ML20132B435

Forwards Proposed Paper Recommending Measures to Improve QA at Reactors.Substantial Changes Have Been Made from Previous Version Forwarded by 820706 Memo.Paper Consistent W/Commission Statements to Congress on QA

ML20132B435
Person / Time
Issue date:	08/10/1982
From:	Deyoung R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE)
To:	Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
Shared Package
ML20132B439	List: ML20132B435 ML20132B560 ML20134F558 ML20134F586 ML20134F600 ML20134F618 ML20134F629 ML20134F644
References
FOIA-84-293 NUDOCS 8209150329
Download: ML20132B435 (46)

Text

{{#Wiki_filter:__ __ 1 ,k( 1 , ~. AUG 1 L M2 m ~- \\ MEMOR'ANDUM FOR: William J. Dircks Executive Director for Operations FROM: Richard C. DeYoung, Director Office of Inspection and Enfcrcement

SUBJECT:

QUALITY ASSURANCE Our proposed paper recomending measures to improve quality assurance at reactors is provided as Enclosure 1. Substantial changes have been made from the previous version which was forwarded by my memoiandum of July 6, 1982. The changes are highlighted belcw. 1. The problem statement has been revised. 2. An initiative for revising the construction inspecticn prograr. has been added. 3. An initiative concerninc designated representatives (FAA approach) has been added. 4. An initiative concerning qualification / certification of CA/QC personnel has been added. 5. An initiative concerning the QA planning and evaluation function (consolidation) has been^added. 6. Management seminars have been recast to indicate tha NRC will coc;erate with industry in sponsoring the seminars. 7. The rcle and importance of followup of allegations in the inspection program has been emphasized in the SECY paper. In other respects the recomendations, initiatives and information are substantially the same. The previous version had concurrence from NRR and RES. However, we have not provided the paper to those offices for concur'rence. - We have reviewed the proposed paper against the Commission's statements to ,ongress on quality assurance and found the paper to be consistent with the Ccmmission's statements. summarizes the review. The long-term review, as contemplated, would be consistent with the Ford Amendment as it has been approved in conference, including the trial programs. ' Original Signed By R. C. DeYou.ng = Richard C. DeYoung, Director Office of Inspectier and Enforcement

Enclosures:

1. Proposed QA Paper y 2. Written Statemen' to Congrest >G A j DEQA:IE E E D E 0: b FC;/Ek r.g p07/ JHj~/82 5/;)/EE / niezek DP Allisen an 5.:.'EE ile ,Q /82 S/g / / ye t A-tH-M 9, 6 M/2

[ UNITED STATES

• ! y y. [',g NUCLEAR REGULATORY COMMISSION

7. :)C

' lr wAssiNcToN. o. c. 2cs55 %.i. -% ' ' u 9..... f For: The Commission From: William J. Dircks Executive Director for Operations

Subject:

ASSURANCE OF QUALITY

Purpose:

To obtain Commission approval of recormendations to improve the assurance of cuality in the desigr. and construction of nuclear projects. Discuriion: The complexity and extent of problems that have been identified in the past few years at 5 of the 32 ur.its now under active construction have caused concerr. regarding the quality of the design and ccnstructier. of r.uclear projects. These problems include nonconforming structural steel welcs at Zimmer', s5ismic design errcrs at Diable Car. yon, inacecuate soil compaction at Midland, voids in concrete structures at Marble Hill and cesign deficiencies at South Texas. summarizes recer.: experierce at each cf these 5 projects. Analysis of the experience at problem sites has resulted in the classification of three primary problem areas: failu,re of the project ranagement team to provide acecuate t.anage ent controls to prevent a significant breakdowr. in quality frer. occurring; failure of the owner's quality assurance program to detect the breakdown in a timely manner and to obtain tnE' appropriate corrective action; and failure of the NRC's programs to recognize the true extent and nature yof the problems. The first two problem areas are fundamentally derived from a lack of total management cormitment to cuality at the nuclear projects inception. This lack of commitment has been exacerbated by the lack cf understanding of the rcle of quality assurance in project management anc the lack cf total understanding of what is recuired by personnei at all levels of the process. CONTACT: E. L. Jordan, IE 492-4545

5 B The Comission - The third problem area is two-fold in derivation. Hi stcri cal'.., the NRC's licensing and ccnstruction inspection programs have not sufficiently examinec the project management controls at sites under construction, but have been oriented towards establishing adequacy within major technical anc functional areas, e.g'., concrete, electrical, etc. The systematic assessment of management perfomance and evaluaticn of all other available infomaticn have not received the same level of effort as operating sites. Second, previous NRC programs have not addressed design quality as specifically and extensively as other areas. In response to the breakcowns ir. quality anc cuality assurance, the Chairman in a hcvember 27, 1951 menorandum directed the staff to determine various apprcaches that could be taken to strengthen cuality assurance, and to provide the Comission a preliminary evaluation of the approaches that appear mest promising. On January 29, 1982 the staff briefed the Commission on initiatives tha appeared to merit further consideration. Incustry represen:E:'ves from the Institute for Nuclear Power Operations (.'NPO) ret with the Commission on February 4,1982 to present their plans for improving the assurance of quality at piants under construction. On July 15, 19E2 the staff again briefed the Com.ission on the actions taker to date anc :he initiatives under consiceration. The staff Fas developed recomrer.dations that should lead te ' effective '.iprovements in cuality anc quality assuran'ce. programs. While many of these recommendations recuire NRC actions, the underlying principle in their development has been that the ultimate responsibility fcr cuality and safety remains with the nuclear industry, Enc none of the recommandations are intenced to transfer this responsibility to the NRC. The reccmmencations are designed to establish additional confidence in the cuality of design and constructicn activities and improve the management control of cuality. The recommendations also are cesigned to improve the NRC capability to evaluate the implementation of licensee programs. Compliance with hRC recuiremer.ts for the cuality assurance program and its implementation is a major censidera-on in establishing this cenfidence. Although a resident inspector is now assigned to every site at which constructicn is more than 15 percent comple e, the NEC is limited in its ability tc assure coc'.ience with all hRC recuirements because of the limited inspection resources. The staff recomends implemen a-'er c' a syster c# cesignatec NEC representatives (analt; cut c the U. syster' it ex ent its inspectier rescu-ces.

The Cc=ission During the development of the recomencations presented in this paper, several additional actions were consicerec. Although an adequate basis to recomend additional actions could not be established at this time, further study is warranted. The staff _recomends a long-tertr. review, utilizir.g knowledce within and outside NRC to evaluate the merits of ~ additional actions, monitor the outcome of ongoing industry and NRC initiatives and recomend chances in crocram directier as appropriate. This review would satisfy the directicn provided the NRC in an amendment acce:ted by the House anc Senate conferees in their jcint corsiceration cf the NRC's FY 82-E3 authorization bill. Some of the actions considerec anc endorsed by the staff ~ are associated with existing acency p*cgrar.s. The fellowuc of allegations is an essential part cf the NRC's inspection procram, and is an effective extensier of irspection resources. Allegations provide an c;portunity fcr non-NRC personnel to enter potential problems into the NRC's probler correction chain. The principal cbjective of the resultant NRC inspectior effort is to obtain sufficient inferratior. throuch inospenden in-depth examinations to establish the sigrificance of the particular allecation anc te effect c rrective action comen-surate with it's significance. Tc en:ourece and previce personnel an opportunity to make an ailecation, NRC ir.soectors wear hard hats that unicuely identify them. Tne resident inspector's office has a telephone answeri.g cevice for 24-hou-response to callers, and the telephcne nurbe-s c' the.re.sicer.t inspector and NRC regional office are listec in the local teie-phone directory. Effective October 12, 1952 NRC pestines at the site will identify the legal protecticn afforded people. who provide allegations. With respect to enforcement, the NRC has sufficient authority to take appropriate enforcement action for inadequate quality assurance. The options extend over a brcac range fror. meeting with a licensee, notices of violatior anc civil penalties tc issuance of orders for medi'icaticn, suspension or revocatien of licenses. The staff intencs to continue to take strong actions in respcnse te significant quality assurance breakccwns and has excressed this in:entior. in the enforcement policy. Actions have been initiated at near-term operating liter.se facilities to improve staff ccnfidence in the cuality c# design and construction activities. These acticns inciuce self evaluatier.s by licensees, arc in rcs: cases, er irds:ence-. desien review. The liritec ex:erience te cate wi:n :ne inceber. cent cesign reviews (E clcsure 4} incit.ces tre

s ~ The Gcmission 4-identification of numerous deficiencies (nonconformances with the original specifications), which have requirec reanalysis. Relatively few of the deficiencies have required hardware changes, and to date, none of the ce'iciencies identified would have prevented safety-related components, systems, or structures from performing their intended function. The initiatives recomended in this paper are directed toward reactor facilities not yet licensec for operation. At this time, the staff believes a reasonable basis fcr not backfitting these initiatives to operating reactors is provided by previous reviews c' the facilities, their operating history, extensive starttp test programs, ar.d the reviews' and upgrades in response to T": anc Euiletin actions. The staff believes that further censideration should be given to operating reactors as part of the long-term review. The staff. re.corrnencations are sumari::ed below. The staff recomends continuing those actions that are already u-derway and implementing the remaincer of the recomendations as soon as practicable. Each initiative is cescribed further in Enclosure 1.

• Measures at Near-Term Ocerating License Facilities The NF.C should continue to employ the measures currertly in use to establish ccnfidence in the quality and eff5ciiseness of utility cuality assurance procrams at near-term cperatirg license facilities ur.til other NRC cr industry procrats are capable of providing this conficence.

These measures include applicant self evaluation, independent design review and regional evaluations.

• Industry Initiatives The NRC should continue to interact with Ih?O in its development of industry initiatives, measure their effective-ness and adjust the correspondinc HEC actions te provide for effective use of both incustry and NFC resources.
• Construction Procrams The NRC shculd complete development anc 1mplement plannec revisions to enhance the effectiveress of its ccnstructic-inspection procecures.

L l The Comission ~ The NRC should ccm:lete develcpment and implement its program for construction assessment team inspections at selected facilities to provide a basis for evaluation of the management performance essential to quality construction. The NRC should cohpiete development and implement the integrated design inspection process tc assess the cuality of design activities including examination cf as-beilt configuratier, at near-ter:r operating licensee f acilities. The NRC should expand its capability to identify generic design and construction deficiencies by cor.puterized analysis of information reper.ed by vendors, construction pemit holders and NRC inspectors. The NRC should pursue revision of its statutory authority to allow implementation cf a system of designated representati analogous to the system empicyed by the Federal Aviation Administration.

• Manaaement Quality management seminars for tcp level managers with facilities under design ard constructicn shoulc be spenscred by industry.

The seminars wouid focus recent experier.ce cf selected managers and recogni2ed experts in the design are construction of nuclear prcjects. TheNRCwillrecuestthateachutilitywithafac'iiityu'ncer construction reevaluate its quality assurance program ard implement improvements in areas where the evaluations identify a need. The NRC should take actions to improve the enforcement of existing standards for cualification of quality assurance and quality control personnel and pursue establishment of a system of third party qualification and certification for such personnel. The MFC should continue to explore with labor and ctber organizations, potential methoos and incentives to assure quality in design and construction relatec productier activities.

1 The Commission

• Lona-Term Review The NRC should commence a long-term review for continuing evaluation of cuality anc quality assurance problems related to design, construction, testing and operations, and potential soTutions to those probier.s and their impact on the adequacy of NRC quality assurance policies and programs.

Expertise outside NRC shoulc be utilized by establishing an advisory panel to the NRC staff.

• Quality Assurance Planning and Evaluation The NRC should rneke creanizational realitnments to combine within a single orgarifation the functicis of research, stancards develcpment and inspection prcgram develcpment for cuality assurance at reactors.

The licensing function should remain in NRR until regionalized. Upon regionalization the program direction and assessr-en will be combined with the rest of the OA overview func-ion. The staff"has developed resource estimates and implementaticr. schedules for the new initiatives. The resource estir.ates, implementation schedules, and staf# respcrsibilities for implementation are discussed in En:'esure 2. The staff responsibilities are assignec consistent with the reccmmer.ded organizational realignment. Tne resource estir.ates tc implement the recommendec initiatives are sunmarizec beicw. ~ 1. Estimated Industry Resources 280 man years new effort reconmended in FY E3 310 man years new effert recommended in FY E4 2. Estimated NRC Staff Resources 15 staff years new effort recommended in FY 83 11 staff years new effort recommended in FY Sa 3. Estimated NRC Contractor Rescurces 52.2 million new effort recorrended in FY E3 S1.4 million new effort recenr. ended in FY E4

The Commission - Recommerdation: That the Commission approve the staff proposals as summarized above. William J. Dircks Executive Director for Operaticr.s Enciosures: 1. Initihtives 2. Resources, Schedules and Staff Responsibilities 3. Examples of Recent Ouality Assurance Problems 4 Indepencent Design Review for Near-Term Cperating License Facilities Distribution: Commissioners Commission Staff Offices Office Directors Regional Acministrators EDO ACRS ASLEF ASLAF SECY OGC OPE R.'C. DeYoung, IE J. H. Sniezek, IE E. L. Jordan, IE J. M. Taylor, IE D. Eisenhut, NRR R. Vollmer, NRR R. Mattson, NRR W..Morrison, RES T. Dorian, ELD J. Scinto, ELD V. Stello, DEDROGR D: DECA:IE ELD DD:IE D:IE EDG ELJerdan i Dorian JHSniezek RCDeYcung WJDircks El 182kie S/ 182 8/ 182 8l lE2 El /E2

~ En:lesure 1 INITIATIVES 1. Introduction The Commission has considered quality assurance to be a key facter in the cesign, construction and operation of nuclear pcwer plants for many years. Proposed versions cf the General Design Criteria usec in 1967 recognize. the interest in cuality assurance. Appercix E to 10 CFF. 50, published in June 1970, describec mandatory criteria for acceptable quality assurance procrams for safety-related features.* Subsecuer.tly, a nucer cf natier.El standards and regulatory guides provicing additional guicance have been issued to upgrade quality assurance programs. In the 1973 time frame, the Atomic Energy Comission expended major effort to cormunicate to industry the framework, e.g., plans, protecures, orgar.ization, of a cuality assurance program that would be acceptable to AEC. This framework is reflectec in current quality assurance programs that have been approved by NRC. II. Problem Statement Examination of the problems,that have been icentified recently indicates that the fundamental cause of most desicr. and ccnstructier. deficiencies is the lack of total management comitment to quality. This lack of comitment has been intensified by the lack of uncerstandinc of the role cf cuality assurance in project manugement anc the lack of totai uncerstanding cf what is required of personnel at all levels of the process. The owner's project management team is responsible fer the everall planning and management of the design, construction, and testing c' the nuclear, pcwer plant. If the senior management has a strong ccmmitment to quality, anc if that commitment is imbuec in a capable prcject management team, then the subsequent actions of this team will corrunicate that ccmitment to all involved parties. The project mar.agerer,t team com unicates and "As useo in this paper and defined in Appendix E, quality assurance comprises all those planned and systematic actions necessary to provide acequate confidence that a structure, system, or component will perform satisfactorily in service. Quality assurance includes quality control, wnich ccmprises these c,uality assurance actions relatec to the physical characteristics of a material, structure, component, or system which prcvide a means to control the cuality cf the material, structure, component, or system tc preceterminec recuiremer.ts.

