ML19301F171
| ML19301F171 | |
| Person / Time | |
|---|---|
| Issue date: | 03/04/1986 |
| From: | Palladino N NRC COMMISSION (OCM) |
| To: | Bevill T HOUSE OF REP., APPROPRIATIONS |
| References | |
| NUDOCS 8603130132 | |
| Download: ML19301F171 (33) | |
Text
L Distribution:
' f ** %
=
%'c;
/
UNITED STATES V$te))o
[j c f( }
NUCLE AR REGUL ATORY COMMISSION JRoe
~ J/
c W ASHINGT ON, D. C. 20555 TRehn
'^ k l JSniezek
/
PRabideau ED3 r/f CH AIRMAN March 4, 1986 HDenton JDavis JShea The Honorable Tom Bevill, Chairman RMinogue Subcommittee on Energy and Water Development GCunninghan Committee on Appropriations RScroggins GWKerr United States House of Representatives Washington, DC 20515 CHeltenes Regions I-V
Dear Mr. Chairman:
As requested by House Report 99-195, enclosed is (1) a report for Calendar Year 1985 on all changes to systems, components, procedures, organization or design of operating plants where such changes are imposed by the Commission or its staff, and (2) a report on steps taken to assure that alterations of licensed facilities are made only pursuant to legal requirements.
Commissioner Asselstine adds the following:
In 1985, the primary initiative of the Commission to control improvements in safety was the promulgation of the backfit rule.
I disapproved that rule for the reasons attached hereto.
The Commission claims that, without the backfit rule, the staff and the Commission could not be disciplined in developing new initiatives to improve nuclear reactor safety.
The backfit rule specifies the
" disciplining iactors" that must be satisfied before the NRC can improve safety.
However, the Commission has decided that the backfit rule does not apply to new rules or staff positions that relax safety standards. I find it curious that the Commission is willing to limit its ability to improve safety through a binding rule on itself but refuses to so limit its discretion to relax safety requirements.
In response to the above comment, the Comnission notes that the only issue is whether or not there is a need for a rulemaking on the subject of relaxation -- the Commission agrees that a disciplined process shall apply to the relaxation of NRC re qui remen t s.
The Commission has directed its staff to assure that appropriate staff processes are in place.
Sincerely, h
),
N l1 \\ c. t.
- d' r
n-~
Nunzio'~J. Palladino h+a
Enclosures:
As stated Originated:
HRR/DEDROGR
% C3/3c/3q
a i
a RE90RT ON CHANGES TO SYSTEMS, CCMPONENTS, STRUCTURES, PROCEDURES, ORGANIZATION, OR DESIGN OF GPERATING POWER PLANTS IMPOSED BY THE NRC 1.
PURPOSE The purpose of this report to Congress is to advise you of any changes to operating nuclear power plants imposed by NRC during CY 1985.
2.
BACKGROUND During CY 1985, licenses of operating reactors were continuin-g to implement NRC requirements promulgated prior to 1985-such as, those contained in the (TMI) Action Plan, rules; e.g.,
fire protection, equipment qualification, anticipated transients without scram, and other approved actions.
Staff resources that were reviewed in order to ascertain changes to systems, components, structures, procedures, organizations, and design of opercting reactors imposed by the NRC in CY 1985 were the plant specific backfitting tracking systems, final rules, generic letters, and bulletins.
The plant specific backfitting tracking system yielded no plant specific backfit information; the remaining sources yielded information as discussed below.
3.
RULES Rules that were approved and published during CY 1985 in Title 10 of the Code of Federal Regulations, Part 50, are " Analysis of Potential Pressurized Thermal Shock (PTS) Events" and
" Hydrogen Control Requirements."
Neither of these rules contains any new requirements that would cause licensees to changes systems, ccmponents, structures procedures, organiza-tions, or design of their operating reactors in CY 1985.
The rule on PTS requires only reporting of materials data and analyses of flux reduction options for all pressurized water reactors and detailed PTS risk analyses for those plants that may reach a specified material condition during plant life.
These required will form part of the basis for any necessary future plant-specific corrective actions at this time.
The total cost to the industry for data reporting and analyses was estimated to be 2.6 million dollars.
The rule on hydrogen control requirements for boiling water reactors with Mark III containments and for pressurized water reactors with ice condenser containments codified Commission intentions and actions that had been documented in transcripts and licensing proceedings beginning in 1980. Licensees with the type of containments cited above had been required to meet the provisions of the rula through prior licensing activities independent from rulemaking.
Therefore, the rule itself posed no new requirements for the affected plants.
The total cost to the industry to meet these hydrogen control requirements was estimated to be 9 million dollars.
4 3.
GENERIC LETTERS The staff issued 22 generic letters in CY 198E.
Only one of these, Generic Letter 85-02, " Staff Recommerded Actions Stemming From NRC Integrated Program for the Resolution of Unresolved Safety Issues Regarding Steam Generator lube Inteurity," was relevant to the review.
This generic letter requested informatien on a voluntary basis from ell licensees on their overall program for stem generator tube integrity and rupture mitigation.
The staff will also use this information to assess the licensees' programs and also as a basis to determine if additional regulatory actions are necessary in plant-specific instances of extensive or severe degradation.
The cost per plant to submit this information was estinated to be 15 thousand dollars.
4.
NRC BULLETINS Bulletins are normally used as vehicles to request information pertaining to inspection and enforcement activities pursuant to the provisions of Section 182a of the Atonic Energy Act of 1954, as amended.
The staff issued 3 bulletins which requested submittal of reports on licensee actions during the latter part of CY 1985:
Inspection ano Enforcemen+ Calletin (IEB) 85-01, " Steam Binding of Auxiliary Feedwater Pumps," IEB 85-02, "Undervoltage Trip Attachments of Westinghouse DB-50 Type Reactor Trip Breakers," and IEB 85-03 " Motor Operated Valve Common-Mode Failures During Plant Transients Due to Improper Switch Settings."
IEB 85-01 recuested the developnent and implementation of procedures for monitoring certain system conditions.
This was estimated at a total cost for all affected plants of eight hundred thousand dollars for this one-time effort.
Continuing ef'scts on these modifications were estimated ft: air affected plante, at a total cost of three nundred thousard dollars per year.
