cycle Alternatives, Draft N'UREG-0332 includes references to WASH-1400 data. Somatic health effects have been con-sidered in numerous forms including hearings and impact statements. Although the format of the documents involved '

has varied slightly, the method of incorporating WASH-1400 has been the same as in NUREG-0332 (draft). No reconsideration of previous licensing actions appears necessary. The final version of NUREG-0332 should include a range of mortality .

values for the uranium fuel cycle that includes a con- -

sideration of a broader range of accident risk estimates.

l-2

g .;

l l

O Synopsis of Category 2 Issues

1. Synopsis: The Safety Evaluation Report for Offshore Power Systems Floating Nuclear Power Plants 1 through 8, NUREG-0054, l 1

issued October 8,1976, referred in its Appendix C to the results of the WASH-1400 study. The WASH-1400 data were used in a comparative sense, and no firm reliance appears to have been placed on the data.

2. Synopsis: The " Estimation of Safeguard-Related Risk Associated with Continued Operation of Existing SNM Processing Facilities" by J. H. Conran in late 1976 and other related earlier documents, compared safeguards-related risk to safety-related risk-(as given in WASH-1400), in an attempt to show that NRC safeguards approach should be more conser-va ti ve'.

2-1

- . ,. i.-

1 l

, )

3. Synposis: Liquid Pathway Generic Study, NUREG-0440, February 1978 and Offshore Power Systems, DES, Part III, NUREG-0127 (Revision 1) uses WASH-1400 methods and numerical values to compare risks of a floating nuclear plant to land-based l plants.

l I

4 l 4. Synopsis: Letter to G. Paulson, Ass'istant Commissioner for Science,

' Department of Envir' omental Protection, State of New Jersey, l 4 and minutes of a meeting in New Jersey on fiarch 21, 1977,

! re: Liquid pathway Study uses WASH-1400 values to compare risks a floating nuclear plant to land-based plants.

5. Synopsis: Commissioner Action Paper, SECY 78-137, March 7, 1978, Assessments of Relative Differences in Class 9 Accident-Risks provides an evaluation of alternatives to sites with high population densities. WASH-1400 consequence l.

2-2

models were used to perform analyses of the differences between the Perryman site and.other alternative sites from the standpoint of accident risks.

o

6. Synopsis: The letter to W. D. Rowe (EPA) dated November 18, 1976, re: nuclear accident risks states that the Reactor Safety Study indicates that the approach to safety as set forth in the Commission's regulations has been successful and that the safety and environmental risks from accidents are lower than the risks from most other natural and man-caused events. This language is patterned after the 1974 Interim General Statement of Policy.

2-3

7. Synopsis: Letter from S. Levine to G. Paulson, New Jersey Depart-ment of Environmental Protection dated November 9,1976, regarding an investigation of the probability of hyop-tehtical catastrophic accidents in the Oyster Creek The use of certain results in the Nuclear Power Plant.

i Reactor Safety Study by the author of the Oyster Creek study is questioned in this letter. The critique includes a discussion of how the results in the Reactor Safety Study were generated. In addition, the extrapolation of failure probabilities over a 30-year time period is

discussed and compared to the 5-year time period extra-polation in the Reactor Safety Study.

j 8. Synopsis: Memo from Buhl to Vollmer dated June 6,1978, provides

^

comments on GSA's DES regarding disposal of Charlestown site. WASH-1400 material used in the DES was discussed  ;

1 and risks described in the DSS were evaluated in the i context comparison of overall risk.

),

m i

4 l 2-4

9- Synopsis: A June 20,1977, letter from S. Levine to G. Paulton, New Jersey Department of Environmental Protection, trans- l mitted our comments on a draft report titled, "An Investigation of Probability of S=,,aious Accidents in the Oyster Creek Nuclear Power Plant". The report used failure probabilities from WASH-1400. -

l 1

I l

I 2-5 l

0 4

1 Synopsis of Category 3 Issues

1. Synopsis: Testimony presented at the Baaver Valley, Unit No.1 hearing

-4 as the base value for probability used a figure of 1x10 of pipe rupture leading to a LOCA. A table on p.15 of the testimony provides ranges of failure rates from various sources.

i

2. Synopsis: In the CRBRP FES (NUREG-0139, Section 7.1.2) the staff compared a number of selected CRBRP accident sequences with the res0lts of similar sequences analyzed in WASH-1400 1

in order to provide an additional basis for gaining perspective on risks of very severe accidents in CRBRp.

3. Synopsis: In addition to their deterministic evaluation of the reliability of the control and shutdown systen for CRBRP, the ' staff utilized WASH-1400 data and analyses of 3-1

t

\ .

3. Synopsis: Certain Westinghouse Topi:a1 Reports rely upon absolute values of probability of accident events as set forth in WASH-1400. These reports currently are under staff review.

Certain of these reports (WCAP-8966, WCAP-8976 and WCAP-9283) are referenced in RESAR-414, and the remainder are expected to be referenced in other applications.

4. Synopsis: Risk assessment has ,been indirectly considered in the Mark I Short Tenn Pool Dynamic Program (NUREG-0408).

The conclusion of the Short Term Progiam (STP) was that, base,d on the demonstration of a minimum safety factor of two against failure, the Mark I plants could continue to cperate during an interim period of about two years while a methodical and comprehensive Long Term Program is con-ducted. This conclusion was based on the use of most l probable loads for the postulated LOCA and without an evaluation of Safety Relief Valve loads. This approach was found acceptable on the basis of the low probability l

'of a LOCA during the nominal two years needed to complete the Long Term Program. Consideration was also given to the low probability of a LOCA in establishing the Mark I technical specification related to d operation which impcses a positive pressure in the drywell relative to the wetwell so that in the event of a LOCA the pool dynamic loads are reduced. 3-2

The conclusions of the Mark I STP are only valid for Mark I .

plants under AP' operating conditions. Plants are allowed to operate in a non-AP mode for the limited periods specified in the Technical Specifications based on the expected low probability of a LOCA during this time limited period.

i S. Synopsis: In discussing the interpretation of Ceneral Design Criterion 19, we noted in NUREG-0138 that the analysis of the Browns Ferry fire in the Reactor Safety Study (WASH-1400) supports the staff position that for an i

event in the control room to lead to serious consequences

~

it would need to inv::ive damage of redundant equipment in the control room (or anywhere else) in such a way that operations at the secondary control stations could not ac-complish long-term cooling of the reactor. The staff concluded that a serious accident resulting from damage to the control room is of sufficiently low probability as not to warrant revision of the current design basis.

The fire damage experience at Browns Ferry involving (among other things) the loss of control of a number of systems helps to verify the many redundant means are available to resnurceful reactor operators to maintain a reactor in safe condition.

3-3

l l

I

6. Synopsis: The staff practice of not requiring that a passive mechanical valve failure be considered .s a single failure ]

following a postulated design basis accident is based on our judgment that such failures have an acceptably low likelihood of occurrence during both the injection (short-t'erm) and recirculation (long-term) phases of a loss-of-coolant accident. Further, analyses of ECCS performance in WASH-1400 indicate that passive mechanical failures of valves were unimportant contributors to ECCS unavailability during both the injection and recirculation modes of operation. Thus the staff does not consider that changes in safety criteria are warranted at this time but studies will seek to compile a more rigorous data base on passive valve failures.

7. Synopsis: An Information Report on the Single Failure Criterion (SECY-77-439) was sent to the Comissioners on August 17, 1977. This report describes current practice on application of the single failure criterion to LHR electrical and fluid syst$

It draws upon WASH-1400, in part, to support the concl.usion that the single failure criterion, as it is currently ap-I 3-4 l

l

l plied, leads to a generally acceptable level of hardware redundancy in most systems important to safety. It also 1

points out that methods such as those used in WASH-1400 will gradually come into increasing use as a supplement to the Single Failure Criterion.  !

I J

i j 8. Synopsis: In considering loss of offsite power subsequent to normal safety injection reset following a LOCA, we stated in NUREG-0138 that the analyses in the Reactor Safety Study, WASH-1400, indicate the likelihood of a LOCA to be about one change in 1000, per reactor year. This was combined i;; .

with the probability of the loss of offsite power in a one-hour period following a LOCA (about one chance in i

50,000) to obtain a combined probability of this sequence of events which was very low.