0 . EnCicsure 1 obtains through contractural and procedural arrangerents with the desicners, fabricators, and constructors a level of cuality comensurate with the owner's comitment. The comitment to cost and schedule must be prcperly balanced with quality through these contractual and procedural arrancements. For example, if the constructor earns contractual credit strictly with the schedule of physical installation, the message frot project management is production. On the other hand, if earned credit is commensurate with the schedule of owner accepted, acecuately dccumentec instaliation, the message is quality production. The latter case previces the procer incentive for getting work accomplished right the first tire. This is then reflected in the policy and procedural direction tc the various organization sub-tiers. Similarly, the role of cuality assurance in the ; reject maragemen: tear. is determined by the senior management's commi:mer; tc cuality. Proper ' implementation of the quality assurance criteria is an impcr: ant elemen; in successful project management. However, cua'ity assurance prcgrars cannot substitute for poor project managemen cr a lack cf ccmitment :c quality. Quality assurance must be an integral cart of ar c' the p-cjec: planning and management activities from the prcjects ince;; ion, and its role must be comunicatec and fully unders: coa by ali participants in the design and construction process (frc serier manacerer: to the craftsman). For example, if the inspection function is planned anc ccncuctec as an integral part of physical installation activities, then early cetec:icr and correction of procedural or ciner inacecuacies will result in er5ancirc quality, cost, and schedule. All participants.nust be acecuately trainec to understanc and obtain these benefits. Weaknesses' in the existing approach to assurinc cuali y are appareht.' They are evidenced by the frequency anc severi:3 of design anc~ constructicr deficiencies, and by the failure or delay cf indus;ry anc NRC reccenition of the extent and nature of the breakdowns. Previous efforts by the NRC to assure prcgram content and structure have not been belanced with comparable efforts to assure successful procram implementation. The NRC's licensing and inspection procrans have not sufficiently examined the project management controls at sites under construction, but have been oriented towards assurine the adecuacy witnir major technical and functional areas, e.g., concrete, electrical, etc. Tne systematic assessment of management performance and evaluation of al' available information at construction facilities cic nct receive the same level cf effort as operating sites. Previous hRC orograts have nc: adcressed desicn quality as specifically and extensively as other areas. In sum the fundamental issues can best be characterized as the lack c# and NRC's'gement commitnent to cuality and the urtertainty in industry's total mana ability to cetect and ccrrect the resultinc ceficiencies. 'he

r. sed to resoive these issues is ne basis fer the followi ; recorrer.ca-ic"s.

O ' Enclosure 1 III. Initiatives A. Measures at Near-Tere Ocerating License Facilities For those plants in the Near-Term Operating License (NTOL) status, the NRC has implemented three interim measures to provide accitional confidence that required quality assurar.ce programs have been successfull. implemented and ccmpleted curing the design and construction c# the nuclear facility. These measures will be continued until replaced by adequate industry programs or permanent charges in the present NRC program. 1. Self Evaluation An applicant for an operating license will perfert a comprehen-sive self evaluation of the effectiveness of the quality assur-ance program for design and constructicn. This recuires an overall description of the project's cuality assurance program for design and constructior.. The self evaluation is a survey o' the overall quaiity assurance program. The survey will describe + the development and histcry of the program, management involve-ment, audits, reviews, significant problems and corrective actions. The NRC staff reviews the sel' evaluatior, anc provides the results of its review te the licensee. Accitional work, such as corrective actions or further audits, ray be required in particular areas. In addition, the Chief Executive Officer or his designee is requirec tc certify that the facility has been designed, constructed,aretestedinacccrdancewitht;ne, Final Safety Analysis Repcrt and ciner licensing commitments. 2. Regional Evaluation On each new operating license, the NRC staff censiders whether there is a need for additional inspections of selected areas based on an evaluation of the project's inspection and enforce-ment history. This assures consideration of the need for a better assessment of perfermance in potentially weak areas. Tne project's inspection and enforcement history is evaluated with particular attention tc the significent problems that have beer noted at other construction sites. Other information corsideret includes known problem areas, results of NRC inspections anc tne Systematic Assessment of Litersee Performance program, and problems notec elsewhere with the same contractors. Accitionai inspections are performec as warrantec in potentially weak } areas.

o 4-Enciesure 1 3. Inde endent Desian Review Based upon results of the self evaluatior anc regiona'. evaluatior., an applicant for an operating license may be requestet te base an independent design review concutted. The criteria for determining which facility; and the sccps and extent cf the design review also incluce the cctbined nu: lear experience cf the licensee, architect-engineer, anc cor ractors. Tne review provides an evaluation of the cuality cf cesign basec on a detaine,. examination ci a smaii sampie. ine stat; spec;;;es a sample area apprcpriate to the particular project. Fcr LaSaile, the mechanical and structural loacs en the residual hea remcvai system under blowdown and operatinc basis ear:hcuake tendititns were specifiec. The incepencen revie, a:: esses prc;racr,atic areas, e.g., classificatien of systers and :om: nents, cesigt and verification records, interface con:rci and 'r.tercisciplinary review, consistency with FSAR, nonconferr.Entes anc cor ective acticns, anc audit findings and rescie:i:ns. The review includes verification of specific design fea ures by ince: enc'e. calcula #:ns and comparison of installations agains as-buii crawings. Tne NRC staff reviews the selection cf the in:erenden. *eview organization and the plan before impleme-a icn, auci s ne wcrk in procress, and reviews the results. E. Ir.custry Initiative The industry initiative is nct ar. NRC sta## ; :::sai. but a prograr T'e Nr.: s a: is m:n;- that.the industry is presently cevelepinc. toring this procram in order to take bes acvar.: ace c :ne ;n:cs: y-efforts. The Institute for Nuclear Pcwer Operations (:NFO) is ces e'c;ir; criteria which will be used ic evaluate cuality assurance for cesign and construction. As with the existing INPC criteria for plan: operation, they will be based on "best practice," rather than minimur standards of acceptability. Licensees wili use the criteria for self-initiatec evaluations (which can be perf rrec ei-her by ar. inoependent group within the utility or a cc ra:.c-l. ne self-initiated evaluations will be submittet to IN:D by the enc of 19E2. During this trial process, the NRC staf# wii'. be invcive; by review-Deta;is cf ing the criteria and cbserving s:me ci :ne e,E'.ca-icrs. the staff invcivement have not ye-been cevelcrec. The industry will decide, by early 1983, or -he cirecticr cf a At present, the primary alternatives a:: ear tc continuing program. INFC will either begin ccncu::ing cuality assurance estica:icrs be: at individual construction facilities, cr a #:rr cf self "ri :a e: evaluaticn will continue.

. Enclosure 1 1921 and October-1982)g management workshops (tiay 1980, September INPO is also conductin with utility chief executive officers and plant managers in an effort to strengthen the utility comtitment te safe operation. NRC will coordinate its quality management seminars (Enclosure 1, Section D.1) with the industry efforts. C. Cor.struction Inspection Program 1. Procedure Chances The staff is presently revisinc the individual inspection procedures for the various techni;al disci: lines. The main purposes of the precedure revisicr. are: (1) to facilitate performance of the procedures by resicent 'nspectors with reduced input from regional-specialist ins:ectors; (2) eliminate redundancies in the procecures; (3) reexamir.e scope or frecuency of scme inspections based on limitations or. inspector resources; and (4) shift emphasis of inspection from record review to observation of work. This staff effort is continuing. The first series of revised procecures which cover inspection of mechanical systems are in the fir.al stages cf issuance. 2. Construction Assessment Team Ins:sctions This initiative will extend the corcept of the NRC's Perfcrmance Appraisal Te2m (PAT) inspectior, prcgram for operating reactors to about four selected plants uncer construction per year. This initiative was directed by tne Corr.ission in respcnse,to SECY 82-150 dated April 8,1982, "The Performan:e Appraisal Team (PAT) Inspection Program." The procedures for performing mar.agement centrcl inspections at nuclear construction sites were revised by the staff in 1981. The procedures covered licensee T.anagement perforinance in the following construttion areas: Quality Assurance, Design Controls, Project 14anagement, Construction Controls, and Procurement Controls. During 1981, eight trial inspections were performed by regional-based inspectors usir; the revisec procedures. i These inspections were effective in identifying management control problems not identified by the routine inspection program. The manpower cemand in these eigb inspections caused the Regional Acministrators to de#er further performance cf this type of inspection. l

h Enclosure 1 The Construction Assessment Team inspectior.s tc be initiated by the IE staff would be similar to the cer.struction inspections performed previously by the Regional Offices. A team of apprcrimately six individuals with skills in the various areas to be inspected, including centractor personnel with appropriate backgrounds, will visit the selected constructicn site for two to three weeks. Acditional site visits will be scheduled if necessary to collect aaditional information or clarify initial cbsersatiors. The first site inspection has been tentatively scheculec for Bellefonte in September / October 19E2. The construction assessment will ccmplement the integra:ec design inspection. The latter is focusec cn a narrcw area cf technical inspection, while the ccr.structicn assessment is designed to assess the broader programmatic ccntrois. Like the integrated design inspection, the scope of ccnstruction assessments will be modified to be responsive to unique conditions at a particular facility. 3. Integrated Design. Inspections The objective of this initiative is to expand NEC examinaticn cf quality assurance into the design prccess. The staff is deveit;ing an inspection approach which provices a cer;rehensive e;.arir,ation cf the design development and implementaticn fcr a selected system and structure on a given ;rcject. Inis evaluaticn wiii encompass the total design process fror. :he formulation of principal design and architectural criteria through the deyelcpmer; and translation of the design anc its revisions. It wili conclude with onsite verification en a sam; ling basis, of the design of the installed system and structure. This inspection will integrate and augment selected activi:ies of NRR, IE, the vendor inspection proaram, and the regional office. Foiiowir.g development of the evaluation methodology, the staff will conduct a trial inspection with centractor assistance. Subsequent inspections will be performed with a substantial amount of contractor assistance. The results will be provided to the appropriate regional anc headquarters offices to be uset as input to the overall NRC assessnent prior to issuance of the operating license, the evaluation will be a multi-disciplined review that will address areas such as mechanical, electricai, structural, instrumentation and control. The evaluation will inciuce checking sample calculations, however, the em;hasis will be on the systematic nanagemen cf the tctal cesign process. The procedures to implement this at;-cach a e cresently unter ceveicoment. A discussion cf the concettual iogic r.ecessa y to evaluate the design process iciicws. i

F . Enclosure 1 The evaluation will start with development of a logic or flow network of the design process. Each functional entity witnin For each of these the design organization will be icentified. entities, internal and external desien interfaces which invcive From this transmittal of design informaticn will be specified. network, critical'oesign areas or areas with the least tolerance for error will be identified. Within each of the design entities, the specific procedures for the verification anc transmittal cf design information will be reviewed for conformance with the overall quality assurance program, and te identify specific Eased on the results cf the weaknesses in the design process. procedure review and the identification cf critical desicr. areas, a specific sancie of the system and structure wili be Criteria will be preestablished for extending or audited. terminating the audit when problem areas are icentified. In examining a system or structure and its specifications, the review will focus on topics such as: (a) Valicity of desigr. inputs and assurptions. (b) Validity of design specifications. (c) Validity of analyses. (d) Identification of system interface requirements. (e) Potential synergistic effects of charges. (f) Proper component classification. (g) Revision control. (h) Documentation control. (i) Verification of as-built concition. The scope of the evaluation car be rodifiec to be responsive tc unique conditions for a particular f aci'.ity, or known or suspected generic problems. This approach will examine all facets of.Ine design management process for a limitec sample. 4 Evaluation of Reported Information Improvements are planned in the current program for systematic review of information pertaininc to design anc constructier. quality that is now reported pursuant to 10 CFR 21, "Reportinc of Defects and Noncompliances," and 10 CFR SC.55(e), "Ccnstructicr. This procram would have cbjectives sinila-Deficiency Reports." to those of the Office for Analysis and Evaluation cf Oceraticral Computerizec Data (AE00) for nuclear power plant operations. diagncsis would be used to enhance icer.tification cf relations"es that may not be evicent in the marcal screeninc that is dere rea. No expansion of reporting recuirements is currently planned be: revisions are expectec it facilita:e ccmputer input cf key information.

. Encicsure 1 ~ 5. Desionated Reoresentatives Federal Aviation Regulations (14 CFR 153) prescribe the recuirerer.ts for the issuance cf designations to private persens to act in the capacity of FAA representatives in the examination, inspection and testing necessary f5r the issuance of aircraft certificates by the Acministrator. Nccinees meeting the recuirements for appointment are authorizec to re; resent the FAA in determining the compliance of aircraf t, aircraf t ccmconents, anc their repai-or alterations with the requirerents of the Federal Aviatien Regulations. They serve as direct representatives of the FAA it the performance of duties and are guided by the same recuirements, instructions, procedures anc interpretations as FAA employees ir the performance of those duties. These pregrams incluce the Designated Manufacturing Inspec: ion Fepreser.tatise (DMIR) ar.c :*e Designated Engineering Representative (DER). The DER represents the FAA in helping to determine that the aircraft cesien ccmclies with the relevant recuirements ci the reculations and the DMIR represents the FAA in certifyinc certain procuct and manufacturing functions.. These cesignations are effective for one year but ray be renewed for additionai periets of one year. A similar technique of using the cesignated representatives would be useful to the NRC ins;ection effort. It wculd increase the number of inspec crs asai'.atie tc implemen: :he inscect'cn program by providing an immeciate scurce of cualifiec excerienced personnel. Using desicnatec NRC representatives it check key , aspects of the design, f abrication and construction cf a piant at the-specific time i: creased inspection effer: is warranted; would significantly raise the NRC's confidence level of qua'.ity assurance in nuclear power piants. For example, during t.he preoperational anc startup testing phases cf a plant, cesignatec representatives could provide the accitior.ai inspection effort so that all tests are monitored rather thar. a selected few. Increased inspection effort cculd also be applied at problem construction sites without having to reduce the routine level c' inspection effort at other construction sites. The aviation industry uses heldpoints in the manufacturinc process that require inspection anc certificaticn by an FAA inspectcr before the process can centinue. The designatec representative, proviced by tne aviation incustry and actinc fc-FAA, can provide that certificaticr when requirec, which allows the process to centinue withcut delays.

is therefere an advantage to the aviation incustry to provide cesicnated representatives and preven ccstly delays in their manufacturi ;

process There is nc anaicgcus situation to that process at

. Enclosure 1 nuclear plants. There are no preestablished holdpoints (other than CP and OL issuance) in the construction or operation of a nuclear plant that require NRC appreval before the process can continue. Therefore, for the designated representative progran to be successful for the NRC, prograr. incentives would have to be developed to encourage the utilities to support the program. Under the Atomic Energy Act of 1954, as amended, and the Energy Reorganization Act of 197a, as amenced, as well as cther pieces of legislation, the Commission does not have the authority to designate licensee personnel as inspectors in a manner similar to the Federal Aviation Administration's authority uncer its legislation and regulations (see a9 U.S.C. 1355, 31 U.S.C. 453a, and 14 CFR 183). The FAA has the authority to issue designations to provide persons to act in the capacity of FAA representatives while remaining in their original employee status as far as receiving pay. To act in a similar way the NRC would have to have the Atomic Energy Act amended and would have to promulgate regulaticrs based on this amendment. The staff proposes to pursue the statutory chances necessary to implement a designated representative program anc to continue development of program specifics. D. Management 1. Managerent and Quality Imereverent Progrars l The objective of this initiative is to irprese attituoe and performance throughout ai'. levels cf licensee and contractor organizations. The problems that have arisen at construction sites are closely associated with management attitudes and practices. Quality in design and construction is invariably associated with the highest level of management being totally comitted to cuality. Senior managers are personally comitted and are unrelenting in their der. ands on their staffs and centractors for a similar ccomitment. The NRC will cocperate with incustry in sponsoring a continuing series of seminars in which top level nuclear managers can cor: unicate the advantages that can be gainec through strong management involvement in their cwn OA programs. The seminars will be conducted with assistance from independent quality professional, utility and certractor representatives and the NRC. The seminars will be highlighted by the participation of managers fror, utilities which have experiencec sericus cualty ~..

l 1 Enclosure 1 assurance problems and those which have managed highly successful programs. The independent quality prciessionals will convey the improvement principles and techniques of implenentation. The utility and contractor representatives will identify incentives for defect prevention based on direct experiences. The expected outcomes of these seminars will be recognition on the part of licensee and contractor management that pcsitive incentives and benefits are achievable thrcugn enthusiastic inDierentation cf aggressive quality assurance programs. While :nis initiative is directed to facilities uncer constructior, par;icipatior. by facilities in testing and operation will be strongly encouraced. 10 CFR 50, Appendix E, Criterion l'. recuires each utility to recularly review the status and acequacy of its cuality assurante program. The extent enc nature of the recent breakdcens in quality assurance programs have incicated that tnis review has not been effective in maintair.ino an adecuate cuality assurance program at several facilities under ccnstructien. Each utility with,a facility under construc: ion will be recuestec to reevaluate its quality assurance pre; ram, anc tc imciement improvements in areas where the evaluations identify a need. Each utility shoule identify a senicr executive wi:P cverail responsibility for the evaluatior. anc irpienentaticn of the necessary improvements. The NRC will monitor :ne evaluaticr. and implementation of the necessary it;rovements. It is exoected that improvements in the quality assurance prograr vi inc:rp; rate ac.tions such as those listet below. (a) Conduct training sessons for its perscnr.el involved in design and construction. These sessions should et;nasize the importance of each individual's centribution tc er.surinc quality and the enhancement to the cost and schecule goals which can be achieved with a positive program. The result of these sessions would be to get supervisors and employees in the habit of talking positively about quality. (b) Provide better evaluation on a routine basis cf status reports to detect both trends and current nonconforrance problems. Based en this infcrmation, meaningfui errrective actions can be promptly taken to prevent recurrence of both the specific problems anc the roo. cause. The result would be defect prevention as a routine par: cf the operatics.

