ML20209E188

Recommends That Commission Forward Encl Proposed Ltrs Containing Final Decisions on Recommendations of Ford Amend Study Re QA Program Implementation Plan (NUREG-1055) to Congress

ML20209E188
Person / Time
Issue date:	02/25/1985
From:	Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
To:
Shared Package
ML20209E194	List: IA-86-605, Recommends That Commission Forward Encl Proposed Ltrs Containing Final Decisions on Recommendations of Ford Amend Study Re QA Program Implementation Plan (NUREG-1055) to Congress ML20209E188 ML20210G820
References
FOIA-86-605, RTR-NUREG-1055, TASK-PINV, TASK-SE SECY-85-065, SECY-85-65, NUDOCS 8504080205
Download: ML20209E188 (115)

Text

_ _ _ _ . _ _ _ . _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ . _ _ _ _ . _ _ _ _ _ _ __.

,e.*.

SA RfCg

.o g,

I q O 1 :- .:  ;

February 25, 1985

\ *****

/ SECY-85-65 POLICY ISSUE For:

The Commissione(Notation Vote)

From: William J. Dircks Executive Director for Operations

Subject:

QUALITY ASSURANCE PROGRAM IMPLEMENTATION PLAN

Purpose:

To provide the Commission with an information report on the staff's plan for implementing the recommendations of the QA report (NUREG-1055) and other QA initiatives. To provide the Commission with proposed letters to Congress containing the Commission's final decisions on the recommendations in response to the Ford Amendment.

Background:

In transmitting the QA report to the Commission (SECY-84-124), the staff committed to developing a comprehensive QA program imple-mentation plan that is task-oriented. In forwarding the QA report

. to Congress, the Commission noted that the QA report was complex and contained many interrelated recommended actions. The Commission indicated that it needed more time to deliberate further before informing the Congress of its final decisions on i

the recommendations in the report.

Discussion: The QA report, by Congressional direction, concentrated on quality and quality assurance in the design and construction phases of nuclear power plants. The QA report identified a number of recommended actions and areas for further analysis for improving design and construction quality programs. However, the report did not include a task-oriented action plan which placed the recommendations in the context of a comprehensive NRC program l for the assurance of quality. The attached QA program imple-l mentation plan describes such a program (the " revised QA program")

l and provides a task-oriented framework for its development and implementation.

This plan cuntains thirty program elements which have been divided into three categories: 1) those which would apply to a future generation of nuclear facilities which may be constructed; 2) those which can reasonably be expected to be implemented at the nuclear power plant projects now in the final stages of construction; and

3) those which can be applied to operating nuclear power plants.

The elements in each category are ranked in order of the staff's priority.

Contacts: G. T. Ankrum/W. D. Altman, IE

_ 492-4774

JThefommissioners The revised QA program consists of several interconnected features designed to address the root causes of past-quality and quality assurance problems. A particular focus is placed on the critical role of utility management in achieving and assuring quality.

Fundamental premises of the revised QA program include the following:

1) licensees, not the NRC, are pr.imarily responsible for achieving I and assuring quality; 2) substantive improvements in quality in the nuclear industry must come from the industry itself -- they cannot be " inspected in" or " regulated in" by the NRC; 3) the emphasis of NRC and industry QA programs on performance should be increased; and 4) NRC and industry QA programs should be oriented toward prevention and early detection of major quality problems.

The revised QA program includes the following key features:

Tighter Screening - A tighter screening of new construction permit applicants for managerial competence, project team experience, and financial capability and stamina.

Performance-Oriented QA Programs - A reorientation of present NRC QA inspection programs, in which the overall results of a QA program would be given greater emphasis, and conformance to the details of its component pieces would be given less emphasis than at present. Licensee trend analysis programs will be integral in this shift in NRC emphasis, as will identification and correction of root causes of quality problems. Performance-oriented QA programs emphasizing trend analysis apply to both the construction and operation phases, and in the revised QA program, they represent the most signifi-cant change in QA for operation.

Readiness Reviews, Master Inspection Plans, and Incremental NRC Review of Completed Work - The routine use of readiness reviews at predetermined stages throughout the design, construction, start-up and operation life cycle to ensure that a) work that should have been performed has been performed; b) the project is ready to proceed to the next step (e.g., transition from construction to operation); c) the project is planned better; d) senior management is formally involved at critical stages of a project; and e) interfaces, coordination and agreement on work progress and status are improved. Readiness reviews would be conducted by the licensee, reviewed and inspected by the NRC, and sequenced with a master inspection plan to provide continual confirmation of project quality at predetermined inspection points. Although readiness reviews can be used in all phases of the life cycle of a nuclear power plant, they are particularly well suited to providing incremental review of construction work in progress. A pilot program testing the combination of the readiness review concept in conjunction with a master inspection plan and incremental NRC review of completed construction work is under way at Georgia Power Ccmpany's Plant Vogtle.

Reduction of Design Changes - Actions to reduce changes and to complement the predetermined master inspectio'1 plan and

• The'Commissicners acceptance criteria described above. These would include:

a) a requirement for a more complete design before construction or a major modification begins (a standardized design would fulfill this requirement); and b) the use by the nuclear industry of configuration. management techniques (developed by the aerospace industry) to better manage project-related change and maintain current design documents.

Inspection Optimization - More efficient use of limited NRC inspection resources, including the acceptance of certain inspections and audits performed by non-NRC individuals or groups such as the National Board of Boiler and Pressure '

Vessel Authorized Nuclear Inspectors in certain situations instead of NRC inspections, thereby reducing or eliminating duplicative inspections. This revised inspection approach

• would be similar in some respects to the German TUV program.

The overview of th.e plan (Chapter 1.0) describes these key features in more detail and explains how they affect the life cycle of a nuclear power plant from the pre-CP stage through operation and major modifications. Subsequent sections of the plan describe specific activities envisoned by the plan for FYs 85, 86, and 87.

Other important program activities included in the plan but not covered in the preceding synopsis of key features include:

recognition and support of promising industry efforts to improve management; increased emphasis on senior-level management meetings between the NRC and utilities; better understanding of the impact of state utility regulatory agencies' actions on safety and quality; improvements to the NRC inspection program; revisions and improvements to QA guidance and standards; QA for computer software; increased attention to the qualifications of QA/QC personnel; better NRC planning for quality programs for the future; and quality assurance for temporary waste storage installations, permanent high-level waste repositories, fuel facilities, and transportation.

To the extent possible, the staff has not let the preparation of the implementation plan delay initiating the most important acti-vities covered by it. The staff is presently at work on over half of the activities described in the plan. For example, the Vogtle pilot program is under way to test how the readiness review concept used by NASA and DOE can be applied to the NRC licensing process. Other licensees have initiated discussions with the NRC  !

staff regarding additional readiness review pilots. These discussions have included a plant on which construction may be restarted after having been suspended. The performance trend analysis concept also has generated considerable interest and discussion within the NRC and between the NRC and the industry.

Preliminary discussions have been held with several licensees regarding potential pilot tests of the performance trend analysis concept. Further progress on this initiative is pending a policy-level decision by INPO and the Nuclear Utility Management and Human Resources Committee (NUMARC) on member utility parti-cipation in pilot projects.

l

\

• O e

e. f The Commissioners l Other implementation plan activities under way include the following:

transfer of lead responsibility for a current rulemaking activity on design changes from RES to IE  ;

pilot tests of increased residents and team inspections in i Region V -

l 1

increased activity in QA standards development l staff and research activity to develop improved guidance for systematically determining the set of plant structuras, systems, components, and activities to which the NRC-required QA program per Appendix B applies (i.e. , the "Q-list")

development of QA programs and guidance for non-reactor activities, including high-level waste repositories, monitored retrievable storage facilities, and independent spent fuel storage installations.

fhese activities dre in addition to ongoing QA program improve-ments initiated in the past two to three years, including the Independent Design Verification Program, the Integrated Design Inspection Program, and the consolidation and integration of NRC headquarters QA activities.

The activities described in the plan were based on the QA resources which were budgeted for FYs 85, 86, and 87. Those budget plans provided for small growths in IE's QA resources during each year. However, the OMB budget mark for FY 86 substantially reduced the resources allocated to QA. As a result, some planned work in the QA program implementation plan will have to be dropped, suspended, or indefinitely deferred. Since NRC must maintain its programs for ensuring the safe operation of current plants, OMB budget actions will make it necessary for NRC staff to cancel planned work which would apply only to a future generation of plants. Implemen-tation plan elements dropped or indefinitely suspended as a result of the OMB mark include:

• all activities associated with future plants only, including work on limiting design changes, third party audits, a CP advisory panel, and development of a revised screening program for new CP applicants

• reduction of the inspection burden on licensees

• some Kist Report follow-up actions related to internal NRC QA program improvements

• QA for transportation and fuel facilities

1 D f The-Commissioners The staff has developed proposed letters to Congress (Enclosure

1) containing the Commission's decisions on the staff's recommen-dations in the QA study. The proposed letters to Congress describe the deliberative process the Commission used (public comments, staff briefings, resource considerations) to reach its decisions. It also indicates that planned work applying only to a future generation of plants will not be undertaken as a result of the OMB budget actions. At present, there are no legislative actions to recommend. All of the planned and ongoing activities in response to the Ford Amendment, as well as the rest of the implementation plan, can be and are being undertaken within the authority of the EDO. The letters to Congress concentrate on policy matters, budgetary considerations, changes to the staff's recommendations contained in the QA report, and Commission conclusions on the recommendations of the QA report.

The letters also note other improvements.to overall NRC programs which will have the effect of improving quality in the nuclear industry. Proposed Commission conclusions on the recommendations of the QA report and staff implementation plan are summarized in Attacnment A to the letters to Congress.

The NRC staff will continue to provide the Commission with quarterly status reports on the NRC QA program activities. Future status reports will concentrate on the progress in implementing the QA program implementation plan. The plan will be revised as necessary to reflect changing priorities and past accomplishments, and those

~

changes will be reported.

Recommandations: That the Commission:

(1) Forward the enclosed letters to Congress containing the Commission's final decisions on the recommendations of the Ford Amendment Study (QA report).

(2) Permit the staff to award a competitive task-order-type contract for assistance in implementing quality assurance initiatives. In the Chairman's memorandum to me, dated October 31, 1984, permission to issue a competitive solicitation was granted, but permission to award a contract was withheld pending final Commission action on the QA report to Congress.

William J. Dircks Executive Director for Operation.s ,

Enclosures:

1. Proposed Letters to Congress

2. Quality Assurance Program Implementation Plan, January 1985

e f 6-Commissioners' comments or consent should be provided directly to the Office of the Secretary by c.o.b. Tuesday, March 12, 1985.

Commission Staff Office comments, if any, should be submitted to the Commissioners NLT Tuesday, March 5, 1985, with an information copy to the Office of the Secretary. If the paper is of such a nature that it requires additional time for analytical review and comment, the Commissioners and the Secretariat should be apprised of when comments may be expected.

DISTRIBUTION:

Commissioners .

OGC OPE ,

OI OCA

_ OIA OPA REGIONAL OFFICES EDO ELD ACRS ASLBP ASLAP SECY

- e ., , , - , e 4 , -

., . ~ <

4 4

e 9 a

9 O

e ENCLOSURE 1

The Honorable George Bush President of the United States Senate Washington, DC 20515 .

Dear Mr. President:

The NRC Authorization Act for fiscal years 1982/83 (P.L.97-415) directed the NRC to conduct a study of quality and quality assurance (QA) in the design and construction of nuclear power plants and to develop improvements to NRC's and the industry's programs for achieving and assuring quality in design and construction. In April 1984, the NRC staff completed their study in response to that Congressional request. The report of that study is entitled Improving Quality and the Assurance of Quality in the Design and Construction of Nuclear Power Plants. The Commission transmitted that report to Congress on April 20, 1984, and it was subsequently published as an NRC staff report, NUREG-1055. In the transmittal letter, the Commission indicated that the staff report was complex and had several interrelated recommendations. The Commission also indicated that it would need to take further time for analysis, including an opportunity for public comments, before informing Congress of its final actions and recommendations on improving quality and the assurance of quality in the design and construction of nuclear power plants.

The Commission has had several meetings with the staff to discuss the report and its implications; public comments on the report have been solicited and analyzed; further review and analysis have been performed on some issues; and the Office of Management and Budget (OMB) has completed its mark up of NRC's budget.for FY 86.

The Honorable George Bush As a result of those activities and other deliberations, the Commission has arrived at its decisions on the staff's recommendations, and they are described in this letter. The following brief discussion of the Commission's regulatory philosophy as it relates to quality assurance is intended to put the decisions into suitable context.

Licensees, not the NRC, are primarily responsible for achieving and assuring quality. Substantive improvements in quality in the nuclear industry must come from the industry itself; they cannot be " inspected in" or." regulated in" by the NRC. The QA report to Congress found that the key to achieving quality and assuring quality lies with utility management. The focus of NRC's and the industry's programs for improving quality in the nuclear industry should be oriented toward prevention and early detection of quality problems, and identifying and correcting root causes of those quality problems. NRC's fundamental task, however, is to assure the safe operation of existing plants and the operation of those presently under construction when they come on-line.

P. L.97-415 directed the NRC to examine and develop solutions to quality problems in the design and construction of nuclear power plants. NRC histori-cally did not devote much direct inspection effort to reactors under construc-tion. This inspection approach was based on the consideration that the potential radiological hazard to public health and safety was posed by operating plants (and not by plants under construction), and therefore the NRC limited resources allocated to plants under construction. These programmatic considera-tions resulted in little forward-looking program development activity and an

The Honorable George Bush inspection program based on the premise that any significant design or construction problems would be found during an intensive start-up and testing program. With the exception of the design problem which occurred at Diablo Canyon, this system worked well in preventing the licensing and operation of plants which were not designed or constructed properly. However, it has not worked well in identifying and correcting problems at the earliest possible stage. It has resulted in large impacts on ratepayers and utility owners and has reduced public confidence in the overall ability of the industry to fulfill the expectations placed upon it.

In response to these lessons identified in the QA report to Congress, the NRC has increased and redirected its design and construction inspection effort for plants currently under construction and is using and testing new approaches for assuring quality at plants currently under construction. In additiun to these improvements to NRC's inspection programs, the NRC has made a number of other improvements to its overall programs which have th^ effect of improving quality in the nuclear industry. These improvements focus largely on the approximately ninety units presently licensed for operation and include such initiatives as improved training and requalification programs for plant operational staff; j increased staff and industry attention to preventive and routine maintenance; research programs on the relationship between plant reliability, safety, and quality; a major new focus on the effectiveness and quality of plant activities related to plant outages; and a thorough reanalysis, for all operating plants, l of the NRC-required technical specifications issued as conditions of operating licenses.

l l

The Honorable George Bush Since the introduction of the legislation that led to the QA study, the industry also has initiated a number of activities that should result in an overall improvement in the quality of nuclear pl-ant construction ar.d operations. These include the Construction Project Evaluation Program of the Institute of Nuclear Power Operations (INPO); the establishment of ?. new senior-level ~ industry group, the Nuclear Utility Management and Human Resources Committee (NUMARC), to address some of the difficult people-oriented issues facing the industry, such as operator training; and the voluntary use of third party management audits by some utilities. In addition, Georgia Power Company has volunteered their Vogtle project for a pilot test of the readiness review concept identified by the QA report to Congress as an area for further analysis, and which may lead to improvements in managing, licensing, and confirming the quality of nuclear construction projects. The pilot test is curr&ntly under way, and it is being followed closely by both the nuclear industry and the NRC.

Although these new NRC and industry programs should, in a broad sense, lead to improved quality in the nuclear industry, the purpose of this letter is to describe the Commission's decisions on the recommendations contained in the staff's QA report to Congress prepared in response to the congressional direction on QA contained b. F. L.97-415. In response to this direction and as a follow-on to the QA report to Congress, the staff has developed a plan to revise NRC's administrative practices pertaining to quality and quality assurance. The plan contains three categories of actions: 1) those actions which would apply to a future generation of nuclear facilities which may be constructed; 2) those which can reasonably be expected to be implemented

o .

The Honorable George Bush at nuclear power plant projects presently in the final stages of construction; and 3) those which can be applied to operating nuclear power plants. Although the plan necessarily deals principally with NRC actions, we believe that it does embody the Commission's previously described QA philosophy which emphasizes the importance of the industry's role in improving nuclear quality. The NRC directly controls only its own activities, but NRC actions can provide great leverage to foster or inhibit licensee actions. The QA report made it clear that simple regulatory compliance is not enough to assure excellence in the construction or operation of nuclear power plants. The planned NRC QA program revisions are intended to create an environment which maximizes the incentives for and ability of utilities to achieve excellence while assuring compliance with the Commission's regulations.

The revised QA program envisioned in the staff's plan, and with which the Commission agrees in principle, includes the following key features:

Tighter Screening - A~ tighter screening of new construction permit applicants for managerial competence, project team experience, and financial capability'and stamina.

Performance-Oriented QA Programs - A reorientation of present NRC QA programs, in which the overall results of the program would be given greater emphasis, and conformance to the details of its component pieces would be given less emphasis than at present. Licensee trend

The Honorable George Bush analysis programs would be integral to this shift in NRC oversight focus, as would identification and-correction of root causes of quality, problems.

Readiness Reviews, Master Inspection Plans, and Incremental NRC Review of Completed Work - The routine use of readiness reviews at predetermined stages throughout the design, construction, start-up and operation life

. cycle. Readiness reviews would ensure that a) work that should have been performed has been performed; b) the project.is ready to proceed to the next step (e.g., transition from construction to operat' ion); c) the project is planned better; d) senior management is formally involved at critical I stages of a project; and e) interfaces, coordination and agreement on work progress and status are improved. Readiness reviews would be conducted by the licensee, reviewed and inspected by the NRC, and sequenced with a master inspection plan to provide continual confirmation of project quality at predetermined inspection points. Although readiness reviews can be used in all phases of the life cycle of a nuclear power plant, they are i

j particularly well suited to providing incremental review of construction work in progress. The pilot program under way at Georgia Power Company's Plant Vogtle is testing the combination of the readiness review concept in conjunction with a master inspection plan and incremental NRC review of completed construction work.