On the basis of our review of this issue as redefined in NUREG-0138, the Offic e of Inspection and Enforcement was I to review the emergency diesel loading for operating pWR's tc assure that all safe shutdown loads (which includes cooling to the diesel generator) are automatically picked 3-5

l l l l

up following an operator action to reset SIS. I&E in-spectors also were to examine emergency procedures to be followed in the event of a LOCA to assure that these pro-cedures do not permit SIS reset by operator action earlier than 10 minutes following the accident signal, unless it can be shown that such action is required in the irrterests l

of safety. However, the staff concluded tha t there is no basis for changes to any operating licenses or for changes of the current staff priority in considering this issue.

~

9. Synopsis: On July 15,.1977, it was stated to the ACRS regarding DC power reliability that, "...a conservative probabilistic assessment of the likelihood of occurrence of Mr. Epler's postulated scenario which is the basis for the concern regarding DC system reliability has been performed."

"The probability for occurrence of unacceptable consequences, i.e., core melt, as a result of this postulated sequence

-9 A comparison with the WASH-1400 core melt is 5 x 10 .

-5 prediction of 5 x 10 indicates that the contribution to core melt of this particular sequence is a fraction of 3-6

one percent. Furthermore, this won'i not change significantly even if it were assumed that there would not be any cap-ability for manual action to restore core cooling; i.e.,

if this number were one instead of 5 x 10"I .

A similar conservative assessment has been made for the postulated sequence initiated by simultaneous loss of both redundant'DC divisions and predicts a core melt probability of <5 x.10-7 . Comparison with the WASH-1400 prediction again shows that the cont!ribution to core melt of the com-mon mode sequence is negligible.

rf's judgment, on the basis of the probabilistic

.s cited, core melt resulting from the simultaneous at .ndependentfailureoftworedundantDCpowerdivisions  ;

is so unlikely as to be incredible; and core melt resulting ,

from common mode failure of these systems is very low in likelihood. We conclude, therefore, that adequate protection of the public health presently exists. However, additional l l

technical studies over the next year should and will be performed to add confidence to this judgment." This issue should be reconsidered in association with the completion of Task Action Plan A-30, including a recheck of the I

anlaysis for use of "the square root method." j l

3-7 1

,. ', .. j

10. Synopsis: As noted in flVREG-0138, in the event of a steam line break inside containment, it is necessary to isolate the main feedwater to the steam generator associated with the failed line to preclude overpressurizing the containment and to limit the reactivity transient. If the single active failure l I

j postulated for this accident is the failure of the appropriate safety grade main feedwater isolation valve to function, then credit is taken for closing the non-safety grade main feedwater control valve. Reliance on this non-safety grade valve in the postulated accident evaluation is pennitted based on the reliability of these valves.

The staff believes that it is acceptable to rely on the non-safety grade main feedwater control valve as a backup because its design and performance is compatible with the accident condition for which it is called upon to function.

The staff position is that utilization of the main feedwater control valve as a backup to a single failure in safety 9rade components adequately protects the health and safety of the public.

This position was taken in the Safety Evaluation Reports for the Erie, Sundesert and San Onofre (2&3) plants.

3-8

11. Synopsis: In a document transmitted to the ACRS in February 22, 1975, regarding grid availability, the staff stated:

"The data base used in the analysis is that provided in .

WASH-1400. The symbology, WASH-1400 numbers with specific references, sample calculations and tabulated results are attached. The conclusions reached is that the improvement in unreliability of offsite power the emergency buses pro-vided by a second immediate access circuit is not significant.

This is true even if the unreliability of the grid, which is the governing factor, were reduced by a factor of 10."

This need not be reconsidered other than a recheck of the i _ .

analysis four use of "the square root method."

12. Synopsis: The Branch input to the proposed response to Congressman Patterson's letter of April 2,1976 re: Postulated Ac-cidents and the Greene County case indicates that Class 9 accidents have been extensively studied and evaluated on a generic basis in WASH-1400.

4 3-9

1 :

13. Synopsis: A letter to Ms. Phyllis Taber' dated May 20, 1976 regarding ,

the safety of nuclear power plants discusses relative occurrences and consequences of non-nuclear and nuclear accidents in the Main Report of the Reactor Safety Study.

14 Synopsis: The letter to Lash and Cotton, NRDC, dated October 4, 1976 relating to proposed generic evaluation of risk acceptability quotes former Chairman Anders on the overall assessment of the Reactor Safety Study. .

15. Synopsis: The Supplement No. 2 to the Staff Safety Evaluation Report on the OPS case, re: accident evaluations states that WASH-1400 results confinn that accident risks are roughly proportional to population density.

3-10

1

16. Synopsis: Development of paper ;n Current Accident Evaluation.

Practices, dated October 3,1977. This draft proposes an in-terim positon that no changes in the safety or environmental regulations pertaining to nuclear power plants is warranted i until a detailed evaluation is made of the draft study.

WASH-1400 statements are used in a confirmatory manner.

l l

17. Synopsis: Section 7.1 of several DES /FES documents contain similar a.

i language relating how WASH-1400 will be used in licensing.

Examples provided include Erie, Allens Creek, Yellow Creek, 1

Arkansas Nuclear One, Unit No. 2, Hatch 2, Zimmer and Montague.

l The Erie document discusses the Reactor Safety Study and

' states that the results of the study will be assessed within the Regulatory process on generic or specific bases as may be warranted.

3-11

18- Synopsis: Responses to comments.on the Allens Creek DES includes the text from the " Introduction and Results" section of the Summary Report of WASH-1400. The Marble Hill response to comments in the DES concludes that the. staff's analysis of accidents did not rely on the Rasmussen report as a basis of its evaluations and conclusions.

19. Synopsis: In the Three Mile Island 2 Hearing, staff witness responses to cross examination in transcript, re: Aircraft crash I hazards made various references to WASH-1400 during testimony.

1 I

20. Synopsis: In external hazards discussions, re San Onofre station in a memo dated October 31. 1978, the probability of a propane explosion was discussed relative to the probability o', c LOCA in WASH-1400.

3-12 1

l l

1

21. Synopsis: Note to J. Lafluer commenting on some EPA studies, dated May 28, 1976. EPA study used data from WASH-1400; NRC was asked to comment on EPA work.

s l

! 22. Synopsis: Letter to W. D. Rowe (EPA, dated April 5,1977) regarding

staff's intent to extend the WASH-1400 methodology to more 1

likely events. This letter states that the NRC intends to extend the detailed assessments reported in WASH-1400

1. .

to more likely events (Class 3-8 accidents). -

1 l

23. Synopsis: Letter to John E. Ward (AIF) dated September 1, 1978 re: SECY 78-137 and the staff's intended use of Class 9 1

accident considerations.

The letter states that we believe that the Reactor Safety Study consequence model can provide useful insights into a few situations but we are aware of the need to be cautious in the direct application of any such analyses.

i 3-13 1

e

• y l

24 Synopsis: Testimony of C. Vernon Hodge and Donald J. Kasum related to radioactivity released as a result of sabotage during shipment of radioactive material, Sterling and Pilgrim hearings. The testimony indicates that no credit is l 1

given for protection afforded by buildings or for evacuation l

)

of the endangered area. WASH-1400 is referenced to in-J dicate that there actually would be a range of mitigating  !

l factors.

25. Synopsis: Response (June 1, 1977) to Congressman Moorhead discusses WASH-1400 to show that risk of accident in excess of

$560 million is extremely remo_te.

Recommendation: No further action is necessary.

26. Synopsis: Response (June 12,1975) to Murphy, JCAE references draft of WASH-1400 in discussion of how small risks from reactors are in evaluating if $560 million is enough of a liability limit.

3-14

27. Synopsis: The NRC response (June 2,1978) to Congressman Hamilton's constituent's letter on nuclear industry subsidies by 1

insuring utilities provide an estimate of annual loss i

based on WASH-1400 consequences.

28. Synopsis: page A-2/6 of Revision i to Task Action Plan A-2, Asymmetric Blowdown Loads On Reactor Primary Coolant System, cites pipe failure probability estimates from WASH-1400.

This information was used to support the staff's engineer-ing judgment for continued operation of the affected plants.