. Enclosure I (c) Provide feedback on the achievements of the cuality assurance program, emphasizing the improvements from all involvec in the program to maintain the concern and enthusiasm en the project toward ensuring quality. (d) Establish a system through which all parties are enccuraged to communicate to management the situations that make it difficult for the employee tc perform cuality work. This information will be included in the system for taking corrective actions. The result of this system would be that employees know that their problems can be heard and addressed. 2. Certification of OA/0C Personnei A significant and prevalent problem in the construction of nuclear power plants is the cualification status of personnel working in the quality control anc quality assurance areas. Some utilities have waived, withcut suitable bases, the ecucation and experience.recuirements fer these peepie. The NRC has net sufficiently enforced these recuirements thrcuct its inspection efforts. Currently, various standards exist for the cualificaticn of QA/QC personnel, for example: (a) ANSI N55.2.6, Qualification cf Testing &. Inspection Ferscnnei (b) ANSI N45.2.23, Qualification of CA Aucit Perscnnel (c) ANSI.N626.3 (Draft), Qualification & Duties of Person'r.ei Engaged in ASME Boiler & Pressure vessei Ccde, Section III, Division 1 & 2 Certifying Activities (d) ASME Section III, Division 2, Appendix VII Cualificaticn of' Concrete Inspection Persennel (e) ASNT, Certification of Level III Nondestructive Testing Personnel (f) AWS QCI-82, Qualification & Certification of Welding Inspectors The staff believes that sigr.ificant improvements can be made in the assurance of cuality at nuclear projects through the certi-fication of CA/QC personnel. Tc achieve these ir.provements the NRC will direct more attentien tc the enfcrcement of the existing standarcs for the qualificatien cf OA/0C perscnnel.

. Enciosure 1 Certification of perscnnel engaged in 0A/0C inspections would provide a cadre of industry personnel that have been certified to meet minimum levels of demonstrated inspection capabilities. Long-term consideration of formal certification of various levels of QA/QC personnel is being given using the following alternatives. (a) Third party certificatior. of persor.nei engaced in OA/CC inspection functions. (b) NRC Licensing of personnel engaged in CA/0C inspection functions. A program for thirc party certificatior. cf hendestructive Testing (NDT) personnel is currer.tiy oncerway with coordina:ict between the Electric Power Research Institute (EPRI), member utilities, American Society cf hondestructive Testing (ASNT) and The program incluces developinc a s:Encard written ;-actice NRC. for the cualifications of the level III NDT inspectors as well as administeri.r.g basic anc method exar.ir.ations anc specific arc practical examinations in the respective areas of nondestructive A registry of personnel noiding the recuired cualificatior testing. and certification would be maintained by the third party cr;ar.izatic Unsatisf actory performance would resul; in remtval frc the Prccrams siniiar :: registry through an establishec procecure. this could be established in other. areas such as weiding, insre::or! QA auditing, concrete inspectors, and labtratory testing personrei. 3.' Craftsranship The staff has initiated discussions with labor unions irvolvec in nuclear constructicn in an effer: te explore the potentiai-methods and incentives te enhance the craf ts rcle in assuring the quality of construction activities. Feedback from the labor unions includec the following points: (a) Craftsmen are not weli informed cf their roic in the QA/0: process. Continuous rework as a result of changes nas a demtralizir; (b) effect on craftsmen and effects the cuality c' the final work.

' Enclosure 1 (c) Utilities and contractors have not provided adecuate training to craftsmen regarcing quality. (d) Utilities are not convinced that cuality assurance is a cost effective approach to construction. Labor perceive: the utilities t5 think OA/QC was a "high cost" item rathe-than a " cost saving" tool. (e) Improved front-end engineering and procurement woulc rec;:e the amount of change and rework. (f) A Nuclear Stabilization Ccrnittee has been established wi-h representatives from labor, utilities and cortractors. tc improve relations between labor anc management. The staff proposes to continue these discussions as part of the long-term review. E. Lono-Term Review Long-term NRC' quality assurance policies and programs will be based t on a review which assesses existinc agency anc industry cuality assurance activities in a broad manner and ther, recemrends an inte; rate long-term agency pian for cuality assurance. Accitionally, the review will focus the viewpoints of various sectors of tre public ar.d the regulated community. The primary function cf the long-term effort wili be tc ccnduct a thorough review of continuing ouality anc ouality assurarice'probiers, anc to propose solutions to improve the cuality assurance programs for design, construction, testing anc operation. This review wil' include a detailed assessment of the problems that develcced at" facilities such as Diablo Canycn, South Texas, Midlanc, Marble Hi'.1, The object of this assessment will be to identify, as and Zimmer. concisely as possible, specific problems that have occurred and tneir root causes, particularly in the area of programmatic deficiencies. Additionally, the review will evaluate existing programs at facilities which have programs that are functioning properly in order to ide-tify the positive aspects of those programs that should be applied gererica Both this review anc the review cf programs at problem facilities will involve site visits by the perscnnel performing the review.

. Enclosure 1 ~ Proposed solutions to generic and plant-specific quality and cuality assurance problems will be reviewed critically tc determine whether the recommended actions would actually resolve the identified problems. The review will develop estimates cf the qualitative and quantitative value/ benefit and impact / cost of proposed solutions, and ways in which they should be implemented for bperating piar.ts, plants presently ur. der construction, or for plants te be constructed in the future. The review will be conductec by the NRC staff anc will include representatives from headquarters, the regicnal offices, and censultants to the NRC. An advisory panel will be establishec :c provice recommenca-ion cn the scope and direction of the review anc ic prcvide a peer review service. The members cf the panel would be selected from gevernmert, industry ard special interest groucs. Their function would be tc provide a broad spectrum of insights anc viewpcints tc the NRC rev'en group. The panel will assure a prcper balance cf censiderations ir. the findings and recommendations cf the review. The Hcuse and Senate in their current jcint consiceration of the NRC's FY E2-83 authorization bill have acceotec in conference an e amendment which directs the NRC tc stuc'y ways tc improve quality assurance programs. Implementatier. cf this review is consistent w':h that directicn. The review will ccmmente shcrtly after Commissien approval cf this proposal. It is expec ed that fir. dings and recccrenda:#cns from the review will be available for Ccmmissien consiceratior, witrin one year of Commission approval. F. Ocality Assurance Plannina and Evaluation The recommendation to form a single organi ational unit dedica:ec' c the various aspects of quality assurance was mace in.a report prepared for the NRC by Sandia Labs in August 1977, entitled " A Stuoy of the Nuclear Regulatory Commission Quality Assurance Program."* 'Page 60, "A Study of the Nuclear Regulatcry Commissicr Ouality Assurance Program," NUREG-0321

.? . Enclosure 1 The centralization of quality assurance functions has been one cf continual interest and ncw more than ever needs to be achievec. The NRC presently views respcnsibility for quality assurance as threefold: first, to determine the adecuacy cf the licensee's quality assurance program descrintion contained in the safety analysis report; second, to ascertain that the licensee has established and adecuately implemented the approved quality assurar.ce prograr and to verify compliance with NRC regulations; and third, to cevelop tne regulations, standards and guides addressing QA in the design, construction and operation of nuclear facilities. The respcnsibility for these three functicns is currertiy divided among three separate offices, NRR, IE and RES, with execution cf the inspection function from five regional effices. These three functional areas are not separate and discrete areas but are highly interrelated, requiring continual interface. For example, the inspec icn experience needs to be continually factorec into the licensing effort, inspection program development and development of regulaticns anc standards. In addition, recent quality assurance issues (e.g., Diablo Canyon, Marble Hill, South Texas) have been highly reactive and have requirec rapid NRC management attention and respcnse from the three separate offices for their various quality assurance functior.a1 areas. The functional cuality assurance areas need to be realigned for the following reasons: (a) To more effectively utilize the limitec sta" rescurces and, expertise in cuality assurance engineering. (b) To establish a more discernible policy anc cosition er cuality assurance issues. (c) To establish unity of control and to provide bech informaticr. and coordination with industry. (d) To bring together the licensing, inspectior anc standarcs functions on interrelated issues. (e) To provide industry a signal that NRC management considers quality a leadirg part of the NRC cperaticn and of sufficient importance to depart from the existing crganizational structure.

. Enclosure 1 It is recognized that most NRC activities are cuality assurance related. The NRC review process is an intercisciplinary functicr. invciving many organizational components. Only these NRC heaccuarters activities which directly relate to the review ano evaluation of the implementation of licensee's OA orccrams and cevelcoment of flRC rules'. standards and guides are# recommended to be consolidated at this time. Such consolidation will occur in the Office of Inspection and Enforcement. At a later point in time, after phasin; out the NTOL reviews, the licensing functier. will be re;ienalizec. Upen regionalization, the program directier. and assessment will be combined wi+.h the rest of the CA overview function. e e e y

O Enciosure 2 Resources, Schedules anc Staff Responsibilities The nuclear industry currently expends substantial resources for quality assurance at power reactors. For example, about 7500 positions are currently devoted to quality control and quality assurance on construction prcjects. ~ The NRC Regional Offices cevote about 130 positions to inspection of peaer reactors under construction and vendors. These ins;ections are concerned, to a great degree, with the effectiveness of the cuality assurance procrams in the various areas that are being inspected. This effort, which provides a measure of the overall effectiveness of the quality assurance programs, is the largest segment of the NRC's efforts relatec to quality assurance at construct'on projects. A smaller part of this inspection effort (abcut 28 positions) is narrowly directed towards inspecting the cuality assurance programs. In a broad sense, the headquarters offices also devote considerable efforts to quality assurance. In the narrow sense, they cevote about 16 positions to direct professional work on cuality assurance programs as fcilews: 4 p:sitiens licensing; 7 positions inspection progran (OA related) cevelopment and develcpment of OA initiatives; 5 positiens research and stancarcs ceveicpment. Much of this effort can be considered applicable tc construction prcjects. 4 NRC contractor work has previously been at a level cf about 5400,000 per year for research/ standards development in quality assurance. The estimated resources for the initiatives discussed in this paper are prcvided in Tables 1 and 2. They can be sumr.arized as felicss: 1. Estimated Industry Resources (a) In FY 83 (i) 2E0 man years new effort recommendec (ii) 420 man years altogether (b) In FY S4 (i) 310 man years new effort reccmmended (ii) 390 man years altogether 2. Estimated NRC Staff Resources (a) In FY E3 (i) 15 staff years new effort recommence: (ii) 36 staff years altogether i (b) In FY 84 (i) 11 staff years new effert reccamence: (i-) 21 staf' years altogether

s -2 3. Estimatec NRC Contractor Resources (a) In FY 83 (i) 52.2 million new effort recommended (ii) $2.2 million altogether (b) In FY S4 (i) $1.4 million new effort recommenced (ii) 51.4 million altogether i.e., New efforts correspond to the new initiatives recommended by the staf#, thcse that are not already underway and well established. Generally, additional efforts are not large in ccmparisor to he rescurces alre The improv'ements will ccte mestiy f rcm recire: tion of existing cevoted to QA.NRC staff and contractor resources can be recirected as recessary te accomplish the initiatives without drcpping any plan resources. NRC staff responsibilities with respect tc developrent and The schecules fcr be affected. implementation of the initiativ.es are incicatec in Table 2. accomplishing the initiatives are provideo in Figure 1. e >b 0

I i 1

• =

=3 '3 L = i a==t ue h 9 .a C: E C: z =l, C C C g C C N C e z -. wl 4 m m e w gg

== g .,,, ee.s

== ,s.

• J C

C 3 .-= = x i W a 3

== C w N C 3* C = y w =.=.e C* N O C-C- ="* w.r{ z aC. > 3l z - atl 4

== == C >l A el 3 W v y -e er G >. =. W zz W D a- >

== e, C C C C g =. 5 a'C C ?J

== zr 3ll c N

== N m 3d = = = =,

== m C g ,=, we m me g A e e W 3 3 w 2.

== =. C .,e w

• J 3

m. u

== . = *. we C "3 C .C e C-C y C C C g z3 C 5 e C e n z N m N m u =y a N CD b =* wa 4 -s g ne= 3

• 3

.b 5

== d 2, 7 = w 33 C g C. 4 N. e. a. w ar

== w 6 g. w yw

.: er==

m C* C'* a ,a

== c e v w all 3== ac! i .= z 3 I A==t e Z %l eet me me .ns b g w e s u u a C. =* eu 3 P = >et C e 3

==

• 9 o

z El u em p= = C Rl C. D C m C ~j e e z O C N C u z E, e -e z=

== .%s N

== w. g = es .=~3e E C to se E= =J 3 CC .sa = e m g

== Se

== == Es

===9 = e w = se C d O

• E e

= w .= sp = = =. .3

==n C D

• w e

= N w C U= w 3 =* .2 = C g S 3 "3== s, e me Q

== == C g ue mm. D a= g

== .es ee = w N v e C C 3 o 7 3 Q Q 3 =

.

lE g-

• ====

e g C

== u ew O = = em C es g 6 = 3 g w

== C D Q 3 .N, "== e Q we

== == Sp me of

== 4 es g C g e C "3 f e .2 C 3 y y e es e 4 6 e

• 3 a

e a# %W

==

a C

== a.s me g 3, = ue

== w C =i em 9 "3 as D = 7 3 e C es O g E= 33 E C 3 "l3 h C 6 3 end C

== u* C la A. w

== 3eO m t = Q -Q.

==2 j

== g g s e 3 C me - aw g.s = C=

== 3 g C EC em 2. g um ** g D 6 es 9 6 *w as u g g a. C O z e 33 3 - 7e =a .= .= z e & 6 Og 5 -C 1 C 3 C "as ee w wg w 3 "es

== == 3 g = l", aC w >Q d

== sa sa 3 p w g 3 e= c z - >== w ee E ts Q e s e== =3 e.e et b es 3 1

== 9 a

== == e a e me w h b g = g

== e a= en MSh O es 3 g C, e 4= ee =8

== =# .,e 3 .e 3

==

g ee

== 5

== .- E es = a= g C

• 8 C

"3 e =

== 9 e = F 3 .C= -= = - v - - b, F = z - = 5 = = .o C-3 33 x y y g i w - - -=.-= g.=.= NA etO

= = =.