IEB P5-01 is intended to assure that auxiliary feedwater (AFW) systers satisfy the requirements of 10 CFR Part 50, Appendix A, General Design Criteria 34 and 44.
IEB E5-02 requested reporting of actions taken to address bulletin concerns and testing of plant equipment.
The total cost of reporting for all affected plants was estinated to be seventeen thousand dollars; the total cost of testing for all affected plants was estimated to be thirteen thousand dollers.
IEB 65-07 is intended to assure that reactor trip breakers satisfy the requirements of 10 CFR Part 50, Appendix A, General Design Criteria 21, 22 and 29.
IEB 85-03 requested determination of certain setpoints and implementation of these setpoints along with confirmatory functional testing.
The total industry cost including dose. labor and averted property damage was estimated at 1.9 million dollars.
IEB 85-03 is intended to implement 10 CFR Part 50, Section 50.55a(g).
L 5.
NRC ORDERS The NRC issued an Order to the Tennessee Valley Authority relating to procedures and training for ersuring that potential safety significant issues are brought promptly to the attention of manager:ent, are promptly evaluated and, if required, corrected (50 FR 26066, June 24, 1985).
This Order was intended to assure proper compliance with reouirements of 10 CFR Part 21 and IC CFR Part 50, Section 50.72.
NRC INITIATIVES TO CONTROL THE ALTERATION OF SAFETY FE E_ MES ON LIED TE G.CILITIES The purpose of this report to Congress is t<
scribe what steps the Fuclear i
Regulatory Commission (NPC) has taken 10 assu e that alterations to safety features of licensed facilities (backtitsj are made only pursuant to legal requirements. The report includes, among other things, procedures to be used, management control mechanisms and speci#ication c# management accountability.
The report is divided into four sections:
(1) Packground, (?) Revision of the Backfitting Pule, (3) Control of Generic Eackfitting, and (4) Control of Plant-Specific Backfitting.
1.
Backgruund As a result of the large volume of new requirements imp'osed en the nuclear industry following the accident at Three Mile Island, Unit 2, the NPC recognized it was necessary to initiate :.ome significant changes within the agency to bring the issuance of new requirements under control and to better order and manage the Drocess by which new requirements wire to be prioritized and imposed on the industry.
Because of the impact generic requirements were having on the incustry at the time, they rec 'ived priority attention.
In October 1981, the Committee to Review Generic Pequirements (CRGD) was established te ensure that proposed requirements applicable to one or mnre classes of power reactors would significantly improve sa#ety and were cost effective to implement.
Section 3 of this report discusses control o# generic back-fitting.
In 1983, the Commission also "..
ized that the CRGR function alonr was not effective in managing a' a of backfitting that occurred on an individual plant basis in t t.
+ of routine licorsing actions and inspection activities.
Tt,
- ' embarked on a program for better management of the ciant-spet"
-kfitting protecs.
Section 4 of this report discusses control of V.ent-specific backfitting.
On September 20, 1985, the NRC publirbed a final rule revising the backfitting process for power reactors. The pur pose cf the revision to 10 CFR 50.109 is to cleerly define what bach #ittina is, what is required of the NRC sto;' when backfits are to be imposed, what exceptions there shall be to apniication nf the rule, end to identify who is respunsible to the Ccamiss10n for proper application of the rule to regulatory practice.
The effective d6te of the rule was October 21, 1985.
Because the recent revision to the backfitting rule is app ~icable to Doth generic and plant-specific backfitting and essentially cedifies termission backfit management initiativer developed over the previous 5 years, the rule will be discussed first in Section 2 of this report.
s 2_
It should be emphasized that some backfitt ng of licensees is a necessary i
part of the regulatory process to ensure puolic health and safety.
To accomplish safety objectives, however, backfitting must be conducted in a formal and disciplined manner. The progranc described in this report focus on ensuring that any proposed changes effecting power reactor licensees are subjected to a thorough regulatory analysis and review by senior NRC management prior to being communicated to licensees.
The re-vised backfitting rule and programs fnr controllina generic and plant-specific backfits described in this report are not intended to ensure that changes are imposed only by formal regulation or order, rather they are ir. tended to ensure that changes provide a substantial increase in safety, that the cost of implementation is justified in light of the safety benefit gained, and that the changes are approved by senior NRC manage-ment. Each of the methods by which new staff positions are communicated to licenset:
~s subjected to either the generic or plant-specific review process fi..
Once a new position is approved for implementation, it may be commun cacea to the licensee by means other than an nrder (e.g.,
generic ietter, bulletin, confirmatory action letter).
If the licensae agrees to imple ment the change or proposer an acceptable alternutive, ro order is needed.
If the licensoe does not agree to comply with the approved position or does not provide a satisfactory alternative, the staft would then impose the requirement by order.
The NRC has commitced the necessary resources to ersure the backfit man-agement program is carried out. During the past year, the NRC has held seninars on backfit management in all major NRC offices ir, Washington and at the five NRC regional o'? ices.
These seminars instructed staff members at all levels in the principles and details of executing the program.
Such seminars will be repeated periodically as necessary.
During early 1986, similar seminars will be conducted with utility staff and managoment to ensure that they understand the backfitting rule and the staff proce-dures governing its implementation.
An agencywide single data base has bcen established, accessible by computer terminal in each office location, to allow ready access to current and accurate information concerning the status of all ongoing backfit actions.
In addition, conformanca with the backfit policies and procedures has been added as an element in the set of performance standards by which key NRC managers are rated in annual per-formance appraisals.
2.
Revision of Backfitting Rule (10 CFR 50.109)
The backfitting rule applies to changes intended to be imposed on only one plant (plant-specific) as well as changes intended to be imposed on all clants or a class or classes of plants (generic).
Backfittino is defined in the rule as the modification of or addition to systems, structures, congonents, or design of a facility; or the design approval or manufacturing license for a facility; or the procedures or organization required to c'esign, construct or operate a facility; any of which may result from a new or amended provision in the Commissinn rules or the imposition of a regulatory staff position interpreting the Commission
f.
rules that is either new or different from a previously appliceble staff position after certain licensing milestones have been achieved.