In the November 17, 1978 memorandum from Stephen Hanauer, it

~

was recommended that the staff reassess the short-term interim ,

acceptance criteria. However, since the information was used only to support the staff's engineering judgment, NRR believes no reconsideration is necessary.

The Safety Evaluation Reports on steam generator operation for l

29. Synopsis:

Surry Unit No. 1 dated February 8, 1977, Turkey Point, Unit l I No. 4 dated February 8, 1977, and Surry, Unit No. 2 dated April 1,1977 used pipe failure probability estimates from WASH-1400. This information was used to support the staff's engineering judgment for continued short-term operation. l These three reactors which were experiencing steam generator tube failures were granted continued operation for 60 days.

3 -15 i

r ',. .

30. Synopsis: For the extension of the ECCS exemption for Dresden, Unit' No.1, the staff constructed simplified fault trees of selected

,ECCS equipment and derived numerical probability estimates ,

using failure rates from WASH-1400. The exemption from 10 CFR 50.46 was extended' from December 31, 1977 to October 31, 1978. The results of the probability logic were not used in the December 29,1977 SER. The information was used to sup-portthestaff'sengineeringjudgment.

In an October 28, 1977 note to I. Wall, Mr. Taylor sent the l

l results of some probabilistic assessments pertaining to an ECCS single failure exemption for Dresden 1. This was done in response to a request from 00R.

1

31. Synopsis: The Conclusion section of all Fire Protection Safety Evaluation Reports such as Amendment 60 to Hatch 1 operating 1

license contains a quote from the review group report on te the fire at Browns Ferry (NUREG-0050). The quote is in part, "the' study (WASH-1400) concludes that the potential for a significant release of radioactivity from such a fire about 20% of the calculated from all other causes analyzed."

,This quote has been part of the staff's bases for allowing continued operation of the facilities until implementation I

of facility modifications.

3-16

l This statement has been used only to support the staff's

] i overall technical judgment. However, an additional l paragraph is being added to the SERs to further clarify -l the staff's bases for allowing continued operation.

l l

! Y2. Synopsis: In the May 9,1978 RSB input to the Safety Evaluation for tt e

Haddam Neck Overpressure Protection System, the sta.ff l i

tentatively accepted tne results of a quantitative fault 4

tree analysis. This analysis was used as a portion of the supporting basis for omitting as a design base transient

inadvertent water injection into the primary system through the high pressure safety injection pump (HPSIP). The fault-tree was constructed primarily of possible operator errors

~

that may combine to cause the event. Failure probabilities were taken from WASH-1400.

J 3-17

'33. Synopsis: The February 13, 1976 Safety Evaluation for Vermont Yankee J authorized continued operation for 30 days until hold-down l devices were installed on the torus. The licensee presented .I as supporting information pipe failure probabilities from WASH-1400. The staff, with more conservative failure esti-mates, effectively endorsed probability values as supporting I i

information to the staff judgment in granting continued J

operation. Other factors affecting staff judgment were the AP mode of operation, recent inservice inspections of affect-ing piping, and short period of tire (30 days).

l l

1

34. Synopsis: RSB's October'18, 1977 Safety Evaluatic n granted a one r

cycle exemption from the Appendix K single failure criteria applied to the Big Rock Point Nozzle Spray System (NSS). The exemption request was made since the licensee could not sub- l stantiate the ability of the Ring Spray Systen alone to provide adeqt. ate core ccoling in light of recent test data, i The staff evaluated the probability of a non-refloodable LOCA 'I and the failure of the NSS, und the probability of a LOCA in the NSS (refloodable LOCA) and the failure of the feed-3.18

{\

> ...,a.. ,, . . - . _ , . - . ,

i

: l water system using the WASH-1400 fault tree techniques. The staff's recomendation that the one cycle exemption be granted was not based on these probability assessments alone. Several other factors related to the BRP ECCS performance and reliabilitp were conside' red by the staff, and our conclusions reflect an l integrated assessment.

35. Synopsis: The April 1,1977 Safety Evaluation granted a six month I exemption from the ECCS single failure criteria to San Onofre.

Component failure rate data from WASH-1400 were used as a portion of the supporting bases for granting the exemption.

~

i

36. Synopsis

Pages A-12/3,4 of Revision 1 to Task Action Plan A-12, Fracture Toughness and Potential for Lamellar Tearing of Steam Generator and Reactor Coolant Pump Supports, cites .

1 pipe failure probability estimates from NASH-1400. This 1 I

information was used to support the staff's engineering a 1

judgnent for continued plant operation. l T

I I

3-19 l

i

', 1 l

l

37. Synopsis: To achieve a level of safety for CRBRP com-parable to that for LWRs as far as residual risks associated with core melt. accidents, the staff utilized WASH-1400 analyses of the times to containment failure to aid in establishing CRBRP containment integrity re-quirements. If the CRBRP review is reactivated, this decision should be reevaluated. In light of the current inactive status of the project, no further action on reconsideration is recommended at this time.

l

38. Synopsis: Reference to WASH-1400 was made by the licensee in providing the justification for not removing the catwalks from the Nine Mile Point, Unit No.1 containment torus for a period of five months. To the best of our recollection, the I

licensee's position was accepted as the basis for con-tinued operation. However, the catwalks have since been removed.

l l

3-20

', \

1 1

= s (

39. Synopsis: WASH-1400 is occasionally used to support reviews of events considered for reporting as abnonnal occurrence.

. . g

- .. )

.- ;1 J

40. Synopsis: In periodic updating of the IE reactor inspection procedures, .

9 a cross-check has been made to determine that WASH-1400 high

• s risk event related procedures, and equipment receive ,

~]

appropriate inspection attynti,on. Although the specific a

values stated in WASH-1400 were used in this evaluation, i -

they were used to make subjec$1ve domparisons and to con- -i 1

~

firm previous conclusions.

4 1

41. Synopsis: IE is studying ways of using risk analysis to improve the inspection program to make resource allocations and to

~

categorize risk related procedures with emphasis on human factors.

I 3-21

' ' * == * * '

• e mik 9 ..we'd ...*'.4 ,%...

• 4

.A

t .

42. Synopsis: Some accident sequences taken from WASH-1400 Were made the basis for. scenarios in developing procedures for the In-cident Response Center. This use is marginal in relation m l

'to the significant question being raised, b.ut it is included ,

here to assure completeness. ,

43. Synopsis: While none of the results or models of WASH-1400 were used

_s in licensing reviews, the consequence,model computer code __

(CRAC)~hasbeenusedbyNySS'.inNUREG-0194,aspecialstudyof a

transportation sabotage, and some data from WASH-1400 has been used in generic environkental' statements on transportation I of radioactive materials (NUREG-0170 and SAND 77-1927).

However, no new regulatory actions or changes to rules have resulted from these efforts. Thus, no regulatory actions or staff positions have been affected by WASH-1400 material.

0 3-22 .

.. 1 "7

, 1

44. Synopsis: In a November 11, 1976 letter from W. J. Dircks to Hon. L. M. Hamilton regarding decontamination processes, _

reference to the probability and consequences of a core

. melt as stated in WASH-1400 was made. Since no l licensing action was take.n no reconsideration.is necessary. ,

However, uncertainties should have been presented. . a

45. Synopsis: Memo from I. B. Wall to R. DeFayette dated August 23, 1976, *

Subject:

Draft Responses f5r Califormia State Energy Resources -2 3

Conservation and Developnent Commission. This memo uses --

' ~~

results from the Reactor Safety Study to illustrate the dis-tinction between the design} basis accident used for preparation l

of emergency plans and the Reactor Safety Study. In addition, -

further clarification was provided.regarding evacuatfon and relocation as used in the Reactor Safety Study.

46. Synopsis: Memo from I. B. Wall to R. W. Houston, dated Septer*2e" 14, 1976,

Subject:

Probability of 10 CFR 100 Doses. This meno , ,

transmits a c,opy of the memo from I. B. Wall to R. DeFayette dated August 23, 1976. This latter memo is covered in item 4,

• 1 e

above.

1 l .

3-23' t - ..

T Y

t ,-

l l 47. Synopsis: Memo from S. Levine to R. G. Ryan, dated October 7, 1976, f Comments on EPA Draft Publication ConcerDing the l subject.:

~

}-

Technical Bases for Dose Projection Methods to be Used as

(

a Basis for Protective Actions for Nuclear Incidents. The .