6 D 0 e

1 I dis la* / - t s t Imaleel itesous tes f or 1)A leilt I at i ves ( Alt nsse t t.f r ) T V t,8) lY tl NH( dil(f". IIHL f4Hi lillMf llt! 5179) LOlilll At I lillHikil4 V %I AII lufillIAi l ilA litill AllVIS _ (ilAll,VH)_,_()lA,il tilj (1 1000;) . _ (, Hall V H} _ ('a l Al l VH) _ _ (} lHtHf ) tilut - SeIf Ivatuatton 10 ? 's 1 >> 0.9 NIOL - Regional Evaluation 0. 11 H.5 0) 1.0 Niut - Independent ihn ign Hewlew 61 5.0 21

1. tl in.lus t s y Initiative 14 7.6 S4 I I leispet t ion Ps o(edur e thanges J. fi 1.0 tonstruction Assessne*nt

. Inspettluns 0.6 6. I' 100 0.6 6.0 f,tF) lait a-alra t ed Eks ign inspections 2.1 3.5 (HN) 4.9 1.4 1/0 I valuat loss of Repor ted leefos niat losi 0.2 250 0.2 ISO Des tesnated Hepresentatives 0.5 84 2.0 flanagemen t.. Progs'ains 210 1.4 100 2tN) 0 fl Cer t i f ic a t ion... Pe rsosuie l 2.1 0.2 20 0. 11 Craftsmanshio 0.4 tong Iesus Hewlew 1.2 150 0.1 100 ljuellty Assurance... E valuation IDIALS: 420 22'y) 390 /l 1404 NCI(5: (l) liesouste estimates for Initiatives related to licensing peak shasply in TV OI and drop sharply lei TV 84. Construction delays may seduce this warlation. (2) NI0t prooraes niay be reduced t>eginnina in mld lY 83 as other NHC and leniustry Initiatives take effect. (3) lotals ase rousuled to two signifl(ant figures. O e g

s i I h Sl tA u A e 4 1 l 1 l I i N V v I Hd i i 1 W 5 H H 5 l l l l I A l l l A V A L 1N A I s e v i' i t t n e i t l l P P P W W' eH a l O O O t t ( G l I t t t t l s t I o k t i l i L i i i i u t P 1 v V V V V H p H t s i A i ti V. p l 8 u i se H 8 d Wl d dI p H P l. 8 i O 6 l D i a 0 i u t I ) D i t l IH A A A A A H A l D t n H A i e A V u u t i l l [ [ d l t f l l ni l' I l i t l P t ( l' i se i t i l l a l i l tH s 0 n l N I u aw k u t I t k I I l' 1 H H u C I t s f I S t l I e 6 5 S l N l l 5 i I i i P I l l f l % i H o a V i f I l P l S D A N t A L I H I 4 A H A l l I a t 5 J e l e tn I snu l tc n e o es i w p u t e s t n a t l i s m a we s n s l u n n lu u H e e l f s o n n t n e n a i y a c I v n v t t i i s u t i i t c s a s l n u d t m s e es e a a e d-u u e n t t r c D e r n g P f l nI u e u t a e s a v t v u e d e n o s s w t u t n i t e s y ic e P e n l e a t s l l s i a a l dis l A s u v a u n p v s n e l n ie f e o N o i e u i i h R s u p in 1 l u n l l n d t t s s f l e i l e e n u t e o e n a n m A a V l g d I l S R I y l c t i t e c a e m e y t u s t a m l l s l t A a n e s l t t r s s c t 9 u n g i t i u e s c l i a t t g l l H n d s n i a s n r a n a L t i u u N l 0 I u n o n v e e s D j C C t t I li li 14 l I C i I t t A )s e e

m-m. b g 2 a = "6 5 s. -s n "J .e u .u e t.d 4 s 3 ~ ( :CC=3Lt.]

• t, hai

.aa.. .e.

• u e

u .c 3 t 144J79 - Z i f a d a 4

• J t.Jc ;

3 .= we f, I$ 8 I*. I) ( I)$ da el) 1 1 4 4 A A A i 1 y 1 ? J f 3 1117., 4 =3 e0*m b he. I ruu t'4 a n 13 3 - t 2::a i t; s

• d l#

t ;u r t : t w - e, vaa;; its Yy t, ice ntas - t i.sas ,a 1 n [ A t 4:uo>attas - : :n 3 t e t t 11 'M j l I

• w d $UIt)fh l

t teattte; b,, j t itas s'4:u s.: 0; - Z p aj 1 => t g A I i t 2n ds - g aa;ng:w = l 1 1 ~ g. I I 1 I c 2.: ua:e, - z aa r m i l .r 1 I b 3 i l l t aawtI '2 I g l c aa;:uo uns - t asaa4 stes l I ,X g i g ] 2 a t sal t; i I [ 8 I [ l I I l l I ara se.;; t i I i I I I l l I t t -6 1 I I r i t r ( gugt.4aamsng. g,a.a n; 1. 1 = =: 7 }(9 ;uI.) M m< b + ,,e

e. k e

w3 3 0 =e 4 .e p'

• C e

4 = .C 3 2

• d 3

W aC. e l =

• 8 9

43 2 g U C son sg me e s = et I' e o o a 3 e o Q h. Z S,

z..

al 4 ..e ql 4,

== =- a b aC 4 (; 3 3" w w 3 G w A g m, .u. =.. e a u 3 C 7 d h V Qi e = u e b e C* r p 3 = ** - Z .=. 3 E "C aC "g" C" w g' 2 ne

== m ..l e* 3 W g' 3 2 d C, A u* E 2' 3 D I-. Vf, ,M h.s.l -l S a 2i u c = w 0 he w p g e p ? .* *= 3 '= "3 3' 4.3

== 8 e' S

• C 3*

=

4 z-g u ne C db =s e 2 s g' 3 -.3 2 Q 3 5 9 3 5 t 4, 2n

• 2

.;y s,e m

• • l e

-s 3= 7 2 "5

• 8 h.I 3

g .e .s 3: 3 e

=

d W 3 b E E 3l w b 21 31 b. w w e w - = . 3 l t, a r,, s s 3 S z a W 2, 9 E E ang to t

• 3

)

== h.

1. 8 ee b

E' $I h 9 3 ""e

== a m. a; e de J E b 9

• =*

.-El e. '4 ; a e s 3

== p g A a, 3 = 9 48 b d As b te S = m u aC =*

1. s a.

3 as O . E 6

• J 4

gl = h {* 8. e -8 I* 7 g =.

~ Examples of Recent Ouality Assurance Problems During the past few years, there base been sericus cuality assurance prcblems at nuclear power plants. Some examples are listed beice: 1. Marble Hill In June and July 1979, NRC confirmed alle;ations :f improperly repaired concrete imperfections at Marble Hill. The imperfections were generally identified as concrete consolidation problems (honeyccmb and voics), anc improper repair (patching) of these imperfecticns. NRC inspections confirmed that; An excessive amount of honeycomb anc air voics had occurred. Apprcx.- a. mately 4000 concrete patches existec. b. In many instances these imperfections were improperly repairec, and/or unacceptable materials were used for the repair. Quality centrol rec'ords traceable tc the re;2 irs were either non-c. existent or inadeouate. d. Personnel responsible for such re; airs were tracecuately trained and supervised. The licensee was not in controi nor suf#iciently aware cf the ab:ve e. ci r,cumstances. These events led to a halting of all safety-related werk at the site ir August 1979. Work was not permittec by URC to resume until December 1980, when the utility's quality assurance pregram anc that of its contractors, had been substantially upgraded and the acecuacy of completec construction work had been verified. 2. Widland Excessive settlement of the diesel generator bu.cinc was cbserved in 1978. The unexpected settling was subsecuently attributed to inacecuate anc poorly compacted soil uncer the building. Cther safety related NFC's ir.sestigaticn determined systems and structures were affected. that design and construction specificatiens had not been fc'.lewed du*ing placement of the soil fill materials and that tnere was a lack of centrol and supe-v'sion of the soil placement activities by the utility and its Extensive rework hes begun, and tre coerating license contracters. a;;1ication is currently being litigated be'cre an NRC Hear'rg Ecard.

- Enciosure 3 S. Zimmer Allegations received in January 1981 prompted an NRC investigaticn of quality assurance problems at the Zimmer site. The investigatien has identified a large number of quality assurance related problens. The majority of the problems identifiec focus on the ineffectiveness of controls implemented by the licensee and its centractors for assuring the quality of work perforced. In that regarc, numerous deficiencies have been found concerning traceability cf materials, har.dlin; cf nonconfor-mance, interface between ecnstruction ar.d quality control, quality reccrds, and the licensee's overview of ongoing work. An extensive review of the as-built plant is being performed. Limited independent measurements were performec by the NRC ir. selected areas cf concern in an attempt to characterize the actual safety significance of these deficiencies. Although a fen problens recuiring corrective acticr. were identified, the majority of the tests and examinations disclosed nc haroware prcblems. The licensee will perforn a comprehensive cuality confirmation program and resolve ider.tifiec pr:biems before an cperatin; license is issued. 4. South Texas In response to allegations that QC inspectcrs were being threatenec if they reported unacceptable items during concrete placer.erts, tne NRC initiatec an investigation through its Regicn IV Office in July 1977. Ten investigations of allegations were performed during the period July 1977 to Ncvember 1979. The retuits of.these investigaticns establishec that the alle~ations of g harassment, intimidation and lack of supptr: of OC ir.spectors were sub-stantiated. The investigation demonstrated shcrtccmings in the management ar.d that the implementaticn of the OA/0C prcgram at the Sou~th Texas Project did not meet the standards required to assure that the facility will be constructed to NRC requirements. Safety-related work was stopped in 1980. NRC allowed restart in designated areas only after QA for that area was upgraded and verified by the NRC. In January 1981, the licensee initiated a cesign review cf those pcrtier.s of the engineering cesign work perforted by Brown and Root, Inc. The Quaorex Corporation assisted the licensee in this review. Briefly, the Quadrex report found that Brcwn and Rect ftiled to properly impler.ent ar overall design consistent with the neecs cf a nuclear pcwer piant. The licensee replaced Brown and Root with Sechtei Power Corpration as architect-engineer in September 1951. NRC is mcnitoring the :erfornance of Bechtel as they resolve the. problems icentified in the Quacrey repor;.

. Enciesure 3 5. Diablo Canyon f At Diablo Canyon, the Pacific Gas & Electric Company (PGLE) providec incorrect information to an expert consultant, who used the infornation in developing the seismic response spectra for the design of certain seismic piping and equipment restraints. NRC investigatcrs have found that there was a lack of rigor anc formality i6 the procedures used for verifying the accuracy of information transferred by PG&E to its censultants. These procecures did not comply with NRC repuirererts calling for verification of design information at each stage ci the process by an indepencer.: person qualified in the pertinent ciscipiines. Proper quality assurance controls were not employed in techr.ical and precurement communications with service-type contractors. Nor were document controis adecuate tc assure that these invcived in design hac reacy access to the ecs: recent information available. Following discovery cf these errers in seismic cesign, the recer.tly issued operating license for Unit I was suspended in late 1921. Pricr to the NRC's reinstatement of the operating license the licer.see will be required to complete an extensive design reverification pregram for these areas in question.

Independent Design Review For Near-Term Operating License Facilities In order to provide further staff confidence in the quality of design and construction at near-term operating license applicants, licensees have been recuested to conduct an independent verification of selected desigr. and cor.struction activities. The independent verification wouic be performed by an independent contractor with qualifications accectable to the NRC. Indepencent verification efforts have been completec at LaSalle Unit 1 and 2 and San Onofre Unit 2 and 3. Reviews are presen:1y in process at Grand Gulf, Suscueha.na, Shoreham, Watts Bar, Palo Verde, Summer and St. Lucie. At LaSalle, the licensee hired an independer.t ccn:racter, accrevec by NRC, to perform a review of the mechanical and structurai cesign cf loop C residual heat removal system excluding all branch piping less than 3 inches, in the fun.ctioning mode of the low pressure injection system usir.g icacs resulting from the actuation of the automatic depressurization system in conjunction With the operating basis earthquake to verify that this syster has been designed and constructed in accordance with the application anc that the NRC requirements have been satisfied. Commonwealth Eciser contractec the Teledyne Engineering Services (Teledyne) to perform this review with HEC approval. The preliminary findings by Teledyne resulted in 21 Error /Ceviation anc 31 open-items reports which were transmitted to the licersee and the NRC staff. Upon submittal of all Teledyne's preliminary fincings, the licensee transmitted it responses to Teledyne and the NRC staff anc, ir acdition, the licensee received permission from the NRC staff te establish a cialogue between Teledyne 'and its Architect-Engineer (Sargent & Lundy) to ciscuss' the. potential errers found in the Teledyne review. Of these 52 repcrts which involved various problems in the design area and none in cuality assurance, 2g were closed by Teledyne based on the acquisition cf accitionai information and/or clarification of existing information. The 13 remaining reports were reviewed by Teledyne's Project Review Internal Committee. This ccmmittee, composed of three senior level Teledyne engineers who together had the expertise to resolve the technical issues, and the Teledyne Project Manager concurrec that none of these reports have the potential for significant safety impact. The NRC staff reviewed those open-itens and error-deviaticns reports submittec to the Project Review Committee and concluded that these reports can be categorized as not having a significant safety ir;ect en '.aSalie. In addition, the NRC staff feels that Teledyne has ;erformed an in-depth review of the analytical procedures and design calculations usec in the piping, equipment, and compcnent support design to assure the adequacy of the design bases, the adecuacy of the oesign implementa:icn, anc the censistency between the cesign documents and the Final Safety Aralysis Reco-t ccaritments.

. Enclosure 4 The independent design verification program conducted by Teledyne on the locp C residual heat removal system indicated that the quality assurance centrol and implementation, design process, procedures and Final Safety Analysis Report commitments are acceptable except in the area of response spectra, which was reviewed by NRC staff. The results of the limited review previce increased assurance that the quality ass 0rance program established anc irplemented by the licensee and its principal contractors did effectively contrcl the everall program and construction activities for the LaSalle County Station. While several design deficiencies were icentified, the overall design anc construction activities were adequately performed se that no adverse impact on safety was found. At San Onofre the licensee contracted with Torrey Pines Techr.ciogy, a subsiciary of the General Atomic Company (GA), to perferr an incependent evaluation of the seismic design and quality assurance program for San Orcfre 2 and 3. The design verification encompassed a review cf the seisr'c design of San Onofre 2 and 3 to: verify that the design pro' cess converted the seismic cesign basis a. specified in the San Onofre 2 and 3 Finai Safety Analysis Report (F5AR) into the design documents that are transmitted to the ccr.structcr er f abricator, and evaluate the SCE ouality assurance (CA) audit plan and its implementaticr. b. at the construction site and the f abricater's shops. The design proc'ess performed by the equipment fabricaters was net part of this-review program. The work was diviced into eight major tasks: Task A. Design Procedure Review Task B. Design Procedure Implementation Review Task C. Seismic Design Technical Review Task D. Audit Pian Review Task E. Processing of Findings Task F. Reperts Task G. Pipe Segment Waikdown Task H. Independer,t Calculations The review was conducted by individuai GA reviewers investigating each area ccvered by Tasks A through D, G, and H. Wher a reviewer #ound a ceficiency that might have safety significance, it was de:umented in a "Fotential Fincir; Report."

. Enclosure 4 After the Potential Finding Report (PFR) was written, it was.sent to the eriginal design organization" that was responsible for the area covered by the PFR. The original design organization (ODO) ther. investigated the PFR and The PFR and the 000 response was then reviewed by a GA responced in writing. committee, and the PFR was classified as (1) Out of sccpe, (2) Invalid, (3) Observation, or (4) Finding. Out of secpe items are these which are beyonc the original scope of the review. For example, the review was oriented towards desigr serification. Procurement items are considered out of scope. Invalid Fincines are the result of apparent deviations, uncovered in the course of the independent verification, that are rescived to the satisfactier. cf project personnel, usually during the Pctential Finding review by the Origir.al Desicn Observations are valid deviaticr.s that are juecec not to have Organizations. the potential for significant impact on the seismic cesicn acecuacy of San Oncfre Units 2 and 3. Findings are valid deviations that ceuid Fave potential fo.r significant impact on the seismic design adequacy. Of the total of 170 PFRs that were initiated, 77 were cetermined to be invalic af ter additional inforsation.was, reviewed. Of the 93 PFFs that were cetermined to be valid, 7 were classified as fincings and SE as cbservaticns. The numbers of findings and observations fcr each of the varicus tasks are as follows: Task Findings Observations A 3 2 B 1 35 C-1 41 D 2 5 G 0 2 H 0 1 Tctal 7 86 The staff has concluded, based on its review of the results of the design verification program, that the GA design verification program has not discovered anything that would cause the staff to charge their previous conclusiens that the San Onofre 2 and 3 quality assurance and seismic cesigr programs are acceptable, and provides additional assurance tnat plant cesign and construction have been appropriately accomplishec.

Statements :c Concress A. Hearing, Ucall Subcommittee, 11/15/El 1. Chairman's opening statement,11/19/E1 Statement - CA shoulc be 'used as a fernal managerert tool. It a. needs managemer.t support to be effective. Paper - The problem descriptions are ccnsisten wi-h this viet.. 2. ED0's opening statement,11/1g/81 a. Statement - Problems fourt at five piar:s are describec. Paper - Enclosure 3 provides generally ccr.sistent descriptions for the same 5 plants. The primary cifferer.ce is for South Texas. The paper emphasi es OA/0C breakcowns, cer.cre e, harrass-ment, etc., and only mentions the Quacrex report. On tne other hand, the ED0's prepared.enarks focus aimcst entirely en the Quadrex report. b. Statement - NRC is placing residents at all ccrstruction sites more than 15; complete. Paper - Page 2 of the SECY paper states tr.is has zeer. cone. c. Statement - NRC enforcemar.t posture ras been tcughened. Paper - Page 3 of the SECY paper states tha tre toudher pcs' ure will continue. d. Statement - Construction Assessmer, Team inspec: ions are ciscussed. Paper - These inspections are recommendec as an initiative. e. Statenent - Construction Inspectier Fregram is being revised. Paper - These revisions are ciscussec as ar, ir.itiatise, f. Statemen: - Systematic Assessment o# Licensee Per#crrance is being concuctec. Paper - SALP itself is nt ciscussec as en ir.itia.ive. It is ar. ongoing program. g. Statemer.: - The mctile van is beir;.Se:. Facer - Tne var is rt; c.tc.ssec as t- ' ':'a:..e.