The licensing milestones specified in the rule are as follows:
(a) After the date of issuance of the construction permit for the facility (for facilities having construction permits issued after October 21, 1985); or (b) Six nonths before the date nf docketing of the operating license application for the facility (for facilities having construction permits issued before October 21, 1985); or (c) The da.e of issuance of the operating license for the facility (for facilities having operating licenses); or (d) The date of issuance of the design approval for design approval /
licenses issued under 10 CFR 50 Appendix M, N or 0 (i.e., Manufacturing License, Duplicate Design Approval or Reference Design Approval, respectively).
The rule specifies that the Commission trill require the backfitting of a facility only when it determines, based upon a systenatic and documented analysis, that there is substantial increase in overall protection of public health and safety or the common defense and security to be derived from the backfit and that the direct and indirect costs of implementation for that facility are justified in view of this increased protection.
There are two situations specified in the rule where the backfit analysis is not required and the substantial increase in overall prntection stan-derd does not apply. These situations arise when the staff finds and declares, with appropriate documented evaluation for its finding, either:
(a) that the modification is necessary to bring the facility into compli-ante with its license or the rules or orders of the Commission, or into conformance with written commitments by the licensee; or (b) that a regulatory action is necessary to ensure that the facility noses no undue risk to the public health and safety.
In reaching a backfit determination, the Connission must consider how the backfit should be prioritized and scheduled in light of other regulatory actiors ongoing at the facility and, in addition, will consider informa-tion available concerning any of the following nine factors as part of the backfit analysis required by the rule:
(a) Statement of the specific objectives that the proposed backfit is designed to achieve; (b) General description of the activity that would be required by the licensee in order to complete the bachfit;
4.
(c) Potential change in the risk to the public from the accidental offsite release of radioactive material; (d) Potential impact on radiological exposure of facility emnlovees; (e) Installation and continuing costs associated with the backfit, in-cluding the cost of the facility downtime or the cost of construction delays; (f) The potential safety impact of changes in plant or operational com-plexity, includir.g the relationship to proposed and existing regula-tory requirements; (g) The estimated resource burden on the NRC associated with the proposed backfit and the availability of such resources; (h) The potential impact of differences in facility tyoe, design, or age on the relevancy and practicality of the proposed backfit; and (i) Whether the proposed backfit is interim or final and, if interim, the justification for inposing the proposed backfit on an interim basis.
The rule states that the Executive Director for Operations (EDC) shall be responsible for implementation of the rule and all analyses recuired by the rule shall be approved by the EDO or his desionee.
The key element of the backfit process is utilizing the reculatory analysis to arrive at a decision about what to do with a proposed backfit.
This entails deciding on whether nr rot there is a " substantial increase in protection" and on whether or not the increased protection is iustified 4r light o# the expected costs.
Ultimately, both are matters of informed jt dgment by the senior NRC officials empowered to mee tne ba9 fit decisions.
3.
Control of Generic Backfitting In October 1981, the Commission approved the establishment of the Committee to Review Generic Requirements (CRGR) and the position of Deputy Executive Director for Regional Operations and Generic r quirenents n
(DEDROGR).
This Committee is chaired by DfnROGR, and consists of one senior mar.eaer from each of the six NRC Offices associated with the reculation of nuclear power plants-Members are appointed by the Executive Director for Operations (EDO). These members are individual contributors and not office representatives. The CRGR was formed to manage the imposition cf fornal, approved generic requirements and eliminate the impositien of informal generic requirements that had been issued at a low level in MPC and not approved by nanagement.
The CRGR has the responsibility tn review and recommend to the EDO approval or disar-oval of proposed requiremenis or staff positions to be imposed by o NRC staff on one or nore classes of power reactors. The imple-mentati.sn of this responsibility is conducted in such a manner so as to assar ? the provisions of 10 CFR 50.109 as pertaining to generic re-quiremern s and staff positions are implemented by the staff. Additionally, the CRGR ' unctions so as to eliminate or remove any unnecessary burdens placed on licensees, reduce the exposure of workers to radiation in imple-menting some of these requirements, and conserve NPC resources, while at the same time assurino adequate protection of public health and safety.
The controls ensure that requirements in place or to be issued (a) do in fact contribute effectively and significantly to the health and safety of the public, and (b) do lead to utilization of both NRC and licensee resources in as optimal a fashion as possible.
By having the Committee submit recommendations directly to the EDO, a singla agencywide point of control is provided.
In addition, while the CRGR focuses primarily on proposed new requirements erd staff positions, it also reviews selected existing requirements and staff positions which may place unnecessary burdens on licensee or agency resources.
The Charter of the CRGR was initially approved by the Commission on June 16, 1982, and changes to the charter have been and continue to be approved by the Commission.
In addition to the controls described above, the charter directs that each NRC office develop internal procedures to assure that the following policy requirements regarding reactor licensees are carried out:
(a) All proposed generic requirenents are submitted for CRGR review.
(b) All documents, letter and communicatians that establich, reflect or interpret NRC staff positions or reqcirements are submitted for re-view by CRGR unless these documents refer enly to approved require-nents or staff pocitions.
(c)
For all other communications with licensees, no statements are to be used which might suggest new or revised generic requirements, staff positions, guidance or recommendations unless the provisionc of the CRGR Charter have been met.
(d) For those rare instances where it is judgad that an emergency action is needed to protect the health and safety of the public, no review by the CRGR is necessary. However, the Chcirnan of the CRGR, will be notified by the office originating the action.
These emergency acticas are reported to the Connission.
Over the past 4 years, the Committee has conducted ovw 30 meetings and reviewed such issues as: severe accident policy, emergency response capa-bility, inadequate core cooling instrumentation, safety goals, operator steffing rule, Salem ATWS task force recommendations, ATWS rule, pressurized thermal shock, the ethod for prioritization of generic issues, insider rule, and WR pipe cracking.
Some of the items that are on the agerda for future consideration are:
quality assurance, reporting and recordkeeping, spent fuel management, operator licensing, radiation protection, shutdown decay heat rennval, containment leak testing, and technical specifications.
Since 1981, the CRGD hu been a catalyst 0r better management oversight and participation by JRC management in the backfitting of power reactors.
The rate of issuance of new generic reovirements has been reduced. The CRGR has been accepted by the NRC staff as a reality ard has the firm backing of the EDO, ACRS and the Commission.
Industry attitude towards the CRGR also seems to be qenerally positive.
4 Control of Plant-Specific Backfitting In June 1983, the Commission recognized the need to address i.et only generic backfits controlled by the CRGR but to control piant-specific backfits as well and directed the staff to develop specific procedures for these situations.