Comments in the memo use results of the Reactor Safety Study }

{ to illustrate points made in the review. ,

l .

. 'a

)

48. Synopsis: Letter from S. Levine to H. B. White, Sacramento County, 5

?

California, dated June 30, 1976. This letter provides some j clarifying information regardi'ng WASH-1400 in terms of estab- -

lishing an appropriate basi) orf which to fonnulate emergency plans .

5 --

i _

, g

49. Synopsis: Memo from S. Levine to B. Rusche dated August 9,1976,

Subject:

Review of Draft Liquid, Pathway Generic Study.

This memo uses WASH-1400 results to support comments on the ,

i draft liquid pathway generic study. ]

l 3-24 l

l

50. Synopsis: A study perfonned by Battelle, Columbus for RES on,the effects of containment venting on LWR meltdown accident risks compared WASH-1400 results with other results calculated with various containment venting schemes.

4

. o

. - - .. -p

51. Synopsis: A Sandia study for research on the value-impact assessment }, a 4,

of alternate containment concepts used the methodology of W i

WASH-1400 to determine the potential risk reduction from w

(

'j 3

various containment designs. --*

i 9 m

},'

4

< 1 y ,

t 1 52. Synopsis: Memos from Buhl to Stolz dated September 8,1978, and i

November 6, 1978, provide a reassessment of the Diablo l l

Canyon analysis of the risk to the public from a seismic u event in light of the comments of the Lewis Committee.

Methodology and absolute values of risk from WASH-1400 ,

1 were compared" to the applicant's recommendations. ,

\

4

)

$ e 9

4 l

3-25 1

f r  :

53. Synopsis: Task Action Plant A-37, " Turbine Missiles" (Revision 1) in Section 3, " Basis for Continued Plant Operation and -

L.icensing Pending Completion of Task," states:

"The basis for allowing continued operation of the existing LWRs, pending completion of this task is the low probability .

bf unacceptable damage to an essential system by turbine missiles. The Reactor Safety Study (WASH-1400) assessed '

the turbine missiles' accident risk and concluded that LWR designs have a considerable degree of protection provided by plant design and layout such that the public risk associated with large turbin'e missiles is insignificant comparedtorisksfromotperaccidentcauses."

a An October 14, 1977 memo frop I'. Wall to S. Pawlicki q

also comments on TAP A-37. -

1 Also memo from M. Taylor to S. Pawlecki dated September 3, j 1976 addresses turbine missile. '

4 1

O e

I 3-26 m

t 54.1 Synopsis: In an October 14, 1977 memo, I. Wall sent J. Stolz coments s on PAB's review of Diablo Canyon Amendment 52. The analyses in support of the Amendment and therefore these comments. refer w c

to component failure prob'abiliti'es, and consequence models and results from WASH-1400. A December 30, 1977 memo.from Wall _ _.'.

.to Stolz.provides.a~ draft SER input supporting Amendment 52; - -

. . . . . . . . - . -..=,,;__.__---

,9

~^

55. Synopsis: In an August 3,1977, memo I. Wall sent J. Knight comments on - e Task A-18, Pipe Rupture Desjgn priteria. The comments were a .

based in part on the results of WASH-1400.

' ' d

. t

~

56. Synopsis: The June 20, 1977 and August'll,1977 memo from S. Levine to R. Fraley transmitted calculations performed by PAB of j

of Control Room Doses for Postulated Core Meltdown Accidents.

i f

The doses were calculated for two accidents as cha acterized i

-l in WASH-1400. l l

3-27 i

~. - - . - ~ + _ . . _ . . . . - . . . . _

1 .~~

57. Syncpsis: The March 28, 1977 memo from Mat Taylor to Ian Wall transmitted viewgraphs on three ACRS generic issues which were to be used in an infonnal presentation to NRR. The viewgraphs used _l results and insights from WASH-1400.

.l

_-..m. ..__ _ . _.. __. .. y

- . _ .: l

58. Synopsis: Memo from I. B. Wall to V. M. Panciera dated July 9,1976,

Subject:

-Estimated Impact upon Public Risk Associated with -

-f a Non-inerted BWR Containment. This memo compares the risk ./

associated with a non-inerted BWR containment to the risk .

associated with the inerted containment used in WASH-1400 and makes reconinendations based on this analysis. -d

}*

i

. \

i

59. Synopsis: Battelle, Columbus prepared a report on the effect of engineered safety feat 0res on LMFBR risk due to accidents. WASH-1400 accident event trees were used in the analyses. .

.m e

e 9

3-28 -

l

• • .4ms.==4+e. *me - we = * - . . - ,..g - . . . . ,, , ___

;. l

60. Synopsis: Memo from Vesely to Staley, DSE, from Vesely to Ayer and from Vesely to Burkhardt dated June 7,1978 providing an _

I analysis of flood frequency of the Kishiminetas River.._ .

using. WASH-1400 methods to develop a frequency curve. ,, ,

. n

.. . y

% i

61. Synopsis: Memo Buhl to Mattson dated September 21, 1978 provides ,

RES comments on Supplement 1 to NUREG-0460. Methodology q and insights from WASH-1400 were used in the recoriimenda- (

tion to NRR.

]9 8 4

a l

62. Synopsis: The March 21, 1977 memo fromi W . Vesely to R. Baer,

. t C. Berlinger, S. Israel, and J.. McGough transmitted a description of the allowed downtime calculational approach used by PAB. Accident probabilities are used in the calculations.

63. Synopsis: The February 25, 1977 memo from S. Levine to B. Rusche -

and R. Minogue transmitted Research Information Letter-10, Pressure Vessel Failure Probability Prediction.

~

The draft report compared the new failure probabilities j .

with those predicted in WASH-1400. The report was only a draft and no licensing action was taken based upon it.

3-29 m _ . . ._. . ...

_ . . . . m

64. Synopsis: Memo from I. B. Wall to File, dated April 5,1976,

Subject:

Minutes of Meeting held ou April 2, 1976.

Memo from I. B. Wall and W. E. Vesely to H. J. C. Kouts ,

dated March 16, 1976, Sugject: Comments,on " Reliability t Assessment of CRBRP Reactor Shutdown Systems" , ,

~ ~

(WARD-D-0118, Riv.1), November 1975. ' These memoranda

[

discuss the role of probabilistic analysis in the licens- -

ing of the Clinch River Breeder Reactor Plant. The _

discussion with memoranda relied on WASH-1400 insights, ,

s data and analyses of similar LWR systems to assess the feasibtiity of the CRBRP Gontrol Sy. stem to meet the . _ ,

numerical goals set fory it.by the applicant.

a 1 g ,

t -

65. Synopsis: Memo from Edison to Novak dated November 7, -1978 provided an assessment,' using WASH-1400 techniques of changing the test frequency of the containment spray recirculation pumps. This assessment was used by the staff in its consideration on alternate testing scheme for the Surry pumps. -

G 3-30

\

66. Synopsis: In September 1976, the Director, RES-testified in a court proceeding related to the constitutionality of the I

Price-Anderson Act. His testimony covered what WASH-1400 was and its results. RES categorized this as a 1. Since absaiute values of risk were act relied upon to make any specific licensing decision in this instance, NRR has ,

classified it as a 3. It should be noted that tpe court ruled against the NRC in this instance, but .ias overruled

~

by the Supreme Court. Further, as we understand it, the

, Supreme Court decision did not depend on the numerical w

4 risk estimates of the RSS.

R

=

67. Synopsis: Memo from Buhl to Mattyon . dated Fiay 18, 1976 comments a

on proposed NRR study of missile impact effects on t -

structured barriers. Memb compares proposed study with an, attached event tree and concludes proposed study only covers a small part of total accident sequence probability.

Memo uses WASH-1400 analyses to confirm RES conclusion on utility of NRR study.

f G

3-31 .

l -

. 68. Synopsis: The May 15, 1977 memo from S. Levine to R. Ryan discussed the Program Plan being developed by,Sandia I Laboratories on Emergency Planning and Response Evalua-

-tion. This work is based in part on the models and

' methodology of WASH-1400. .