's alreacy tein; use: cu-ir; *:u.' e ' s:e:-':*s.

e 3. Discussions, 11/19/81 a. Statement - In Commissioner Gilinsky's cpinion we should cesigt plants before building them. Pacer - This point is not Explicitly discussed. It would be an appropriate subject for the icng-term review. E. Hearing, Moffett Subcommittee, 12/14/51 1. Chairman's opening statement,12/la/81 a. Statement - A new inspection proc am has beer. ir.stitutec for FY 82. Paper - Changes are in progress anc continuing. Program revisions in FY 83 are recomrended as an initiative. 2. Discussions, 12/14/E2 a. Statement - The five example plants are named. Paper - The paper discusses the sare five c'. ants as examples, b. Statement - In the Chairman's cpinion stancardi:atien is one c' the ideas with the ecst promise. Paper - This subject is net ciscussed. would be an accrop,*iate subject for the long-term review. C. Fearing, Simpson Subcommittee, 2/22/S2 1. Hearings,2/22/82 2. Kammerer letter to Stafford, 4/1/82, followup, a. Statement - The NRC is evaluatin; the initiatives in the EDC's memo of 1/21/82 to the Chairman. Paper - Specific actions are new recomrenced. b. Statement - NRC is encouraging licensees tc perforr third tar:y audits. NRC views INPO pregrar as su;:ler.e-ting but not suc-planting NRC efforts. NRC and industry initiatises anc descrittiens are Pater consistent with ttis view. I L

L -- L D. Hearing, Ottinger Subcommittee, 3/3/82 1. Kamerer letter to Ottinger, 3/1/82, responses to cuestions a. Statement - FY 83 budget recuest includes about a0 professicnal staff and $200k for quality assurance. Accitional resource estinates would be premature while assessing opticns for imprcve-ment. Paper - Enclosure 2 states that the NRC currently cevotes 16 positions in HQ to 0A and 28 positions in the regions to QA for construction reactors. These are the FY 83 numbers from the Chairman's recent budget crosscut. The pacer aisc states that, in the past, funding has been abcut 5400k per year. This corresponds roughly to what was expended ir. FY E2 anc what is intended in FY 83 (aside from initiatives). b. Statement - The NRC is evaluating the initiatives in the ED0's memo of 1/21/82 to the Chairman. Paper - Specific a'ctions are now recommended. c. Statement - Specific evaluat. ion criteria for initiatives have not been erO.oiished. Initiatives are beir; evaluated for ccs effective impact. Paper - No criteria are stated in the paper. It car be said that the staff generally consicerec cost-effective impact ir., developing initiatives. d. Statement - Organizeitonal hanges are being cor,sidered. Paper - Such changes are now recommended as an ir.itiative. e. Statement - QA paper is scheduled for 3/31/82. Paper - Timing of paper is not discussed. j 2. Hearings, 3/3/82 l I a. Statement - An important lesson from Diabic Canycr is that URC had a gap in its treatment of desigr. Paper - Several initiatives adcress this ga:. b. Statement - There was a general c'scussier C ircecerdert a uc'i ts. They are nct necessarii) re c ai re c # c r a ', '. :i a r.t s.

4-Paper - The initiative ciscu'ssicns are consistent with the views presented. 2. Chairman letter to Moorhead, 4/19/E2, followup. a. Statement - Standardizec desigr helps CA and lack of it makes QA more difficult. Paper - This subject is not discussed. It wouic be an apertpriate subject for the long-term review. b. Statement - Criteria for independent accitors are described. Paper - No criteria are discussed. The paper is not that detailed in this area. Statement - Applicants are respcnsible for the safety of their c. plants. However, the NRC sta#f determines if that responsiblity has been met. Paper - Page 2'of the SECY paper states that none of the initiatives is intended to detract from licensee respor.sibility, d. Statement - Over reliar.ce by applicants on third parties tc manage QA is nct as effective as strong applicant involvemer.t. Paper - This point is not explicitly ciscussed. It can be considered implicit in the orcblem statements where the importance ' cf management commitment and integratien of QA into eff'ective' project management teams are discussed. E. Hearings, Bevill Subcommittee, 3/22/52 1. Hearings,3/22/82 a. Statement - Responsibility remains with plar.ts' owners. Paper - Page 2 of the SECY paper states that ncne of the ir.itiatives is intended to detract fror.14censee responsibility. 2. Kammerer letter to Sevill, 3/26/S2, follo.up Statenent - The NP.C is evaivating the initiatives ir. the ECO's a. meme of 1/21/62 to the, Chairr.ar. Paper - Specific actier.s a e r: recenrencec. d

. b. Statement - The QA paper is scheduled for 3/31/82. Paper - Timing of the paper is not discussed, Statement - Independent audits strengthen utility CA and c. supplement NRC inspections. No pitfalls have been identified. Paper - The initiatives reflect this view. d. Statement - Independent audits oc not detract f rom licensee responsibilty. Paper - Page 2 of the SECY paper states that none of the initiatives is intended tc detract from licenses respcnsibilit;'. Statement - NRC supports the INPO program and will monitor arc e. review it. NRC will design new efforts tc minimize duplicaticr. and overlap, to take advantage of INPO while still providing acecuate independent review. If the INPO program appears effectise anc vigorous,.it.may substitute fcr some NRC initiatives. Paper - These views are reflected in the initiatives. f. Statement - INPO criteria coulc be usec by thirc ; arty auditors. Paper - This is stated in the incestry initiative description, Statement - The staff will review INPC criteria and results o g. achieve coordination. Paper - The industry initiative cescription states that durirg the trial program the staff wiii review tre criteria and cbserve some evaluations.

~ Statements to Concress A. Hearine, Udall Subcommittee, 11/15/81 1. Chairman's opening statener.t,11/19/S1 Statement - QA shoulc be tsed as a fcrmai management tool. It a. needs management support to be effective. Paper - The problem descriptions are cer.sistent with this vieve. 2. ED0's opening statement, 11/19/81 Statement - Problems fourd at five plants are described. a. Paper - Enclosure 3 provides generally censistent descriptions for the same 5 plants. The primary difference is for South Texas. The paper emphasizes QA/QC breakdowns, concrete, harrass-ment, etc., and only mentions the Quadrex report. On the other hand, the ED0's prepared remarks focus aimest entirely on the Quadrex repor;., l l b. Statement - NRC is placinc residents at all censtructior, sites more than 15% complete. Paper - Page 2 of the SECY paper states this has been der.e. Statement - NP.C enforcement posture has been tcughened. c. Paper - Page 3 of the SECY paper states that the tougher ~pciture will continue. 6, itstement - Construction Assessment Team inspections are ciscussed. Paper - These inspections are recomrended as an initiative. Statement - Construction Inspection Program is beinc revised. e. Paper - These revisions are discussed as ar. initiative. f. Statement - Systematic Assessment of Licensee Performance is being conducteo. Paper - SALP itself is not discussec as an initiative. It is an ongoing program. g. Statemert. - The mobile van is being used. Paper - The var is nct disc ssec as an dritiative. 's alread.s teirc used curir.g r:u. ire i sre:-4:t s.

I 3. Discussions, 11/1g/81 a. Statement - In Commissioner Gilinsky's opinion we should desien pler.ts before building them. Paper - This point is not Explicitly discussed. It would be an appropriate subject for the icng-term review. E. Hearing, Moffett Subcommittee, 12/14/81 1. Chaiman's opening statement, 12/14/81 a. Statement - A new inspection progra-has been institutec for FY 82. Paper - Changes are in progress and continuing. Program revisiens in FY 83 are recommended as an initiative. 2. Discussions, 12/la/82 a. Statement - The fi've example plants are namec. Paper - The paper discusses the same five plants as examples. b. Statement - In the Chaiman's cpinion stancardi:atien is one of the ideas with the mest promise. Paper - This subject is not discussed. It wcule be an appropriate subject for the long-term review. ~ ' C. Hearing, Simpson Subccmmittee, 2/22/82 1. Hearings,2/22/82 2. Kamerer letter to Stafford, 4/1/82, followup. a. Statement - The NRC is evaluating the initiati\\es it the EDO's mem: of 1/21/82 to the Chairman. Paper - Specific acticns are ncu re:em encec. b. Statement - NRC is entcuragin; liter. sees tc per#crr thirc ; arty audits. NRC views INPC prograr as sup;iementing but r.ct sup-planting NRC efforts. Paper - NRC and industry -! nit'atives ar.c cestricticr.s are censister.: with this view.

e** o ? i D. Hearing, Ottinger Subcommittee, 3/3/82 1. Kammerer letter to Ottinger, 3/1/82, responses to questions a. Statement - FY 83 budget recuest ir.cluces about 40 professional staff and $200k for cualitf assurar.ce. Additional rese.rce estinates would be premature while assessing options for improve-ment. Paper - Enclosure 2 states that the NRC currently devotes 16 positions in HQ to QA and 28 positions in the regions to QA for construction reactors. These are the FY 83 numbers from the Chairman's recent budget crosscut. The paper also states that, in the past, funding has been abcut $400k per year. This corresponds roughly to what was expended in FY 82 and what is intended in FY 83 (aside from initiatives). b. Statement - The NRC is evaluating the initiatives in the ED0's memo of 1/21/82 to the Chairman. Paper - Specific a'ctions are now recommended. c. Statement - Specific evaluation criteria for initiatives have not been established. Initiatives are beinc evaluated for cost effective impact. Paper - No criteria are stated in the paper..It can be said that the staff generally considered cost-effective impact i,n, de'veloping. initiatives. d. Statement - Organizaitonal changes are being considered. Paper - Such changes are now recommended as an initiative. e. Statement - QA paper is scheduled for 3/31/82. Paper - Timing of paper is not discussed. 2. Hearings, 3/3/82 a. Statemert - An important lessen frcm Diablo Canycn is that NRC had a g3p in its treatner. of design. Paper - Several initiatives address this gap. b. Statement - There was a oe'nerai discussior. cf indecender.t aucits. They are not necessarily -eccired fer all clatts.

4-I ' Paper - The initiative ciscussions are censistent with the views presented. 3. Chairmar, letter to Moorhead, 4/19/E2, followup. a. Statement - Standardized design helps OA and lack of it makes QA more difficult. Paper - This subject is not discussed. It woulc be an apprc;riate subject for the long-term review. b. Statement - Criteria for independent accitors are described. Paper - No criteria are discussec. Tne pa;er is not that detailed in this area, c. Statement - Applicants are responsible fer the safety of their pl a r.ts. However, the NRC staf# cetermir.es if that responsiblity has been met. Paper - Page 2'of*the SECY paper states that none of the iritiatives is intended to detract from licensee respersibility. d. Statement - Over reitance by acolicants en thirc parties te manage QA is nct as effective as stronc ac;1icant involvement. Paper - This point is net explicitly ciscussed. It can be considered implicit in the problem statemer.ts where the importance of management ccmmitment and integraticn of QA intc effectivi prcject managerer.t teams are discussed. 4 E. Hearings, Bev'il Subcommittee, 3/22/E2 1. Hearings, 3/22/82 a. Statement - Responsibility remains with plar.ts' owners. Faoer - Pace 2 of the SECY paper states that r.cne of the initiatives is intended to detract from liter.see respcr.sibility. 2. V.ammerer ietter te Sevili, 3/26/52, #c'icnu: a. Stater.ent - The NP.C is evaluatint tne initiatives ir. tne ED0's me. c cf 1/21/82 to the Chairr.an. Paper - Specific actions a're rce recomre cec.

..<,e b. Statement - The QA paper is scheculed for 3/31/82. Paper - Timing of the paper is not discussed, c. Statement - Independent audits strengthen utility 0A and supplement NRC inspections-No pitfalls have been identified. Paper - The initiatives reflect this view. d. Statement - Independent audits dc not detract from licensee responsibilty. Paper - Page 2 of the SECY paper states that none of the initiatives is intended to detract from licensee respctsibility. e. Statement - NRC supports the INPO program and will monitor anc review it. NRC will design new efforts to minimize duplicaticr. and overlap, to take advantage of INPO while still providing acec;uate independent review. If the INPO prograr appears effective and vigorous, it may substitute for some NRC initiatives. Paper - These views are reflected in the initiatives. f. Statement - INPO criteria could be usec by third party aucitors. Paper - This is stated in the industry initiative description. g. Statement - The staff will review INP0 criteria and results tc achieve coordination. Paper - The industry initiative description states that during the trial program the staff will review the criteria and cbservet some evaluations. 4

[', ,p u% August 20, 1982 SECY-82-352 w....+j POLICY ISSUE For: The Comis(Notation Vote) From: William J. Dircks Executive Director for Operations

Subject:

ASSURANCE OF QUALITY

Purpose:

To inform the Comission of staff initiatives approved within the authority of the Executive Director for Operations, to improve the assurance of quality in the design and con-struction of nuclear projects; and to obtain Comission toproval to pursue revision of the NRC's statutory authority to allow implementation of a system of designated representatives analogous to the system employed by the Federal Aviation Administration. Discussion: The complexity and extent of problems that have been identified in the past few years at 5'of the 32 units now under active construction have caused concern regarding the quality of the design and construction of nuclear projects. These problems include nonconfoming structural steel welds at Zimer, seismic design. errors at Diablo Canyon, inadequate soil compaction at Midland, voids in concrete structures at Marble Hill and design deficiencies at South Texas. sumari,zes recent experience at each of these 5 projects. Analysis of the experience at problem sites has resulted in the classification of three primary problem areas: failure of the project management team to provide adequa.te management controls to prevent a :;1gnificant breakdown in quality from occurring; failure of the owner's quality assurance program

• Mt to detect the breakdown in a timely manner and to obtain the ew appropriate corrective action; cnd failure of the NRC's

. programs to recognize the true extent and nature of the t-problems. ,The first two problem areas are fundamentally derived from Q I a lack of total management comitment to quality at the b h0 nuclear projects inception. This lack of comitment has h l{0 , been exacerbated by the lack of understanding a' the role of quality assurance in project management and the lack of total understanding of what is required by personnel at all I levels of the process. F0I A-84-2M CONTACT: E. L. Jordan, IE M[3 492-4848 ~

s The Ccmmission The third problem area is two-fold in derivation. Historically, the NRC's licensing and construction inspection programs have not sufficiently examined the project management controls at sites under construction,'but have been oriented towards establishing adequacy within major technical and functional areas, e.g., concrete,. electrical, etc. The systematic assessment of management perfomance and evaluation of all other available infomation have not received the same level of effoit as operating sites. Second, previous NRC programs have not addressed design quality as specifically and extensively as other areas. In response to the breakdowns in quality ar.d quality assurance, the Chairman in a November 27, 1981 memorandum directed the staff to determine various approaches that could be taken to strengthen quality assurance, and to j provide the Commission a preliminary evaluation of the approaches that appear most promising. On January 29, 1982 the staff briefed the Comission on initiatives that appeared to merit further consideration. Industr from the Institute for Nuclear Power 0.perations (y representatives ~ INPO) met with the Comission on February 4,1982 to present their plans for improving the assurance of quality at plants ender construction. On July 15, 1982 the staff again briefed the Comission on the actions taken to date and the initiatives under consideration. The staff has developed initiatives that should lead to effective improvements in quality and quality assurance programs. While many of these initiatives require NRC l actions, the underlying principle in their development has .. been that the ultimate responsibility for quality and safety remains with the nuclear industry, and none of the initiatives are intended to transfer this responsibility to the NRC. The initiatives are designed to establish additional confidence in the quality of design and construction activities and improve the management control of quality. The initiatives are also designed to improve the NRC capability to evaluate the implementation of licensee programs. Compliance with NRC requirements for the quality l assurance program and its implementation is a major consideration in establishing this confidence. Although a resident inspector is now assigned to every site at which construction is more than 15 percent complete, the N9C is limited in its ability to assure compliance with all NRC requirements because of the limited inspection resources. The staff recomends implementation' of a" system of designated NRCrepresentatives(analogoustotheFAAsystem)toextend its inspection resources.

