ML19257B006
| ML19257B006 | |
| Person / Time | |
|---|---|
| Issue date: | 03/29/1979 |
| From: | Edison G NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | John Hickman SANDIA NATIONAL LABORATORIES |
| References | |
| REF-GTECI-A-30, REF-GTECI-EL, TASK-A-30, TASK-OR NUDOCS 8001150061 | |
| Download: ML19257B006 (11) | |
Text
.
j
.S, UNITED STATES NUCLEAR REGULATORY COMMISSION
^
p
- ),j WASHINGTON, D. C. 20555 i4x(
g_
,.~gi e
' \\.d o
MAR 2 91979 Mr. J. W. Hickman Sandia Laboratories Albuquerque, New Mexico 87185
Dear Jack,
The purpose of this letter is to provide you with the Commission's policy statement on the use of WASH-1400 in the development of regulatory actions and NRC staff positions. This topic is particularly germane to the DC Power System Study which is relying heavily on probabilistic risk assessment mc; hods and may require the use of some data or results from WASH-1400. As the DC Power study is intended to add confidence to NRC licensing positions and r.egulatory actions, I am providing you with specific guidance on this matter.
Enclosed for your use is the Commission press release dated January 19, 1979, entitled " Nuclear Regulatory Commission Issues Policy Statement on Reactor Safety Study and Review by Lewis Panel" and memorandum from the Executive Director for Operations dated January 18, 1979, subject, " Staff Actions Regarding Risk Assessment Review Group Report." This press release and memorandum will provide you with general guidance on the use of WASH-1400 in light of the criticisms raised in the Risk Assessment Review Group Report (NUREG/CR-0400).
With regard to the DC Power Systems Study, some specific emphasis is required.
I would be particularly concerned with the use of WASH-1400 data and results if they have not been critically reviewed prior to appli-cation in the DC Power System Study.
It would not be satisfactory, for example, to use data or results out of WASH-1400 without first examining their source and uncertainty. As pointed out in the enclosures to this letter, " component parts of the Study (WASH-1400)...(may be used) as appropriate, that is, where the data base is adequate and analytical techniques permit" (see enclosure 1).
However, "such techniques should not be used to estimate the absolute values of probabilities of failure of subsystems unless an adequate data base exists, and it is possible either to quantify the uncertainties or to support a conservative analysis."
(see enclosure 2) 1767 291 8001150Q(o[
Mr. J. W. Hickman.
If any questions on this matter arise during the course of the study, I will be happy to discuss them with you.
Sincerely, Gordon E. Edison, Head Systems Engineering Section Probabilistic Analysis Staff Office of Nuclear Regulatory Research
Enclosures:
As Stated cc w/ encl:
P. Baranowsky A. Buhl S. Hanauer P. McGrath J. Pittman F. Rosa 1767 292
UNITED STATES c
' ['.:jg.
NUCLEAR REGULATORY COMMISSION
,%, D [#
~
WASHINGTON, D.C. 20555
- c.. !
OFFICE OF PUBLIC AFFAIRS No.
79-19 FOR IMMEDIATE RELEASE Contact :
Frank L.
Ingram (Mailed - January 19, 1979)
Tel.
301/492-7715 NUCLEAR REGULATORY COMMISSION ISSUES POLICY STATEMENT ON REACTOR SAFETY STUDY AND REVIEW BY LEWIS PANEL The Nuclear Regulatory Commission today issued a state-ment of policy concerning its views of the Reactor Safety Study (WASH-1400)* in light of criticisms of the study made by a special review group headed by Dr. Harolc Lewis of the University of California at Santa Larbaia.
The Commission said it accepts the findings of the review group and is taking actions to respond to them.
A copy of the Commission's policy statement is attached.
The policy statement noted that while praising the study's general methodology and recognizing its contribution to assess-ing the risks of nuclear power, the Review Group was critical of the Executive Summary, the procedure followed in producing the final report, and the calculations in the body of the report.
The Review Group also criticized, in some cases severely, varicus calculational techniques in the study as well as its lac} of clarity.
The Review Group indicated the Executive S __....c _ y i s a poor description of the contents of the report and should not be portrayed as such, does not a'dequately indi-cate the full extent of the consequences of reactor accidents, and-does not sufficiently emphasize the uncertainties invclved in che calculation of their probability.