The Honorable George Bush Reouction of Design Changes - Actions to reduce and control changes and to complement the predetermined master inspection plan and acceptance criteria des.cribed above. These would include a) a more complete design before construction or a major modification begins (a standardized design would meet this need), and b) the use by the nuclear industry of configura-tion management techniques (developed by the aerospace industry) to better manage project-related change and maintain current design documents.

Inspection Optimization - More efficient use of limited NRC inspection resources. This would include the acceptance of certain inspections and audits performed by non-NRC individuals or groups such as the National Board of Boiler and Pressure Vessel Authorized Nuclear Inspectors in certain situations instead of NRC inspections, thereby reducing or elimi-nating duplicative inspections.

The above program features are intended to improve quality and quality assurance over a broad range of nuclear activities falling under NRC oversight, including.

nuclear plant operation, maintenance, modification, waste management, and transportation. Some of these areas can and will be implemented without sub-stantial further development work. However, some will require additional study or extended rulemaking proceedings.

The Honorable George Bush Impact of Budget Cuts-on Planned QA Improvements As originally drafted, the staff's plan was based on previously planned growth in resources assigned to the NRC quality assurance program in FYs 86 and 87.

As a result of recent OMB actions affecting NRC's budget for FY 86, a number of the activities contained in the plan will be dropped, suspended, or indefinitely deferred. Since the Commission must maintain its programs for ensuring the safe operation of current plants, OMB budget action will make it necessary for the NRC staff to drop planned developmental work which would apply only to a future generation of plants.

The Commission recognizes that the cancellation of this future-oriented work is not consistent with the forward-looking nature of the Congressional guidance on quality assurance contained in P. L.97-415. However, resource reductions for NRC's QA program have led the Commission to conclude that its near-term QA responsibilities for operating plants require termination of the planned future-oriented work which does not also lead to improvements in the quality of

~

plants currently under construction or in operation. Specifically impacted by the budget cuts will be future-oriented developmental work related to the screening of new applicants, the degree of design completion and configuration management techniques.

_- _ _ _ - - - - - . - - - - - - - . - - - - - - - - _ _ _ _ _ _ _ 9

The Honorable George Bush The Commission's conclusions and exceptions regarding the QA report to Congress which was previously transmitted as a staff report are contained in Attachment A. Administrative actions that are under way or planned by the Commission ~to address the findings of the QA report are summarized in Tables 1 and 2 of l l Attachment A.

l As stated in its 1984 policy and planning guidance document to the staff, the Commission believes that NRC's fundamental task is to ensure that existing _

nuclear facilities and those coming on-line in the future operate safely.

-Consequently, the highest priority will be given to assuring that operating facilities maintain adequate protection of public health and safety. Following operating plants in decreasing order of priority are, respectively, plants currently under construction, and future plants for which construction applications have not yet been received. The decision to drop planned develop-mental work that applies only to a future generation of plants was made consistent with this prioritization.

At this time, no changes to NRC authorizing legislation are recommended. Each of the-Commission's proposed administrative actions can be implemented within the NRC's current statutory authority. However, after further analysis, some issues may result in legislative proposals at a future date.

Sincerely, Nunzio J. Palladino Chairman

Enclosure:

Attachment A, Commission Conclusions on the Staff QA Report to Congress

i ATTACHMENT A COMMISSION CONCLUSIONS ON THE STAFF QA REPORT TO CONGRESS The Commission is in general agreement with the findings and conclusions of the QA report (which was previously forwarded to Congress as a staff report) and has approved the staff's action plan (which takes into account public comments received on the QA report and current budget limitations) to implement the' j findings of the report. The Commission endorses the staff's earlier findings,

, conclusions and recommendations contained in the QA report, with the following i

modifications:

1) Construction permits (cps) for future nuclear power plant applicants need not be conditioned on post-CP demonstration by the applicant of its capability and effectiveness in implementing its quality assurance program.

1 The NRC has ample authority within its current regulations to take

appropriate action if a licensee does not capably and effectively implement

its QA program. Changes in the Commission's practices are being undertaken,

, however, to improve NRC and licensee ability to assess the effectiveness of QA programs at the earliest possible stage in the construction process.

Licensees should demonstrate the effectiveness of their programs continually throughout '.he construction process, but the CP need not be conditioned on such demonstration.

2) It is not clear that a new, special advisory board needs to be established to supplement staff advice on the qualifications of new applicants. This is one alternative that would have to be further analyzed before imple-mentation to see if an additional source of advice is needed or whether an ,

l already existing body (e.g., the Advisory Committee on Reactor Safeguards)

! could perform that function. Efforts by industry-sponsored groups such as the Nuclear Utility Management and Human Resources Committee (NUMARC) and the Institute of Nuclear Power Operations (INPO), having the effect of a "self-screening" process, will also affect any decision in this area.

, Planned further analysis of the advisory board concept will not be undertaken l

at this time.

j 3) Other feasible alternatives to mandatory third party audits are available to accomplish the QA program goals of improved management and increased prevention and detection capability. Readiness reviews, now being tested and evaluated at the Vogtle nuclear plant, provide one possible alternative.

Increased NRC use of team inspections and resident inspectors and voluntary use of third parties by some utilities provide other alternatives. As a result of these considerations, the staff has proposed, and the Commission has agreed, to defer rulemaking activities to require periodic third party audits for plants under construction.

l

TABLE 1. Commission Action on Staff Recommendations Contained in the QA Report to Congress (see NUREG-1055, Table 2.1)

QA Report Recommendation Commission Action

1. Enhanced Pre-Construction Permit Planned developmental action (CP) Review of Experience and cancelled due to budget priority Managerial Quality -

of operating plants Establishment of an Advisory Dropped pending exploration of Board potential industry action on this topic

2. Post-CP Demonstration of Dropped in favor of Readiness Review Management Effectiveness Program

3. QA Program Performance Develop for potential implementation Objectives at operating plants and plants under construction. Cooperative NRC/

industry performance objective and trend analysis pilot programs will be explored

4. Management Appraisals as an Drop-in favor of. readiness reviews and Adjunct to Construction other management initiatives. Limited Appraisal Team (CAT) management process reviews implemented Inspections as part of CAT inspections

5. Inspection Prioritization All plants currently under construction of Plants Currently Under are receiving a-" graded" inspection Construction (Problem Plant approach. Planned long-term analysis Identification) (for application.if construction of a new generation of nuclear power plants should begin) indefinitely suspended

( due to budget priority of operating plants

6. Improved Diagnostic Capability / Implement as part of QA Performance Trend Analysis Objectives Program I

l 7. Senior Management Meetings Implement

8. Enhanced Vendor Program Being implemented

9. Third-Party Audit / Interim Recommended third party rulemaking i

CAT Inspections / Interim deferred pending outcome of readiness j Independent Design Verification review program; no activity planned

! Program (IDVP) due to budget priority of cperating

! plants; perform CAT inspections at 4/ year (no expansion in program);

continue IDVP on case-by-case basis

10. Regional Team Inspections Increasing use of team inspections.

j Pilot under way in one NRC region l _ ._.

LTABLE 2. Commission Action on Staff Recommendations for Further Analysis Contained in the QA Report to Congress (see NUREG-1055, Table 2.2)

Area Recommended For Further Analysis Commission Action

1. QA Report-Type Study for Plants Limited study efforts under way in Operation -

or planned

2. Prioritization of QA Measures: Analysis is in progress Guidance on " Safety-Related" vs.

"Important to Safety"

3. Measuring Effectiveness of QA Research projects focusing on plants under construction and in operation under way

4. Essentially Complete Design Combined with #5 below at CP Stage

5. Configuration Management Implementation planned by subsuming into rulemaking on control of post-CP design changes (transferred to the Office of Inspection and Enforcement from the Office of Nuclear Regulatory Research in September 1984). Planned developmental work related to future plants dropped due to budget priority of operating plants

6. Readiness Reviews Pilot program being initiated for plants under construction

7. Quality Engineering Further analysis and developmental work not planned due to budget priority l of operating plants i

l 8. Alternate Ownership and Efforts redirected to explore improve-Management Arrangements ments within current arrangements in

! an all operating rea'ctor i environment

9. Feasibility of Designated Further consideration deferred pending l Representatives a licensee request for a pilot program.

l Budget priorities preclude any activity in this area

10. Limiting Construction Permits Analysis of the problems faced by NRC in

the event of resumption of nuclear plant construction is included as a possible i contingency in the study described in 8.,

above l

l l

i  ?

l

. c QA Report Recommendation Commission Action

11. Expanded Resident Program More residents being assigned to sites. Pilot test under way in one NRC region to reduce regional insp~ections in favor of resident inspections .

12. Improved Licensee Detection Develop self-detection capability Capability for potential implementation as part of the QA Ferformance Objectives Program recommendation  ;

13. Independent Audits of NRC QA Implement Programs

The Honorable Thomas P. O'Neill Speaker of the United States House of Representatives .

Washington, DC 20515

Dear Mr. Speaker:

The NRC Authorization Act for fiscal years 1982/83 (P.L.97-415) directed the NRC to conduct a study of quality and quality assurance (QA) in the design and construction of nuclear power plants and to develop improvements to NRC's and the industry's programs for achieving and assuring quality in design and construction. In April 1984, the NRC staff completed their study in response to that Congressional request. The report of that study is entitled Improving Quality and the Assurance of Quality in the Design and Construction of Nuclear Power Plants. The Commission transmitted that report to Congress on April 20, 1984, and it was subsequently published as'an NRC staff report, NUREG-1055. In the transmittal letter, the Commission indicated that the staff report was complex and had severai interrelated recommendations. The Commission also indicated that it would need to take further time for analysis, including an opportunity for public conments, before informing Congress of its final actions and recommendations on improving quality and the assurance of quality in the design and construction of nuclear power plants.

The Commission has had several meetings with the staff to discuss the report and its implications; public comments on the report have been solicited and analyzed; further review ano analysis have been performed on some issues; and the Office of Management and Budget (0MB) has completed its mark up of NRC's budget for FY 86.

9

The Honorable Thomas P. O'Neill ~

As a result of those activities and other deliberations, the Commission has 1

arrived at its decisions on the staff's recommendations, and they are described in this letter. The following'brief discussion of the Commission's regulatory philosophy as it relates to quality assurance is intended to put the decisions into suitable context.

Licensees, not the NRC, are primarily responsible for achieving and assuring quality. Substantive improvements in quality in the nuclear industry must come from the industry itself; they cannot be " inspected in" or " regulated in" by the NRC. The QA report to Congress found that the key to achieving quality and assuring quality lies with utility management. The focus of NRC's and the industry's programs for improving quality in the nuclear industry should be oriented toward prevention and early detection of quality problems, and identifying and correcting root causes of those quality problems. NRC's fundamental task, however, is to assure the safe operation of existing plants ar 'he operation of those presently under construction when they come on-line, P. L.97-415 directed the NRC to examine and develop solutions to quality problems in the design and construction of nuclear power plants. NRC histori-cally did not devote much direct inspection effort to reactors under construc-

., tion. This inspection approach was based on the consideration that the

! potential radiological hazard to public health and safety was posed by operating plants (and not by plants under construction), and therefore the NRC limited resources allocated to plants under construction. These programmatic considera-tions resulted in little forward-looking program development activity and an

-____--_'_ . _ _ _ _ _ _ . . _ . ~ - _ _ . _ _ . _ _ _ _ _ _ _ . _ _ _ __ __ _ _ , _ _ _ ,_ ,

l .

The Honorable Thomas P. O'Neill inspection program based on the premise that any significant design or constructicn problems would be found during an intensive start-up and testing program. With the exception of the design problem which occurred at Diablo Canyon, this system worked well in preventing the licensing and operation of plants which were not designed or constructed properly. However, it has not worked well in identifying and correcting problems at the e'arliest possible stage. It has resulted in large impacts on ratepayers and utility owners and has reduced public confidence in the overall ability of the industry to fulfill the expectations placed upon it.

In response to these lessons identified in the QA report to Congress, the NRC has increased and redirected its design and construction inspection effort for plants currently under construction and is using and testing new approaches for assuring quality at plants currently under construction. In addition to these improvements to NRC's inspection programs, the NRC has made a number of other improvements to its overall programs which have the effect of improving quality in the nuclear industry. These improvements focus largely on the approximately ninety units presently licensed ~ for operation and include such initiatives as improved training and requalification programs for plant operational staff; increased staff and industry attention to preventive and routine maintenance;

, research programs on the relationship between plant reliability, safety, and quality; a major new focus on the effectiveness and quality of plant activities related to plant outages; and a thorough reanalysis, for all operating plants, of the NRC-required technical specifications issued as conditions of operating licenses.

i

i 4 f . 3 The Honorable Thomas P. O'Neill Since the introduction of the legislation that led to the QA study, the industry also has initiated a number of activities that should result in an overall improvement in the quality of nuclear plant construction and operations. These include the Construction Project Evaluation Program of the Institute of Nuclear Power Operations (INPO); the establishment of a new senior-level industry group, the Nuclear Utility Management and Human Resources Committee (NUMARC), to address some of the difficult people-oriented issues facing the industry, such as operator training; and the voluntary use of third party. management audits by some utilities. In addition, Georgia Power Company has volunteered their Vogtle project for a pilot test of the readiness review concept identified by the QA report to Congress as an area for further analysis, and which may lead to improvements in managing, licensing, and confirming the quality of nuclear construction projects. The pilot test is currently under way, and it is being followed closely by both the nuclear indastry and the NRC.

Although these new NRC and industry programs should, in a broad sense, lead to improved quality in the nuclear industry, the purpose of this letter is to describe the Commission *s decisions on the recommendations contained in the staff's QA report to Congress prepared in response to the congressional direction on QA contained in P. L.97-415. In response to this direction and as a follow-on to the QA report to Congress, the staff has developed a plan to revise NRC's administrative practices pertaining to quality and quality assurance. The plan contains three categories of actions: 1) those actions which would apply to a future generation of nuclear facilities which may be constructed; 2) those which can reasonably be expected to be implemented

. - __ ,- ,. .- __. . -~

g, w . . . _ _ _ _ , . _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ . _ _

s O The Honorable Thomas P. O'Neill 4 at nuclear power plant projects presently in the final stages of construction; and 3) those which can be applied to operating nuclear power plants. .Although the plan necessarily deals principally with NRC actions, we believe that it does embody the Commission's previously described QA philosophy which emphasizes the importance of the industry's role in improving nuclear quality. The NRC directly controls only its own activities, but NRC actions can provide great leverage to foster or inhibit licensee actions. The QA report made it clear

~

that simple regulatory compliance is not enough to assure excellence in the construction or operation of nuclear power plants. The planned NRC QA program revisions are intended to create an environment which maximizes the incentives for and ability of utilities to achieve excellence while assuring compliance with the Commission's regulations.

The revised QA program envisioned in the staff's plan, and with which the ,

Commission agrees in principle, includes the following key features:

Tighter Screening - A tighter screening of new construction permit applicants for managerial competence, project team experience, and financial capability and stamina.

Performance-Oriented QA Programs - A reorientation of present NRC QA programs, in which the overall results of the program would be given greater emphasis, and conformance to the details of its component pieces would be given less emphasis than at present. Licensee trend

The Honorable Thomas P. O'Neill analysis programs would be integral to this shift in NRC oversight focus, as would identification and correction of root causes y of qualit problems.

Readiness Reviews, Master Inspection Plans, and Incremental ew NRC Revi of Completed Work - The routine use of readiness reviews at predhterm stages throughout the design, construction, start up and operation life cycle.

Readiness reviews would ensure that a) work that should performed has been performed; b) the project is ready to proceed to the next step (e.g. , transition from construction to operation); c) the project is planned better; d) senior management is formally involved cal at criti 1

l stages of a project; and e) interfaces, coordination and agreement on wo l progress and status are improved.

l Readiness reviews would be conducted by l

l the licensee, reviewed and inspected by the NRC, and sequenced a with master inspection plan to provide continual confirmationa of yproject qu lit at predetermined inspection points.

Although readiness reviews can be I used in all phases of the life cycle of a nuclear power plant , they are particularly well suited to providing incremental review of construction work in progress.

.The pilot program under way at Georgia Power Company's I

Plant Vogt1E is testing the combination of the readiness review concept in conjunction with a master inspection plan and incremental w NRC revie of completed construction work, l

l

The Honorable Thomas P. O'Neill Reduction of Design Changes - Actions to reduce and control changes and to complement the predetermined master inspection plan and acceptance These would include a) a more complete design criteria described above. i before construction or a major modific& tion begins (a standardized des gn figura-would meet this need), and b) the use by the nuclear industry of con tion management techniques (developed by the aerospace industry) to bet manage project-related change and maintain current design documents.

Inspection Optimization - More efficient use of limited NRC inspection This would include the acceptance of certain inspections and resources. l audits performed by non-NRC individuals or groups such as the Nationa Board of Boiler and Pressure Vessel Authorized Nuclear Inspectors in certain situations instead of NRC inspections, thereby reducing or elim nating duplicative inspections.

The above program features are intended to improve quality and quality ass including over a broast range of nuclear activities falling under NRC oversight, d

nuclear piant optration, maintenance, modification, waste management, an Some of these areas can and will be implemented without sub-transportation.

However, some will require additional stantial further development work.

study or exte.nded rulemaking proceedings.

1

The Honorable Thomas P. O'Neill Impact of Budget' Cuts on Planned QA Improvements

~

As originally drafted, the staff's plan was based on previously planned growth in resources assigned to the NRC quality assurance program in FYs 86 and 87.

As a result of recent OMB actions affecting NRC's budget for FY 86, a number of the activities contained in the plan will be dropped, suspended, or indefinitely deferred. Since the Commission must maintain its programs for ensuring the safe operation of current plants, OMB budget action will make it necessary for the NRC staff to drop planned developmental work which would apply only to a future generation of plants.