l I The Comission During the development of the initiatives presented in this paper, several additional actions were considered. Where we were not able to establish an adequate basis to initiate additional actions at this time, further study is warranted. A long-term rev hw will be conducted, utilizing knowledge within and outside NRC to evaluate the merits of additional actions, monitor the outcome of ongoing industry and NRC initiatives and initiate changes in program direction as appropriate. This review will satisfy the direction provided the NRC in an amendment accepted by the House and Senate conferees in their joint consideration of the NRC's FY 82-83 authorization bill.. Some of the actions considered and endorsed by ' he staff t l are associated with existing agency programs. The followup of allegations is an essential part of the NRC's inspection program, and is an effective extension of inspection resources. i Allegations provide an opportunity for non-NRC personnel to enter potential problems into the NRC's problem correction chain. The principal object <ve of the resultant NRC inspection effort is to obtain sufficie; t inform 4 tion through independent in-depth examinations to estaalish the significance of the particular allegation and to effect corrective action comen-surate with it's significance. To encourage and provide personnel an opportunity to :ake an allegation, NRC inspectors wear hard hats that uniquely identify them. The resident inspector's office has a telephone answering device for 24-hour i response to callers, and the telephone numbers of the resident inspector and NRC regional office are listed in the local tele-phone directory. Effective October 12, 1982 NRC postings at the site will identify the legal protection afforded people who provide allegations. With respect to enforcement, the NRC has sufficient authority to take appropriate enforcement action for inadequate quality assurance. The optiens extend over a broad range from meeting with a licensee, notices of violation and civil penalties to issuance of orders for modification, suspension or revocation of licenses. The staff intends to continue to take strong actions in response to significant quality assurance breakdowns and has expressed this intention in the enforcerrent policy. Current rules are not specific on whether or not a licensee or permit holder is required to notify the NRC of changes to the quality assurance program description previously accepted bytheNRCintheSafetyAnalysis. Report (SAR). Additionally, current regulations do not e'xalicitly r6 quire licensees or permit holders to implement tie accepted NRC SAR quality assurance

The Comission. program description. Rulemaking action is currently in progress which will clarify the NRC staff position regarding the types of changes to the licensees' and applicants' quality assurance program descriptions that can be made without informing the NRC and clarify, in the regulations, the requirement to implement the accepted quality assurance program description. Actions have been initiated at near-term operating license facilities to improve staff confidence in the quality of design and construction activities. These actions include self evaluations by licensees, and in most cases, an independent design review. The limited experience to date with the independent desfgn reviews conducted at LaSalle and San Onofre (Enclosure 4) includes the identification of numerous deficiencies (nonconfonnances with the original specifications), which-have required reanalysis. Relatively few of these i deficiencies have required hardware changes, and to date, none of the deficiencies identified would have prevented safety-related components, systems, or structures from perfonning their intended function. The initiatives in this paper are directed toward reactor facilities not yet licensed for operation. At this time, the staff concludes a reasonable basis for not backfitting these initiatives to operating reactors is provided by previcus reviews of the facilities, their operating history, extensive startup test programs, and the reviews and upgrades in response to TMI and Bulletin actions. Further consideration j will be given to operating reactors as part of the long-term t review. The initiatives sumarized below have been approved within the authority of the Executive Director for Operations. The staff will continue those actions that are already underway and implement the remainder of the initiatives as soon as practicable.. Each initiative is described further in Enclosure 1.

• Measures at Near-Tenn Operating License Facilities The NRC will continue to employ the measures currently in use to establish confidence in the quality and effectiveness of utility quality. assurance programs at near-tenn operating license facilities until other NRC or industry programs are capable of providing this confidence.

These measures include applicant self evaluation, independent design review and regional evaluations. 9 --,-------,--.w_ --_--.e,---,._

The Comission *

• Industry Initiatives The NRC will continue to interact with INP0 in its development of industry initiatives, measure their effective-ness and adjust the corresponding NRC actions to provide for effective use of both industry and NRC resources.
• Construction Programs The NRC will increase the resources allocated to the inspection of reactors under construction by an additional 0.3 (FY 83) and 0.5 (FY 84) staff years per unit under construction.

The NRC will complete development and implement planned revisions to enhance the effectiveness of its construction inspection procedures. The NRC will complete development and implement its program for construction assessment team inspections at selected facil'ities to provide a basis for evaluation of the management performance essential to quality construction. The NRC will complete development and implement the integrated design inspection process to assess the quality of design activities including examination of a:;-built configuration at near-term operating licensee facilities. The NRC will expand its capability to identify generic design and construction deficiencies by computerized analysis of information reported by vendors, construction permit holders and NRC inspectors.

• Management Quality management seminars for top level managers with facilities under design and construction should be sponsored by industry. The seminars would focus recent experience of selected managers and recognized experts in the design and construction of nuclear projects.

. The NRC will request that each utility with a facility under construction reevaluate its quality assurance program and implement improvements in areas where the evaluations identify a need. The NRC will take actions to improve the enforcement of existing standards for qualificat' ion of quality assurance and quality control personnel and pursue establishment of a system of third party qualification and certification for such personnel.

l ~he Commission. I The NRC will centinue to explore with labor and other organizations, potential methods and incentives to assure quality in design and construction related production activities.

• Long-Term Review The NRC will commence a long-term review for continuing evaluation of quality and quality assurance problems related to design, construction, testing and operations, and potential solutions to those problems and their impact on the adequacy of NRC quality assurance policies and programs.
• Quality Assurance Planning and Evaluation The NRC will make organizational realignments to combine within a single organization the functions of research, l

standards development and inspection program development for quality assurance at reactors. The licensing function will remain in NRR until the current backlog of licensing actions is completed. The Office of Inspection and Enforcement is designated lead office with responsibility for development of NRC policies and programs for quality assurance and for implementation of the quality assurance initiatives. The following staff recommendation is provided for Commission consideration and approval.

• Designated Representatives The NRC should pursue revision of its statutory authority to allow implementation of a system of designated representa-

. tives analogous to the system employed by the Federal Aviation Administration. The staff has developed resource estimates and implementation schedules for the new initiatives. The resource estimates, implementation schedules, and staff responsibilities for implementation are discussed in Enclosure 2. The staff responsibilities are assigned consistent with the organizational realignment. The resource estimates to implement the initiatives are consistent with the NRC FY C3-84 budget. The NRC resource estimates are summarized below. e ,,,,--.-----,,,-,.e-e,---- ,,..-----,-,--------._--._-----,_,,.-,-.,.-,,--...-.-,--..---,,_-,,,--n n e----- ...--w-

1 i The Ccmission 1 1. Estimated HRC Staff Resources 25 staff years new effort in FY 83 24 staff years new effort in FY 84 2. Estimated NRC Contractor Resources $2.2 million new effort in FY 83 $1.4 million new effort in FY 84 Recomendation: That the Comission approve the staff proposal as sumarized'above. Willi . Dircks Executive Director for Operations

Enclosures:

1. Initiatives 2. Resources. Schedules and Staff Responsibilities 3. Examples of Recent Quality Assurance Problems 4. Independent Design Review for Near-Term Operating License Facilities

mmissioners' comments should be provided directly to the office of

• h3 Secretary by c.o.b. Wednesday, September 8, 1982.

ommission staff' office consnents, if any, should be submitted to the Iommissioners NLT Tuesday, August 31, 1982, with an'information 2 py to the office of the secretary.. If the paper is of such a 13ture that it requires additional time for analytical review and Iomment, the Commissioners and the Secretariat should be apprised

>f when comments may be expected. )ISTRIBUTION: camissioners )GC IPE lECY LEGIONS ICA IIA IPA l 'DO l lLD SCRS iSLBP LSLAP

9 P, t e o O ENCLOSURE 1 O e 1 i e M e g e e e e

INITIATIVES I. Introduction The Comission has considered quality assurance to be a key factor in the design, construction and operation of nuclear power plants for many years. Proposed versions of the General Design Criteria used in 1967 recognized l the interest in quality assurance. Appendix B to 10 CFR 50, published in June 1970, described mandatory cr i;eria for acceptable quality assurance programs for safel;y-related features.* Subsequently, a number of national standards and regulatory guides providing additional guidance have been issued to upgrade quality assurance programs. In the 1973 time frame, the Atomic Energy Comission expended major effort to comugicate to industry the framework, e.g., plans, procedures, organization, of a quality assurance program that would be acceptable to.AEC. This framework is reflected in current quality assurance programs that have been approved by NRC. II. Problem Statement -.] Examination of the problems that have been identified recently indicates that the fundamental cause of most design and constryction deficiencies is the lack of total management comitment to quality. This lack of comitment has been intensified by the lack of understanding of the role of quality ~ assurance in project management and the. lack of total understanding of what is required of personnel at all levels of the process. Ine owner's project management team is responsible for the overall planning and management of the design, constniction, and testing of the nuclear power plant. If the senior management has a strong comitment to quality, and if that comitment is imbued in a capable project management team, then the subsequent actions of this team will comunicate that comitment to all involved parties. The project management team comunicates and "As used in this paper and defined in Appendix B, quality assurance comprises all those planned and systematic actions necessary to provide adequate confidence that a structure, system, or component will perform satisfactorily in service. Quality assurance includes quality control, which comprises those quality assurance actions related to the physical characteristics of a material, structure, component, or system which provide a means to control the quality of the material, structure, component, or system to predetermined requirements. d

' Enclosure 1 obtains through contractural and procedural arrangements with the designers, fabr'icators, and constructors a level of quality comensurate with the owner's comitment. The comitment to cost and schedule must be properly balanced with quality through these contractual and procedural arrangements. For example, if the constructor earns contractual credit strictly with the schedule of physical installation, the me3 sage from project management is production. On the other hand, if earned credit is comensurate with the schedule of owner accepted, adequately documented installation,.the message is quality production. The latter case provides the proper ) incentive for getting work accomplished right the first time. This is then reflected in the policy and procedural direction to the various organization sub-tiers. Similarly, the role of quality assurance in the project management team is determined by the senior management's comitment to quality. Proper J 1mplementation of the quality assurance criteria is an important element in successful project management. However, quality assurance programs cannot substitute for poor project management or a lack of comitment to quality. Quality assurance must be an integral part of all of the project planning and management activities from the projects inception, and its role must be comunicated and fully understood by all participants in the design and construction process (from senior management to the craftsman). For example, if the inspection function is planned and conducted as an i integral part of physical installation activities, then early detection and correction of procedural or other inadequacies will result in enhancing quality, ccet, and :chedule. All participants must be adequately trained to understand and obtain these benefits. Weaknesses in the existing approach to assuring quality are apparent. They are evidenced by the frequency and severity of design,and construction deficiencies, and by the failure or delay of industry and NRC recognition of the extent and nature of the breakdowns. Previous efforts by the NRC to assure program content and structure have l not been balanced with comparable efforts to assure successful program implementation. The NRC's licensing and inspection programs have not 'sufficiently examined the project management controls at sites under construction, but have been oriented towards assuring the adequacy within major technical and functional areas, e.g., concrete, electrical, etc. The systematic assessment of management performance and evaluation of all available information at construction facilities did not receive the same level of effort as operating sites. Previous NRC programs have not addressed design quality as specifically and extensively as other areas. I In sum the fundamental issues can best be characterized as the lack of total management comitment to quality and the uncertainty in industry's and NRC's ability to detect and correct the resulting deficiencies. The need to resolve these issues is the basis for the 'follbwing initiatives.

. Enclosure 1 i .II. Initiatives A. Measures at Near-Tem Operatino License Facilities For those plants in the Near-Term Operating License (NTOL) status, i the NRC has implemented three interim measures to provide additional confidence that required qJality assurance programs have been successfully l implemented and completed during the design and construction of the nuclear facility. These measures will be continued until replaced by l adequate industry programs or permanent changes in the present NRC program. 1. Self Evaluation l An applicant for an operating license will perform a comprehen-I sive self evaluation of the effectiveness of the quality assur-l ance program for design and construction. This requires an overall description of the project's quality assurance program for design and. construction. The self evaluation is a survey of the overall quality assurance program. The survey wt 1 describe the development and history of the program, managemen : involve-ment, audits, reviews, significant problems and corre.:tive actions. The NRC staff reviews the self evaluation c,d provides, the results of its review to the licensee. Additionet work, such as corrective actions or further audits, may be equired in particular areas. In addition, the Chief Executive C/ficer or l his designee is required to certify that the facility has been designed, constructed, and tested in accordance with the' Final i l Safety Analysis Report and other licensing comitments. 2. Regional Evaluation On each new operating license, the NRC staff considers whether there is a need for additional inspections of selected areas based on an evaluation of the project's inspection and enforce-ment history. This assures consideration of the need for a better assessment of performance in potentially weak areas. The project's inspection and enforcement history is evaluated with particular attention to the significant problems that have been noted at other construction sites. Other information considered includes known problem areas, results of NRC inspections and the Systematic Assessment of Licensee Performance program, and problems noted elsewhere with the same contractors. Additional inspections are performed as warranted in potentially weak i areas. l { I

i Enclosure 1 3. Independent Design Review Based upon results of the self evaluation and regional evaluation, an applicant for an operating license'may be requested to have an independent design review conducted. The criteria for detennining which facility, and the scope and extent of the design review also include the combined nuclear experience of the licensee, architect-engineer, and contractors. The review provides m evaluation of the quality of design based on a detailed examination of a small sample. The staff. specifies a sample area appropriate to the particular project. For LaSalle, the mechanical and structural loads on the residual heat removal system under blowdown and operating basis earthquake conditions were specified. The independent review addresses programatic areas, e.g., classification of systems and components, design and verification records, interface control and interdisciplinary review, consistency with FSAR, nonconfonnances and corrective actions, and audit findings and resolutions. The review includes verification of specific design features by independent calculations and comparison of installations against as-built drawings. The NRC staff reviews the selection of the independent review organization and the plan before implementation, audits the work in progress, and reviews the results. 8. Industry Initiative The industry initiative is not an NRC staff proposal, but a program that the industry is presently developing. The NRC staff is moni-l toring this program in order to take best advantage of the industry efforts. The Institute for Nuclear Power Operations (INPO) is developing criteria which will be used to evaluate quality assurance for design and construction. As with the existing INPO criteria for plant operation, they will be based on "best practice," rather than minimum standards of acceptability. Licensees will use the criteria for self-initiated evaluations (which can be perfonned either by an independent group within the utility or a contractor). The self-initiated evaluations will be submitted to INPO by the end of 1982. During this trial process, the NRC staff will be inw1ved by review-ing the criteria and observing some of the evaluations. Details of the staff involvement have not yet been developed. The industry will decide, by early 1983, on the direction of a continuing program. At present, the primary alternatives appear to be: INPO will either begin conducting quality assurance evaluations at individual construction facilities, or a fonn of self-initiated evaluation will continue.

. Enclosure 1 INFO is also conductin 1981 and October 1982)g management workshcps (May 1980, September with utility chief executive officers and plant managers in an effort to strengthen the utility cennitment to safe operation. NRC will coordinate its quality management seminars (Enclosure 1, Section D.1) with the industry efforts. C. Construction Inspection Procram 1. Procedure Changes The NRC does not have sufficient inspection resources to fully i implement all of the existing procedures in the reactor construction inspection The FY 83-84 NRC budget allocates an additional 0.3 (FY 83) program.and 0.5 (FY 84) staff years per construction project i j to execute the construction inspection procedures. The staff is i presently revising the individual inspection procedures for the various technical disciplines to better match the budgeted resources. l The main goals of the procedure revision are: (1) to facilitate performance of the procedures by resident inspectors with reduced 4 I input from regional-specialist inspectors; (2) eliminate redundancies in the procedures; (3) reexamine scope or frequency of some l inspections based on limitations on inspector resources; and (4) shift emphasis of inspection from record review to observation ,{ of work. This staff effort is continuing. The first series of I revised procedures which cover inspection of mechanical systems are j in the final stages of issuance. 2. Construction Assessment Team Insoections This initiative will extend the concept of the NRC's Performance. 1 Appraisal Team (PAT) inspection program for operating reactors to about four selected plants under construction per year. l This initiative was directed by the Connission in response to SECY 82-150 dated April 8,1982, "The Performance Appraisal Team (PAT)InspectionProgram." The procedures for performing management control inspections at nuclear construction sites were revised by the staff in 1981. i { The procedures covered licensee management performance in the following construction areas: Quality Assurance, Design controls, j Project Management, Construction Controls, and Procurement Controls. During 1981, eight trial inspections were perforwed by regional-based inspectors using the revised procedures. These inspections were effective in identifying management' l control problems not identified by the routine inspection i program. The manpower demand in these eight inspections caused the Regional. Administrators to defer. further pe.formance of this type of inspection. l 8 - m.