The Review Group ccncluded the Executive Summary has lent itself to misuse in the discussion of reactor risks.
The Review Group criticized the peer review process, pointing out that in some cases cogent from 'ritics either were not acknowledged or were ccmments c
e.aded.
The Review Group concluded that the error bounds on accident probabilities were greatly understated.
1767 293
% 79-19 The Reactor Safety Study was begun under the auspices cf the former Atomic _ Energy Commission and a draft version was circulated for comment in April, 1974.
The final report was made public on October 30, 1975 by the Nuclear Regulatory Commission, which assumed -the regulatory functions of the former AEC in January, 1975.
Criticism of the study has centered on the method of treating peer comments on the draft report as well as on the substance of the final report.
Following letters from Congressman Morris Udall, Chair-man of the House Committee on Interior and Insular aff airs expressing misgivings about the Reactor Safety Study, and in particular about the Executive Summary published with the main report, the Commission established a Risk Assessm.ent Review Group in July, 1977.
The Commission said it expected that the review group, headed by Dr. Lewis, would assist th e Ccmmission in establishing a policy regarding the use of risk assessment in the regulatory process and would clarify the achievements and limitations of the Reactor Safety Study.
The Review Group submitted its report on September 7, 1978.
In its policy statement, the Commission said:
~s_
'(1)
It_ withdraws any explicit or implicit past endorse-ment of the Executive Summary.
(2)
It agrees th at the peer review process followed in publishing WASH-1400 was inadequate and that proper peer review is fundamental to making sound tech-nical decisions.
The Commission will take whatever corrective action is necessary to assure that effec-tive peer review is an integral featu a of the NRC's risk assessment program.
(3)
It accepts the Review Group Report's conclusion that absolute values of the risks presented by WASH-1400 should not be used uncritically either in the regulatory process or for public policy purposes and has taken and will continue to take steps to assure that any such use in the past will be corrected as appropriatn.
In particular, in light of the Review Group's conclusions on accident probabilities, the Commission does not regard as reliable the Reactor Safety Study's numerical esti-mate of the overall risk of reactor accident.
0 1767 294
. 79-19 (4 )'
It has directed that a review be made of Commission correspondence and statements involving WASH-1400 and corrective action as necessary will be taken.
The Commission also said that with respect to the com-ponent parts of the Reactor Safety Study, it expects its staff to make use of them as appropriate, that is, where the data base is adequate and. analytical techniques permit.
The Cc= mission also said that, taking due account of the reserva-tions expressed by the Review Group, it supports the extended use of probabilistic risk assessment in regulatory decision-making.
It said that the NRC staff has been provided with additional detailed instructions concerning continued use of risk assessment techniques and results.
, 'WASB-14 00 also is known as the "Rasmussen Report. "
Attachment s_
l e
1767 295 e
i IIRC STATEMEi!T Ol1 RISK ASSESSMEf1T AfiD THE REACTOR SAFETY STUDY REPORT (UASH-1400) lii LIGHT OF THE RISK ASSESSMcIlT REVIEW GROUP REPORT The Risk Assessment Review Group, chartered by the I RC in July,1977 to " provide advice and information to the Commission on the final 1/
repdrt of the Reactor Safety Stud,, UASH-1400," and related matters, Tew The Rev submitted its report to the Commission on September 7, 1978.
Group, chaired by Professor Harold Lewis of the University of California at Santa Barbara, 2/ uds formed in response to letters from Congressman Udall, Chairman of the House Committee on Interior and Insular Affairs, expressing misgivings about the Reactor Safety Study (UASH-1400), and in particular about the " Executive Summary" published with the Main Report.
It was expected that the Review Group's report would " assist the Commission in establishing policy regarding the use of risk assessment in the regulatory process" and that it would " clarify the achievements and limitations of the Reactor Safety Study."
In August,1972, the Chairmen of the Atomic Energy Commission informed the Chairman of the Joint Committee on Atomic Energy that the Atomic Energy Commission had undertaken an in-house study "to provide a basis for submitting recommendations to the Congress regarding the extension or modification-of the Price-Anderson Act."
A draft version of the study report was circulated for comment in April,1974.
On October 30, 1975, the !!uclear Regulatory Commission 3/ announced that the final report had been completed.