The Commission recognizes that the cancellation of this future-oriented work is not consistent with the forward-looking nature of the Congressional guidance on quality assurance contained in P. L.97-415. However, resource reductions for NRC's QA program have led the Commission to conclude that its near-term QA responsibilities for operating plants require termination of the planned future-oriented work which does not also lead to improvements in the quality of plants currently under construction or in operation. Specifically impacted by the budget cuts will be future-oriented developmental work related to the screening of new applicants, the degree of design completion and configuration management techniques.

The Honorable Thomas P. O'Neill The Commission's conclusions and exceptions regarding the QA report to Congress whichwaspreviouslytransmittedasastaffreportarecontaInedinAttachment A. Administrative actions that are under way or planned by the Commission to address the findings of the QA~ report are summarized in Tables 1 and-2 of Attachment A.

As stated in its 1984 policy and planning guidance document to the staff, the Commission believes that NRC's fundamental task is to ensure that existing nuclear facilities and those coming on-line in the future operate safely.

Consequently, the highest priority will be given to assuring that operating facilities maintain adequate protection of public health and safety. Following operating plants in decreasing order of priority are, respectively, plants currently under construction, and future plants for which construction applications have not yet been received. The decision to drop planned develop-mental work that applies only to a future generation of plants was made consistent with this prioritization.

At this time, no changes to NRC authorizing legislation are recommended. Each of the Commission's proposed administrative actions can be implemented within the NRC's current statutory authority. However, after further analysis, some issues may result in legislative proposals at a future date.

Sincerely,

~

Nunzio J. Palladino Chairman

Enclosure:

Attachment A, Commission Conclusions on the Staff QA Report to Congress

ATTACHMENT A COMMISSION CONCLUSIONS ON THE STAFF QA REPORT TO CONGRESS The Commission is~in general agreement with the findings and conclusions of the QA report-(which was previously forwarded to Congress as a staff report) and has approved the staff's action plan (which takes into account public comments received on the QA report and current budget limitations) to implement the findings of the report. The Commission endorses the staff's earlier findings, conclusions and recommendations contained in the QA report, with the following modifications:

1) Construction permits (cps) for future nuclear power plant applicants need not be conditioned on post-CP demonstration by the applicant of its capability and effectiveness in implementing its quality assurance program.

The NRC has ample authority within its current regulations to take appropriate action if a licensee does not capably and effectively implement its QA program. Changes in the Commission's practices are being undertaken, however, to improve NRC and licensee ability to assess the effectiveness of QA programs at the earliest possible stage in the construction process.

Licensees should demonstrate the effectiveness of their programs continually throughout the construction process, but the CP need not be conditioned on such demonstration.

2) It is not clear that a new, special advisory board needs to be established to supplement staff advice on the qualifications of new applicants. This is one alternative that would have to be further analyzed before imple-mentation to see if an additional source of advice is needed or whether an

. already existing body (e.g., the Advisory Committee on Reactor Safeguards) could perform that function. Efforts by industry-sponsored groups such as the Nuclear Utility Management and Human Resources Committee (NUMARC) and the Institute of Nuclear Power Operations (INPO), having the effect of a "self-screening" process, will also affect any decision in this area.

Planned further analysis of the advisory board concept will not be undertaken at this time.

3) Other feasible alternatives to mandatory third party audits are available to accomplish the QA program goals of improved management and increased prevention and detection capability. Readiness reviews, now being tested and evaluated at the Vogtle nuclear plant, provide one possible alternative.

Increasea NRC use of team inspections and resident inspectors and voluntary use of third parties by some utilities provide other alternatives. As a result of these considerations, the staff has preposed, and the Commission has agreed, to defer rulemaking activities to require periodic third party audits for plants under construction.

~

TABLE 1. Commission Action on Staff Recommendations Contained in the QA Report to Congress (see NUREG-1055, Table 2.1)-

QA Report Recommendation Commission Action

1. Enhanced Pre-Construction Permit Planned developmental action (CP) Review of Experience and cancelled due to budget priority Managerial Quality of operating plants Establishment of an Advisory Dropped pend.ing exploration of potential Board industry action on this topic

2. Post-CP Demonstration of Dropped in favor of Readiness Management Effectiveness Review Program

3. QA Program Performance Develop for potential implementation Objectives. at operating plants and plants under construction. Cooperative NRC/

industry performance objective and trend analysis pilot programs discussions currently in progress will be expl.ored

4. Management Appraisals as an Drop in favor of readiness reviews and Adjunct to Construction other management initiatives. Limited Appraisal Team (CAT) management process reviews implemented Inspections as part of CAT inspections

, 5. Inspection Prioritization All plants currently under construction l of Plants Currently Under are receiving a " graded" inspection I

Construction (Problem Plant approach. Planned long-term analysis Identification) (for application if construction of a new generation of nuclear power plants should begin) indefinitely suspended due to budget priority of operating plants

6. Improved Diagnostic Capability / Implement as part of QA Performance Trend Analysis Objectives Program

7. Senior Management Meetings Implement

8. Enhanced Vendor Program Being implemented

9. Third-Party Audit / Interim Recommended third party rulemaking CAT Inspections / Interim deferred pending outcome of readiness Independent Design Verification review program; no activity planned Program (IDVP) due to budget priority of operating plants; perform CAT inspections at 4/ year (no expansion in program);

continue IDVP on case-by-case basis

10. Regional Team Inspections Increasing use of team inspections.

Pilot under way in one NRC region.

4 2-QA Report Recommendation Commission Action

11. Expanded Resident Program More residents being assigned to sites. Pilot test under way in one

~

NRC region to reduce regional inspections in favor of resident inspections

12. Improved Licensee Detection Develop self-detection capability Capability for potential implementation as part of the QA Performance Objectives Program recommendation

13. Independent Audits of NRC QA Implement .

Programs

_-.,w * -.w.- e -- - -..~ _ - . , , - - -,y3 y,,. -- --.----

TABLE 2. Commission Action on Staff Recommendations for Further Analysis Contained in the QA Report to Congress (see NUREG-1055, Table 2.2)

Area Recommended For Further Analysis Commission Action

1. QA Report-Type Study for Plants Limited study efforts under way in Operation -

or planned

2. .Prioritization of QA Measures: Analysis is in progress Guidance on " Safety-Related" vs.

"Important to Safety"

3. Measuring Effectiveness of QA Research projects focusing on plants under constructiop and in operation under way

4. Essentially Complete Design Combined with #5 below at CP Stage

5. Configuration Management Implementation planned by subsuming into rulemaking on control of post-CP design changes (transferred to the Office of Inspection and Enforcement from the Office of Nuclear Regulatory Research in September 1984). Planned developmental work related to future plants dropped due to budget priority of operating plants

6. Readiness Reviews Pilot program being initiated for plants under construction

7. Quality Engineering Further analysis and developmental work not planned due to budget priority of operating plants

8. Alternate Ownership and Efforts-redirected to explore Management Arrangements improvements within current arrangements in an all operating reactor environment

9. Feasibility of Designated Further consideration deferred pending Representatives a licensee request for a pilot program.

Budget priorities preclude any activity in this area .

10. Limiting Construction Permits Analysis of the problems faced by fikC in the event of a resumption of nuclear plant construction is included as a possible contingency in the study described in 8., above.

O 9

• 9 e

0 e

9 0

ENCLOSURE 2 l

4 5

+

5 4.~.

U.S. NUCLEAR REGULATORY COMMISSION QUALITY ASSURANCE PROGRAM IMPLEMENTATION PLAN:

PROGRAMS TO IMPROVE QUALITY AND THE ASSURANCE -

0F QUALITY IN THE DESIGN, CONSTRUCTION AND OPERATION OF LICENSED NUCLEAR ACTIVITIES' 0FFICE OF INSPECTION AND ENFORCEMENT JANUARY 1985 i

e 1

n 2

l

l QUALITY ASSURANCE PROGRAM IMPLEMENTATION PLAN

, TABLE OF CONTENTS GLOSSARY OF ACRONYMS / ABBREVIATIONS..........................................v INTRODUCTION................................................................vii i

RESOURCE ESTIMATES.....................................................viii PRIORITIZATION OF QA PROGRAM IMPLEMENTATION PLAN ACTIVITIES............ix ,

NRC QA P0LICY..........................................................xi l

! QA PILOT PR0 GRAMS......................................................xii

! ORGANIZATION OF THE QA PROGRAM IMPLEMENTATION PLAN.....................xii i

1. 0 OVERVIEW OF THE REVISED QA PR0 GRAM.................................... 1-1 1.1 STATEMENT OF THE QA PR0B LEM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1 ,

l 1.2 THE REVISED QA PR0 GRAM........................................... 1-2 l 1.3 APPLICATION OF THE REVISED QA PR0 GRAM............................ 1-3 1.3.1 Pre-Construction Permit Stage............................. 1-4 1.3.2 Design Phase.............................................. 1-4 1.3.3 Construction Phase........................................ 1-5 1.3.4 Transition from Construction to Operation................. 1-8 l 1.3.5 0 p e ra t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1- 10 2.0 UPDATE OF NUREG-1055 RECOMMENDATIONS................'.................. 2-1 3.0 QUALITY ASSURANCE FOR OPERATING REACTORS AND PLANTS CURRENTLY UNDER CONSTRUCTION.......................................................... 3-1 3.1 QA PROGRAM IMPLEMENTATION........................................ 3-2 3.1.1 Performance Objectives and Trend Analysis. . . . . . . . . . . . . . . . . 3-4 3.1.2 Measurement Indices / Effectiveness of QA Programs.......... 3-6 3.2 MANAGEMENT ISSUES................................................ 3-7 3.2.1 Recognition and Support of Industry Efforts............... 3-9 3.2.2 Readiness Reviews, Master Inspection Plans, and Incremental NRC Review of Completed Work.................. 3-10 3.2.3 Configuration Management and Design Completion............ 3-12 3.2.4 Management Meetings....................................... 3-14 3.2.5 Impact of Actions of SURAs on Safety and Quality.......... 3-14 3.2.6 Integration of Human Factors Work in Management and O rga n i z a t i o n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3- 15 3.2.7 Third-Party Audits........................................ 3-16 3.3 NRC ' S INS P ECTION P R0GR AM. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-17, 3.3.1 Enhanced Management Review in SALP........................ 3-18 .

3.3.2 Kist Report Follow-Up/0ther Improvements to NRC QA P ro g ram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 - 19 3.3.3 Problem Plant Identi fication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-20 3.3.4 Increased Emphasis on Team Inspections / Contractor Support / Resident Inspectors............................... 3-20 3.3.5 Improved QA Training for Inspectors and NRC Staff......... 3-21 3.3.6 Reduction of Inspection Burden on Licensees and Vendors... 3-22

?.3.7 Integrated Design Inspections and Independent Design Verification Program...................................... 3-23 iii

3.4 NRC PARTICIPATION IN THE STANDARDS DEVELOPMENT PROCESS........... 3-24 3.4.1 Regulatory Guidance on QA Programs and Activities:

Endorsement of ANS-3.2, NQA-1 and NQA-2................... 3-25 3.4.2 Computer Software QA for Nuclear Applications............. 3-27 3.4.3 Improved Q-List........................................... 3-28 3.4.4 Alternative QA Requirements for Subtier Suppliers......... 3-29 3.4.5 NRC Participation in Consensus Standards Committee

. Activities................................................ 3-31 3.5 PERSONNEL ISSUES IN ACHIEVING AND ASSURING QUALITY............... 3-31 3.5.1 Qualification / Certification of QA/QC Personnel............ 3-32 3.5.2 Line Organization Responsibilities for Achieving and Assuring Quality...................................... 3-34 4.0 REVISED QUALITY ASSURANCE PROGRAM FOR FUTURE PLANTS 4.1 CRITERIA FOR PROJECT TEAM QUALIFICATIONS, EXPERIENCE, CA PA8 I LI T I E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4- 1 4.2 ADVISORY BOARD TO REVIEW CP APPLICANTS........................... 4-2 4.3 QUALITY IMPLICATIONS OF THE CHANGING UTILITY-CONTRACTOR-VENDOR-SUPPLIER INFRASTRUCTURE.......................................... 4-4 l

5.0 QUALITY ASSURANCE FOR NON-REACTOR ACTIVITIES.......................... 5-1 5.1 MONITORED RETRIEVA8LE STORAGE FACILITIES AND INDEPENDENT SPENT FUEL STORAGE INSTALLATIONS....................................... 5-1 5.2 HIGH-LEVEL WASTE REPOSITORIES.................................... 5-3 5.3 FU E L FAC I LIT I ES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4 5.4 TRANSPORTATION................................................... 5-5 APPENDIX A INTEROFFICE COMMENTS ON THE QA PROGRAM IMPLEMENTATION PLAN......A-1 APPENDIX B SECY MEMORANDUM DATED OCTOBER 19, 1984..........................B-1 iv

GLOSSARY OF ACRONYMS / ABBREVIATIONS ACRS Advisory Committee on Reactor Safeguards A/E architect-engineer ANI- Authorized Nuclear Inspector ANS American Nuclear Society ANSI American National Standards Institute

, ASME American Society of Mechanical Engineers ATWS anticipated transient without -scram ,

CAT Construction Appraisal Team CCB change control board CFR Code of Federal Regulations CP construction permit DHFS Division of Human Factors Safety, Office of Nuclear Reactor Regulation

-DOE U.S. Department of Energy DR designated representative-EPA U.S. Environmental Protection Agency FSAR Final Safety Analysis Report FY fiscal year GPC Georgia Power Company HLW high-level waste IDI Integrated Design Inspection IDVP Independent Design Verification Program IE Office of Inspection and Enforcement IEEE Institute of Electrical and Electronics Engineers INPO Institute of Nuclear Power Operations ISFSI independent spent fuel storage installation MRS monitored retrievable storage NASA National Aeronautics and Space Administration NB National Board of Boiler and Pressure Vessel Inspectors NMSS Office of Nuclear Material Safety and Safeguards NQA nuclear quality assurance NRC U.S. Nuclear Regulatory Commission NRR Office of Nuclear Reactor Regulation NSSS nuclear steam supply system NT0L near-term operating license

! NUMARC Nuclear Utility Management and Human Resources Committee NWPA Nuclear Waste Policy Act OL operating license OMB Office of Management and Budget PAT Performance Appraisal Team PORC Plant Operations Review Committee PRA probabilistic risk assessment QA quality assurance i QC quality control Office of Nuclear Regulatory Research RES SAA system assurance-analysis SALP Systematic Assessment of Licensee Performance SECY Office of the Secretary SER Safety Evaluation Report SQA software quality assurance SRP Standard Review Plan (NUREG-0800) l SURA state utility regulatory agency l SY staff year TMI Three Mile Island TUV Technischen IIberwachungs-Vereine v

INTRODUCTION The purpose of this document is to establish goals and objectives underlying the Nuclear Regulatory Commission's (NRC's) quality assurance (QA) program in a task-oriented framework of program elements, projects and activities designed to achieve those goals and objectives. The document defines the NRC QA program for FYs 85, 86, and 87 for nuclear power plant design, construction, and operation and for nuclear material safety and safeguards. This 1-atter category includes NRC quality-related activities -for transportation, fuel facilities, monitored retrievable waste storage installations, and high-level waste repositories.

This QA progrgg) implementation plan updates and refines the recommendations of the QA report to Congress. This update and refinement are based on public comments on the QA report to Congress, further analyses of some issues and concepts, and guidance from the Commission and Advisory Committee on Reactor Safeguards (ACRS). Some recommendations of the QA report have been dropped, others have been modified, and some new recommendations have been incorporated.

(Chapter 2.0 of this plan discusses the status of the QA report recommendations.) i T5is plan has a broader scope than the QA study (which was limited to design and construction of nuclear power plants) and addresses many other aspects of NRC's QA program.

This document outlines a regulatory development program that addresses some issues not normally associated with " quality assurance" or " quality control."

These issues include management and the process by which a nuclear plant is determined to have been constructed correctly. This broader than traditional view of QA is necessary to meaningfully address the underlying causes of .

auality problems in the nuclear industry. As the QA study demonstrated, the root causes extend far beyond traditional QA programs and organizations to some basic institutional premises of the U.S. nuclear industry.

This document is intended to be a working-staff-level document, subject to modification and change as planning assumptions, priorities and operational needs change. As a working-le, vel document, the plan will necessarily be too detailed for some audiences and not detailed enough for others. Although describing the overall NRC QA program for the next three years and identifying specific work products (e.g., reports, regulatory positions, decisions, regulatory guides, rule changes, etc.), it does not detail the sequence of steps involved in reaching each of these work products. There are several reasons for this.

The intermediate level of detail at which this document is written was deliber-ately selected to appeal to the widest possible audience: specific enough for senior NRC management to see the program's direction and anticipated products, yet general enough to give the responsible management and staff flexibility in its implementation. More detailed plans will be necessary for each project described in this plan, but they are not included here for the following reasons: 1) they would make an already long document unnecessarily cumbersome (a) NUREG-1055, Improving Quality and the Assurance of Quality in the Design and Construction of Nuclear Power Plants, May 1984, hereafter referred to as the QA report or the QA study.

vii

and reduce its usefulness to a general audience; 2) some project plans cannot be prepared until the results of other more preliminary projects are known; and

3) including them would unnecessarily limit the flexibility of the staff to respond to new information, new understandings of problems, and changing priorities. Since this plan is a living document, comments or suggestions from interested persons or organizations will.always be considered and are welcomed.

Project managers have been or will be appointed for each of the projects or subelements of the implementation plan. The project managers are responsible for developing project plans, which include milestones, schedules, intermediate products, decision points, contractor needs (if applicable), etc. The project plans are more detailed than the implementation plan, and they provide the working-level staff with implementation details for individual projects.