. Enclosure 1 The Construction Assessment Team inspections to be initiated by the IE staff would be similar to the construction inspections performed previously by the Regional Offices. A team of approximately six individuals with skills in the various areas to be inspected, including contractor persennel with appropriate backgrounds, will visit the selected construction site for two to three weeks. Additional site visits will be scheduled if necessary to collect additional information or clarify initial observations. The first site inspection has been tentatively scheduled for Bellefonte in September / October 1982. The construction assessment will complement the integrated design inspection. The latter is focused on a narrow area of techn,1 cal inspection, while the construction assessment is designed to assess the broader progrannatic controls. Like the integrated design inspection, tie scope of construction assessments will be modified to be responsive to unique conditions at a particul'ar facility. 3. Integrated Design Inspections The objective of this initiative is to expand NRC examination of quality assurance into the design process. The staff is developing an inspection approach which provides a comprehensive examination of the design dcvelopment and implementation for a selected

y:tc= :nd :tructure on a given project.

This evaluation will i encompass the total design process from the formulation of principal design and architectural criteria througn the development i and translation of the design and its revisions. It will conclude with onsite verification on a sampling basis, of.the design of the installed system and structure. This inspection will integrate and augment selected activities of NRR, IE, the vendor inspection program, and the regional office. Following development of the evaluation methodology, the staff will conduct a trial inspection with contractor assistance. Subsequent inspections will be performed with a substantial amount of contractor assistance. The results will be provided to the appropriate regional,and headquarters offices to be used as input to the overall NRC assessment prior to issuance of the operating license. The evaluation will be a multi-disciplined review that will address areas such as mechanical, electrical, structural, instrumentation and control. The evaluation will include checking sample calculations, however, the emphasis will be 1 on the systematic management of the total design process. The procedures to implement this approach are presently under development. A discussion of the' concep'tual'Togic necessary l to evaluate the design process follows.

l Enclosure 1 The evaluation will start with development of a logic or flow network of the design process. Each functional entity within the design organization will be identified. For each of these entities, internal and external design interfaces which invvive transmittal of design information will be specified. From this network, critical design areas or areas with the least. tolerance for error will be identified. Within each of the design entities, the specific procedures for the verification and transmittal of design information will be reviewed for conformance with the overall quality assurance program, and to identify specific weaknesses in the design process. Based on the results of the procedure review and the identification of critical design areas, a specific sgle of the system and structure.will be audited. Criteria will be preestablished for expanding or terminating the audit when problem areas are identified. In examining a system or structure and its specifications, the review will focus on topics such as: Validity of design inputs and assumptions. Validity of design specifications. Validity of analyses. Identification of system interface requirements. Potential synergistic effects of changes. Proper component classification. Rcticien centrol. Documentation control. Verification of as-built condition. The scope of the evaluation can be modified to be responsive to unique conditions for a particular facility, or known or suspected generic problems. This approach will examine all facets of the design management process for a limited sample. 4. Evaluation of Reported Information Improvements are planned in the current program for systematic review of information pertaining to design and construction quality that is now reported pursuant to 10 CFR 21. " Reporting of Defects and Noncompliances," and 10 CFR 50.55(e), " Construction Deficiency Reports." This program would have objectives similar to those of the Office for Analysis and Evaluation of Operational Data (AE00) for nuclear power plant operations. Computerized diagnosis would be used to enhance identification of relationships that may not be evident in the manual screening that is done now. No expansion of reporting requirements is currently planned but revisions are expected to facilitate computer input of key information. _-______m.

I 4 Designated Representatives Federal Aviation Regulations (14 CFR 183) prescribe the requirements for the issuance of designations to private persons to act in the capacity of FAA representatives in the examination, inspection and testing necessary for the issuance of aircraft certificates by theN authorized to represent the FAA in determining the compliance of aircraft, Administrator. aircraft components, and their repair or alterations with the requirem of the Federal Aviation Regulations. of the FAA in the perfomance of duties and are guided by the same requirements, instructions, procedures and interpretations as FAAThese pro employees in the perfunnance of those duties.the Designated M The DER represents the DesignatedEngineeringRepresentative(DER). FAA in helping to determine that the aircraft design co These in certifying certain product and manufacturing functions I periods of one year. l A similar technique of using the designated representatives would beIt l useful to the NRC inspecdon effort. inspectors available to implement the inspection program by providingUsing d an imediate source of r;alified experienced personnel. NRC representatives to cneck key aspects of the des is warranted, would significantly raise the NRC's confidence level of For example, during the quality assurance in nuclear power plants. preoperational and startup testing phases of a plant, all tests are monitored rather than a selected few. effort could also be applied at problem construction sites without having to reduce the routine level of inspection effort at other construction Under a statutory regime and regulations like those of the FAA, l NRC could ensure that t!e designated representative would not be subjec sites. to harassment as he would be under the same protection as NRC employees. The aviation industry uses holdpoints in the manufacturing process that require inspection and certification by an FAA inspector before the process The designated representative, provided by the aviation industry and acting for FAA, can provide that certification when required can continue. It is therefore an which allows the process to continue without delays. advan There is no and prevent costly delays in their manufacturing process.There are no analogous, situation to that process at nuclear plants. l l --w-- ,,-,--.a. n,_.,,.,_,---,c-_ ,_,n.,_, - -, ,a-,

.. Enclosure 1 preestablished holdpoints (other than Cp and OL issuance) in the construction or operation of a nuclear plant that require NRC approval before the process can continue. Therefore, for the designated representative program to be successful for the NRC, program incen'..ives would have to be developed to encourage the utilities to support the program. Under the Atomic Energy Act of 1954, as amended, and the Energy Reorganization Act of 1974, as amended, as well as other pieces of legislation, the Cormission does not have the authority to designate licensee personnel as inspectors in a manner similar to the Federal Aviation Administration's authority under its legislation and regulations (see 49 U.S.C.1355, 31 U.S.C. 483a, and 14 CFR 183). The FAA has the authority to issue designations to provide persons to act in the capacity of FAA representatives while remaining in their original employee status as far as receiving pay. To.act in a similar way the NRC would have to have the Atomic Energy Act amanded anc would have to promulgate regulations based on this amendment, The staff proposes to pursue the statutory changes necessary to implement a designated representative program and to continue development of program specifics. E. Managewent 1. Man!; ment and Quality' Improvement Programs The objective of this initiative is to improve attitude and performance throughout all levels of licensee and contractor organizations. The problems that have arisan at construction sites are closely associated with management attitudes and practices. Quality in design and construction is invariably associated with the highest level of management being totally committed to quality. Senior managers are personally committed and are unrelenting in their demands on their staffs and contractors for a similar connitment. The NRC will cooperate with industry in sponsoring a continuing series of seminars in which top level nuclear managers can communicate the advantages that can be gained through strong management involvement in their own QA programs. The seminars will be conducted with assistance from independent quality professional, ' utility and contractor representatives and the NRC. The seminars will be highlighted by the participation of managers from utilities which have experienced serious quality e

i 1 . Enclosure 1 i l I assurance problems and those which have managed highly successful programs. The independent quality professionals will convey the improvement principles and techniques of implementation. The utility and contractor representatives will identify incentives I for defect prevention based on direct experiences. The expected outcemes of these seminars will be recognition on the part of licensee and contractor management that positive incentives and benefits are achievable through enthusiastic implementation of aggressive quality assurance programs. While this initiative is directed to facilities under construction, participation by facilities in testing and operation will be strongly encouraged. l 10 CFR 50, Appendix B, Criterion II requires each utility to regularly review the status and adequacy of its quality l assurance program. The extent and nature of the recent breakdowns in quality assurance programs have indicated that this review has not been effective in maintaining an adequate quality l assurance program at several facilities under construction. i Each utility with a facility under construction will be requested o to reevaluate its quality assurance program, and to implement improvements in areas where the evaluations identify a need. Each utility should identify a senior executive with overall responsibility for the evaluation and implementation of the necessary improvements. The NRC will monitor the evaluation and implementation of the necessary improvements. It is expected that improvements in the cuality assurance program will incorporate actions such as those listed below. 1 4 (a) Conduct training sessens.for its personnel involved in t design and construction. These sessions should emphasize l the importance of each individual's contribution to ensuring quality and the enhancement to the cost and schedule goals which can be achieved with a positive program. The result of these sessions would be to get supervisors and employees J in the habit of talking positively about quality. (b) Provide better evaluation on a routine basis of status i reports to detect both trends and current nonconformance problems. Based on this information, meaningful corrective actions can be promptly taken to prevent recurrence of both i the specific problems and the root cause. The result would be defect prevention as a routine part of the operation. t l I i

, Enclosure 1 (c) Provide feedback on the achievements of the cuality assurance program, emphasizing the improverents from all involved in the program to maintain the concern and enthus'iasm on the project toward ensuring quality. -(d) Establish a system through which all parties are encouraged to comunicate to management the situations that make it difficult for the employee to perform quality work. This information will be included in the system for taking corrective actions. The result of this system would be that employees know that their problems can be heard and addressed. 2. Qualification and Certification of QA/QC Personnel A significant and prevalent problem in the construction of nuclear power plants is the qualification status of personnel working in the quality control and quality assurance areas. Some utilities have waived, without suitable bases, the education and experience requirements for these people. The NRC has not sufficiently enforced these requirements through its inspection efforts. Currently, various standards exist for the qualification of QA/QC personnel, for example: a ANSI N45.2.6 Qualification of Testing & Inspection Personnel b ANSI N45.2.23, Qualification of QA Audit Personnel c ANSI N626.3 (Draft), Qualification & Duties of Personnel Engaged in ASME Boiler & Pressure Vessel Code, Section III, Division 1 & 2 Certifying Activities l (d) ASME Section III, Division 2 Appendix VII Qualification of Concrete Inspection Personnel (e) ASNT, Certification of Level III Nondestructive Testing Personnel (f) AWS QCI-82, Qualification & Certification of Welding Inspectors NRC will direct more attention to the enforcement of the existing standards for the qualification of QA/QC personnel. Certification of personnel engaged in QA/QC inspections would provide a cadre of industry personnel that have been qualified to minimum standards and certified to have demonstrated inspection capabilities. l _ _, ~..

i

Enclosure 1 A program for third party certification of Nondestructive i

Testing (NDT) personnel is currently underway with coordination between the Electric Power Research Institute (EPRI), member utilities, American Society of Nondestructive Testing (ASNT) and NRC. The program includes developing a standard written practice for the qualifications of the level III NDT inspectors as well as administering b.asic and method examinations and specific and practical examinations in the respective areas of nondestructive testing. A registry of personnel holding the required qualifications and certification would be maintained by the third party organization. Unsatisfactory performance would result in removal from the registry through an established procedure. Programs similar to this could be established in other areas such as welding, inspectors, QA auditing, concrete inspectors, and laboratory testing personnel, Fonnal certification of various levels of QA/QC personnel will be i considered as part of the long-term review (Enclosure,1, Section III.F). i J g i 3. Craftsmanship The staff has initiated discussions with labor unions involved in nuclear construction in an effort to explore the potential methods and incentives to enhance the crafts role in assuring. the quality of construction activities. Feedback from the labor unions included the following points: (a) Craftsmen are nat well informed of their role in the (A/QC process. (b) Contiruous rewo.k as a result of changes has a demoralizing effect on craftsmen and effects the quality of the final work. (c) Utilities and contractors have not provided adequate training to craftsmen regarding quality. (d) Utilities are not convinced that quality assurance is a cost effective approach to construction. Labor perceived the utilities to think QA/QC was a "high cost" item rather than a " cost saving" tool.- E s

Enclosure 1

~ (e) Improved front-end engineerir.g and procurement would reduce the amount of change and rework. (f) A Nuclear Stabilization Committee has been established with representatives from labor, utilities and contractors to improve relations between labor and management. t 'The staff proposes to continue these discussions as part of the long-term review. F. Long-Term Review Long-term NRC quality assurance policies and programs w,i.11 be based i on a review which assesses existing agency and indus'try quality assurance activities in a broad manner and then recomends an integrated long-term agency plan for quality assurance. Additionally, the i review will focus the viewpoints of various sectors of.the public and the regulated comunity. The review will be conducted by the NRC staff and will include representatives from headquarters, the regional offices, and consultants to the NRC. The primary function of the long-term effort will be to conduct a thorough review of continuing quality and quality assurance problems, ' and to propose solutions to improve the quality assurance programs for design, construction, testing and operation. This review will include a detailed assessment of the problems that developed at i facilities such as Diablo Canyon, South Texas, Midland, Marble Hill, and Zimer. The object of this assessment will be to identify, as' concisely as possible, specific problems that have occurred and their root causes, particularly in the area of pro ramatic deficiencies. Additionally, the review will evaluate exist ng programs at facilities which have programs that are functioning properly in order to identify the positive aspects of those programs that should be applied generically. Both this review and the review of programs at problem facilities will involve sita visits by the personnel performing the review. Proposed solutions to generic and plant-specific quality and quality assurance problems will be reviewed critically to determine whether the recomended actions would actually resolve the identified problems. The review will develop estimates of the qualitative and quantitative value/ benefit and impact / cost of proposed solutions, and ways in which / 'they should be implemented for operating plants, plants presently under 3 construction, or for plants to be constructed in the future. 4 a f 9

, Enclosure 1 ~ The House and Senate in their current joint consideration of the NRC's FY 82-83 authorization bill have accepted in ccnference an amendment which directs the NRC to study ways to improve quality assurance. programs. Implementation of this review is consistent with that direction. G. Quality Assurance Planning and Evaluation. The recommendation to form a single organizational unit dedicated to the various aspects of quality assurance was made in a report prepared for the NRC by Sandia Labs in August 1977, entitled "A Study of the Nuclear Regulatory Commission Quality' Assurance Program."* The centralization of quality assurance functions has been one of j continual interest and now more than ever needs to be achieved. The NRC presently' views responsibility for quality assurance as threefold: first, to determine the adequacy of the licensee's quality assurance program description contained in the safety analysis report; second, to ascertain that the licensee has established and adequately implemented the approved quality assurance program and to l Verify compliance with NRC regulations; and third, to develop the l regulations, standards and guides addressing QA*in the design, construction and operation of nuclear facilities. The responsibility for these three functions is currently divided among three separate offices, NRR, IE and RES, with execution of the inspection Tunction from five regional offices. These three functional areas are not separate and discrete areas but are highly interrelated, requiring continual interface. For example, the inspection experience needs to be continually factored into the licensing effort, inspection program development and development of regulations and standards. In addition, recent quality assurance issues (e.g., Diablo Canyon, Marble Hill, South Texas) have been highly reactive and have required rapid NRC management attention and response from the three separate offices for their various quality assurance functional areas. l l Page 60, "A Study of the Nuclear Regulatory Conmission Quality Assurance 'rogram," NUREG-0321 l --._.-----a.----,.-,--,-----.,..,.,.w..----,,,n v v --w.e-n---- - - - -, - -,- - - - - -,, - - - - - - - - - - - ~ - _ _,,, - -, - -

15 - The functional quality assurance areas need to be realigned for the following reasons: (a) To more effectively utilize the limited staff resources and expertise in quality assurance engineering. (b) To establish a more discernible policy and position en cuality assurance issues. (c) To establish unity of control and to provide both information and coordination with industry. (d) To bring together the licensing, inspection and standards functi'ons on interrelated issues. (e) To provide industry a signal that NRC management considers quality a leading part of the NRC operation and 'of sufficient importance to depart from the existing organizational structure. It is recognized that most NRC activities are quality assurance g related and that the NRC review process is an interdisciplinary function involving many organizational components. NRC headquarters activities which relate to the development of HRC policy, rules, standards and guides, and review and evaluation of the implementation of licensee's QA programs are to be consolidated at this time. The censolidation will occur in the Office of Inspection and Enforcement. Ine licensing function will remain in NRR until the current backlog of licensing actions is completed. e. 9 4 6 8 9 '*

p.__ n s

• ge e

e ENCLOSURE 2 9 l l O .,n___ _

Resources, Schedules and Staff Responsibilities nuclear industry currently expends substantial resources for quality trance at power reactors. For example, about 7500 positions are currently sted to quality control and quality assurance on construction projects. NRC Regional Offices devote about 130 positions to inspection of power

tors under construction and vendors.