Criticism of ti e document following relea.se centered on the method of treating peer comments on the draft report as well as on the substance of the report.
The fiRC press release
_ 'N accompanying publication of WASH-1400 praised the report, describing it as a " realistic assessment..., providing an objective and meaningful estimate of the present risks associated with the operation of present day light water reactors in the United States," gave several comparisons to show that the risk from nuclear power was much less than from other ran-made activities, and included a statement that "the final report is a soundly based and impressive work....
Its overall conclusion is that the risk attached to the operation of nuclear power plants is very low compared with other natural and man-made risks." 4/
In view of the importance attached to the Reactor Safety Study, within and outside the Commission, both prospectively and after it was rade public, the Commission has reexamined it's views regarding the Study in light of the Review Group's critique.
0 1767 296
y\\ \\
- While praising the study's general nethodology and recognizing its contribution to assessing the risks of nuclear power, the Review Group was critical of the Executive Summary, the procedure followed in producing the final report and the calculations in the body of the report.
Among the major failings of the study, the Review Group cited:
The Executive Sum: nary:
The Review Group concluded that "the Executive Summary of the RSS is a poor description of the contents of the report, should not be portrayed as such, and has lent itself to misuse in the discussicn of reactor risks."
The Review Group indicated the Executive Summary does not adequately indicate the ' full extent of the consequences of reactor accidents and does not sufficiently emphasize the uncertainties involved in the calculations of their probability.
As a result, the reader may be left with a misplaced confidence in the validity of the risk estimates and a co-e favorable impression of reactor isks in comparison with other risks than warranted by the study. 5/
The Peer Revieu Process:
The Review Group Report criticized the RSS staff response, pointing out that in some cases cogent conments from critics either were not acknowledged or were evaded and that, in general, the record of response to valid criticism
^s was weaker than it should have been.
The Report points out that the. lack of clarity of WASH-1400 itself led to major diffi-culty in tracing a line of thought through the study and crippled many efforts to accomplish responsible peer reviews.
Accident Probabilities:
The Review Group was unable to deter-mine whether the absolute probabilities of accident sequences in UASH-1400 are high or low, but believes that the error bounds on those estimates are, in general, greatly understated.
This, the Report said, is true in part because there is in many cases an inadequate data base, in part because of an inability to quantify common cause failures, and in part because of some questionable methodological and statistical procedures.
The Review Group also criticized, in some cases severely, various of the calculationa1 techniques in the Study as well as its lack of clarity.
The Review Group cited the following as major achievements of the study:
" WASH-1400 was a substantial advance over previous attempts to estimate the risks of the nuclear option.
1767 297 me==-
"l? ASH-1400 was largely successful in at least three vtays; in making the study of reactor safety mere r'ational, in cstablishing the topology of many accident sequences, and in delineating procedures through v:hich quantitative estimates of the risk can be derived for those sequences for vhich a data base exists.
"Despite its thrtccminr;s, WS.SH-1403 provider, at ".".is tim.c the most cc;glete single picture of tccident prcbabilitics associated with nuclear reactors.
The fault-tree / event-tree approach coupled with an adequate data base is the best available tool with which to quantify these probabilities.
"UASH-1400 made clear the importance to reactor safety dis-cussions of accident consequences other than early fatalities."
The Commission accepts these findings and takes the following actions:
Executive Summary:
The Commission withdraws any explicit or implicit past endorsement of the Executive Summary.
The Peer Review Process:
The Commission agrees that the peer review process followed in publishing WASH-1400 was inadequate and that proper peer review is fundamental to making sound, technical decisions.
The Commission will take whatever corrective action is necessary to assure that effective peer review is an integral feature of the NRC's risk assessment program.
Accident Probabilities:
The Commission accepts the Review Group Report's conclusion t' hat absolute values of the risks presented by UASH-1400 should not be used uncritically either in the regulatory process or for public policy. purposes and has taken and will continue to take steps to assure that any such use in the past will be corrected as appropriate.
In particular, in light of the Review Group conclusions on accident probabilities, the Commission does not regard as reliable the Reactor Safety Study's numerical estimate of the overall risk of reactor accident.
Ccamunication with the Congress and the Public:
Commission correspondence and statements involving UASH-1400 are being reviewed and corrective action as necessary will be taken.