Depending on the specific project, working groups involving cognizant staff from other offices and the regions may be formed to assist (or take the lead) in developing and/or implementing a particular project. For example, Region II has the agency lead for the Vogtle readiness review pilot program, with licensing support from the Office of Nuclear Reactor Regulation (NRR) and program development support from the Office of Inspection and Enforcement (IE).

Performance trend analysis is another program area in whicn substantive input from regional office personnel is anticipated, e.g., in pilot programs and in the development of performance indicators. These working groups will be formea as needed, with the consen't and advice of regional and other divisional management, and disbanded when their work is complete.

RESOURCE ESTIMATES This implementation plan describes an ambitious QA program for the next three years. It identifies a number of activities and end products, including possible legislative changes, rule changes, regulatory guides or other guidance, regulatory positions or decisions, policy statements, reports, and~new or revised inspection procedures. The programs described in this document were planned based on the IE QA resources for FYs 86 and 87, requested of and approved by the Comuission during the FY 86 budget process.

Reductions from those approved budget plans will cause the cancellation or indefinite suspension of some planned activities. As a result of actions by the Office of Management and Budget (OMB) in December 1984 affecting NRC's budget for FY 86, a number of the activities described in the plan are being dropped or suspended. The NRC must place its highest priorities on the operational safety of existing plants and plants coming on-line (see 1984 Policy and Planning Guidance, NUREG-0885, Issue 3, p. 3). As a result, QA resources allocated to this plan will be prioritized as follows: first, measures to improve the quality of operating plants; second, measures to improve the quality of plants under construction; and third, measures that apply only to a future generation of plants. The following section, "Prioritization of QA Program Implementation Plan Activities," provides a rough prioritization of activities within these three categories. It should be noted that because of the budget priority of operating plants and plants presently under construction, no work that applies o_n_ly to future plants is being undertaken. In addition, as a result of the budget action, the level of effort for some projects that apply to current plants (e.g., reduction of the inspection burden on licensees, some Kist Report follow-up, QA for fuel facilities and transportation) is being reduced.

viii

i This implementation plan also has a slight budget impact on presently budgeted  ;

totals for other IE programs, the NRC regions, and other NRC program offices. l The total overall impact on these other programs is approximately three staff 4

years per year. 'IE's level of effort needed for the Vogtle pilot program (see section on QA Pilot Programs below) is included in its budget plans, but other agency resources that have been or may be reprogrammed for the Vogtle pilot were not included in the IE request. .

[ PRIOP.ITIZATION OF QA PROGRAM IMPLEMENTATION PLAN ACTIVITIES

.The reader should note that this plan was prepared prior to OMB budget cuts

in the NRC QA program for FY 86. So that the revised QA program envisioned by 1

the plan could remain intact in concept, few changes to the body of the plan

. have been made to reflect the budget cuts. This section and the proceeding one are the only places in the plan where the impact of the OMB budget cuts is comprehensively discussed. The reader should refer to these sections (and j particularly,the following category listings) to determine the priority and

viability of a given implementation plan activity. Tables 2.1 and 2.2 of

! Chapter 2.0 are also helpful in this regard. Due to the budget cuts, the j delivery dates of some of the planned end products, including higher priority items, may slip (e.g., from FY 86 to FY 87).

I 4 This plan contains thirty program elements which can be divided into three

! categories. The first category contains those activities which can be applied to operating nuclear power plants. The second includes the' activities which can reasonably be expected to be implemented at the handful of nuclear power plant projects in the final stages of construction. The third category is those actions applying to a future generation of nuclear facilities which may be constructed. The activities within each of these three categories are ranked in order of budget priority, with some elements falling into more than i one category. Because of the applicability of some elements to more than one category and the specificity of the elements on the list, the categorization of the elements is not strictly consistent with the conceptual current vs. future split delineated in Chapters 3.0 and 4.0. In addition, budget cuts will postpone certain activities (which the staff originally thought might apply to current plants, e.g., third party audits) for so long that they realistically can have an impact only on future plants. The prioritization of the elements in each category is as follows:

1. Elements Which Apply to Operating Plants i

Performance Objectives and Trend Analysis NRC Participation in Consensus Standards Committee Activities 1

Enhanced Management Review in Systematic Assessment of Licensee Performance (SALP)

Management Meetings

)

l Recognition and Support of Industry Efforts i

Measurement Indices / Effectiveness of QA Programs

)

• Impact of Actions of State Utility Regulatory Agencies (SURAs) on Safety '

i and Quality

)

l ix

~_

• Problem Plant Identification Quality Implications of the Changing Utility-Contractor-Vendor-Supplier Infrastructure Improved Q-List Kist Rep rt Follow-Up Improved QA Training for Inspectors and NRC Staff Reduction of Inspection Burden on Licensees and Vendors Alternative QA Requirements for Subtier Suppliers .

Qualification / Certification of QA/ Quality Control (QC) Personnel Integration of Human Factors Work in Management and Organization Line Organization Responsibilities for Achieving and Assuring Quality

2. Elements Which Can Apply to Plants Currently Under Construction Readiness Reviews, Master Inspection Plans, and Incremental NRC Accep-tance of Completed Work (applies only to those few plants with scheduled completion dates in'1987 or beyond and which volunteer for a pilot project)

Integrated Design Inspections and Independent Design Verification Program Increased Emphasis on Team Inspections / Contractor Support / Resident Inspectors Enhanced Management Review in SALP Management Meetings Recognition and Support of Industry Efforts Problem Plant Identification Kist Report Follow-Up

3. Elements Which Will Apply to Future Plants Readiness Reviews, Master Inspection Plans, and Incremental NRC Accep-tance of Completed Work

• Configuration Management and Design Completion (a)

(*) Denotes element which applies only to future plants. Planned work that applies only to future plants will not be undertaken due to the budget priurity of operating plants.

x I

Performance Objectives and Trend Analysis Quality Assurance for Non-Reactor Activities (Monitored Retrieveable Storage Facilities and High-Level Waste Repositories)

Enhanced Management Review in SALP l Quality Implications of the Changing Utility-Contractor-Vendor-Supplier Infrastructure -

Criteria for Project Team Qualifications, Experience, Capabilities (a)

Measurement Indices / Effectiveness of QA Programs Impact of Actions of SURAs on Safety and Quality Quality Assurance for Non-Reactor Activities (Fuel Facilities and l Transportation) l Qualification / Certification of QA/QC Personnel l Line Organization Responsibilities for Achieving and Assuring Quality Advisory Board to Review Construction Permit (CP) Applicants (a)

• Third-Party Audits (a)

NRC QA POLICY As mentioned earlier, this document comprises an implementation plan for NRC's QA programs and policies for FYs 85, 86 and 87. It also reflects the Commission's statement of NRC's QA policy in the 1984 Policy and Planning Guidance (NUREG-0885, Issue 3). NRC's QA program and this QA program implementa-tion plan are based on the following basic tenets:

! 1. Licensees, not the NRC, are primarily responsible for achieving and assuring quality. In particular, licensee management must assume j responsibility for achieving and assuring quality, and management must l be held accountable for quality related failures, as well as for quality

! successes.

2. Substantive improvements in quality in the nuclear industry must come from the industry itself; they cannot be " inspected in" or " regulated in" by the NRC.

1 (a) Denotes element which applies only to future plants. Planned work that applies only to future plants will not be undertaken due to the budget priority of operating plants.

xi

3. The focus of the NRC's and tha industry's. programs for QA should be more on performance than has been the case in the past;'they should adopt the policy that the achievement of quality and the demonstration of that achievement is more important than the method of achievement.

4. The NRC's and the industry's QA programs should be oriented toward prevention and early detection of major quality problems.

5. Activities and programs to provide assurance of quality and safety should not become so burdensome that they interfere with the ability of the performing organization to achieve quality and safety.

6. A 0.A program is a tool to be used by line management for monitoring work activities and for confirming that they have been performed correctly. It is not a substitute for management. The level of performance of the QA program is highly correlated with the level of performance of other line management activities.

7. Greater predictability should be brought to the licensing and regulatory process. Lack of predictability in licensing hampers planning, prolongs construction, increases costs, lowers morale, and leads to design and construction shortcuts, all of which adversely affect plant quality.

QA PILOT PROGRAMS This plan describes a number of task-oriented activities currently under way or planned by the NRC to improve the assurance of quality in the design, construction and operation of nuclear power plants. These actions include pilot programs to further develop and test new QA program concepts; analyses leading to decisions on major. policy issues; support of ongoing or new industry self policing or.self-improvement activities; research and standards develop-ment; cost / safety-benefit analyses of proposed actions; rulemaking activities; licensing; and inspection and enforcement. Consistent with ACRS' guidance [ACRS letter to the Commission dated March 21, 1984 (NUREG-1055, Section 10.3)] the staff has not let preparation.of the QA program implementation plan delay action on the most important of the recommendations in the QA report to Congress. For example, one of the activities described in the plan, readiness reviews, has progressed to the point that a pilot program is being initiated to test its feasibility and benefits. This pilot program, an operational readiness review program at Georgia Power Company's Plant Vogtle, will test many of the key features of the QA program implementation plan. These features include developing master inspection plans and incremental NRC acceptance of completed work. Preliminary discussions on establishing pilot program (s) to demonstrate, test, and refine the concept of performance-oriented QA programs and trend analysis have been held with other licensees.

ORGANIZATION OF THE QA PROGRAM IMPLEMENTATION PLAN The QA program implementation plan is divided into five chapters:

1.0 Overview of the Revised QA Program 2.0 Update of NUREG-1055 Recommendations 3.0 Quality Assurance for Operating Reactors and Plants Currently Under Construction 4.0 Revised Quality Assurance Program for Future Plants 5.0 Quality Assurance for Non-Reactor Activities

.-_ ___. I

Chapters 3.0 through 5.0 are further broken down into elements and subelements with a discussion of the planned activities, products, benefits and associated issues of each.

Chapter 1.0, " Overview of the Revised QA Program," describes how the major features of the implementation plan fit together and how they affect the life

}

cycle of a nuclear power plant. The overview is the only place where this unified description appears, and it should be used as a reference when reading the ensuing, more detailed discussion of- the plan's elements and subelements.

~However, not all of the plan's elements /subelements appear in the overview. In most cases, this occurs because the element does not represent as fundamental a change to present practice as do the features highlighted in the overview, or because the element /subelement was adequately discussed in the QA report.

h Chapter 2.0 briefly summarizes and updates the staff actions iecommended in the QA report to Congress.

Chapter 3.0 contains the staff actions for improving the quality of the design, construction, and operation of existing nuclear power plants. That chapter reflects the lessons learned from the QA study and identifies staff actions, including ongoing or planned industry actions. Specifically, the following elements are addressed: a) QA program implementation (including performance objectives and trend analysis), b) management issues (including readiness reviews), c) NRC inspection program (including inspection optimization), d)

NRC participation in the standards development process, and e) personnel issues in achieving and assuring quality. These elements of the plan require the majority of the staff QA resources budgeted over the next three years.

Chapter 4.0 identifies the staff activities planned for future nuclear power plants and quality issues. In these activities, the lessons learned from the QA study would be translateo into new or revised programs.for NRC pre-construction permit (CP) review of the experience of the applicant's project management team and its capability for effectively controlling the design and construction of a nuclear power plant. Also, a long-range planning capability is envisioned for l anticipating and dealing with future quality issues.

Chapter 5.0 examines quality and QA issues in the balance of the fuel cycle.

The Office of Nuclear Material Safety and Safeguards (NMSS) has licensing responsibility for fuel facilities, transportation and waste repositories

[ including permanent high-level waste (HLW) repositories, monitored retrievable rtorage (MRS) installations, and independent spent fuel storage installations (ISFSIs)]. The Office of Inspection and Enforcement, which has lead agency responsibility for QA, will provide QA support and expertise to NMSS in its areas of responsibility.

Appendix A addresses the comments received from the other program offices and the regions on the draft implementation plan and how they have been incorporated into the plan.

Appendix B provides a copy of a SECY memorandum referenced in the plan.

xiii

In the interest of conciseness, it is assumed that the reader is generally familiar with the QA report (NUREG-1055). That study's findings and recommendations are often cited and described only to the extent necessary to support program plan elements and subelements. To gain the maximum under-standing of the motivations behind many of the elements of this QA program implementation plan, the_ reader should refer to the QA report.

xiv  !

1. 0 OVERVIEW 0F THE REVISED QA PROGRAM (a)

The NRC report to Congress on quality assurance (NUREG-1055) identified a number of shortcomings in the NRC's and licensees' programs for assuring quality.in the design and construction of nuclear power plants. Although that report recommended several improvements to existing NRC and licensee programs and identified several areas for potential improvements requiring further analysis, it did not develop a task oriented action plan which drew the recommendations into a cohesive program.

The purpose of this document is to describe such a quality assurance (QA) program (hereafter referred to as the " revised QA program") and to provide a task-oriented framework or plan for developing and implementing the revised QA program. The revised QA program described in this document is based on the QA study, public comments on the QA study, guidance from the Commission and the Advisory Committee on Reactor Safeguards (ACRS), and further staff analyses.

The revised QA program is broader than the QA study's scope, covering the operational phase of nuclear power plants as well as design and construction, and covering QA for other non-reactor areas within NRC's responsibility (monitored retrievable storage facilities, independent spent fuel storage installations, high-level waste repositories, fuel facilities, and transportation).

1.1 STATEMENT OF THE QA PROBLEM The QA study found that the fundamental problems and ultimate solutions in achieving and assuring quality rest with management within the muclear industry.

Although informed, involved, experienced management at all levels is important in solving the nuclear QA problem, it is particularly important at the most senior levels (e.g., vice president, executive vice president, chief executive officer). Improved management in the nuclear industry is essential for main-taining an acceptable level of nuclear quality and safety. An important factor contributing to improved management is the need for increased predictability and stability in the regulatory process. This factor requires improvements by both industry management and the NRC. Even the most capable of management teams can be frustrated and defeated by constantly changing requirements and acceptance criteria.

In the QA study, other factors cited as contributing to repeated cases of major quality-related failures included the following:

insufficient prior nuclear experience of the project team (a) In order to show the interrelationships of the major concepts of the QA program implementation plan and how they affect the life cycle of a nuclear power plant, this chapter describes the revised QA program as envisioned by the staff based.on previously-approved resource levels.

Recent Office of Management and Budget (OMB) budget action has affected the staff's plans, particularly those plans for work that would apply only to future plants. See the introduction of this document for a prioritization of implementation plan activities and a discussion of the impact of reductions in budgeted resources.

1-1

.- _, ---. - . . , . m - -e_ -

_ . _ . _ . , _ - _ . _ . - - - . , _ _ , _ . - - - - , , . . _ . _ . _ - . _ _ . _ _ ~ _ ,

failure of the project team to translate individual nonconformances or problem areas into symptoms of project-wide problems in project management inaccurate, incomplete, or irrelevant project information being provided to management failure of the project team to take adequate corrective action for identified problems failure of the project team to adequately plan for project activities premature transition from one project phase to the next failure of the NRC to assimilate scattered, diverse inspection findings into a comprehensive and accurate picture of a project's status excessive internally and externally generated rates of change which over-whelm otherwise adequate management systems ineffective implementation of licensee QA programs limited personnel resources that constrained NRC inspections of.QA program implementation a view that QA is just another NRC requirement rather than a tool to help exercise effective management oversight of a project.

1.2 THE REVISED QA PROGRAM The revised QA program described in this document consists of several intercon-nected features designed to address the root causes of the identified QA problems. The key features of the revised QA program are described below. It is emphasized that the following features represent areas of continued staff investigation and development, not final actions.

Tighter Screening - A tighter screening of new applicants to determine managerial competence, project team experience, and financial stamina

.is planned.

Performance-Oriented QA Programs - A reorientation of present NRC QA programs is planned. In this reorientation, the overall results of the program will be given greater emphasis, ai;d conformance to the details of l

its component pieces will be given less emphasis than at present.

I Licensee trend analysis programs will be integral in this shift in NRC l emphasis, as will identification and correction of root causes of quality problems. Performance-oriented QA programs emphasizing trend analysis apply to both the construction and operation phases, and in the revised QA program, they represent the most significant change in QA for operation.

i

• Readiness Reviews - Readiness reviews can potentially be used at l

l predetermined stages throughout the design, construction, start-up and l operation life cycle to ensure the following: a) work that should have been performed has been performed; b) the project is ready to proceed to the next step (e.g., transition from construction to operation); c) the 1-2

. - . - - , . , - - _m. - . . --. _ - _ . , , -- -

, . _ ~--.--.-- .,. -%_ y... ___-___m _, m,

project is planned better than in the past; d) senior management is formally involved at critical stages of a project; and e) interface, coordination and agreement on work progress and status are improved from the past. Readiness reviews would be conducted by the 1.icensee, reviewed and inspected by the NRC, and sequenced to mesh with the inspection plan.

• Incremental NRC Review of Completed Work - Planned, incremental, in process NRC reviews could be used whila a project is under construc-tion, rather than at the end of a construction project. A large licensing decision at the end of a construction project can be reduced to a series of smaller, incremental decisions during the construction process.

Readiness reviews and inspections performed by the licensee according to the' inspection plan will form the basis for the in process NRC review of completed work.

Reduction of Design Change - To reduce design changes and to complement the predetermined inspection plan and acceptance criteria above, the following actions are planned: a) a requirement for a more complete design before construction or modification begins (a standardized design would fulfill this requirement); and b) the use by the nuclear industry of configuration management techniques (developed by the aerospace industry) to better control and manage project-related change.

• Inspection Optimization - Limited NRC inspection resources need to be used more erficiently. This feature involves.the potential' acceptance of certain inspections and audits by non-NRC individuals or groups such as the National Board of Boiler and Pressure Vessel Authorized Nuclear Inspectors in certain situations in place of NRC inspections. Using non-NRC groups could reduce or eliminate duplicative inspections anc could enable all parties involved [ utilities, architect-engineers (A/Es),

vendors, and outside inspectors, as well as the NRC] to better use their resources.