~ *

" ~ '

. c

-" ~

The NRC/ EPA Task Force has used information in the RSS

~

^ as a basis to perform calculations which illustrate the . -,q  ;

' likelihood of certain offsite dose levels given a core melt Q

accident. The results derived from the RSS based work servedH i

to confirm the Task Force judgment that offsite i'ianning ]1 for a generic distance around nucleaf power plants is H l

prudent and useful.

I a

. t k

' ~

flemo from Levine to Rya'n, SP, dated May 22, 1978 provides coments on dra'ft NUREG-0396.

e.

O o

e 4

3-32 m- w-y - w y wm ,m , q s,

.e ,a ==w-$ e m sp me mes.o

....h ,

69. Synopsis: The May 17, 1977 memo from I. Wall to S. Eilperen transmitted coments on Judge McMillan's de' cision con- H cerning Carolina Environmental Study Group, Inc. ,

et al. , v. United States A'tomic Energy Commission, N, et al., U. S. D. C... W. D. N. .

C., No. C _C-73-139. The decision and comments relied on the WASH-1400 methodology .

and results. .

~

9

70. Synopsis: As part of staff efforts regarding Seismic Scram, UCRL k 3

performed a study (UCRL-52156, " Advisability of Seismic

-l Scram") which relied upon some NASH21400 data regarding H accident probabilities iIs j'means of evaluating relative core melt probability with any w,ithout seismic scram. i

- \ 1 The staff has, as yet, taken no_ final action regarding i 1

this matter. .

Memo from I. B. Wall to V. Panciera, dated April 15, 1976, l

Subject:

Coments on the Advisability of Seismic Scram. y These comments were based on WASH-1400 irsights and results. ..

m 9

e 9

3-33

. . - . ~ . . - .- . . .. ..

J

71. Synopsis: In the development of Branch Technical Position RSB 5-1 on Residual Heat Removal, (attached to SRP 5.4.I), the results of WASH-1400 were used to show the potential need for increased requirements for RHR systems. Neither the -

numerical data nor the methodology of WASH ,1400 was used. }

72. Synopsis: Efforts are underway to modify the existing NRC-FCI code and  ;

s use it to calculate probabilities and consequences of steam

.+ '

explosions. Calculations of' steam explosion consequences _,

(but not probabilities) w9re performed for the FNP's docu-mented in NUREG-0440; this study assumed that the steam t i '

explosion probabilities to be bounded by the WASH-1400 results.

73. Synopsis: Probabilistic techniques similar to those of WASH-1400 were used to perform a study of allowed outage times for ECCS components for incorporation in plant Technical Specifications. ,

~

Data basically were used in a comparative sense.

3-34

. . . - . - - _ =;, . .; . ...n.. -

l i

s ^:

~

sco

74. Synopsis: In considering Task Action Plan B-68, the WASH-1400 T

probability values and analysis were used to determine .- l the overall probability of core melt resulting from a =PWR )

4

~' - ~

reactor coolant. pump flywheel missile ifr.pacting on an ~1 ECCS line due to pump overspeed following a cold leg .

]

break. Furthermore, based on the PWR design assessed, missile impact during a LOCA.would contribute less than 2% relativ[ .

4

.,m__..__.

to khe overall PWR core melt probability. 9

-I

75. Synopsi's:' The probability of an SS was eitracted f. om WASH-1400 3

for use in an enclosure to the R C working paper on over-pressure protection while operating at low temperatures.

This probability was used to suggest that the probability

, c:

of an overpressure event caused by an SSE while operating at low temperatures may not be a significant contributor ,

to the overall frequency of overpressure events (as deter-mined from actual operating data) and therefore should not .

be considered.

3-35

j J l R3 C recommended that the overpressure protection system must be designed to withstand the operating basis earthquake I (OBE). While the data from WASH-1400 was considered when determining the seismic requirements, it was not the 3

primary basis for the R C decision.

1 i

f

76. Synopsis: WASH-1400 was examined for justification of the, staff's proposed RHR Shutdown position (single failure / safety grade / seismic, etc.) to see if it did reduce the probability of core melt. It was found that the RHR position would not affect the WASH-1400 results since hot standby was con-sidered to be a success path in WASH-1400. As noted in a January 19, 1978 memo, NRR concluded that: "No quantitative assessment was made of the reduction in risk that would result from the proposed improvement in the RHR system (SRP 5.4.7), and the effect of a loss of the RHR cooling on risk was considered small and hence not evaluated."

In conclusion, the staff recommended implementation of the "RHR shutdown position."

r 3-36.

, j s  : . 1 l

l l

I

77. Synopsis: In considering whether mechanical failure of isolation valve in RHR suction line would preclude activating RHR system in Diablo Canyon, a comparison was made of the probability l of mechanical valve failure and SSE with the probability of core melt calculated in WASH-1400. We considered the valve failure probability acceptable because it was small compared to the WASH-1400 value. Moreover, steam generators provided alternate means of long term decay heat removal.

~

78. Synopsis: Using WASH-1400 values, we noted that the probability of a loss of offsite power at the time of the large loss-of-coolant accident is extremely unlikely (with a median value on the order of 10~7 per reactor year) and indeed is much less likely than several other scenarios con-sidered in WASH-1400. Based on this low probability of occurrence, we concluded that the Shoreham response re-garding recirculation pump trip was acceptable.

3-37

t

79. Synopsis: WASH-1400 is referenced twice regarding BWR rod drop accidents in a June 17, 1975 memo from H. Richings to D. Ross. In the first reference, the absolute values of accident probabilities for severe BWR accidents were used in a relative way to support the choice of a probability criterion such that the occurrence of the accident need no be considered a design basis event. It should be pointed out, however, that the pritnary basis for the choice of the criterionwasWASH-1270(ATWS). The reference to WASH-1400 was only supplementary in character.

The second reference to WASH-1400 was with respect to the probability 'of human error. Again the reference was.

supplementary in character and primary reliance for the estimate of the probability for human failure was not based

'on the reference to WASH-1400.

l 3-38

80. Synopsis: In considering grid frequency decay, we stated in NUREG-0138:

"Considering,the likelihood of occurrence of excessive

~

frequency decay and the release to atmosphere that would result from release of a portion of the total gap activity ,

to the primary coolant system, an accident such as that l postulated would represent a negligible portion of the reactor accident risk predicted in the Reactor Safety Study (WASH-1400) . "

l l

81. Synopsis: The staff relied on a probability analysis in developing its position regarding containment purging. No WASH-1400 results were incorporated in the analysis thus this is not a Category 1 item.

3-39

, t

82. Synopsis: The justification of the need for Regulatory Guide 1.139,

" Guidance for Residual Heat Removal," is based in part on the WASH-1400 result that showed the probability of core melt due to system and equipment failures that result in the inability to remove fission product decay heat is higher than the probability of core melt in the event of a large LOCA. Additional bases for the regulatory position of Regulatory Guide 1.139 are provided in the discussion, and it is the view of the staff that the position would be unchanged if the WASH-1400 results had not been considered.

[ Note that the use of WASH-1400 results is a conservative action; i.e., the need for increased safety is demon-strated.]

83. Synopsis: WASH-1400 estimates for fission product gap activity (Ap-pendix VII) were used to affirm the use of Regulatory Guide 1.25 source terms in Regulatory Guide 1.89 to determine the radiation environment for qualifying electrical equipment.

The more conservative source term of Regulatory Guide 1.25 was used in developing Regulatory Guide 1.89.

3-40

1

. l 1 .

84. Synopsis: WASH-1400 was used to provide an estimate of the conse-1 quences of sabotage. However, the decisions to implement reactor sabotage regulations were not based on the WASH-1400 results but rather on the knowledge that sabotage could 1

cause releases that would be harmful to the public. WASH-1400 is referenced in:

(1) " Safety and Security of Nuclear Power Reactors to Acts of Sabotage," SAND 75-0504 Sandia Laboratories, March 4

1976; (2) Memo R. B. Minogue thru L. V. Gossick to B. Huberman, Director of Policy Evaluation transmitting a discus-sion of design threat levels entitled, " Basis and Rationale for Selections of a Design Threat Level for Power Reactors Sabotage-Protection" prepared by SD staff, January 3,1977; (3) Transcript of the public hearings on the Material Access Authorization Program "Rulemaking in the Matter of 10 CFR Parts 11, 50 and 70, Docket Rm-50-7, July 10,11, and 12,1978."