[767298~
gik o
-4
..ith respect to the component parts of the Study, the Commission expects the staff to make use of them as appropriate, that is, where the data base is adequate and analytical techniques permit.
Taking due account of the reservations expressed in the Review Group Report and -in its presentation to the Commission, the Commission supports the extended use of probabilistic risk assessment in regulatory decisionmaking.
The Commission has provided additional detailed instructions to the NRC staff concerning continued use of risk assessment techniques and results in respense to specific criticisms raised by the Risk Assessment Review G roup.
~
,e
~
e 1767 299 G
e e
NOTES 9
1/
Its charter reads:
"The Review Group will piovide advice and information to the Commission regarding the final report of the Reactor Safety Study, WASH-1400, and the peer comr.ents on the Study, advice and recont.endations on developments in the field of risk assessment niethodology and on future courses of action '.;hich sh::uld be taken to improve this methodology and-its cpplication.
This advice and information will ascist the Commission in establishing policy regarding the use of risk assessment in the regulatory process, in improving the base for the use of such assessments.
It will also clarify the achieve-ments and limitations of the Reactor Safety Study."
2/
The other members were Dr. Robert J. Budnitz (Lawrence Berkeley Laboratory, University of California), Dr. Herbert J. C. Kouts (Brookhaven National Laboratory), Dr. Walter Loewenstein (Electric Power Research Institute), Dr. William Rowe (Environ-mental Protection Agency), Dr. Frank von Hippel (Princeton University) cnd Dr. Fredrik Zachariasen (California Institute of Technology).
Dr. Budnitz is presently on leave from the University of California and is serving (since August 1978) as Deputy Director of the GRC's Office of Nuclear Regulatory Research.
3/
The Nuclear Regulatory Comnission tras established on January 19, 1975 to carry out the regulatory functions of the Atomic Energy Commission, which was abolished on that date.
--4/'
The press release at the time of publication said that the report is "the culcination of the most comprehensive risk assessment of nuclear power plants made to date.
The objectives of the study were to make a realistic assessment....
The overall conclusion...is that the risks attached to the operation of present day nuclear power plants are very low compared to other natural and man-made risks....
Nuclear power plants are about 10,000 times less likely to produce fatal accidents than man-rade non-nuclear activities....
Hon-nuclear accidents involving comparable lcrge dollar value damage are about 1,000 times more likely than nuclear power plant accidents....
The chance that a person living in the general vicinity of a nuclear power plant will be fatally injured in a reactor accident is one in five billion per year....
In the event of an unlikely reactor accident with a probability of one in a million per reactor per year, latent health effects except for thyroid nodules would be such a small percentage of the normal incident rates that they v ould be difficult to detect...."
1767 300
The !!RC Chainc.an was quoted as saying, "The Corunission bel-ieves that the Reactor Safety Study Rep 6rt provides cn objective end meaningful estimate of the public risks associated uith the operation of present day light water reactors in the United States....
The final report is a soundly based and impressive work....
Its overali ccrclusion i, that the risk atttched to the caration of nuclear :nt:ce ple.nts is very lael ccreared with other natural and man-made risks." The press release went on to say that more thcn 1800 pages of comments were received from a br.oad spectrum of people and all were carefully considered in preparing the final report.
5/
Professor Lewis, in reporting to the Commission, said that the Executive Summary ~was not a summary of the repo rt.
He concluded it was written as a public statement that reactors were safe compared to other risks to which the public is exposed and he stated it should not have been attached to the report and described as a part of it.
~_
9#
~
x e
t 1767 301
~
c*::: '.hi...
.ic........... :". It*" m : 5 2 m r.
?
ll UNITED STATES
"'8vg NUCLEAR REGULATORY COMMISSION g
WASHINGTON, D. C. 20555
,y i,
g
'a January 18, 1979 n
Sg m.<
'OFFICT Of' THe SECRETARY MEMORANDUM FOR:
Lee V. Gossick Executive Director for Operati s
FROM:
Samuel J. Chilk Secretary of the Comissfor c
SUBJECT:
~ STAFF ACTIONS REGARDING RIgK '
ESSMENT REVIEW GROUP REPORT
~
Attached is.a policy statement issued by the Comission on January 18, 1979.
In addition, the Comission has provided the following instructions for the staff.
{
kend copie's 'of the Risk Assessment ~ Review Group' Report (NUREG/CR-.