Thus, in April 1984, NRC Manual Chapter 0514, staff policy and procedures for the management of plant-specif'c backfitting of nuclear power plants, was developed and published in the Federal Register.
The manual chapter is expressly intended to prcvide a machanism to deal with backfit matters unique to single plants. These matters arise mainly from routine and special inspection activities and routine licensing actions such as periodic reviews associated with licensees' requests for technical specification changes or applic6tions for approval of core reloads.
Since the manual chapter was issued prior to the publication of the final back-fitting rule, the chapter was subseouently revised to incorporate the soecific provisions of the backfit rule. The revised manual chapter is presently being reviewed by the Commission.
The manual chapter defines the objectives, authorities, responsibilities, and establishes the basic requirements for cctions to be taken where the NRC staff considers inposing a new plant-specific regulatory staff posi-tion or e nuclear power plant licensee. The management process is intended to provide a disciplined NRC review of new or changed positions that may be imposed at plants that have already implemented or otherwise invested heavi?y in meeting previously applicable requirements or positions which were considered by the staff to provide acceptable levels of safety.
The plant-specific backfit management process described in the manual chapter will enhance regulatory stability by assuring that changes in regulatory staff positions do in fact provide substantial additional safety protection and at the same time result in optimal use of NRC end licensee resources. The basis for a plant-specific backfit decisico is similar to that being applied by the CRGR for generic backfits.
Key elements of the plant-specific process include:
(a) A clear directive tc the staff that the NRC is responsible for iden-tifying potential plant-specific backfits when the staff position meets the "backfit" definitirn in 10 CFR 50.109.
_7_
(b) A requirement that proposed plant-speci#ic backfits are communicated to the licensee only after the regulatory analysis racuired by 10 CFR 50.109 is completed and approved by the appropriate NRC Program Office Director or Regional Administrator.
(c) A recognition that a licensee may claim that an NRC-imoosed position is a backfit if it was not identified as such by the NRC.
(d) An appeal procedure that licensees may use to obtain reconsideration of NRC decisions under the backfit program. Appeals focus on whether or not safetv is substantially improved at a reasonable cost, as described in the ragulatory analysis.
The EDO is responsible to the Commission for plant-specific backfit actions.
NRC Program Office Directors and Regional Administrators have the authority to review cases and make decisions for the agency on individual actions.
Under the program, the Director, Regional Operations and Generic Requirements (ROGR) Staff is responsible to the EDO for assuring that process controls for overall agency management of the program are developed and implemented.
These include: assuring plant-specific procedures are developed by each office and region associated with reactor licensing issues, training of NRC staff, developing and implementing a progress monitorir,g system, and obtaining and evaluating both staff and industry views on the conduct of the plant-specific backfit program.
SEPAPATE VIEWS OF CCMMISSIONER ASSELSTINE In adcptirg this backfitting rule, the Commission continues its inexorable march down the path tcward non-regulation cf the reclear indus-t ry.
In two previcus decisicns the Commission fcunc acceptable the present level of risk of a severe accident at the most highly pepriated site for an operating nuclear plant in this country.
See, Consolidated Edison Comcany of New York (Indian Point, Unit No. 2), CLI 85-6, 21 NRC 1043 (19E5).
The commission's decision was made withcut an adequate explaratico cr raticrale; it was nade without an adequate analysis of the issues; and it was made by ignoring the enortcus uncertainties in our tretbeds fcr estinat re risks. The Commission then extended that decision to all i
nuclear plants through its Severe Accident Policy Staterent.
Ece, "Folicy Staterert er revere Reactor Accidents Regarding Future Designs and Existing Plants," 50 Fed. _P;epq. 32138 ( August 8,19E5). This backfitting rule is e
arother part of the Conmissicn's withdrawal from active regulatirr cf the industry.
The Commission's rule in effect says that t.ucleer reector risks are so acceptable and so well understood that the burden of proof for icwering the risk to the public must be placed on the prcpcr.cnt rf irpreved safety even if that proponent is tre Ccmission itself. This optimistic view of the risks posed by r.uclear power plants is unjustin d.
The Cermission's e
adopticr. of this rule 's truly an unprecedented step in the arenals of regulation.
I can think of no other instance in which e recuietory agency has been so eager to stymie its cwn ability to carry cut its
e responsibilities.
Indeed, the adoption of this rule is the nost corpelling evidence to date of the Commission majority's open hcstility te tre regulatory mission of 'his agency.
I do not believe that there is e reed for a formal Ccar,ission regula-tion restrictir.g the Connission's ability te require safety improvements for nuclear power reacters.
My opinion is shared by cur rea'. tor safety technical staff, including the Connission's chie# safety of#icer. The Commission's purported reeser for promulgating this rule is to add " disci-pline" to the backfitting process. The Commissier can add discipline to the backfitting prccess without at the same time unnecessarily limiting its own discretion to impose new safety requirenents by fettering itself with this rule.
Further, by adding layer upcn layer of procedures to the backfitting process the Centission has created a lawyer's partdise in which litigatier ever procedural irregularities may hamper the Commission's ability Ic irpose needed safety recuirercr.ts.
Even if I felt that a backf.: rul t vere appropriate, I would not support a rule as poorly thcught out as is this one. This rule sets a threshold standard for improvements to scfety v bich is much too high given cur present understanJing of risk erd the uncertainties associated with cur methods of estimating risk.
Further, tFe factors to be considered in determining whether a backfit should be irposed are skewed against imposing new requirements.
In additice, tre Crnaission's determinatien of risk in the cost-benefit balance is to be based er urreliable risk analyses.
The consequer.ce of this rule is tn lirit the NF.C staff's anc even the Commission's ability to idertify erd correct safety weakresses at the nuclear pcwer plants in cperation and urter ccr.struction in this country.
As a result, these weaknesses are likely to persist until they cause sericus operating events or acciderts which pose a direct threat to the health and safety of the public. This rule, then, further ensures a continuation of the piecemeal, reactive approach to safety which has been responsible for many of the fai'ures cf the past.
By this step, the Cornission is moving in tne wronc dirertion -- a direction that will likely result 4-Turther serious operating everts, more accidents, and a lower level of safety than thtt achieved in rary rcre forward-thinking countrias in the world.
I discuss each cf ry reasons for opposing this rule in more detail below.