3-41

, t i

85. Synopsis: In denial of PRM 50-19, the calculated consequences of core meltdowns in PWR and BWR reactors were used to estimate the potential effectiveness of an evacuated containment to mitigate the effects of a Class 9 core meltdown accident.

Risk assessment results and models (i.e., probability of the events)werenotused.

86. Synopsis: ,In their responses of December 15, 1977 and July 6,1978, to the Commission on the UCS petition for emergency

' and remedial action, the staff utilized the work of the Browns Ferry Review Group as reported in NUREG-0050. This group utilized the models of WASH-1400 to prcvide additional support to the staff position.

f

87. Synopsis: Supplemental Testimony of Darrell Eisenhut on Contention I-10, in the matter of Kansas Gas and Electric Company

]

and Kansas City Power and Light Company, (Wolf Creek Generating Station, Unit No. 1), Docket No. 50-482, January 6, 1976. The contention is similar to the Callaway contention in Itsn 88 below. The conclusion regarding the draft WASH-1400 report is also the same as in the Callaway testimony.

3-42

1  :

88. Synopsis: Supplemental Testimony of the NRC staff on Contention I-7 and on Contention I-29, in the Matter of Union Electric Company (Callaway Plant, Units 1 and 2),

Docket Nos. STN 50-483 and STN 50-486. Contention I-7 alleges that the staff's analysis of the environmental impact for the proposed facility is inadequate because Class 9 loss-of-coolant failure of ECCS core melt accidents are dismissed without detailed analysis, in spite of the probabilities for such an incident being onein17,000perreactoryear(WASH-1400). The staff testimony concluded that the draft WASH-1400 report did not present any information concerning the frequency of o'ccurrence of the accident sequence described in Contention I-7 that alters the conclusion that the environmental risk of such an accident can be considered to be negligible and need not be considered further.

3-43

S,YNOPSIS OF CATEGORY 4 ISSUES

1. Synopsis: The core melt evaluation for the CRBRP, FFTF and FNP are being reviewed and utilizing the molten core-concrete ,

penetration evaluation models, data and results of WASH-1400 as guidance in the core melt evaluation assessments for FFTF, CRBRP and the FNP.

2. Synopsis: With regard to water hammer, there is no specific reference to WASH-1400 in Section 3 " Basis for Continued Plant Operation and Licensing Pending Completion of Task" of TAP A-1. However, the WASH-1400 estimates of pipe rupture probabilities have been considered along with data on pipe cracking or rupture obtained during the staff re-view of water hammer events. In view of the low probability of piping failure due to water hammer and the corrective actions being taken,with respect to water hammer ir. PWR steam generators, continued operation and licensing of plants can proceed while Task A-1 is being conducted.

3. Synopsis: With regard to intersystem LOCA, WASH-1400 identified the intersystem LOCA in a PWR as a significant contributor to the risk resulting from core melt. The staff has analyzed this and other similar scenarios using the general methodology 4-1

_ _ - - - - - - 4

l and the data of WASH-1400. Memo dated July 3, 1978 from Buhl to Novak providing minor coments on NRR Irtersystem LOCA I Analysis. Minor changes in terminology and definition of terms were recomended. The staff analyses are limited to I those sequences which are significant contributors to risk in relation with the WASH-1400 results. Using these analyses, the staff plans to determine leak testing frequency. l

4. Synopsis: With regard to the use of probabilistic assessments of l reliability, we stated in. NUREG-0138 that: #

"The staff agrees that present technology does not pennit

-7 a rigorous demonstration of the WASH-1270, objective of 10 per reactor year. As shown by the Reactor Safety Study (WASH-1400),*however, the use.of a reactor protection system with a low unavailability, plus additional cap-ability provided by other systems to limit transients, prevents anticipated transients without scram (ATWS) from being the predominant contributor to core melt probability for light water reactors (LWRs). The conclusion supports the staff position that an acceptable level of safety can be achieved by use of reliable transient-limiting systems in conjunction with a highly reliable reactor protection system."

4-2

, i i

i 5. Synopsis: With regard to protection against ringle failures in

l reactivity control system, we stated in NUREG-0138 that- 1 A

"The release to the environmen* resulting from such re-lease of gap activity to the primary cooling system would represent a negligible contribution to the react.or accident risk predicted in the Reactor Safety Study (WASH-1400).

i An in depth review of the analyses has not been carried out since the transients have not been generally judged to be

a Condition II event and the reviews have been commensurate i with the apparently small safety significanca of the event.

The analyses which have been submitted, however, have been reviewed and nor:e have been found unacce'ptable."

Synopsis: The RSS consequence model (CRdC code) was used to calculate 5.

• consequences of a core melt at the GETR. Results were transmitted infonnally to and at the request of PSS/NRR. l Not documented and approach abandoned.

4-3

f .

i y

7. Synopsis: As part of evaluation of Diablo Canyon for interim license (which has not been used) the Probabilistic Analysis Staff i prepared a summary evaluation of the risk of operation of Diablo Canyon for a range of probabilities of a seismic l

event.

l

8. Synopsis: Memo from I. B. Wall to E. G. Case, dated June 29, 1976,

Subject:

Proposed Regulatory Guide 1.108, " Periodic Testing of Diesel Generators Used as Onsite Electric Power Systems at Nuclear Power Plants. This memo provides  !

comments on the proposed Regulatory Guide from the stand-point of overall public risk based on diesel generator

! unavailability.

)

1 4-4

, t

9. Synopsis: In Exhibit A. Section 6, Part IV of the Nuclear Energy Center evaluation an accident risk analysis is provided utilizing the methodology of WASH-1400 and data modified by the staff to reflect the specific design considerations of a nuclear park.

10. Synopsis: Memo from Edison to Novak dated November 7,1978 providing com-ments on the probability of a LOCA plus loss of offsite power.

Comments of the Lewis Committee were available and-reflected in the memo when the response was prepared.

11. Synopsis: Memo from M. A. Taylor and W. E. Vesely to I. Wall, dated August 6,1975,

Subject:

_ BWR Rod Drop Accident. This memo uses WASH-1400 methodology to analyze the rod drop accident for the ten oldest BWR reactors.

4-5 h

e _ _ _ - -

I ':

I 1

12. Synopsis: The staff is presently reevaluating the effectiveness of existing transportation reg'1ations in protecting the health and safety of the public. To a very great extent, that reevaluation is depending on quantitative risk assess-ment. There is, of course, little in common between re-actor accident probabilities and transportation.. accident probabilities. But there is some similarity in accident l consequences and post-accident cleanup between the two.

Therefore, the staff is using the consequence analysis portions of W.SH-1400 in the transportation analyses.

These uses are documented at this time in NUREG-0170 (Vol.1) and a Sandia contractor report SAND 77-1927.

The Sandia report is a precursor of a staff environmental s tatement. -

The st. *f use of quantitative risk assessment in general, and k ,H-1400 material in particular has been cautious and ciitical. Some aspects of the staff's questions on the validity of this risk assessment are addressed specifically e in the overall sunnary and conclusions of HUREG-0170 (Vol.1, p.ix). No rulemaking action has yet been taken on the basis of these risk assessments.

4-6

. t SYNOPSIS OF CATEGORY 5 ISSUES

1. Synopsis: The staff utilized the event tree / fault tree methodology .

of WASH-1400 to evaluate the reliability of the CRBRP Shutdown Heat Removal System. This evaluation was used in parallel to the staff's detenninistic approach (i.e., diversity, redundancy, etc) and provided additional insight on design changes and their contribution to achieving the required diversity and redundancy to meet the applicable General Design Criteria.

6

2. Synopsis: A study of comparative risk evaluations for advanced reactors is being done utilizing WASH-1400 type methodology. The objective of this work is to provide early guidance on the licensability (i.e., confonnance with the well-established regulatory criteria and practices) of a given advanced reactor relative to the present generation of LWRs.

5-1

3. Synopsis: Section 7.1.2.5 of the Report to ACRS on RESAR-414 des-cribes the Westinghouse design verification program for the Integrated Protection System (IPS). The program will include a system reliability analysis based upon techniques similar to those in WASH-1400. Staff reviewers should be alert for reliance on absolute values from WASH-1400.

l l

4. Synopsis: A study of systems interactions in advanced reactors uses event and fault trees and involves an evaluation of methods and techniques available for a qualitative and quantitative 1

study of systems interactions and common mode failures.