• Third-Party Audits - The NRC will retain its option of supplementing its inspections by requiring audits by qualified third parties whose perfor-mance will, in turn, be verified by the NRC. The NRC is trying other approaches (as described above) which, if successful, may preclude the need for a program of periodic third party audits as recommended in the QA report.

1. 3 APPLICATION OF THE REVISED QA PROGRAM Impleinenting the key features by the NRC and licensees will result in some significant departures from current practice in how plants are designed, constructed, licensed and operated. The following discussions describe how the revised QA program would affect nuclear power plants, from the pre-construction permit (CP) stage through the operational phase.

1-3 l l

i 1.3.1 Pre-Construction Permit Stage Before a CP is issued, an applicant would be subjected to a screening that considers in more depth than at present.the issues of its proposed construction team's previous-nuclear experience, its managerial experience and capability, and its financial resources and stamina. All of these factors have been found to significantly affect plant quality and the prospects that a project team, i

once issued a CP, would be able.to successfully complete a nuclear power plant that meets NRC requirements. -

Implementing a CP review process expanded to address managerial and financial qualifications would complement the adoption of a one-step licensinn orocess.

The enhanced managerial review before a CP is issued is applicable whether or not one-step licensing is approved. The tighter screening of new CP applicants would apply whether the applicant is a utility or some new form of applicant (e.g., an nuclear steam supply system (NSSS) manufacturer, an A/E, or a

! consortium organized to build and/or operate a nuclear power plant).

The Commission recently took action to re-institute a financial qualifications

! review for CP applicants. The financial review would~be incorporated as part of a broader review of managerial, experience, planning and resource issues conducted by the staff. Public comments on NUREG-1055 were mixed on l establishing a special board of financial, utility, ar.d management experts who

would provide advice independent of the staff on a CP applicant's prospects for successful project completion. Some commenters against this special board stated that introducing another board, panel, or committee into the licensing process would bring additional uncertainty to the licensing process and result in more delays. The QA program implementation plan describes further work leading to a decision on this concept. Alternative proposals for obtaining advice [e.g., via the Institute of Nuclear Power Operations (INPO) or the i

Nuclear Utility Management and Human Resources Committee (NUMARC)] to supplement the staff's analysis of an applicant's qualifications will be reviewed as part of this decision process.

1.3.2 Design Phase

In the revised QA program, new CP applicants would be required to more completely define their plant design at the CP stage (i.e., submit a more complete design). After the CP is issued, fewer design changes would be

permitted by either the applicant or the NRC. Similar to the discussion on the tighter screening of new applicants, this limitation on design changes would apply to both one-step and two-step licensing programs.

As part of the CP licensing process, the applicant would commit to and establish a comprehensive design process management program including an inspection pian, readiness reviews, and configuration management. The licensee would establish a change control board (CCB) which would act as a clearinghouse for the manage-l ment of changes, from whatever source. These management tools will induce better planning and, by reducing the level of design change, will make the projects easier to manage and to complete.

1-4

1. 3. 3 Construction Phase As part of the CP approval in the revised QA program, new applicants would commit to establishing a project management program which incorporates a performance-oriented QA program with built-in diagnostic and trend analysis capabilities.

Check points or milestones would be defined and agreed to at the outset and documented in a master utility inspection plan. Construction progress could be assessedprospectivelyandretrospegvelythroughreadinessreviews. The NRC, in return, would shift the emphasis of its construction inspection program from checking compliance with codes and regulations to include the following:

1) reviewing the performance of the applicant's project management program in achieving its self-established quality goals and objectives, including the quality of completed work and hardware, 2) reviewing'the applicant's effec-tiveness in determining and correcting root causes of quality problems identi-fied through its diagnostic and trend analysis programs, 3) reviewing the results of the readiness reviews at each of the predetermined construction inspection points, 4) providing NRC review of work plans and of completed work, and 5) accepting certain non-NRC inspections in place of certain NRC inspections, thus reducing the applicant's overall inspection burden while permitting more efficient use of limited NRC inspection resources.

The master inspection plan would specify the acceptance criteria and certain checks that must be made by the licensee or its agent at certain points in the construction process. Based on prior knowledge of the inspection points in the master inspection plan, the NRC would allocate its resources to cover planned licensee inspections in a manner consistent with the safety significance of the inspections being performed and other priorities. The master inspection plan would define inspections of completed work that would be performed by the l_icensee or its agent and that may or may not be overseen by the NRC. Although activities covered by the master inspection plan would be considered in planning NRC inspections, NRC inspections would focus on, but would not be limited to, activities and inspection points delineated in the master inspection plan.

The master inspection plan would be a public document but not subject to a public comment process. The intent is to adopt a procedure which would require any contentions relating to the inspection plan to be raised at the time of the plan's adoption. Issues which could have been raised at that time would not be admitted at a later point in the licensing process.

(a) A certain amount of NRC checking of compliance with codes, standards, commitments, and regulations will be necessary under the revised QA program to provide assurance of quality independent of the licensee's

, programs. However, the mix of compliance-oriented to performance-oriented inspections would be altered from present practice.

l 1-5

The construction QA program described above incorporates two major philoso-phical and policy shifts in emphasiri between past and present NRC practices:

a change in emphasis from prescription to performance, and a change in emphasis from post-construction inspection and review of completed work to incremental insper:tions and review. This construction program would also reduce duplication of inspection effort. Each is discussed in more detail below.

A Performance-Oriented Approach The first philosophical and policy shift is to strike a balance that complements past approachs to QA with a performance-oriented approach. In a performance-oriented program, the desired quality attributes of the completed construction project would be defined without NRC detailing the specific r:ays to achieve them. This change would allow the licensee greater flexibility in developing approaches that can be tailored to the specific project, corporate philosophy, local work conditions, etc. It would also encourage innovation. The NRC would not prescribe the performant;e objectives, although the NRC might participate in developing associated guidance. Each licensee would define its own performance objectives and set up a management information and data analyses program that would monitor, perform trend analysis, and evaluate key construction performance indicators.

Project performance determined by the trend analysis program would be frequently reviewed by management and compared against project goals and objectives.

Corrective action to address performance shortcomings and failt.re to meet project goals would be oriented toward eliminating or correcting the oroblems' root causes.

NRC's revised inspection program for the project would complement routine QA program inspections with performance-oriented inspections that emphasize direct observation of work in progress; evaluation of the quality of completed work; and assessment of the licensee's performance both in evaluating its own performance and in identifying and correcting root causes of quality problems.

The ability of licensee management to identify the key construction performance parameters, to collect and analyze the performance data, and subsequently to act on the data would be an important indicator of the licensee's implementation of its quality assurance activities. NRC's balanced inspection program would also provide a useful source of information for management meetings and a basis for any necessary enforcement action.

Incremental NRC Inspections and Review of Completed Work Under the revised QA program, plant quality would be determined and evaluat.>d on an incremental basis to a greater extent than in the past. If plant quality were found to be acceptable, the NRC would confirm while construction was still in process that previously completed work met NRC requirements.

This change would bring greater stability and predictability to the licensing and construction process, thereby increasing public and investor confidence and defusing some construction pressures that act to diminish the achievement and assurance of quality. That change would also contribute to potentially lower 1-6

D lower plant costs by avoiding rework and delays, and increased public confidence in the quality of a given plant by providing a way for QA issues and conten-tions to be detected and resolved while the plant is being built, rather than after it is completed or nearly completed.

The licensing of several plants has been delayed significantly by many last-minute allegations concerning the plant's quality. Resolving the allegations has been time-consuming and costly to both the applicants and the NRC.

Resolving allegations requires intensive personnel resources, and the lateness ,

of the allegations has made resolving them more difficult because the long-since-completed work is often inaccessible (e.g., rebar in containment concrete) and because the construction workers who built the plant may have left the area. An NRC progr am change is needed in which completed work could .

be reviewed by NRC earlier in the project and in which the ,(U4 contentions could be resolved earlier in the project, less ambiguously, and at lesser cost to licensees, to the NRC, and to public confidence in the nuclear power plant.

To achieve this program change, several recommendations from the QA report need to be combined. The quality of completed work could be periodically appraised through readiness reviews, third party audits, or increased NRC inspections. The NRC would give tentative approval of the quality of the work done to date. NRC actions and decisions would be public documents. The public would be able to identify issues, allegations, or contentions pertaining to the particular area under review at that time. Later allegations concerning the work woulo not be considered if they could have been raised at the time the area was under review. However, the NRC would retain the ability to reopen any issue where new information is presented.

Under this process, the applicant, stockholders, state utility regulatory agencies (SURAs), the NRC, and the public would have periodic affirmations of the quality, safety, and management control of the project. This process has several important benefits: it would provide positive incentives to the appli-cant to build quality into the plant; it would provide the financial community the confidence it needs to provide continued financial support to the project; and it would increase NRC and public confidence in the quality and safety of the plant.

This program would require a change in NRC licensing practices to gear the Safety Evaluation Report (SER) to licensee-submitted readiness review modules.

While no substantive changes to the methodology or content of these SERs is envisioned, there would necessarily be changes in the timing and organization of the SER. The NRC in process review would have to be based on objective measurement of the quality of the completed work. Of the ways that have been suggested for confirming quality, the concept of readiness reviews is preferred over inspections by the NRC and audits by third parties or designated representa-tives. Aeadiness reviews are preferred because they better convey the philosophy cf putting responsibility for quality with the performing organization, not with an inspection group. Readiness reviews look forward as well as backwards.

They can be used at several stages (in addition to construction), including start up, cperation, and refueling, throughout the life of a plant. Readiness reviews ave discussed under Section 1.3.4, " Transition from Construction to Operation."

1-7

Reducing Ouplication of Inspection Effort

. Nuclear plants in general and construction projects in particular are subject to inspection audits or surveys from several organizations, including the NRC, INPO, the American Society of Mechanica AuthorizedNuclearInspectionAgencies)yngineers(ASME)Codesystem(ASME, and possibly third parties (hired by the licensee with or without the NRC's encouragement). Some of the inspection efforts are duplicative, resulting in both an unnecessary inspection burden on licensees and a less-than-optimum use of- resources by the inspection groups, including the NRC. Under the revised QA program, NRC inspection credit would be granted for certain inspections performed by non-NRC groups having proper technical qualifications and programs. In particular, the revised QA program anticipates an inspection approach similar in some respects to the German TUV (Technischen Uberwachungs-Vereine) program, in which, e.g., NRC would not normally inspect the construction of the pressure boundary of commercial nuclear power plants but would retain the NRC option to inspect. Given proper safeguards, this responsibility could be delegated to a group such as the National Board, and the National Board inspections could carry with them NRC

" regulatory credit." This option is identified as one which the NRC would examine; other options such as ANI activities would also be reviewed for applicability.

l As part of its comments on the third party audit concept contained in the QA report., the National Board volunteered to assume more inspection responsi-bility for primary pressure boundaries. Until recently, the ASME Boiler and Pressure Vessel Code's QA program requirements have not adequately addressed the requirements of Appendix B to 10 CFR Part 50 for certain (non-hardware) requirements; however, the differences are close to being resolved. Both the National Board and the NRC recogniza that the National Board inspections would need to be augmented to substitute for NRC inspections for installing safety-related equipment, and that safeguards would be needed to ensure the adequacy of the National Board program to meet NRC requirements. The National Board is one of several groups whose inspections would be considered for regulatory credit in place of NRC inspections.

1.3.4 Transition from Construrtion to Operation Several projects, including some in which the construction phase was well managed, have experienced difficulty in making the transition from construction to operation (e.g., Palo Verde). In contrast, some other projects made the Jnder the ASME Code system, the ASME accredits organizations doing' nuclear work and issues Code symbol stamps. It does this by performing audits and surves. The organizations doing nuclear work under the ASME system are required to engage the services of an Authorized Nuclear Inspection Agency to perform certain verifications and certifications. These agencies are either jurisdictions (e.g., state governments) or insurance companies.

Their employees who are qualified to perform these activities are certified by the National Board of Boiler and Pressure Vessel Inspectors (National Soard) and are called Authorized Nuclear Inspectors (ANIS). The National l Board also may be called in by the states or the ASME to investigate problems. The ASME, the National Board and the ANIS are all part of the ASME Code system.

1-8 i - . _ - . - _

l transition smoothly (e.g. , St-. Letie 2). The essential factors that underlie a project's success in effecting a smooth start-up have been demonstrated to be

1) early involvement of start-ur and operations personnel in construction planning, 2) the sequencing of construction activities to support single-system completions (rather than a bulk-completion philosophy), 3) planning and scheduling for turnover, and 4) identification ano timely correction of problems as systems are turned over, rather than waiting for the entire plant to be completed. At St. Lucie 2, construction planning and scheduling were structured to complete reactor systen and to turn them over to the start-up group at the earliest possible time. The QA study and NRC experience indicate that difficulties in transition from construction through start-up to operation generally are more of a reflection en the utility's management programs and operational readiness than they are on the competence of its contractors.

An operational readiness review program is a management tool which can help utilities better manage their progress through the transition points. A readiness review is a formal, structured assessment of a program or project to determine the preparedness of the personn'el, the plant and hardware, and the management programs to proceed beyond some milestone. The review is based both on the current state of readiness and on performance to date (work completed and plant status). Readiness seviews have been used as a formal planning and evaluation tool for many years by other industries and government agencies [e.g., the National Aeronautics and Space Agency (NASA) and the Department of Energy (00E)] with considerable success, including application in nuclear construction projects (e.g., Fast Flux Test Facility, Gaseous Centrifuge Enrichment Plant, Tokomak, PUR6X reprocessing plant).

In the life cycle of a nuclear power pl. ant, many opportunities are available to enhance quality by using the disciplined approach to planning and project management embodied in the master inspection plan and readiness review concepts.

The routine use of readiness reviews as a planning and project management tool could help to achieve quality in each phase in the life cycle of a nuclear power plant: design, construction, start-up, operation, refueling, outage, modification, and decommissioning. The beriefits of readiness reviews may be realized particularly in the difficult-to-manage transition from construction to operation. For major modifications and plant start-up after a lengthy outage, comprehensive readiness reviews, together with an inspection plan to er,sure the smooth execution of planned activities, could provide a structured, thorough, and rigorous way to demonstrate that applicable quality and safety requirements have been met. In addition to ensuring better planning and hence a lesser probability of the repetition of past quality problems caused or exacerbated by poor planning (e.g., South Texas in 1979), readiness reviews offer the significant benefit of assuring that senior management is formally involved at critical stages of a project.

In this revised QA program, readiness reviews are encouraged in each of these phases as well as in the transition between successive phases. For example, during construction, readiness reviews (monitored by the NRC for scope, depth and technical content) would be an acceptable way to request regulatory (licensing and inspection) review for work completed according to an established inspaction plan and for regulatory agreement on readiness to continue construction based on preparedness of personnel, plant, and programs to proceed.

1-9

The operational readiness reviews pilot program under way at Georgia Power Company's Vogtle project is a test of the benefits and feasibility of readiness reviews in conjunction with master inspection plans and incremental NRC review of completed work.

1.3.5 Operation For the operational phase of nuclear power plants, the revised QA program embodies several of the features already discussed under other plant life-cycle phases, including performance-oriented QA programs, trend analysis and readiness i reviews.

Conceptually, these features could be applied to the operating phase similarly as they were applied to the design and construction phases; however, these features must be implemented differently. For example, in a performance-oriented QA program for plant operation, a method might be developed to measure and analyze operational indicators of plant QA program performance, such as mistakes in surveillance testing, mistakes in valve or system lineups, repetitive unplanned reactor trips from similar causes and amount of rework in safety system maintenance. NRC's role would not be to prescribe what performance indicators are to be measured and analyzed, but to review the technical merit and completeness of the indicators that are selected, the performance of the trend analysis and diagnostic programs in providing relevant management information in a timely fashion, and the effectiveness of the licensee in root cause analysis and in making program corrections and improvements based on information provided by its diagnostic and trend analysis programs.

Readiness reviews would be used to improve planning in the operation stage during a transition from one operational status to another (e.g., for major repairs or modifications). Some of the structured planning process inherent in l

readiness reviews is already embodied in the operational phase (e.g., start-up from cold shutdown).

In the operation phase, the already established INP0 evaluations are an important example of improved allocation of NRC inspection resources that have reduced the inspection burden on licensees. In recognition of the effective-

! ness of INP0 assessments at operating plants, the NRC reduced its Performance Appraisal Team (PAT) inspections by 80%. The remaining level of NRC inspection is designed to monitor and overcheck INP0 evaluations to ensure that they are adequate to meet NRC needs. This serves as both a model and precedent for the inspection recognition proposed to be given for inspections by other organiza-tions.

1-10

(

i

P 2.0 UPDATE OF NUREG-1055 RECOMMENDATIONS In the transmittal of the QA report to the Commission (SECY-84-124), the NRC staff asked for approval of staff action to begin the regulatory analysis process possibly leading to the following regulatory actions:

1) a post-construction permit (CP) demonstration rule

2) establishment of an advisory panel for CP applications

3) a third party audit rule. -

The staff believes that implementing the revised QA program will provide sufficient improvements that some of these actions may not be necessary and should be deferred. In fact, some may result in duplication and unnecessary additions when considered in conjunction with existing programs and other QA study recommendations being implemented. For example, if the readiness review program described in this plan is successfully implemented, it will, in effect, accomplish the management improvements the staff hoped to achieve through the post-CP demonstration rule and the third party audit rule. Readiness reviews place the responsibility for the project where it belongs -- with licensee management. In addition, readiness reviews do not carry the nonconstructive implications of another layer of inspection, which third party audits carry and which was objected to heavily in the public comments on NUREG-1055.

Because no new CP applications are imminent and because public reaction was mixed on the proposal for establishing an advisory panel for CP applicants, the staff concluded that the request for this rulemaking should be deferred until alternatives to this proposal can be further analyzed. In particular, the staff wishes to explore the feasibility of an existing industry organization

[e.g. , the Institute of Nuclear Power Operations (INPO), the Nuclear Utility Management and Human Resources Committee (NUMARC)] providing a self-screening rather than establishing an additional committee providing advice to the Commission.