The Nature of the Eeckfitting Froblen When asked to descrite the backfitting problem, most of our licensees point to the substantial number of hardware nodi 34cetiers, procedural changes and human factors inprovements which have been required by the NRC in recent years.
The bulk of these new requirerents can be traced to three sources:
the Cammissior's fire protection rule; its rule requiring the environmental qualification of electrical ec.uipment; and the Commissicr's response to the Three Mile Islerd accident.
?.t is worth noting that each of these broad safety initiatives res edepted by the Commission itself in response to the identification of significant areas of safety vulnerability within the industry.
i
/
- /
Typicall'y, the industry does not challenge the need for improvererts in fire protectica or the neet to assure that safety-relcted electrical ecuipment will.: able to function under serious accident conditions. fer tces the industrj deny the need to address the numerous safety weakresses brcught to light by tSe Three Mile Island accident.
Rather, the industry largely focuses its criticisr.s on the process used to translate th,se broad areas of needed irprovenent into specific modifications to plant hardware, procedures and cperatier.s.
The industry raises five specific complaints.
First, our licensees trgue that new requirements ofter fail te define clearly what is expected of the industry. Second, they centend that the implenentation of these di ectives are translated requirements -- the process by which more generel r
into specific codi'icetions -- is not well managed.
In support of this argument, the incustry points to scre past failures ir decurenting proposed modifications, ir ersuring consistency in rakinc piant-specific implementaticn cecisions, in providing effective meragar.ent oversight of plant-speci#ic decisions, and in providinc a fair opportunity to appeal cbjectionable staff-proposed modifications. Third, our litersecs assert thct specific plant modificaticns are proposed by staff members who have a single narrow area of interest, ano little consideration is given to the cverall safety impact of the prcposed change.
fourth, the industry argues that the staff's implementation process all too often fails to provide a final decision from the staff en the adequacy of the licensee's ef#rrts to corpl," with a requirement until after the licensee has made the rodification to its plant. This prccess, they contend, exposes the
_/
-E-industry to s'econd-guessing by the NRC staff and sometimes lecds te rekire repeated redifications to address the same problem.
Firally, the licensees argue that the Commission sometimes adopts arbitrary cod urrealistic deadlines ft. tre inglenentation of new requirements. More than anything else, these ccrplaints focus on the management of the he.ckfitting process.
In the literally ttcusands of backfitting decisions made by the fJRC over the past several years I am sure that some exarplcs can be found to support eech c' these complaints. However, I believe tb3t each of the industry's valid corcerrs is adcressed by the administrative beckfitting manecerert irprovements already adopted by the Cornission.
Why A Rule?
The Conrissic.n clcins that this rule is necessary to provide a " nan-agement process" for the adoption of Commission regulatirrs and staff interpretations cf these regulations.
Implicit in the tcmmission's expla-nation of this rule is a feeling on the part of the Crrrission that it is necessary to add discipline to the backfitting process.
The Ccmnission never clearly explains why if added discipline is recessary they nust enbed that discipline or managerent process in a formal agency rule. The Cemnis-sien also refuses to explain the connectier tetween discipline and the recessity for a substantici thresho'd that must be surmounted before safety can be improved.
Further, the Cenmitsicr never adecuately explains why it cannot accceplish improved management of the tccLfitting process without at
.l
-E-the same time limiting its cwn discretion to reauire safcty improzenents by requiring that Ccnmission rules comply not erly with the Acministrative Procedure Act, but alsc with this rule.
By chcosing to accpt this rule the Cor=issicn adrits feilure.
The Commission adnits that it has been incapable of solving administratively whatever probl'ms it sees with the management c# the backfitting process.
e By adopting this rule, the Cennission says to the world that it so nis-trusts its own ability to act in a sensible mar.rer erd it so mistrusts its ability to certrel the f1C staff that it must have a formal rule which linits the Commissicn's discretion and which can be used as a bludge7n to control the staff.
The irony of this all is that the adr4ristrative actions taken by the Commission to formali:e the backfittirr prccess have already been success-ful in adding discipiire to the precess and in addressing the valid con-cerns of the industry.
A senior officici cf cre cf the utilities involved most actively in the backfit debate recently told me that wFer. his ccmpany first expressed its ccncerns about backfittirc el' it wanted was to get NRC management to pay attentien to backfitting problens. The company simply wanted a brief written statenent for each preposed beckfit describing the proposed change ard the NRC staff's reasons for recuiring the change, together wi+h the right to appeal to upper NRC marecercrt the decisions being made at the 'crer levels of tne NRC staf#. Their obiective, this official said, is already being achieved by the Crrrission's internal management directiros to the NRC staff.
'.,/
-i-Whatever "backfittirg prcblen" exists is really a managenert prcblen.
The Commissicn's statement of consideratiers erkrewledges that. And, the problem is being corrected irdependently of this rule.
Why then must this rule be promulgated without delay? Apparertly it is because the rcverent to put in place a backfit rule is much like an avalanche -- once it starts rcliing it cannot be slowed down or charced in course.
Lawyer's Paradise By embedaing the Commission's backfit maragement process in the form of a rule, the Corrnission has chosen to formalize a process which ought to be a purely internal management tccl.
Ir doing so, the Comnission has icposed upon itself a particular raregcr:ent process as a legal requirement which cannot be ignored, adapted to circumstances, or changed without once again going thrLugh formal rulemaking prccedures. The rule provides a myriad of opportunities for licensees to invoke procedural irregularities in challenging the Ccanission's rules and its backf4tting decisions. The rule lends itsei' rcadily to being used as a delaying tactic by uncoopera-tive licensees, and it has the potential for hamstringirg tFc Ccmmission's ability to impese needed safety improvenents while the legal wrangling sces en e"t en and on.
The Commission's decision to include Commission rulemakirg witFir the ccverage of the backfit rule is an excellent example of the cverpreceduralization of the backfit prccess. The Administrative Precedere Act End cases interpreting it set out requirements for rulemakinc.
/
ll
-F-Irterested pa'rties are given an opportunity to corrent en the preposed agency action before it goes into effect. The Commissien rust **cr take those ccments into account before promulgating a final rule. The courts can then review Comission action and test it for reasonableress erd raticrality. The Commission wishes to add on to these legal requirements a very high standard cr threshold the Commission must meet befcre it can inst 4tute safety improvements.