5. Synopsis: References to WASH-1400 were made on page 65-4 of the testimony on ATWS for the Black Fox hearing. WASH-1400 also is mentioned on pages TAP-38 of the testimony regarding Task B-34 and on page A-37/9 of the Task Action Plan for

. Task A-37. In none of these cases was specific information 9

from WASH-1400 relied upon.

5-2

)

6. Synopsis: WASH-1400 methodology was used for a preliminary analysis of the ANO-2 core protection calculator system. The analysis was not used in the final decision on AND-2 Similar methodology was used in evaluation of reliability of B&W RPS-II and Westinghouse IPS. None of these analyses has been used or referenced in a licensing action.

7. Synopsis: Operator error data was extracted from WASH-1400 to assist in evaluating the potential for an overpressurization event to occur while the DHR relief valves were isolated.

However, the use of the WASH-1400 data was not the basis for the acceptance of any design.

5-3 1

4 . . . . . - . . _ _ _ _ _ _

> ': 1 i

8. Synopsis: A WASH-1400 type analysis was used as a partial basis for recomending only manual seismic fire protection capability in new plants and for not backfitting operating plants or plants' under construction.

9. Synopsis: In the staff response to h Board question (North Anna, j

Units Nos. 1 and 2), reference was made to Regulatory Guide 1.120, which includes the following statement:

l "Although WASH-1400, Reactor Safety Study, An Assessment of Accident Risks in U.S. Cominerical Nuclear power Plants, dated October 1975, concluded that the Browns Ferry fire did not affect the validity of the overall risk assessment, the staff concluded that cost-effective fire protection measures should be instituted to significantly decrease the frequency and severity of fires and consequently l initiated the development of this guide." l l

5-4

. i is

10. Synopsis: Probability was used as a rationale to:

1) justify break exclusion for " super pipe," .

2) determining failure mode difference between high and moderate energy piping 1.e., breaks vs. cracks, and,

3) justify exemption of single active failures for certain piping systems.

Probability was also used as a partial basis for excluding $

certain primary piping breaks from consideration as CDA initiators in Clinch River and FFTF.

11. Synopsis: Diesel generator reliability operating experience was used as a probability data base coupled with probability of loss of offsite power to support the staff position on requiring diverse power supplies for auxiliary feed systems.

(Note: Draft of proposed ANS 51.1 references WASH-1400 as .

basis for two hour maximum period for loss of offsite power.)

5-5

1 ,

/3

12. Synopsis: The working paper for Regulatory Guide 1.63 regarding electrical penetrations for pump power supplies in contain-ment included the following statement: q "We have performed a probatiilistic analysis using the above failure data (failure rate calculated at the 95%

confidence level); the established LOCA probability of 10-4 per reactor year; and conservative assumptions regarding the time intervals during which the pump penetrations would be subject to failure (while energized) given that a LOCA occurred first, or during which a plant is subject to a LOCA(whilenotacoldshutdown)giventhatakm) penetration failure occurred first. Our detailed calculations are shown in Enclosure 2. The results of this analysis indicate that

- the probabil'ity of a LOCA concurrent with a pump penetration short circuit failure is less than 3.6 x 10-9 per year. This is considered to be an insignificant risk to the public health and safety. In our opinion a regulatory requirement directed toward reducing this risk cannot be justified, and may in fact have a negative impact on safety by diverting both applicant and staff resources form matters of greater safety significance."

5-6

,i

13. Synopsis: In the description of Generic Issue Task Action Plan A-25, the following statement is included:

"The approach selected for problem resolution is that of a reliability analysis of typical plant onsite Class IE J

power systems."

1

14. Synopsis: The " break exclusion region" for piping systems penetrating containment contained in Stnadard Review Plants 3.6.1 and  ;

3.6.2 is based on the premise that probability of pipe rupture in this region has been reduced when compared with

~

that of a "non-break exclusion region."  ;

h a

15. Synopsis: In our study to assess the effects of postulated event and devices (snubbers) on normal piping system operation,

' the probability of deleterious interaction of such devices l 1

with the piping system will be quantified, f 5-7  :

l l

1.

1

16. Synopsis: During the' period in which generic activity on Task Action Plan A-2 regarding asymmetric loads on RV supports was progressing, several plants were licensed prior to the completion of our complete evaluation based on scoping calculations, design conservatisms and the low probability .

for pipe rupture. This represents a subtle qualitative use of WASH-1400 without a definite value being stated, that the probability of a primary loop pipe rupture is low. l l

l

17. Synopsis: It is expected that our future work dealing with responses l to dynamic loadings will use probabilistic techniques for combination methods, or as the rationale for decoupling.

18. During a general review of the turbine missile problem, we performed a risk assessment review of the valves which are part of the turbine control system. Based on data which was available, a failure probability as a function of valve

' inspection frequency was determined for use in the overall turbine missile study. .

5-8 t

. .t I

f 4

3 l

19. Synopsis: Letter to Senator Case dated October 2, 1978 referencing l

~

. low probability of core melt accidents.

l s

In a talk by Dade Moeller of ACRS, re: Containment Spray

20. Synopsis:

System Failures, LER data were compared to WASH-1400 failure data by a present' AAB member, although prior to' his joining the Branch.

1

21. Synopsis: LASL under technical assistance contract to the NRC is using fault tree and event logic in analyzing nuclear plant vital areas as part of the security plant review. Fault trees from WASH-1400 have been used as part of the overall logic 4

structure. No numerical estimates from WASH-1400 have been

used. The results of the evaluation are transmitted from I LASL to RSLB in a letter report that is withheld from public disclosure in accordance with 10 CFR 2.790(d). The site specific fault trees / event trees are classified as Confidential y NSI and are kept in approved security repositories at either LASL or RSLB.

5-9

l

22. Synopsis: In SECY 77-388A, the staff proposed guidelines for the preparation of Value-Impact analysis. In an example of 9

where further action may be needed, WASH-1400 techniques were referenced as the type of analysis that could be i conducted. .

23. Synopsis: Memos from I. B. Wall to G. A. Arlotto dated June 30, 1975 l

and July 3,1978.

Subject:

IEEE/NPEC/P577, Draft 1, l

" Reliability Requirements in the Design and Operation of Nuclear Power Generating Stations." This memo presents detailed connents on the above cited draft. The comments relied on insights from WA5H-1400.

24. Synopsis: Letter from H.J.C. Kouts to W.6. Rowe, EPA dated July 7,1978, regarding Emergency Response Protective Action Guides. This letter forwards connents to EPA on the Protection Action Guides.

The connents relied on insights from WASH-1400.

5-10

25. Synopsis: Memo from S. Levine to V. Stello dated June 24, 1976,

Subject:

' D0R Re-review Program for Operating Nuclear Power Plants. This 1

' memo discusses the difficulty of applying risk assessment to the l re-review program. The memo relied on WASH-1400 insights.

26. Synopsis: Memo from S. Levine to H. Lowenberg dated July 23, 1976,

Subject:

Review of GESMO Chapter IV, Section C. This memo provides co..iments on the environmental. risks associated with Class 1-9

, accidents.

1

. 27 Synopsis: Meno from I.B. Wall to T.R. Wilson dated December 13, 1974,

Subject:

Statistical Analysis of Diesel Failure Data. This memorandum encloses a report on statistical tests performed on data obtained on diesel generator performance. The methods used are simile.r to those that were used to evaluate data in WASH-1400.

1 5-11

l

. .,l I 1

l l

28. Synopsis: Memo fror. W.E. Vesely to A.C. Thadani dated September 23, 1976, l l

Subject:

Review of EPRI Report "ATWS Reappraisal" (EPRI NPe251).

The memo relies on techniques.similar to those in WASH-1400 to.

criticize the EPRI report.

l l

l

29. Synopsis: Memo from S. Levine to R. Boyd dated October 8,1976,

Subject:

Responses to NRDC et al Fourteenth Set of Interrogatories in CRBRP proceeding. This memo relies on insights from the Reactor Safety Study to respond to interrogations.

5-12

,t i

30.. Synopsis: In a January 19, 1977 memo, S. Levine sent comments to G. Arlotto 4

on the Environmental Impact Statement on the Transportation of i

Radioactive Material by Air and Other Modes. In the memo l 4

reference was made to the risk assessment contained in the EIS.