.l.0400) and of the' January 18, 1979 Comission policy statement to all known domestic and international recipients of the RSS.
In the future, copies of the RSS Executive Sumary and the complete RSS will be distri-buted only when accompanied by a copy of th? Review Group's report and a copy of this. statement.
Quantitative risk assessment techniques and results can be used in 2.the licensing process if proper consideration is given to the results of the Review Group. The staff should use the following procedures re-garding the use of quantitative risk assessment techniques and results pending development of further guidance:
In comparisons of risks from nuclear power plants with a.
other risks, the overall risk assessment results of the RSS (i.e., curves or tables of the probability of occurrence of various consequences) shall not be used without an indication of the wide range of uncertainty associated with those estimates. Any such use should note the difficulty of placing high confidence on estimates that are well below the values set by experience.
t~
1767 302 el eua
} } ' ' l Z j' (
(
.'5
i-
~~b.
Quantitative risk assessment techniques may be used to estimate the relative.importance of potential nuclear power plant accident sequences or other features where sufficient similarity exists so that the comparisons are not invalidated by lack of an adequate data base. Such techniques should not be used to estimate absolute values of probabilities of failure of subsystems unless an adequate data base exists, and it is possible either to quantify the uncertainties or to support a
~
conservative analysis.
The quantitative estimates of event probabilities in the c.
~ RSS should not be used as the principal basis for any regu-latory decision. However, these estimates may be used for relative comparisons of alternative designs or requirements provided that explicit considerations are given to the criti-cisms of those estimates as set forth in the Report of the Risk Assessment Review Group.
d.
The RSS consequence model shall not be used as the basis s
.for licensing decisions regarding indiYidual nuclear power plant sites until significant refinements and sensitivity tests are accomplished. However, the consequence model may be used for relative comparisons provided that such estimates are not the primary basis fo'r such reviews 2nd p'iovided that explicit consideration is given to the criticisms of the various elements of that model as set forth in the Report of the Risk Assessment Review Group.
The staff shall prepare'and submit by June 30, 1979, detailed pro-cedures to ensure the proper and effective use of risk assessment theor, methods, data development and statistical analyses by the staff. Pendi review by the Commission of these detailed procedures and the bases and rationale supporting them, the Office Directors will obtain the advice of the ED0's Regulatory Requirements Review Committee should questions arise regarding the implementation of the above instructions.
3.
The staff shall review the extent to which past and pending 11-censing or other regulatory' actions, including Comission, ACRS and li-censing board ~ actions and statements, have relied on the risk assessmen models and risk estimates of the RSS. The Commission will examine the results of.this review to detemine whether the degree of reliance identified was and continues to be justified and to decide whether regulatory modifications are appropriate.
1767 303
= + -
- - r.::
... r. ".- _......_
2.-
4 The staff shall give special attention to those activities identified by the Review Group as being especially amenable to risk assessment, i.e., dealing with generic safety issues, fomulating new regulatory requirements, assessing and re-validating existing regulato:
requirements, evaluating new designs, and fomulating reactor safety research and inspection priorities.
5.
The staff shall prepare a review of current NRC practices and pro-cedures in two areas of particular concern to the Review Group:
a.
the peer review process for risk assessment developments, and b.
the coordination among the research and probabilistic analysis staff and the licensing and regulatory staff, in order to promote the effective use of these techniques.
The Comission will make whatever changes are necessary to assure that effective peer review and interoffice coordination are integral '
features. of NRC's risk assessment. program. =n :
.. : _m 6.
The staff shall examine the significance of the technical issues raised by the Review Group and the appropriate courses of action for dealing with them. These issues include questions about statistical methods, data base quality and availability, consequence modeling, human factor considerations, earthquakes, fires, and comon cause failures. The Comission will address what changes should be proposed in the approved FY 79 and proposed FY 80 research program to improve the data base, including that on human behavior.
As an addditional action, the staff shall undertake a review of statistical methods and human factor considerations used in risk assessment.
s.
Attachment:
As stated cc:
Chairman Hendrie Comissioner Gilinsky Comissioner Kennedy Comissioner Bradford Comissioner Ahearne James L. Kelley, 0GC Kenneth Pedersen, OPE 17'67 304 Joseph J. Fouct)ard, OPA Carlton C. Kamerer, OCA
..