Further, the rule requires a strict cost-benefit balance, scrething the courts have not found is required by the Atenic Energy Act or the Administrative Procedure Act. Contrary to Chairman Falladino's assertion, the Commission cannot decice vbether to follow these row, nore stringent requirements in incividual rulemaking proceedings on a case-by-case basis. This regulation rcw teccmes binding on the Cecrission, and must be followed in all future rulemakings.
In addition to this, the backfit rule applies to staf# interpretations of Comissior rules.
By the rule's literal terms, any staff interpretation of a Comnissicr. rule would also have to meet the requirererts of the backfit rule. Thus, the Commission would be required to neet a high threshold and perform a cost-benefit balance for any rule it itseer. ard the staff vculd then have to again meet the same high ihreshnid and perform a new cost-benefit l'alance before it could interpret that rule. That is absurd, but that is what the rule appears to require.
Even if rulcraking were not to be included within the scope of this rule, staff interptetaticns of Comission rules would be.
This presents an intercstire dilemma. The Commission's 1965 Policy and Planning Guidance
9 states:
"The Commission intends to shif t its regulatory emphasis a;ay frrr detailed, prescriptive requirements toward performance criteria." See.
NUREG-0885 Issue 4, "U.S. I;uclear Regulatory Commission Policy erd Plann:ng Guidarce 19EE." This means that new rules will tend to be ganeral in nature.
It is difficult for safety reviewers and inspectors te review ard inspect generalities. They need to develop positions on acceptance crite-ria, hardware t'equirements, applicable quality assurarce provisions, technical specificaticrs, etc. The rule would require that these interpre-tations meet the requirements of the backfit rule. Thus, if rulemakings are outside the scope cf tFe backfit rule, but interpretations of those rules are not, it may create a situation where the staff cannot adequately interpret the rule l'ecause the interpretations would not meet the backfitting requirements, even though the Cervission's rule has been otherwise legally prcmulgated.
If these are tFe results the Conmission intended, then the Commis-sion's backfit rule nakes no sense whatsoever.
If this is not what the Commission intendcd, then the Commission should nake that absolutely clear.
Such ambiguities do nothing but provide fodar for litigaticn. These prob-lems illustrete further how poorly written and how poorly explained is this rule. These are the kinds of issues one shou'd not have to nake guesses about.
It is interesting to note that my colleagues constantly complain abcut the length and over-legalization of the licensing process, often arguirg that there are too nany procedures and that the process is too forne.l.
_ 10 Yet, ti.cse sine Commissioners who want to make the licensing process as simple and informal as possible have here added layer upon layer r.' f rece_
dures to the backfittiro process.
In fact, they have added procedures far beyond those which a
" ally required, and in the prccess have added new cpportunities for litigation.
Barrier co Improved Scfety The Commission's rule sets up a threshold standard that the Commissior must meet before it can adopt new safety requirements.
Under this ru'c, the Cenrission cannot reduce the radiological risks to the public unless it first determines that a prcpcsec safety improvement provides a " substantial increase ir the everall protection of the public health and safety or the common defense and security." Section50.109(a)(3). Thus, tFe Commission creates e E4cr.ificant barrier to re;ctor safety irpreverents.
The Commission's explanation in tre statement of corsideratiens of what it ' ears by " substantial increase" is sc unclear as to be useless.
However, an indication of what the Commi ssico really interds ctr be fcurd in the Comnission's recent indian Point decision.
21 tiRC 1043.
In that proceeding, the Conmission's technical staff and Liter. sir.g Prard urged the Commission to require a set of safety improvements fnr the reactors sited at the most highly populated locations in this cour.try.
I!ren learning that those improvements erly cut the risks to the public in half, the Comnission rejected them as not providing a "substant'.ti" ircrease in protection.
21 NRC 1043.
I would' prefer a standard which does rnt set such a high threshold for the imposition of safety inprcvererts.
In fact, I proposed sutb a stan-dard.
Under my prcposed standard the Agency wculd require improved safety upon a determination tFst e prcresed reasure provides a net ircrease in the protection of the public health and safety ar.d that the costs of this improvement are not inccrrcrsurate with the increased prctecticr.
This standard wculd allow more imnrovements in sefety than would the Commis-sion's standard but would still exclude proposed changes which would result in only trivial safety improvenents. The Commission rejected this standard because it dces rct present a high enough barrier to block new safety requi rements. Apparently, the Comission is interested ir rcre than simply bringirg discipline to the backfirting process.
Rather, it is re":lly int. rested in tying its cwn and the staff's hands to restrict the rumber of safety irprevements.
If the Commissinn really cnly wanted discipline and scund thinking to be brought to the backfi+ ting process, it wcult' ret fcci the re e to prcoose such a stringent threshcid for safety improvements.
Tipping the Scale Against Safety The ruie specifies nine factors that "ere tn be used as balancing riechanisms in the decisierrakir,g process for backfitting".
See, Section 50.1C9(c) for list of factors.
If one cuts thrruch the extraneous ratter in that section of tFe ru'e, cne finds that the Commissicn recuires cost-benefit analyses to be performed on all prrpesals for backfits.
Of course, in cost-benefit ert. lyses the bottom line depends oa what facters cre chocses to put on the scale.
t.
f40t satidied that a high threshold standard will sufficiertly lirit the number of backfits, the Conmission has also decided ^.] stack the cost-benefit balance. The only benefit the Commission is able te icentify as being appropriately considered in decisiens on whether safety should te improved is the " potential change in the risk to the oublic frer the accidertel cff-site release of " iicactive material." Section 50.109(c)(3).
Risk is typically defined as the probability cf an accident multiplied by tFe consequences, with the latter expressed as the collective dose to the public (person-rem). However, even here the Agency's typical pre.ctice is to ignore societal deses beycnd a 50 mile radius. As calcula-tiers cf accident consequences indicate, this procedure captures less than half of the health censequences of core meltdown acc; dents.
The Commission refuses to include among the " balancing facters" the averted ccsts of off-site property damage resulting from radiological releases. The Commission dces not seem to realize that core -altdcwn acciderts cer certaminate off-site property to hazardous radiation levels and that there is a real benefit in preventing that from occurrirg.
Avertirg tre recessity to decontarinate such prcperty is a real benefit cf backfits which lessen the likelihocd of off-site releases of radicective rate rial s.