In addition to the three actions discussed above, the QA report and SECY-84-124 contained many other recommendations and identified several topics for further analysis (NUREG-1055, pps. 2-33 and 2-38). This plan incorporates all of these recommendations as areas for further study, with the following exceptions:

1) management appraisals as an adjunct to Construction Appraisal Team (CAT) inspections

2) expanded interim CAT inspections

3) NRC Designated Representatives (DRs).

2-1

Management appraisals as an adjunct to CAT inspections were dropped in favor of other management initiatives (see Section 3.2). The CAT inspection program was not expanded because of NRC budget and personnel resource limitations. During a meeting.on October 16, 1984 to discuss the QA report to Congress with the staff, the Commission decided not to go forward with a DR pilot program for a plant under construction unless requested to do so by a licensee. Consistent with this decision, further DR program activities have not been included in this plan.

Tables 2.1 and 2.2 summarize.the status of the QA report recommendations as i presented in this plan. Changes to the original recommendations have been l

based on public comments on the QA report, recent Commission and Advisory Committee on Reactor Safeguards (ACRS) meetings on QA, further staff analyses, and budget considerations.

i a

2-2

TABLE 2.1. Staff Analysis of QA Report Recommendations (see NUREG-1055, Table 2.1)

QA Report Recommendation Final Staff Recommendation

1. Enhanced Pre-CP Review of Planned developmental action Experience and Managerial cancelled due to budget priority Quality - of operating plants Establishment of an Advisory Dropped pending exploration of potential Board industry action on this topic

2. Post-CP Demonstration of Dropped in favor of Readiness Review Management Effectiveness Program

3. QA Program Performance Develop for potential implementation

• Objectives at operating plants and plants under construction. Cooperative NRC/

industry performance objectives and trend analysis pilot programs will be explored

4. Management Apprai als as an Drop in favor of readiness reviews Adjunct to CAT Inspections and other management initiatives.

Limited management process reviews implemented as part of CAT inspections

5. Inspection Prioritization All plants currently under construction of Plants Currently Under are receiving a " graded" inspection Construction (Problem Plant approach. Planned long-term analysis Identification) (for application if construction of a new generation of nuclear power plants should begin) indefinitely suspended due to budget priority of operating plants

6. Improved Diagnostic Capability / Implement as part of QA Performance Trend Analysis Objectives Program

7. Senior Management Meetings Implement

8. Enhanced Vendor Program Being implemented

9. Third-Party Audit / Interim. Recommended third party rulemaking CAT Inspections / Interim deferred pending outcome of readiness Independent Design Verification review program; no activity planned Program (IDVP) due to budget priority of operating plants; perform CAT inspections at 4/ year (no expansion in program);

continue IDVP on case-by-case basis.  ;

2-4

a

-g.

QA Report Recommendation F,inal Staff Recommendation

10. Regional Team Inspections Increasing use of team inspections.

Pilot under way in one NRC region

11. Expanded Resident Program .

More residents being assigned to sites. Pilot test under way in one NRC region to reduce regional inspections in favor of resident inspections

12. Improved Licensee Detection Develop self-detection capability Capability for potential implementation as part of the QA Performan'ce Objectives Program recommendation

13. Independent Audits of NRC QA Implement Programs 2-5 I

l l

TABLE 2.2. Status of Areas Requiring Further Analysis (NUREG-1055, Table 2.2)

Area Recommended For Further Analysis Final Staff Recommendation

1. QA Report-Type Study for Plants Limited study efforts under way in Operation -

or planned .

2. Prioritization of QA Measures: Analysis is in progress Guidance on " Safety-Related" vs.

"Important to Safety"

3. Measuring Effectiveness of QA Research projects focusing on plants under construction and in operation under way

~4. Essentially Complete Design Combined with #5 below at CP Stage

5. Configuration Management Implementation planned by subsuming into rulemaking on control of post-CP design changes (transferred to the Office of Inspection and Enforcement from the Office of Nuclear Regulatory a Research in September 1984). Planned developmental work related to future plants dropped due to budget priority of operating plants

6. Readiness Reviews Pilot program being initiated for plants under construction

7. Quality Engineering Further analysis and developmental work not planned due to budget priority of operating plants

8. Alternate Ownership and Efforts redirected to explore Management Arrangements improvements within current arrangements in an all operating reactor environment

9. Feasibility of Designated Further censideration deferred pending Representatives a licensee request for a pilot program.

Budget priorities preclude any activity in this area

10. Limiting cps Analysis of the problems faced by NRC in the event of a resumption of nuclear plant construction is included as a possible contingency in the study described in 8., above.

2-6

3.0 QUALITY ASSURANCE FOR OPERATING REACTORS AND PLANTS CURRENTLY UNDER CONSTRUCTION The QA study focused primarily on examining past quality-related problems in the design and construction of nuclear power plants and on recommending improve-cents to NRC and industry programs and approaches. Although it is clear that significant improvements can be made to NRC's and industry's quality programs for design and construction, it is also. clear that 1) no new applications for construction permits (cps) are likely to be submitted for some time; 2) the number of plants currently under active construction is decreasing, and many under construction are nearly complete; 3) the number of operating plarts is increasing, and the current operating plant population is aging; 4) the health

~

-and safety implications of quality shortcomings in operations are more immediate and offer less time for correction than do quality shortcomings in design and construction; and 5) much less NRC QA staff effort has been directed at improving quality programs for operation than for design and construction.

Based on these considerations, this NRC QA program implementation plan places heavy emphasis on the quality of nuclear power plant operation. NRC's fundamental mission, as stated in the 1984 Policy and Planning Guidance (NUREG-0885, Issue 3), is "to make sure that existing nuclear facilities and those coming on line operate safely. Consequently, the nighest priority will be given to assuring that operating facilities maintain adequate levels of protection of health and safety." Although this chapter emphasizes QA for operating reactors, most program elements in the chapter include some activities directed at improving the quality of design and construction as well.

Many of the lessons learned and recommendations from the QA study apply to the operational phase, e.g., the importance of management, prior nuclear experience, root cause analyses, and qualifications.of QA/ quality control (QC) personnel.

This plan is tailored 1) to adopt from the QA study those lessons and recommendations that can be immediately transferred to operations, 2) to determine if other QA study results can be transferred, the applicability of which may not be immediately obvious, and 3) to determine imprevements to quality for nuclear power plant operations which may not have a parallel or counterpart in design and construction (or are not findings or recommendations of the QA study).

The QA study identified several areas where the NRC and utilities have failed to maintain or achieve the desired level of performance as required by NRC regulations and as expected by the public and the Congress. For example, problem areas included cases where the licensees failed to implement all aspects of their programs as required, cases where licensee management neither took a sufficiently active role nor provided sufficient resources to assure plant quality, and cases where the personnel responsible for design, construc-tion, and management of the project lacked the requisite experience and quali-fications. The Kist Report (Appendix B to the QA study) pointed out several

~

shortcomings in the NRC's own QA program, especially the NRC inspection pr'ogram.

Inadequate, incomplete or insufficient NRC (or industry) guidance in some 3-1

areas and shortcomings in vendor programs were also highlighted. The QA program elements in the following five sections are structured to address these problems for plants in operation as well as for those in design and construction stages. The program elements are also structured to develop an approach to assure that the required levels of performance and protection of health and safety can and will be achieved.

3.1 QA PROGRAM IMPLEMENTATION The QA study confirmed the common belief that many of the nuclear industry's QA problems do not rest with the adequacy of the QA programs themselves, as described in Final Safety Analysis Reports (FSARs), QA manuals, etc., but with the effective implementation of those programs. Many cases can be cited where licensees have failed to take advantage of the benefits of a fully implemented QA program. In such cases, NRC's_QA program requirements were generally perceived simply as items to be " checked off" in a long list of regulatory requirements, rather than recognized as useful management tools. The activities planned under this program element are directed toward improving the overall implementation of the QA program by increasing its usefulness as a management control system and assuring that the required level of performance is achieved.

The goal of this QA program element is to balance the attention of NRC's and industry's QA programs between programmatic aspects of QA programs (written program descriptions, QA manuals and other paperwork) and the quality of work activities, including the end product (i.e., program performance). The QA report agrees with some in industry who claim that a large part of the QA problem in the nuclear industry is that too much attention has been paid to the programmatic details of quality assurance and too little attention has been paid to the program's results.

The performance-oriented QA programs envisioned by the plan would have as their foundation the existing QA requirements in Appendix B; consensus standards such as NQA-1, which provide guidance on how to meet Appendix B; and licensee commitments to these consensus standards. That is, the existing regulatory base, which is necessary for compliance inspection and enforcement action, would remain intact. What would change is the emphasis of NRC's inspection effort. Several data points suggest that an inspection strategy'that balances compliance inspection with performance evaluation is preferable to a strategy that overemphasizes either at the expense of the other. An inspection approach biased too heavily toward compliance tends to suffer from a " missing the forest for the trees" syndrome. An inspection approach biased too heavily toward performance may be unable to establish the legal basis upon which to take necessary corrective action.

3-2

The existing NRC inspection program provides for both compliance and performance-oriented inspection activities. Construction Appraisal Team (CAT) inspections provide one of the best examples of a performance-oriented inspection method. CAT inspections concentrate initially on the quality of the end product or hardware. If the hardware is of acceptable quality, then the licensee is assumed to have executed an effective program for achieving quality.

If the hardware is deficient, the CAT inspection then looks for the programmatic shortcomings that may have led to the hardware deficiencies. This programmatic look for root.causes of identified shortcomings in performance includes an analysis of whether licensea commitments were followed. . Enforcement actions then may.be taken based on violations of commitments that led to substandard performance.

One objective of this QA program implementation plan element is to make greater use of the general inspection approach used by the CAT inspection program in the overall NRC inspection program. This approach looks first at performance.

It then determines root causes of poor performance, and bases enforcement action, if it is necessary, on violations of requirements and commitments that were determined as part of the root cause analysis. This approach starts with an identified quality or safety problem and works backwards from there to the root causes. This approach has certain obvious advantages over other inspection approaches. It focuses NRC and licensee attention on situations where real quality or safety problems have been identified, not on hypothetical problems; it emphasizes identifying and correcting the root causes of quality problems; and it conveys the message that end results are what is important, not just the methods of obtaining those results.

. In the revised program envisioned by this QA program implementation plan, the overall NRC inspection program would strike a balance between performance-oriented inspections, such as described above, and compliance-oriented inspections. As pointed out by one Regional Administrator, "we should not abandon our reviews of a good prescriptive compliance-oriented QA program.

Program reviews of this type do provide the opportunity to identify and correct weaknesses or deficiencies before they result in hardware problems."

The NRC inspection program provides sufficient guidance for " compliance-oriented" reviews. An objective of this program element is to expand and sharpen program guidance for the complementary " performance-oriented" reviews. Essential to developing such guidance are 1) describing what the licensees' QA programs should accomplish (i.e., what the performance objectives are), 2) developing methods to determine the extent to which these objectives are met, and 3) puttir.g in the hands of the NRC inspectors the tools for measuring the extent to which QA program objectives are met. The subelements in this section collectively address these essential steps.

3-3

The benefits which would result .from accomplishing the goal of this program element are as follows:

More attention will be focused on what is done rather than on how it is done, resulting in more time for the NRC and industry to focus on the quality of completed work.

More NRC and licensee attention will be paid to identifying and correcting the root causes of quality problems.

Diagnoses of developing quality problems will take place sooner.

With an increased focus on end product quality, QA can be applied more cost-effectively, and the capability for more timely detection of major quality and safety problems will improve. ,

The issues associated with this program element include the following:

difficulty with an industry and its regulators making the transition from the quality assurance framework which has developed over the past fourteen years to a more performance-oriented framework development of useful and measurable indices of performance.

This QA program element is addressed by two subelements - p'erformance objectives and trend analysis, and measurement indices / effectiveness of QA programs -- discussed in Sections 3.1.1 and 3.1.2 below.

3.1.1 Performance Objectives and Trend Analysis A three pronged approach to developing performance-based inspections is planned.

The Institute of Nuclear Power Operations (INPO), the Nuclear Utility Management and Human Resources Committee (NUMARC), and other organizations have already completed some work on performance goals and objectives for operating plants. This work will be reviewed, and discussions with cognizant representatives of the industry will be held to assess whether any NRC action is necessary or appropriate. In addition, an interoffice advisory group of NRC staff will be established to develop any changes which may be usefully made in NRC's inspection methodology to enhance the results of industry activities in this area. Should it prove desirable, joint NRC/ industry pilot programs with cooperating licensees to test, refine, and improve the performance-oriented approach will be undertaken.

Preliminary discussions are under way with one operating plant licensee to initiate a pilot program to test, develop, and refine the concept of a performance-oriented QA program. Other cooperating pilot program candidates are being sought. The pilot program under discussion will involve developing performance objectives, identifying and monitoring relevant performance 3-4

p parameters, performing trend analysis of QA program performance with respect to-the objectives and parameters, determining root causes of poor program per. form-ance, and taking corrective action to address root causes. The present QA commitments for the candidate plant will be applicable during the pilot program in parallel with the testing and development of the performance-oriented QA approach. Although the first pilot program may be performed at an operating plant, the basic concepts to be tested can be applied to plants under construction.

Licensees that have expressed an interest in participating with NRC in pilot programs see such participation as providing certain benefits for themselves and the industry. Performance-oriented programs give licensees more flexi-bility in developing their QA programs to best fit their particular corporate and plant situation and needs. By giving the licensee more flexibility in determining how its QA program is defined and implemented and by focusing increased licensee management and NRC attention on the quality of the end ,

product rather than on the details of the program the licensee uses to monitor work activities, the effectiveness of licensee QA programs may be improved. To achieve this goal, general guidelines on determining site-specific performance goals and objectives should be established, while allowing the licensee latitude to determine its own specific approach for achieving the performance objectives.

Conceptually, the performanca-oriented QA program envisioned for the future would have the following characteristics:

• Licensees would develop the site-specific performance objectives and criteria for the QA program at their plants. The NRC would provide general guidance.

• Licensees would select parameters and perform trend analyses on them.

Skill in selectivo the parameters and in addressing root causes of problems reflectu n. trends would be an element in NRC's periodic Systematic Assessment of Licensee Performance (SALP) evaluation.

• The results of plant performance trend analyses would indicate the status of ongoing activities and their acceptability.

• The NRC program for inspecting QA programs would be revised to provide more emphasis on analysis of plant trends.

NRC performance-oriented QA inspection efforts would consider the quality and validity of the information going into the trend analysis reports, the directions of the observed trends, the significance of those trends, and utility management's reaction to them, including identification and correction of root causes of quality problems.

The NRC would analyze root causes in selected incidents (e.g., to determine the cause of significant quality or safety problems), and the results would be discussed in senior management meetings.

• The current prescriptive requirements for hardware or operating procedures are not expected to change. Rather, trends in meeting those requirements or procedures would be used in measuring the effectiveness of the licensee's QA program implementation.

l 3-5 l

1 _ _ _ _ __ _ _ _ _ _ .

To develop a performance-oriented QA program, the following activities are planned:

Review performance objectives of work completed or in progress by INP0 and others. Hold discussions with cognizant NRC staff.

Establish an NRC interoffice advisory group to integrate approaches for performance objectives and to assist in developing and implementing the program, including participating in pilot program activities.

Develop methods, including trend analysis programs, to evaluate the extent to which QA program goals have been achieved.

Conduct pilot program (s) to develop and test the QA program performance objectives and trend analysis measurement concept.

Monitor pilot programs and program results. Revise candidate program approaches to reflect experience and lessons learned from the pilot program.

End Products If the performance objectives and trend analysis approach proves successful, the end products of these activities would include the following:

NUREG or other report describing results of pilot program (s) and guidelines for performance-oriented QA programs (FY86) revision of Chapter 17 of the Standard Review Plan (SRP), NUREG-0800, if appropriate (in conjunction with routine revision of the SRP) revised inspection modules to provide inspection guidance for gathering information on performance-oriented QA programs (FY87) revised SALP procedures to use the information gathered on licensee performance in implementing their performance-oriented QA programs and trend analysis programs (FY87).

3.1.2 Measurement Indices / Effectiveness of QA Programs To determine the effectiveness of a performance-based QA program, measurement indices must be developed. The existing techniques for evaluating QA programs, which monitor compliance with written commitments and procedures, would need to be modified to permit the NRC evaluator (license reviewer or inspector) to view the QA program in a more comprehensive manner and to determine if the overall program .is effective.

The NRC staff currently has under way a modest research program to develop measures of QA program effectiveness. This research effort was begun partially in response to questions raised by the Advisory Committee on Reactor Safeguards (ACRS) and others to determine the cost-effectiveness of Appendix B QA programs.

The current research efforts are oriented toward addressing both the cost-effectiveness of Appendix B quality requirements and the larger issue of devel-oping effectiveness measures for QA programs in general.

3-6

The activities planned to assess the feasibility of developing measurement indices far QA program effectiveness include the following:

Perform feasibility studies on developing candidate performance measures for QA program effectiveness at the component, system, and plant levels (initial QA research efforts are presently under way).

Conduct follow-on efforts based on the feasibility studies to develop, test, and refine candidate effectiveness measures for QA programs. These efforts would include identifying and determining data requirements for the candidate measures and assessing the burdens and benefits associated with those measures.

End Products The end products of these activities include the following:

input to developing performance-oriented QA programs and program guidance and also input to the end products listed under Section 3.1.1 (FY85, FY86) performance measures for the cost / safety-benefit analysis of QA program requirements. (FY87).