These inspections are concerned, to a at degree, with the effectiveness of the quality assurance programs in the tous areas that are being inspected. This effort, which provides a measure the overall effectiveness of the quality assurance programs, is the largest nent of the NRC's efforts related to quality assurance at construction facts. A smaller part of this inspection effort (about 28 positions) is owly directed towards inspecting the quality assurance programs. a broad sense, the headquarters offices also devote considerable efforts to lity assurance. In the narrow sense, they devote about 16 positions to act professional work on quality assurance programs as follows: 4 positions ' ansing; 7 positions inspection program (QA related) development and slopment of QA initiatives; 5 positions research and standards development. i of this effort can be considered applicable to construction projects. contractor work has previously been at a level of about $400,000 per year research/ standards development in quality assurance. estimated resources for the initiatives discussed in this paper are provided Tables 1 and 2. These estimates are consistent with the NRC FY 83-84 get. They can be suninarized as follows: Estimated Industry Resources I (a) In FY 83. ~ (i)) 280 man years new effort (ii 420 man years altogether (b) In FY 84 (1) 310 man years new effort (ii) 390 man years altogether Estimated NRC Staff Resources (a) In FY 83 (i) 25 staff years new effort (ii) 46 staff years altogether (b) In FY 84 (1) 24 staff years new effort (ii) 34 staff years altogether

3-3. Estimated NRC Contractor Resources ) (a) In FY 83 (i) $2.2 million new effort (ii) $2.2 million altogether (b) In FY 84 (i) $1.4 million new effort (ii) $1.4 million altogether New efforts correspond to the new initiatives developed by the staff, i.e., those that are not already underway and well established. Generally, additional efforts are not large in comparison to the resources already devoted to QA..The improvements will come mostly from redirection of existing resources. NRC staff and contractor resources can be redirected as necessary to acccmplish the initiatives without dropping any planned accomplishments although the depth and schedule of some planned accomplishments will necessarily be affected. NRC staff responsibilities with respect to development and implementation of the initiatives are indicated in Table 3. The schedules for accomplishing the initiatives are provided in Figure 1. 4 t

R4 I En a gas. i !! E E s-4 s s 2. x E. h :* 3! O d -d 4 6 * =. e. ~e j g =; e. i 1 i 5 r r. 3 = z --lli ee agn iT* i t 0 a. t E s ra 3 .I' 3no. ~. 0 E3 I. i EEE E f- " -t =. 4 :. 1 a =

• I, 1 e_

, M. h.i 4 4 e4 4 4.: 4 4

o.s.

a =l -s s ' i. t1 u., r.- 2 e ~. s- "gg. ge._ E

• a

?: I ...e ~ i.- i.): a 2

3 I.,

t -a s ft,. g W e a 5:s.. .-= 1 s z t = -2 a g a s av g }7 g ~: h s, t -. =. 3Il

,. s1 of.* j

= 2 a.3. :. 1 z i.g. 2 l I C 3 ~ i s. _.s .ga i l = 3 a. _5

• a*
  • t2

.r .s W 7 4 3e-y3 3.: 31" E ~ ~ l a 'I

1. s"I. '#'

aa1 a I. t E ~ w .35E3- $.Y..$ $.---55 m h h b $.:- E~~: t E

e g!!,; gR g s il .s - .e 1 4 m E = 1me,i.e. e. t .~.-- v 3; = -. ~.. x c 4 2" ~ ll

• R
• m 2 E 2

I g2 g 2 % .s' - 3 g t 5 e: 4 5 g$=5 a E]1 s_ ggg s a w E.e ~ ~ Cs{ 2 & ~; 2 gE $$g....e ~ .. ~ =.... e 4 4 4 4 4 4.i d 4

4 4.:

e 3 .2 g e I 4E 3 5 r. I ll ~ 2 i

i

• A

_h f R A R = 2 i{ 8 x e l ~ S 5 2 z-e I E E t 2: O I E I

.3 E E

2 2 s -2 a IE 2 1 8 g.2 3 t j 2 [' 0I" 1. Ie -5 g 8 -5 2 8 3 .2 2 8 "!2 2 2 *,* E F' i' y 3 s 3: g =::

8

.y

g]

5 a. eg-E 1: a8 a! .e r.e m3gi = i

i I m% l., a

!. z.. e:*ys: s v- !.I: & g r 3 8 5% a a m -{ 2

• 1 aeae 82 : a J J a:

g2 2 =- a a sc

8 9 4 d e 8 t c 0= 5 z E E E a j j2 = c = = E E t: to E I % % % % k lt z 1 m g 7 g

• 3 g
• w I

$kYIIII 5 a i s = = a a a a g =al a i

=======.

== .2= SD 3 I I 5 E E I I l I' m E E = EEU W r a a a i _g g I I 3 s=u s e M i. I I E E E I 3.El = a a I 1 3 (.: :8 2 g .I E a w

Ie sa;

. "s.3 3% O ~ a8 !2 ~ f _i { e g. :.! a y 32 5i8 tia

3;g?2

=

i: l.e 2 e: y: : 2 n s e < { E I s E. - ~ z

=~

E i s :: g-2 32 : &nn-1 a O

e ,i e ! s e l g-I 4 t I ~ 2 0 l_ ~ g 900mpleJg = l pwag aaatg I-! = g, t.. e....ee,ou,ue.g. L 13ldeWLi t eQa el e* j d dk h Jb Jk d db d gi I Je$ IIIeM C g euvevenesns - t vo2utts t

0 t puelptu - t Moess ston t 40sagtes - t taaeg f4 E

s l g t stuo;ettel

• 2 (NM t Jeg *12 em a

1 2 puetp6W t tenettes 3 l 14eeg ewsunaseg - 2 gweg ~ 3 iL L i E l [ g, t usalg. g saingsw I ( } l I I l C pae peten = Z ettesel lb j i i t.ie -,z y I l l t SJasuo 'es - t spasA oted g l g d [ l I I 2 atsn1 tg l I 8 I l l e I i t i i l l neweaows 3 I I : I l 1 [ }. 1 ' l.i. 11 1 .l. 1 t ouvewenesas - t a mans I >tm puest ~ 2 5 3 i 1-si a = bl m_. i g E n. 3 -j I M i.I ~' l ] s = E ] a j h!* -g E r s I 2 5 1*e E & t 4* I = = I I2 - 3 3 J I2 I "li; E 1 I2 : ~ s' 2 E 5 E 3

?

1 3 3 w

• =

ii '. 1 s i l e s. s i _i = s s = $ 5n =_ 3 :* I 1 i K ~y y ' L E E ?a(:E 3 = m.

ENCLOSURE 3 O e e I r l l l l 1 i i 1

k Examoles of Recent Quality Assurance problems During the past few years, there have been serious quality assurance problems at nuclear power plants. Some examples are listed below: 1. Marble Hill In June and July 1979, NRC confimed allegations of improperly repaired concrete imperfections at Marble Hill. identified as concrete consolidation problems (honeycomb and voids), improper repair (patching) of these imperfections. i NRC inspections confimed that: An excessive amount of honeycomb and air voids had occurred. a. Approxi-mately 4000 concrete patches existed, b. In many instances these imperfections were improperly repaired, and/or unacceptable materials were used for the repair. Quality control records traceable to the repairs were either non-c. existent or inadequate. ~ d. Personnel responsible for such repairs were inadequately trained and supervised, The iicensee was not in control nor sufficiently aware of the above a. circumstances. These events led to a halting of all safety-related work at theJite in August 1979. when the' utility's quality assurance program and that of its c e had been substantially upgraded and the adequacy of completed construction work had been verified. 2. Midland Excessive settlement of the diesel generator building was cbserved in 1978. The unexpected settling was subsequently attributed to inadequate and poorly compacted soil under the building. Other safety-related systems and structures were affected. NRC's investigation determined that design and construction specifications had not been followed during placement of the soil fill materials and that there was a lack of control and supervision of the soil placement activities by the utility and its contractors. Extensive rework has begun, and the operating license application is currently being litigated before an NRC Hearine Board.

i Enclosure 3 3. Zimer i Allegations received in January 1981 prcmpted an NRC investigation of quality assurance problems at the Zimer site. The investigation has l identified a large number of quality assurance related problems. The majority of the problems identified focus on the ineffectiveness of controls implemented by the licensee and its contractors for assuring the quality of work performed. In that regard, numerous deficiencies have been found concerning traceability of materials, handling of nonconfor-mance, interface between construction and quality control, quality records, and the licensee's overview of ongoing work. An extensive review of the as-built plant is being performed. Limited independent measurements were perfomed by the NRC in selected areas of concern in an attempt to characterize the actual safety stgnificance of these deficiencies. Although a few problems requiring corrective action were identified, the majority of the tests and examinations disclosed no hardware problems. The licensee will perform a ecmprehensive cuality confirmation program and resolve identified problems before an operating license is issued. a 4. . South Texas In response to allegations that QC inspectors were being threatened if they reported unacceptable items during concrete placements, the NRC initiated an investigation through its Region IV Office in July 1977. Ten investigations of allegations were perfomed during the period July 1977 i.o Nuves er 1979. The results of these investigations established that the allegations of harassment, intimidation and lack of support of QC inspectors were sub-stantiated. The investigation demonstrated shortcomings in the management and that the implementation of the QA/QC program at the South Texas Project did not meet the standards required to assure that the facility will be constructed to NRC requirements. Safety-related work was stopped in 1980. NRC allowed restart in designated areas only after QA for that area was upgraded and verified by the NRC. In January 1981, the licensee initiated a' design review of those portions of the enginearing design work performed by Brown and Root, Inc. The Quadrex Corporation assisted the licensee in this review. Briefly, the Quadrex report found that Brown and Root failed to properly implement an overall design consistent with the needs of a nuclear power plant. The licensee replaced Brown and Root with Bechtt.1 Power Corpration as architect-engineer in September 1981. NRC is monitoring the performance of.Bechtel as they resolve the problems identified in the Quadrex report.

. Enclosure 3 5. Diablo Canyon At Diablo Canyon, the Pacific Gas & Electric Company (PG&E) providec incorrect information to an expert consultant, who used the information in developing the seismic response spectra for the design of certain seismic piping and equipment restraints. NRC investigators have found that there was a lack of rigor and formality in the procedures used for verifying the accuracy of information transferred by PG&E to its consultants. These procedures did not comply with NRC requirements calling for verification of design info'mation at each stage of the process by an independent r person qualified in the pertinent disciplines. Proper quality assurance controls were not employed in technical and procurement communications with service-type contractors. Nor were document controls adequate to assure that those involved in design had ready access to the most recent information available. Following discove.ry of these errors in seismic design, the recently issued operating license for Unit I was suspended in late 1981. Prior to the NRC's reinstatement of the operating license the licensee will be required to complete an extensive design reverification program for those areas in question. e l I f l I

O 4 g g o o O e ENCLOSURE 4 e 0 9 6 9 t e t [ l I l t l l e

Independent Design Review For Hear-Term Operating License Facilities In order to provide further staff confidence in the quality of design and construction at near-term operating license applicants, licensees have been requested to conduct an independent verification of selected design and construction activities. The independent verification would be performed by an independent contractor with qualifications acceptable to the NRC. l Independent verification efforts have been completed at LaSalle Unit 1 and 2 and San Onofre Unit 2 and 3. Reviews are presently in process at Grand Gulf, Susouehanna, Shoreham, Watts Bar, Palo Verde, Sumer and St. Lucie. At LaSalle, the licensee hired an independent contractor, approved by NRC, to perform a review of the mechanical and structural design of loop C residual heat removal system excluding all branch piping less than 3 inches, in the functioning mode of the low pressure injection system using loads resulting from the actuation of the automatic depressurization system in conjunction with the operating basis earthquake to verify that this system has been designed and constructed in accordance with the application and that the NRC requirements have been satisfied. Comonwealth Edison contracted the Teledyne Engineering Services (Teledyne) to perform this review with NRC approval. The preliminary findings by Teledyne resulted in 21 Error / Deviation and 31 open-items reports which were transmitted to the licensee and the NRC staff. t Upon submittal of all Teledyne's preliminary findings, the licensee trMtted it rc:pc.:c: to Teledyne and the NRC staff and, in addition, the licensee received pemission from the NRC staff to establish a dialogue between Teledyne and its Architect-Engineer (Sargent & Lundy) to discuss the potential errors found in the Teledyne review. Of these 52 reports which involved various problems in the design area and none in quality assurance, 39 were closed by Teledyne based on the acquisition of additional information and/or clarification of existing information. The 13 remaining reports were reviewed by Teledyne's Project Review Internal Comittee. This comittee, composed of three senior level Teledyne engineers who together had the expertise to resolve the technical issues, and the Teledyne Project Manager concurred that none,of these reports have the potential for significant safety impact. The NRC staff reviewed those open-items and error-deviations reports submitted to the Project Review Comittee and concluded that these reports can be categorized as not having a signific' ant safety impact on LaSalle. In addition, the NRC staff feels that Teledyne has perfomed an in-depth review of the analytical procedures and design calculations used in the piping, equipment, and component support design to assure the adequacy of the design bases, the adequacy of the design implementation, and the consistency between i the design dccumepts and the Final Safety Analysis Report comitments. l 4 - _.. _. _ _ _, _, _... _ _ _ _ - _ ~.

O .. Enclosure 4 The indapendent design verification program conducted by Teledyne on the loop 1 C residual heat removal system indicated that the quality assurance control 1 and implementation, design process, procedures and Final Safety Analysis 4 Report commitments are acceptable except in the area of response spectra, which was reviewed by NRC staff. The results of the limited review provide increased assurance that the quality assurance program established and implemented by the licensee and its principal contractors did effectively control the overall program and construction activities for the LaSalle County Station. While several design deficiencies were identified, the overall design and construction activities were adequately performed so that no adverse impact on safety was found. At San Onofre the licensee contracted with Torrey Pines Technology, a subsidiary of the General Atomic Company (GA), to perform an independent evaluation of the seismic design and quality assurance program for San Onofre 2 and 3. The design verification encompassed a review of the seismic design of San Onofre 2 and 3 to: O a. verify that the design process converted the seismic design basis specified in the San Onofre 2 and 3 Final Safety Anal,ysis Report (FSAR) into the design documents that are transmitted to the constructor or fabricator, and b. evaluate the SCE quality assurance (QA) audit plan and its implementation at the construction site and the fabricator's shops. The design prccess performed by the equipment fabricators was not part of this review program. The work was divided into eight major tasks: Task A. Design Procedure Review i Task B. Design Procedure Implementation Review Task C. Seismic Design Technical Review Task D. Audit Plan Review Task E. Processing of Findings Task F. Reports ~ Task G. Pipe Segment Walkdown Task H. Independent Calculations The review was cenducted by individual GA reviewers investigating each area covered by Tasks A through D, G, and H. When a reviewer found a deficiency that might have safety significance, it was documented in a " Potential Finding ' Report." e /

,. Enclosure 4 After the Potential Finding Report (PFR) was written, it was sent to the "criginal design organization" that was responsible for the area covered by the PFR. The original design organization (000) then investigated the PFR and responded in writing. The PFR and the 000 response was then reviewed by a GA committee, and the PFR was classified as (1) Out of scope, (2) Invalid, (3) i Observation, or (4) Finding. ~ Out of scope items are those which are beyond the original scope of the review. For example, the review was oriented towards design verification. Procurement items are considered out of scope. Invalid Findings are the result of apparent deviations, uncovered in the course of the independent verification, that are resolved to the satisfaction of project personnel, usually during the Potential Finding review by the Original Design Organizations. Observations are valid deviations that are judged not to have the potential fer significant impact on the seismic design adecuacy of San Onofre Units 2 and 3. Findings are valid deviations that could have potential for significant impact on the seismic design adequacy. Of the total of 170 PFRs that were initiated, 77 were determined to be invalid after additional information was reviewed. Of the 93 PFRs that were date: ained to be valid, 7 were classified as findings and 86 as observations. The n;mbers of findings and observations for each of the v.arious tasks are as follres: l Task Findings Observations l A 3 2 8 1 35 C 1 41 0 2 5 G 0 2 H 0 1 Total 7 86 The staff has concluded, based on its review of the results of the design verification program, that the GA design verification program has not l disecvered anything that would cause the staff to change their previous conclusions that the San Onofre 2 and 3 quality assurance and seismic design programs are acceptable, and provides additional assurance that plant design l and construction have been appropriately accomplished. e 0 -,,,,-..,,--n- ,,,e , - -}}