Since these costs in scme instances substantially exceed tre ncretired value of averted health effects resulting from accidents, the Ccmmission has no defensible basis for omitting from the "balar,cing fac-tors" eff-site property decontamination costs.
13 -
The Cordission also rejects the inclusion of the benefits derived from averting darage to the plant itself.
The TMI-E accident, which apparently did not result in extensive melting of the reactor core or substential offsite releeses of radioactivity, resulted in billions of dollars in plant danege, plant clean-up and power replacement costs.
The Comissior's rc'e feils to recognize that preventing such costs has a public benefit. Tte Cerrissicn chooses to ignore averted replacement power costs associated with safety inprovenents that prevent accidents. However, in order to inflate the costs side of the equation which weighs against backfitting, the rule requires corsideration of the replacemant power costs fer the facility downtime associated with implenenting a backfit.
/.t tFe sere time the Comission ignores the benefits of backfits, the Ccnnission tries to include every conceivable " cost" of the backfit in the be'erce.
The rule includes costs such as installation and other costs rsscciated with physically changing the plant; the cost of facility dev:n-tire, e.g. replacement power costs, etc; the cost of construction delay; erd, radiological impact on facility enployees. The Conmission has ever thrown the cost to the f;RC (resource burden on the IGC) into the balcrce.
Obviously this stacking of the deck against safety improvements indicates once again that the Comission is interested in more than just edding disciplire to the backfitting process.
The Commission majcrity tries te argue that the balance of costs and benefits is not slanted because other benefits beycnd those enumerated in
. the rule can le considered. Their own actions ccotredict this argument.
In adopting this rule, the Commission majority expressly rejected prcposals to include additional health and safety and economic benefits cf prcposed back#its that would have resulted in a fair and even-handed censideraticr, of all relevant costs and benefits.
Given its own actions, the true inter +
of the Commission majority is beyond doubt.
Reliarce cr. Indefensible Analyses The Commission's rule places great reliance on Probabilistic Pisk Aralyses (PRA's).
In determining the " change in risk" as required by this rule, the Commission intends to rely on the bottom-line results cf PRA's.
Unfortunately, numbers praduced by these analyses amount really to only esu, mated guesses; yet, the Commission intends to rely bcevily on these numbers, which rearly all PRA practitioners agree are unreliable, in determining whether to require improvements in safety
' reparation of a PRA requires that the analyst calculate the cere meltdcwr probability. Given a particular ccre reltdown scenario, the analyst must then calculate the containment failure mode, the cuentity cf radioactive fission products released from the containment (the "scurce term"), the dispersion of the fission products in the atmosphere and firally the radiation doses to the public.
- he calculated probabilities frer all of the above are multiplied by the aggregate doses to the public.
This is the risk to the public.
To calculate the change in risk, as raquired by this rule, the analyst must first calculate the risk to the putlic before a proposed sa'ety improvement is implemented, ard then calculate the risk assuming tFe improvement is made.
Unfnrtunately the recessary calculations carrct be made based cn data, and scientifically accepted principles and methodolo-gies. Because of major inadequacies in the data base, because cf the vast complexity of nuclear plants, because e trcmendous number of assumptions must be made, because all centributors to risk cannot be quantified and because core mel:dewn phenomena are poorly understood, no one calculation of risk yields a remote'y rearingful value of risk.
I discussed the meaningless nature of these risk estimates in more detail in my separate views on the Severe Accidtrt Prlicy Statement.
Our experierce with the Davis Besse plant provides an excellent example of the inadequacies of PRA's for truly predicting risk.
It also illustrates tFe shortcomings of a system which relies heavily on strict cost-benefit balarces for making decisions on safety irprevements. The Davis Pesse plant has one of the most (if not the most) unreliable emergency feedwater systems (EFS) of any nuclear plant in this crurtry.
The NPC staff has been trying to require Cavis Besse to upgrade its EFS reliability. However, for the last several years, the licenste tes been using reliability and cost-benefit analyses to argue that substantial upgrades should not be required. Two independent reliability eralyses (one by the utility erd tre by the NRC staff) were performec on the EFS at Davis Besse.
The results of these two studies differed by c facter of 100 in their estimate of the reliability of the systems.
The studies also
~
differedenEhatwasthemostcost-effectivewaytoupgradethesystem.
The utility argued that its ccst-benefit analyses shcwcd thet erly some low-cost mirer changes were justifiable while the staff argued that its cost-benefit analyses supported more significant rcdifications.
Because the Commission-recuired ccst-benefit analyses could not demonstrate the necessity of a particular way to upgrade the EFS reliability, the steff could not require a substantial upgrade of that system even though the plant continued to operate with an unrelicble but crucial safety system during the several ycars of the PRA debate between the staff and the utility.
The Jur.e 9,1985 Davis Besse event demcrstrated that the PRA analyses were wrong. Davis Besse had t icss of all feedwater that involved the failure of 14 separate pieces of equipment.
(See. NUREG-1154, " Loss of f>ain and Auxiliary Feedwcter Event at the Davis-Besse Plant on June 9, 1985"). The event led the Agency's chief safety efficer to observe:
"I believe that the recert Davis-Besse event illustrates that, in the real world, system and component reliabilities ctr decrade below those we and the industry routinely assure in estimating core melt frequencies."
(See memorandum from Harold R. Denton to 1.'illiar 1 Dircks, dated June 27, 1985).
Further, it appears that the steam and feedwater rupture control system had a significant role in causing the 1 css of emergency feedwater.
Yet that system was not ever ircluded as a possible contributor to the unreliability of the emergency feedwater system in either of the independent reliability studies. Despite this clear evidence of the weaknesses in FRA studies and their poter,tial fcr reripulation and
distortion, the Commission persists in using them erd in requiring their use by the staff in this rule as the basis for deciding on safety improvements.
Although this rule will have a negative irpact en all aspects of the Commission's reactor s2ftty activities, its effects are likely to be greatest in the area of improving human perferrarce.
Pecent operating experience indicates that rcughly half of all significant operating events can be traced to inadequate human perfcrrarce in such areas as reactor operations, surveillerce testing and maintenance. A number of the Commission's post-TMI requirements havc #ccused en human performance, but recent operating experience dererstrates the need for further improvements in this area.