3.2 MANAGEMENT ISSUES In recent years, many studies, reports, speeches, workshops, and proposals have addressed the role of quality, quality assurance, and management in the nuclear industry. These activities usually emphasize the need for experienced management to be actively involved in and supportive of a fully implemented quality program. The most recent of these reports, the.NRC's QA report, discusses in detail a number of recent quality-related breakdowns at nuclear power plants. These recent examples demonstrate the cost and schedule conse-quences of lack of management involvement and support in assurance of plant l quality. The identified issues and causes of these quality problems can be l summarized as follows:

l l

• The central cause of major quality problems in the design and construction of nuclear power plants was the failure or inability of some utility managements to effectively implement a system that ensured adequate control over all aspects of their project.

l NRC's past licensing and inspection practices did not adequately screen CP applicants for an overall capability to manage or provide effective management oversight of the construction project.

l Effective use has not been and is not being made of current management, engineering, safety, reliability, and quality assurance practices exercised in other industries and federal regulatory agencies where safety and reliability are also critical concerns.

The first two summary conclusions are from the QA report and address design and construction. The last finding is from the Kemeny Commission staff report on QA (October 1979) and focuses on the operation of nuclear power plants. It is j noteworthy that all three conclusions relate to management effectiveness.

! 3-7 l

l The importance of management's commitment to quality is reflected in the NRC's f

which Issue 3, p.can 5 . be found in the NRC's 1984 Policy and planning G places on mana)gement The followingand excerpts highlight theinvolvement:

management importance that the NRC

"...NRC's goal is to assure a high level of quality in management

[maintenance.

emphasis added] of reactor design, construction, operations, and

...The theme of "do it right the first time" should be adopted [by management]

operate safely.to ensure [that] plants are built properly and can

... Assure [that] utilities provide the appropriate management

[ emphasis added] framework and capability...

...The necessity for highly trained and qualified professionals...

to manage [ emphasis added]..."

Improved as the resultmanagement of NRC regulation. in the industry cannot realistically be expected.to develop It must come as the result of industry initiatives, areas needingstemming improvement. from the industry's self-awareness of shortcomings and The NRC staff is aware of the complexity and difficulties of the management issue, and the thrust of this section is to improve management in the industry through industry-initiated actions , peer pressure, utility sharing of experiences, and incentives for utilities to improve management.

1 The primary quality throughgoal of utility batter this QA program element is to improve assurance of plant management.

The subelements' discussed below goal.

present new opportunities for both NRC and industry activity for a demonstrate and to share the benefits of an implemented QA program, actively supported by a management that provides adequate resources.

Other subelements identify ar 3as where rules and regulations could be revised or developed to facilitate and/or to remove impediments to improved management and management programs.

A second but related goal of this program element is to provide the NRC with an improved understanding of the effects of State Utility Regulatory Agencies' The SURAs' regulatory authority over utilities for allowing a costs is now a major factor affecting utility decisions.

Recent events at Midland, Zimmer, and Marble Hill vividly demonstrate the financial impact of .

SURAs'power nuclear actions on utility plant decisions to suspend, continue or modify their activities.

Recent SURA actions in California and Colorado of nuclear power plant availability to produce electrical power.for in impact of such incentives and of other SURA activities needs to be fullyThe safety understood nonsafety issues. by the NRC, without encroaching on traditional state authority in 3-8

The benefits which would result from accomplishing these goals include the following:

• improved appreciation by nuclear utilities of the merits of a fully implemented QA program

• more_ direct involvement of senior utility management in and resource commitment to implementing the QA program

• improved NRC' understanding of the nuclear utility perspective of QA and increased appreciation of how NRC activities may be tailored to help improve overall QA performance

• improved NRC understanding of SURAs and SURA policies; greater NRC appreciation for the rationale behind SURA actions and the response or actions by utilities to SURA (SURA / utility relations)

• improved NRC understanding of the safety implications of SURA actions.

The issues associated with this program element include the following:

• Utility management is an area in which NRC traditionally has not had significant involvement.

• NRC decisions and judgments on management and management issues are inherently more subjective than-detailed engineering analysis, and both potentially result in contention within the NRC and with licensees.

• NRC interest in SURA actions and policies may incorrectly be viewed as a forerunner to federal involvement in traditional state areas of responsibility, and as such could be perceived as NRC overstepping its bounds of authority.

This QA program element is addressed by seven subelements in Sections 3.2.1 through 3.2.7 below.

3.2.1 Recognition and Support of Industry Efforts The NRC and the nuclear industry have recognized the role that management must play in assuring that the QA program is fully implemented and that plantTo realize these quality and safety are achieved and maintained.

utility industry has initiated major efforts to share information and lessons learned on the need for direct management involvement and commitment to quality In addition, INP0 is conducting site assessments which and quality assurance. In early include a review and report of management programs and practices.

1984, the utility industry formed a new committee, the Nuclear Utility Management and Human Resources Committe'e (NUMARC), to coordinate The industry initiatives for improving utility management and management programs.

objective of NUMARC is for utilities to identify for themselves the utilities prescribe remedial or new initiative actions that are needed The QAto raise the stanciard of quality and excellence in utility management.

must come from the nuclear industry itself; they cannot be " r 3-9

a i NRC. The NRC should encourage the nuclear industry to police and improve its own activities and the NRC should be supportive of the industry's efforts.

1 However, the NRC must exercise caution in not relying too heavily on or in i

delegating NRC's responsibilities for monitoring management tc these industry programs. The NRC has the statutory responsibility for nuclear safety. As

such, the NRC must have licensing and inspection methods for. evaluating the i effectiveness of industry initiatives and must take or force action to be taken when industry programs are ineffective.

The goal of this subelement is to ensure that the NRC. recognizes and does nothing to inhibit promising industry programs to improve the effectiveness of management in the nuclear industry in a manner and to a degree that is consistent with NRC's statutory responsibilities.

The activities to be conducted under this subelement include the following:

Assess the effectiveness of industry programs to improve quality and quality assurance in the nuclear industry, and work with industry groups for improvements or new initiatives where appropriate.

Recognize and support proven industry programs to improve management effectiveness as it relates to quality; support promising programs under development and share information on both successful and unsuccessful m agement programs and practices.

Coordinate and integrate NRC activities to the extent appropriate so that 4

the NRC does not unnecessarily duplicate industry activities or take

, actions which would inhibit industry initiatives.

End Product

! The end product of these activities will be:

i recognition and endorsement, as appropriate, of proven industry programs for improving quality through better utility management (as appropriate).

j 3.2.2 Readiness Reviews, Master Inspection Plans, and Incremental NRC Review of Completed Work i

The NRC is exploring methods for providing greater assurance that management programs and practices affecting quality are being implemented as required on a timely basis and that management is assuring that the workforce, programs, procedures, and resources required are available and ready to proceed. In the QA study, the staff identified some Department of Energy (00E) and National Aeronautics and Space Administration (NASA) practices which require' project line management to perform the following: 1) review and evaluate a project or activity to determine its performance to date; 2) assess the.workforce, program, procedures, and resources; and 3) make a definite management decision on the readiness to proceed. NASA and DOE refer to these review activities as

" readiness reviews". A positive aspect of the readiness review concept is that

~

the process forces the active involvement of the licensee and licensee i management at critical stages of the project. Such active involvement of key licensee management makes them better able to discharge their management responsibility for all aspects of the project, including quality.

3-10

1 l

The NRC staff is developing a readiness review pilot program with Georgia Power Company (GPC) at its Vogtle project. Plant Vogtle is approximately 75% complete and is scheduled for fuel loading [ operating license (0L)] in two years. GPC has proposed the readiness review program as a way to assure themselves and the i NRC that all NRC design, construction, and operation requirements and GPC commitments to the NRC have been met and that the plant is ready to operate safely. GPC believes that the readiness review program can help them achieve this goal, and at the same time, minimize delays in operation by providing a mechanism for identifying and resolving-problems, issues, and contentions early. This program also gives the NRC an opportunity to review licensee activities in a more timely and structured manner, thereby giving increased credibility to NRC findings and decisions on issuing an operating license. The readiness review program being developed by GPC and the NRC staff includes developing a master inspection plan with predetermiaed inspection points and provisions for early agreement on acceptance criteria and for incremental NRC inspection and review of completed work. In addition, the readiness review pilot program will provide the added assurance that has been sought through the Independent Design Verification Program (IDVP) process for other licensees that the design fully complies with all regulations and licensing commitments.

These concepts were described in Chapter 1.0, " Overview of the Revised QA Program," of this document.

I The goal of this readiness review subelement is to test the feasibility and l

potential of a system to improve management performance through better planning, I scheduling and more direct management involvement in evaluating plant performance and status. The objective of the activities included in this subelement is twofold. One objective is to assess the feasibility and practicality of readiness reviews, inspection plans, and incremental NRC review for nuclear power plant design, construction, and operation. The second objective is to provide the NRC with sufficient information to~make a decision on changing current NRC practices to accommodate readiness revicws.

The activities included in this subelement are as follows:

• Obtain information, including industry input, on the readiness review i concept and past experiences with readiness reviews for both nuclear and non-nuclear applications.

• Develop a prototype of the parameters of a readiness review program for NRC and define requirements for implementation.

• Monitor GPC's pilot readiness review program for Plant Vogtle and develop l NRC procedures for participation in the pilot program.

t

• Prepare periodic status reports to the Commission on the Vogtle pilot program and on developmental efforts for routine NRC application of readiness reviews.

• Prepare a report to the Commission on the results of the pilot program at Vogtle and present staff recommendations for NRC proposed action (rule, regulatory guide, voluntary by industry, etc.).

3-11

9 End Products The end products of these activities include the following:

reports on the status of the Vogtle pilot program and a final report on the project (status reports beginning in February 1985) a rule, policy statement or guidance on the acceptability of a readiness review program based on a jointly-developed master inspection plan with incremental NRC review of completed work (FY87) revised SRP elements for quality assurance programs accommodating the readiness review concept, if appropriate (FY87) revised NRC inspection procedures to accommodate master inspection plans, readiness reviews and incremental NRC review of completed work (FY87). ,

3.2.3 Configuration Management and Design Completion (a)

The Browns Ferry fire, the TMI-2 accident, NRC regulatory changes, increased public involvement, increased SURA scrutiny a~nd involvement, rising costs, and construction delays have introduced a continuous stream of changes and chal-lenges to the control and management of the design, construction, and opera-tion of nuclear power plants. The environment for managing a nuclear power plant project and the associated complex technical issues requires a strong management control program to manage design changes. The QA report identified two topics for further analysis to address the problem of managing change in the nuclear industry: more complete design at.the CP stage and configuration management.

The QA report cites several examples that support the premise that the more advanced a plant's design at the beginning of construction, the more likely the plant is to be completed without developing major quality-related problems.

Design completion is measured in several ways, and establishing a numerical percentage standard for design completion before construction is started does not appear to be meaningful. Licensees generally believe that requiring 100% 4 design completion prior to CP issuance imposes too great a cost prior to I receiving a construction permit, since there is a risk that a CP may not be granted. Standardized designs will deal in part with this problem, but there is still the potential for submittal of a CP application for a customized plant design. The staff currently believes that such an application should contain no less detail than that contained in an approved standardized design. l

)

The staff also believes that goals should be established for maintaining design '

sufficiently ahead of construction (e.g., 18 to 24 months ahead). Some architect / engineer (A/E) and nuclear steam supply system (NSSS) firms have generallf indicated that the design should always stay at least 18 months ahead of construction until design completion.

The QA report also identified situations in which the sheer volume of changes overwhelmed an otherwise adequate management system. Configuration management represents a disciplined approach to applying technical direction and surveillance (a) Because of the future-oriented nature of this topic, no work is planned in this area due to the budget priority of operating plants. ,

3-12 l

to project design activities and to associated documentation (procedures, drawings, etc.)'to assure that decisions affecting the design are carefully managed. This management tool assures that actions and related documentation are in agreement. A benefit of a fully implemented configuration management program is that in addition to managing and controlling the changes, it limits the number of changes. Changes are limited because a mechanism for screening changes as they arise is provided, and significant and high priority changes are implemented while less-important or less-significant changes are catalogued for future action. .

An important element of a configuration management program is a change control board (CCB) which reviews and approves changes which exceed a certain dollar value or schedule impact. Serving as a focal point, this board would ensure that individuals or subtier groups do not unilaterally make or negotiate unreviewed changes and that all changes are properly coordinated. The configur-ation management arrangements that will be considered in this subelement will employ a CCB, staffed by senior licensee management staff and the licensee's contractors' engineering, construction and project management staff. The CCB will review changes exceeding a predetermined threshold, from whatever source.

The CCB would be empowered to question the-necessity and appropriateness of all proposed changes on technical and economic grounds, although the NRC will make the final decision on changes necessary to meet health, safety or environmental requirements. The CCB functions could be included in the duties of the plant operational review committee (PORC) at operating reactors.

The CCB would report to the utility's executive officer who is responsible for nuclear activities and will periodically issue reports on changes, which will be subject to review by the NRC staff. This arrangement will help control the level of change as well as provide focused, structured management of project change by the licensee.

l The goal of this subelement is to determine what improvements to present i

practice could best ensure that licensees' systems for managing design changes enhance the likelihood of achieving an adequately controlled process.

The activities to be conducted under this subelement include the following:

Survey selected A/Es, NSSS manufacturers and licensees to determine their experiences with design completion and configuration management, past and current industry practices, and plans for the future.

Integrate configuration management and design completion considerations l

into the existing rulemaking activity to limit post-CP design changes.

End Products The end products of these activities incFide the following:

• a decision on incorporating more complete design requirements and l

configuration management considerations into the rulemaking on limiting l post-CP design changes (FY86) 1 completion of the rulemaking on limiting post-CP design changes (FY87).

3-13 l

3.2.4 Management Meetings Principally through the SALP process, the NRC staff periodically conducts senior-level meetings with utility management to discuss the status, progress, and problems of ongoing construction or operational activities. Such meetings normally cover the adequacy of resources, a discussion of significant problems, and the status of corrective measures (including root causes and timeliness).

These management meetings are separate and distinct from enforcement or event-oriented meetings or conferences, which are convened to address specific enforcement or noncompliance issues. The management meetings are intended to ensure that both the NRC's and the licensee's senior management become personally aware of and involved in ongoing activities to assure continuity and completeness of management's (utility's and NRC's) understanding of actual plant status and activities. The Commissioners have expressed their willingness and desire to participate in such meetings where the Executive Director for Operations (ED0) believes that their early involvement might head off potentially significant problems. Currently, other NRC. offices' participate as appropriate in these management meetings.

Management meetings are considered to be the primary vehicle for NRC management to communicate to utility management generic safety and quality matters as well as plant- or utility-specific matters. Periodic management meetings are an integral part of the SALP process. They provide an effective means for discussing SALP results, for establishing lines of communication among key managers, and for focusing the attention of top NRC and licensee management on specific problems. Management meetings provide particularly effective forums for NRC management to convey the results of NRC root cause analyses of licensee safety or quality lapses. These problems often point to management shortcomings and the need for licensee action to address the root causes. The goal of this subelement is to expand the existing practice for senior-level management meetings and to provide guidance on the reasons for such meetings, their con. tent, the level of the participants-(e.g., vice president nuclear, plant manager), the scope, the circumstances under which they-should be conducted, and the considerations related to the frequency of the meetings.

The activity included in this subelement is as follows:

Develop guidance on management meetings including the reasons, content, participants and frequency, and the relationship of non-SALP and SALP management meetings.

End Product The end product of this activity will be:

a revised NRC Manual Chapter which describes senior-level SALP and non-SALP management meetings (FY36).

3.2.5 Impact of Actions of SURAs on Quality and Safety l The NRC needs to develop a better understanding of the associations, interfaces, r and relationships between SURAs and utilities, particularly as they may affect plant operational quality and safety. The potential exists for SURA / utility I

3-14

decisions and actions to directly impact plant quality and safety. For example, 3

SURA incentives or decisions.which have the effect of maintaining a plant in operati.on when the~ plant's continued operation may jeopardize the public health

- and safety would constitute concern to the NRC. SURA actions in allowing or disallowing certain contract costs or requiring certain types of contractual agreements between utilities and contractors while being fiscally responsible, could be detrimental to plant quality and safety. The QA study analyzed the j

types of. contracts used by utilities and identified the value (and disadvantages) of various contractual arrangements. The NRC needs to better understand SURA policies and decisions, in the areas of allowable (and ditallowable) cost reimbursements to utilities, that may dictate utility decisions in contractural arrangements and requirements for design, construction, and operation.

l The goal of this subelement is for the NRC to gain a better understanding and

! appreciation of SURA actions that may have an impact on nuclear plant quality and safety. .

The activities to be conducted under this subelement are as follows:

Review and analyze organizational and institutional interfaces and relationships among licensees, SURAs, and the NRC to develop an NRC

, understanding of the SURA actions, policies and decisions which may have an impact on nuclear power plant quality and safety.

^

Prepare a report on the study's findings and staff recommendations for

NRC action in the area of SURA impact on plant quality and safety. V((*

W t

i End Products The end products of these activities include the following:

a NUREG or other report on the impact of SURA decisions on plant safety, j including recommendations for NRC action and involvement with SURAs (FY86) rule changes or legislative proposals b3 sed on the above end products, if I

appropriate (FY86 and later) 3.2.6 Integration of Human Factors Work-in Management and Organization The Division of Human Factors Safety (DHFS) of the Office of Nuclear Reactor Regulation (NRR) has sponsored research in the area of management and organization, including projects to develop and understand correlations between certain plant performance characteristics and a plant's organizational structure. DHFS and QA concerns in the area of management, although from a different perspective, have a common focus on management, resources,

[ experience, and actions that affect plant safety and quality. The Office of i Inspection and Enforcement (IE) QA staff will maintain an active interface with the DHFS staff on DHFS management and organization projects. IE will ensure j that QA projects and studies do not duplicate the DHFS efforts, but supplement r them as appropriate. For several months, the IE staff and the DHFS and the

[ Office of Nuclear Regulatory Research (RES) staffs have been working together on management issues of mutual interest. The goal of this subelement is to l

ensure that the QA staff integrates the DHFS and RES management and organization i efforts into the QA projects and studies.

l i

! 3-15

The activities to be conducted under this subelement include the following:

Maintain active liaison with the DHFS and RES staffs on projects dealing with management issues.