4 Also, use of data from WASH-1400 inste'ad of from the BEIR report was criticized.

d I

)

31. Synopsis: Memo from I.B. Wall to S.H. Smiley, dated July 30, 1976,

Subject:

Review of " National Security and Accident Recovery Considerations i

ofNuclearEnergyCenter(NEC) Siting,"byG.A.Cristy,C.V.Chester,l l

and R.0. Chester, ORNL-5036. This memo provides comments on the above cited report and relied on insights from WASH-1400.

i 1

-l N

i l

5-13 l

l l

1

.  ?

32. Synopsis: The June 16, 1977 meno from S. Levine to E. Case and R. Minogue transmitted Ril-12, Modifications to Pressure Vessel Failure Probability Prediction. The draft reports contained sensitivity studies on the effects of the new modifications and updated failure probabilities.

33. Synopsis: In a June 14, 1977 memo I. Wall sent to D. Skovholt the results of PAB's review of the Study of NRC QA Programs by Sandia Laboratories. The comments deIt with the reliability analysis  !

and probabilistic techniques used in the study.

5-14

34 . Synopsis: The f,ovember 9,1977 memo from S. Levine to E. Case transmits

! Rll-18 on the FRANTIC Computer Code. The code calculates system unavailability.

35 . Synopsis: In a November 17, 1977 memo I. Wall sent I.C. Roberts comments on

' I N-635, Draft 3, Guidelines for Combining Natural and External Man-Made Hazards at Power Reactor Sites. PAB criticized the probability and risk assessments used in the draft Standard.

1 4

36 . Synopsis: In a July 26, 1977 memo M. Taylor sent S. Pawlicki comments on l

a paper by S. Bush titled, "A Reassessment of Turbine Failure l t

J Probability. " No specific mention of WASH-1400 is made.

1 a

s 5-15

i, **l l

l

37. Synopsis: In a July 27, 1977 memo I. Wall sent R. Moore comments on a proposed contract with Control Analysis Corporation. The study would furnish methods for predicting the probability of the coincident occurrence of several natural or man-made hazards to  ;

nuclear power structures, systems ano components.

38. Synopsis: In an August 23, 1977 memo W. Vesely transmitted information on 1

probabilistic analyses of test interval effects to V. Nerses.

The information addressed system unavailability and relied on WASH-1400 insights.

i 5-16 i 1

39. Synopsis: Memo from Buhl to Mattson dated February 3,1978 provides comments on Draft III of Appendix 2 of the NRR report on ATWS .

i Specific comments related to the scram failure l synthesis models. .l l

4 l

40. Synopsis: Memo from Buhl to Mattson dated March 20, 1978 provides comments on ATWS Draft III. Principal remarks deal with the conservatisms i used in the analysis as well as models used.

41. Synopsis: Memo from Buhl to Kehnemuyi dated April 20, 1978 provides comments on criteria contained in ANSI-N658 on single failures. Comments discuss the use of probabilistic technology and recommend concurrence in proposed ballot.

Y l

5-17

i .' ' h

42. Synopsis: Memo dated January 23, 1978 from S. Levine to E. Case providing RES coments on the draft working paper of the Liquid Pathwaj Generic Study. Principal coments related to WASH-1400 methods used in the LPGS.

43. Synopsis: An April 12, 1978 report to Congress on research to improve LWR safety utilized the methodology to help establish wnat research should be accomplished to improve reactor safety.

44. Synopsis: In a November 23, 1977 memo S. Levine sent E. Case comments on a proposed information paper, Use of RSS Consequence Model in Evaluations of Alternatives to Sites With High Population Densities. j The comments relied on insights gained from WASH-1400.

I 5-18 l

l

i ' ' . '.

4

45. Synopsis: Memo from W.E. Vesely to G.S. Vissing dated December 18, 1975,

Subject:

Regulatory Guide " Periodic Testing of Diesel Generators Used as Onsite Electrical Power Systems at Nuclear Power Plants."

Evaluations were perforned to determine the reliability and risk 2

implications of the proposed testing scheme. Analytical techniques 4

were used that are similar to those used in WASH-1400.

46. Synopsis: Memo from I.B. Wall to R.B. Minogue dated March 4,1976,

Subject:

Minutes of Meeting Held on 3/1/76 to Discuss Degree of Conservatism in the Draft Environmental Impact Statement on the Transportation Comments were based on techniques and j i

of Radioactive Materials.

insights from WASH-1400.

l I

i

47. Synopsis: Memo from S. Levine to R.E. Heineman dated March 26, 1976,

Subject:

Examination of the Seismic Design Basis for Fire Protection Systems.

This memo provides an analysis directed to the question of ,

whether fire protection systems should be designed to seismic  ;

1 Category I systems. Improved data obtained since publication could l i

modify results and widen error bounds but the general conclusions would be expected to remain valid. j 5-19 ,

...)

SYNOPSIS OF CATEGORY 6 ISSUES

1. Synopsis: " Report on the TVA Seismic Issue by NRC Staff Working Group" considered, but recommended against, use of .

WASH-1400 as an aid in determining seismically-induped core melt sequences. The use of WASH-1400 was considered, but rejected.

2. Synopsis: Additional remarks by ACRS member Dr. Okrent in the Com-mittee's Report on Perkins/ Cherokee (April 14,1977)in-cluded a comment about the estimates of the contribu' tion of earthquakes to overall nuclear reactor safety risk, as l

given in the Reactor Study (WASH-1400). The Hearing Board I then requested written materia'l that addresses the reservations i

of ACRS member Okrent. Written material pertaining to quantification of inherent safety margins on seismic design was provided. During the hearing, the Board pursued the question of how the staff rationalizes their position on setting the design basis earthquake against the probabilities.

As staff witness, C. Moon stated that the staff id review a draft of WASH-1400 and did make coments, but that the 6-1

'c ' ' ' . ,

1 a

staff has not then (July 21,1977) adopt that report )

1

- or any similar procedure on its licensing review actions, l

, 1 l

! 3. Synopsis: In the rulemaking hearing for 10 CFR 11 held in Washington, 1 D. C., on July 12, 1978, the staff referred to the " con-sequence tables" in WASH-1400 during presentation of l

testimony. The staff also referred to data in WASHJ1400 l which compares the consequences of other disasters to

)

postulated events at a nuclear plant. (See pages 422-557 of transcript.)

In responding to Mr. Gossick's request, NMSS stated, ". .. the NMSS staff believes that they have taken no licensing or regulatory actions which have i relied on the risk assessment results and models of WASH'-1400." They did identify the following two issues that made " remote" reference to WASH-1400. 1 6-2

. 1 ,' *. *

4. Synopsis: Basic data referenced in the draft WASH-1400 concerning natural gas pipe line failure rates was used in the pre-paration of the environmental statement on the Bear Creek Project of Rocky Mountain Energy Company, Docket No. 40-8452.

However, such data would have been available and might have been used by the NMSS staff whether or not it had also been used in WASH-1400. .

5. Synopsis: Draft input in the Seabrook alternative site review con-tains results of limited studies that led the staff to conclude that population density is a sufficiently crude indicator that relatively large differences in population densities between two sites would be required before sig-nificant differences in residual risks at these sites could reasonably b'e expected. ,

6-3

7 '. , 2 l

6. Synopsis: Com.missioners information cards contain inform,at, ion related to risks from various non-nuclear and nuclear accidents. Data used was compared to WASH-1400. ,

7. Synopsis: The Annual Reports for 1975,1976,1977 and 1978 discuss ,

1 WASH-1400 and 'some uses of the results.

,q

' a

8. Synopsis: An extract from the November' 18,1978 -issue of National __

Journal discusses the Rashussen Report.

9. Synopsis: A December 8,1978 memo from', Levin'e to Denton provides three l

additional items identified by RES that utilized the insights of WASH-1400. They are a letter to Senator J. Glenn dated December 9,1976 and copies of NUREG-0138 and NUREG-0153. The letter to Senator Glenn provides responses to questions about the discussions by NRR of issues in NUREG-0138. Specific is-sues of NUREG-0138 and NUREG-0153 are discussed elsewhere in this enclosure. The letter to Senator Glenn is considered as a Category.2 issue not deserving reconsideration.

6-4

. - -