Indeed, virtually ell members of the Conmission have advccated further measures to imprcve tFE qualifications, experience and training of plant personnel. Specifically, members of the Commission have spoken in faver of increasing the engineering krtwledge and skills of plant operators and requirirc the use of plant-reference sinulators for operator training and testing.
Common sense and sound engineerirc judgment tell us that such rcaseres will have a positive effect in improving plant performance. Yet, it will be especially difficult to assess how such proposed requirements will reduce the risk of a core melt accident which might result in harm to the public. Thus, the practicci effect of this rule will be to thwart the efforts of the NRC staff to develop new safety requirements in the crea of human performance where such requirements could be of the greatest safety benefit.
~
Ignoring Uncertainties The Commission also fai'.s te deal with the huge uncertainties associ-atec with the risks of nuclear reactors. TFe actual risks could be up to 100 times the value frequently picked by the Commission.
One would think that the uncertainties about the level of safety echieved at the operating reactors would have a beerirg on whether reactor safety should be improved.
I proposed that the Comnission articulate its expectaticr.s cn the handling of uncertair. ties in the Dackfitting decisicnmaking process before allowing this rule to become effective. The Connission rejected my proposal.
There is re reference in this rule to uncertainties in reactor risks or te hcw crcertainties are to be factored into safety decisions. The Commiss4cr's silence simply reaffirns its practice of ignoring the enormous uncertair-ties in r.uclear risks when deciding whether to irprove the protection of the public health and safety.
Selective Application of the P,ule The Commission's stated manner of applying this rule is also trou-bl i r.g.
First, according to the statement of Considerations, a licensee mcy request an amendment to its license and the NRC staff is not renuired tc corsider whether the amendment represents a " substantial increase in the overall protection of the public health and safety." However, if the NPC staff wants to amend a license to establish a more stringent standctd, the staff rust first demonstrate that the amencment meets that backfittirp
_ 19 stardard.
Th'us, the rule stacks the deck in favor of the industry and against the NRC staff.
But more troubling is the Commission's apparent intent to apply this backfit rule with its high threshold and cost-benefit analysis only to these new Commission requirements which are irterded to improve safety.
If one reads the Commission's rule literally, it applies to any change in Commission requirements, both a charge to trake requirements nore stringent and one to relax requirements. Furti er, the Comission states in its statement of considerations:
"(Tihe e is no intent on the part of the Commission to include within the sccre cf the except ons (to the rule) new or modified interpretations cf vibet co tstitutes no undue risk to the public health and safety.
In such a case, the ruic applies." All of this seems to indicate that the backfit rule epplies across the board to new Ccmmis-sion regulaticrs and interpretations.
However, the Comission's actions and rhctoric would seen te irdicate othervisc. The Commission has been devoting and centinues to devote considerable agency resources to relaxing the current emergercy ccre cooling regulations and the emergency planning regulations. For example, the staff is developing new and relaxed (relative to the current staff position) acceptance criteria for emergency core cooling systems that would effectively allow the licensees to increase the pcwer level of the operat-ing reactors. Likewise, the Ccmission assign? resources to work on a rule that would allow less comprehensive evacurtion planning.
Both activities involve new or modified interpretaticr.s of what constitutes compliance,
n.
involve a modified ir.terpretation of what constitutes "ro undue risk," and do not fall within any of the exemptions to the backfit rule. Thus, if cne reads the backfit rule literally, the Commission must determir.e that increasing the pcwer level of reactors and dininishirg the level of ener-gency preparedness result in a " substantial increase in the overall protec-tion of the public health and safety or conren deferse and security."
It would take quite a bit of corveluted argument to find that relaxing safety standards meets the rule's substantial ircrease reonirement. One can only conclude that either the Conmissien is vasting resources on these activi-ties or that it does not intend tc apply the backfit rule to actions which relex existing safety standards.
This problem of i.iterpretation is another example of how pccrly thought out and how pcorly written is this rule. The Commission shculd rake clear exactly what is the sccpe cf this rule, and should revise the rule ar.cordingly.
Otherwise, this apparent anbiguity once again prcduces rething but fuel for litigation.
Conclusion I might be as sanguine as is the Commissior ettut the current state of reactor safety, and I might be villing to restrict the Conmission's ability to require safety improvements if there were a cleer ur.derstanding of the level of safety a'. reedy echieved at plants and if that understanding demonstrated that tne potential for severe accidents is indeed very remote.
.. ' Unfortunately, the Ccamissier d:es not tiave a clear understarc'rg cf the level cf safety cf current reactors.
The Cct.:missicn does not know where we are c* the learning curve for reactor perforrarce, and there is a distinct possibility of cne or rcre severe accidents in the foreseeable future. Cperetirc experience indicates that a total Toss of a safety systen is not a rare event, that multiple independent failures oo occur, that there tre corporent and reliability prcblens, that operating practices are frequently deficient, and that there are a wide range of adverse systens ir.teractions. The Commission is reluctant to face these facts erd to denand improved safety because that might suggest to the public that tnc cx't+4rc reactors are unsafe ard tight hinder the further develtprent of the nuclear industry.
In my view, another severe accident may wel' tr ng te e Falt further i
development of the ruc' car industry and, if people are injured, rey jeepar-dire continued operation of existing reacters. TFe (crmission has said there is about a 50-50 cFence of another severe accident in the reyt twenty years. The Ccamission finds that risk so accepteble that it can new, through this rule, put readblocks in the way of further safety irprove-ments.
I find the Commission's actions to Le nct cr.'v erwise but harmful to the public interest and pctentially hazardous to the public health and safety.
The Ccmmission will next turn its attertier te the forth and final ection that will complete the 'ramework for deciding whether the f2C ard
C,
- ^2 -
the industry 'will pursue safety issues t.e'cre accidents occur, i.e. the Safety Goal Policy Statement. That wi' the the final cpp rtunity c ccre to grips with the pivotal isst'.- the Cerrissicn has steadfastly avoiced cver the last several years. As I wrote in my separate views en the 5etere Accident Policy Statement, it is erccertging that there appears to be an emerging consensus within the f:F.C senior technical staff and within the ACES in favor of safety improverents te reduce severe acudent risks.
Ecwever, it is disnaying that the Crrrission, having lest all sight of the broadest lessons learned from the TMT-2 accident, has chosen to hincer enhancing the protection of the public health and safety through tnis backfit rule.