Integrate DHFS project findings into QA projects and studies as appropriate.

Conduct a series of regional office visits jointly with personnel from the OHFS and'IE Operating Reactors Program Branch to discuss management issues and to provide a firsthand basis for responding to regional office needs in the areas of management and organization.

End Products The end. products of these activities include the following:

incorporation of human factors work in management and organization into IE inspection modules (to be determined) revised Performance Appraisal Team (PAT) inspection procedure to incorporate human factors management and organization considerations, if appropriate (to be determined) revised SALP procedure for management and organization, based on human factors management and organization considerations, if appropriate (FY86) revised SRP elements for management and organization for applicants of operating licenses based on human fact g management and organization considerations, if appropriate (FY85).

3.2.7 Third-Party Audits (b)

The Ford Amendment to the NRC Appropriations Bill for FY82/FY83 directed the NRC to develop and test the concept of third party audits. The QA study concluded that third party audits could be a useful adjunct to NRC programs and recommended proceeding with a third party audit rule. The report also concluded that other opportunities may exist to achieve the objectives of third party audits: improved management of nuclear power plant design and (a) This is an NRR end product appearing in NUREG-0985, Revision 1, U.S. Nuclear Regulatory Commission Human Factors Program Plan.

(D) The request by Commissioner Bernthal (see SECY memorandum dated October 19, 1984, p. B-3) to " continue to perform a careful long-term analysis of the concept of third party audits that could be applied should ever construction of a new generation of nuclear power plants begin" cannot be accommodated under the FY86 OMB budget mark.

3-16

construction, enhanced capability to present quality problems, and more timely detection of quality problems. Readiness reviews could provide a viable alternative with many of the benefits and few of the negative aspects'of third party audits. The NRC staff believes that if the QA program enhancements of.the readiness review described in this plan are successful, a third party audit rule will not be necessary.

3.3 NRC'S INSPECTION PROGRAM The NRC has not consistently detected quality problems and breakdowns in a timely manner at some nuclear power plants, nor has it always grasped in a timely manner the significance of such problems. This program element includes actions to place greater emphasis on management issues in the inspection program, to provide better mechanisms (such as team inspection efforts) to aggregate inspection information into more timely diagnoses and detection of quality problems, to improve NRC oversight of its own inspection program (QA of the NRC inspection program), to make better use of available inspection resources, to improve inspector training, and to provide increased inspection coverage of design.

The goal of this program element is to provide increased assurance that NRC inspection programs will help prevent major quality-related problems or will assist in their timely detection. This clement includes the following subgoals: that NRC's inspection programs are clearly defined and prioritized to cover the most important activities from a safety perspective and are staffed with inspectors properly trained for the inspection functions they perform; that the written inspection program is effectively implemented in a timely fashion; and that identified deficiencies are promptly corrected.

Improved oversight of NRC's inspection activities will provide greater confidence in the quality of the licensees' actions.

3-17

The benefits which would result from accomplishing this goal are as follows:

The NRC inspection program would strike a better balance between examining QA program details and examining the quality of licensee program implementation (performance).

Greater assurance would be provided that NRC inspection efforts are directed to those areas of greatest safety significance.

Assurance would be provided that NRC inspection findings are coordinated and integrated in a timely manner to assure correct and timely analysis and to identify problem areas.

Assurance would be provided that NRC inspection programs are being implemented as planned and that resources are being used efficiently.

Greater assurance would be provided that major quality-related problems do not remain undetected for extended periodssof time by both the licensee and the NRC.

The issues associated with this program element include the following:

Significant program development effort has been expended over the past two 2 years to reorient the inspection program t, assure priority coverage of those areas of greatest safety significance. Ongoing efforts to apply risk analysis techniques to this effort are needed.

This QA program element is addressed by seven subelements in Sections 3.3.1 through 3.3.7 below.

3.3.1 Enhanced Management Review in SALP The QA report identified the involvement, capabilities, and prior nuclear experience of utility management as key factors in the quality success of a nuclear project. The goal of this subelement is to improve NRC's capability to avoid the need to improve management after a problem has occurred. This would 1

be done by enhancing the existing SALP process to help detect and correct the

management shortcomings that may lead to serious quality problems before they j fully develop. Among other factors, the performance of utility management is periodically assessed in the SALP program.

Activities under this subelement will focus on examining and testing alterna-

. tive approaches to assessing the effectiveness of licensee management controls that can be integrated into the SALP process. One alternative could be to establish special evaluations or inspections to evaluate management and management effectiveness (e.g.,'a Management Aopraisal Team evaluation).

Another alternative might be to further develop and refine incident evaluation-or the management performance indicators (developed in the QA study and the DHFS/RES research) to assist in the SALP evaluations. Another approach for i

consideration is to incorporate selected information developed as part of the enhanced trend analysis program (see Section 3.1) into the SALP process.

3-18 i

a f

i J

The activities to be condui:ted under this subelement include the following:

• Develop alternative approaches to improving NRC's ability to monitor and evaluate the effectiveness of licensee management control systems.

Determine how information developed from these various alternative approaches can be integrated-into the SALP program.

Conduct pilot tests of the most promising alternative approaches at sites in two or more NRC regions to test-the feasibility and effectiveness of the various alternatives.

Based on test results and on NRC regional and licensee experience with the approaches, revise the SALP program to accommodate the best alternative approach (es) and prepare new guidance to implement these revisions. '

End Products

~

The end products of these activities include the following: .

1 *

a revised SALP program that more directly addresses management issues, '

t f4enagement control systems, and management for quality (FY86) revfjed and'new IE inspection procedures to provide input to the SALP process to support the increased emphasis on management issues (F'r86).

3.3.2 Kist Report Follow-Up/Other Improvements to NRC QA Program (a)

' The Kist Report, an independent management audit and appraisal of NRC's QA program conducted as part of the QA study,.contains 19 recommendations for NRC action-to improve the assurance of quality of both licensee and NRC QA programs.

Some of the Kist rec.ommendations address actions that are presently under way

-by the NRC. Others will require the NRC to review and evaluate their merits and determine the appropriate actions. The goal of this subelement is to

! as;ure that all of the Kist Report recommendations for improving NRC's quality assurance programs are adequately addressed and that appropr.iate NRC action is taken. Further independent analyses of the NRC QA program are also planned.

i The followir@ activity will be conducted under this subelement:

i Systematically review, analyze, and address the Kist Report's 19 1 recommendations. l l

j End Produch The end products of this activity include the following:

• to the Director of IE, a report which explains how each of the Kist recommendations has been addressed, and actions planned as a result of the i Kist recommendations. For any recommendations that are not adopted, the report will explain why they were not adopted (FY85)

' further independent analyses of the NRC QA program, such as the Kist Report (on a continuing, periodic basis).

(a) -Because of budget considerations, minimized effort is planned in this area.

3-19

. . ~

3.3.3 Problem Plant Identification Given limited NRC resources, the NRC must assure that its inspection program is directed to the areas of greatest safety concern and to those areas which experience and analysis show to be the most important on either a generic or plant-specific basis. To achieve this goal, NRC inspection programs might be able to make greater use of structured, systematic safety and reliability analyses, such as probabilistic risk assessment (PRA) or systems assurance analysis (SAA), trend analysis, lessons -learned from inspection experience, the QA study, and special investigations to identify facilities or areas that should receive priority attention. Cost-effective improvements that may develop from the insights gained from ongoing and planned safety, reliability, PRA, and trend analyses, as well as information from other sources such as nuclear insurers, should be used to help focus and direct these efforts. The goal of this subelement is to assure that limited NRC inspection resources are allocated and prioritized with assistance from appropriate analytic techniques which prioritize activities according to their importance to public health and safety. Allocations could be either to potential problem plants or to specific systems that should receive increased inspection attention within all plants.

The following activities will be conducted under this subelement:

Develop, test, and refine appropriate analytical techniques to help allocate and prioritize inspection activities. Conduct pilot tests of these techniques, if appropriate.

Revise the inspection program and inspection procedures, as appropriate, to implement changes.to the allocation and prioritization of inspection efforts.

End Product The end product of these activities will be:

analytical tools for use by the regions and IE headquarters to assist in selecting plants or safety concerns for special team inspections or other inspection efforts (FY86 and later).

3.3.4 Increased Emphasis on Team Inspections / Contractor Support / Resident Inspectors One of the conclusions of the QA study was that multidisciplinary team inspections offer significantly more detection and diagnostic capability than inspections by only one or two inspection specialists. Given limited NRC personnel and the need to provide greater assurance through NRC inspection of the quality of plant design, construction, and operation, the NRC must expTore alternatives to improve and optimize the use of current resources and to expand those rese rces through contractual support for special expertise and peak workloads. The goal of this subelement is to re-examine NRC's current mix of regionally based inspections, resident inspections, regional team inspections, and headquarters inspections 1) to identify areas for further emphasis, such as increased use of regional team inspections or more resident inspectors, and 2) to determine how inspection staff and inspection efforts can best be augmented by contractual support.

3-20

The activities to be conducted under this subelement include the following:

Evaluate the NRC pilot programs on the expanded use of regional team inspections and selectively increased resident coverage.

Increase the use of contractor support on a trial basis, especially in areas requiring special expertise.

Based on analyses and pilot tests,. recommend changes to better optimize the mix of NRC inspections, including the increased use of contractors.

End Products ,

The end products of these activities include the following: ,

an IE decision on increased use of team inspections (FY85) an IE decision on relocating more regionally based inspectors to plant sites (FY85) an IE policy statement and guidelines on using contractors to assist.in NRC inspections (FY86) revised inspection procedures for regional team inspections to incorporate the lessons learned from the PAT, CAT, IDI, special team inspections, and the QA study (FY87).

3.3.5 Improved QA Training for Inspectors and NRC Staff

, The quality of NRC inspection program implementation not only depends on the

( quality, completeness, and depth of the written inspection program and the competence and experience of the inspectors, but also on the inspectors' full understanding of the philosophy, objectives and requirements of the inspecti program. This is true for all aspects of the inspection program, including

! reactor engineering, health physics, and quality assurance. The NRC must ensure that both new and experienced inspectors receive sufficient training on current inspection philosophy and procedures (including the new or modified procedures developed from other subelements of this program plan). Such training is needed to assure thet the pregram is understood and can be implemented effectively. Moreover, the NRC must assure that the inspector l

training curriculum remains current with the changes in policy and philosophy in the inspection program, such as the changes discussed in Sections 3.3.3 and l 3.3.4. The goal of this subelement is to assure that NRC's training program for QA matches the technical, policy, and programmatic needs of NRC inspectors, i the NRC inspection program and other NRC programs.

The activities to_be conducted under this subelement include the following:

Survey QA training needs and desires of NRC inspectors, IE staff, and the staff of other offices.

• Revise NRC training practices to improve and strengthen inspector training in areas of inspection program implementation and QA inspection.

l 3-21

Monitor and update inspector QA training to ensure that it provides adequate coverage of NRC requirements and implementing guidance, as well as of NRC QA goals, objectives, policy and philosophy.

End Products The end products of these above activities include the following:

modification of existing QA training courses to more directly address the technical, programmatic and policy needs of NRC inspectors (FYS5 and continually thereafter) development of new QA training courses (or modification of existing training courses) tailored to meet the needs of a broader NRC audience, including managers (FY86, FY87) increased patticipation of knowledgeable NRC managers, inspectors and outsioe experts in teaching NRC QA courses (FY85 and continually thereafter). -

3.3.6 Reduction of Inspection Burden on Licensees and Vendors (a)

Several organizations have established requirements or performance objectives' for quality and quality assurance in the nuclear industry. Bodies responsible for the requirements often audit or inspect to verify that their requirements are actually met. Several organizetions concerned with nuclear safety have inspection or audit programs focusing on the licensee's compliance with a certain set of requirements. During the year,.a nuclear power plant licensee or major vendor will likely undergo several inspections, audits, or assessments by the NRC, INPO, the ASME Code system [i.e., the American Society of Mechanical Engineers (ASME), Authorized Nuclear Inspection Agencies], and possibly third parties (hired by the licensee with or without NRC's encouragement). Or, in the case of a vendor, it may undergo those inspections, audits, or assessments by several licensees. During one year, some organizations performing nuclear grade work may undergo as many as 50 separate inspections and audits, many by d_ifferent bodies. It is not uncommon for a licensee or vendor to undergo more than one QA inspection at the same time by different groups. Moreover, the areas inspected by the different groups overlap and the inspections are often duplicative.

The 11censee'.s QA organization is particularly impacted by these audits and inspections. This overlap results in an inefficient use of total inspection resources by the inspection groups and a diversion of licensee or vendor resources away from their safety functions.

The goal of this subelement is to establish mechanisms to coordinate NRC activities and industry inspection activities to the extent appropriate. The purpose of this coordination is to minimize the burden on licensee and vendor organizations and the possible adverse impact on attaining quality. Some initial, tentative steps have been taken in this direction. Over the past several years the NRC, the ASME, and the National Boaro of Boiler and Pressure Vessel Inspectors (the National Board) have exchanged a series of letters and have discussed the feasibility of the NRC accepting the ASME/ National Board system of nuclear accreditation instead of other duplicative inspections.

(a) Because of budget considerations, no work is planned in this area.

3-22

In February 1984, the ASME sponsored a forum to estaolish better communications within the inspection community. The scope of the forum was not limited to the ASME Code but addressed the need of all inspection parties. The NRC plans to continue participation in those efforts and to extend all possible cooperation toward achieving the goal.

The activities to be conducted under this subelement include the following:

Perform a comprehensive analysis of the inspection burden on licensees and vendors, and assess the costs associated with overlapping or duplicative inspections.

Perform a comprehensive analysis of the QA requirements, inspection scope and depth, inspector qualificatirns, and inspection practices of different inspection groups to characterize differences, similarities, and areas in which one group's inspections may be accepted instead of another's without compromising the legal or statutory charters of the organizations.

Continue discussions and coordination with the ASME, the National Board, and others to develop approaches to reducing inspection duplication.

End Products The end products of these activities include the following:

an arrannement wherein the NRC could accept ASME Code system inspections in certain clearly delineated areas in place of NRC inspections, subject to certain provisions for NRC oversight and qualifications of the ASME Code system of inspection programs and inspectors (FY86) formal recognition by the NRC of the inspection function performed by l

certain qualified industry groups and the granting of credit to a licensee for vendor inspections performed by such groups instead of direct inspection by the licensee (FY86).

3.3.7 Integrated Design Inspections and Independent Design Verification Program After several significant design deficiencies and design quality program breakdowns were identified, the NRC initiated two new programs to increase the NRC's level of assurance that plants have been designed correctly and that the quality programs for design have worked correctly. One new initiative is the IDI, which is an NRC-conducted inspection of' selected aspects of the plant l

design. The second initiative is the IDVP, under which a licensee-sponsored audit of selected aspects of the design process is conducted by a third party, l independent of the licensee and the original designer. Historically, the NRC

has not expended significant resources in the area of design, because it I assumed that design errors would generally be detected during construction and l pre-operational testing. Recent discovery of design or design quality program errors and problems at a late stage in construction has resulted in expensive modifications to some plants, delays in plant completion, and reduced public confidence in the overall ability of the NRC and its licensees to properly discharge their responsibilities. The goal of this subelement is to assure that the quality of design activities is verified at the earliest practical time.

3-23

The activities to be conducted under this subelement include the following:

Conduct IDIs at plants as early in the construction process as possible, consistent with other programmatic considerations, including past plant, utility, A/E, constructor and NSSS vendor performance, proximity to near-term operating license (NTOL) status, and results of other inspection activities. Currently, three IDIs per year are scheduled.

, Continue case-by-case to encourage WTOL licensees to provide additional assurance that the design process has implemented all regulatory requirements and licensing commitments. Review and critique IDVPs (carried out by licensees' contractors) or other licensee means of providing additional assurance. Prepare appropriate input to Safety Evaluation Reports (SERs).

End Products The end products of these activities include the following:

IDI inspections at a frequency of three per year (FY85,.FY86, FY87)

IDVP reviews on a case-by-case basis for plants for which the licensee has used this method to provide added design assurance (in conjunction with NT0L licensing) reviews of other licensee means of providing assurance of design quality (as required).

3.4 NRC PARTICIPATION IN THE STANDARDS DEVELOPMENT PROCESS One of the problem areas identified in the QA report was the proliferation of various industry QA standards and the NRC practice of adopting some parts of industry QA standards as acceptable practices and modifying other parts of standards to meet NRC requirements. Different issuance dates of the standards and varying dates of licensee commitments to implement those standards in part or in whole have resulted in different QA/QC standards, requirements, and practices in the nuclear industry.

The number of different requirements has resulted in a lack of uniformity in QA programs from site to site, and in some cases, among different contractors at the same site. It has also resulted in confusion over exactly what require-ments do apply at a given site. One key to the construction quality success of-St. Lucie 2 was the licensee's decision to allow only one QA program (their own), with only one set of QA requirements, and one kind of QA program documentation project-wide. This uniform project-wide QA program gave greater assurance of program implementation and eliminated QA program differences as an interface issue among contractors on the project.

A related problem is that in implementing licensee commitments to various standards, standards provisions sometimes are interpreted far more literally than was intended by the consensus group who wrote the standard. Standards writers cannot possibly anticipate every situation or question that may arise in applying a standard, but they can indicate in the wording of the document that its use should be accompanied by judicious use of flexibility and common sense. The key to making these choices lies in undarstanding the purpose of the document and the overall objectives of the QA program and its component parts.

3-24

A goal of this subelement is to achieve more uniformity in the use and content of various industry quality standards, better statement of QA program goals and the purposes of prescribed QA measures in the standards themselves,'and more flexibility in interpreting and applying the standards. Another goal of this subelement is to improve the availability, quality and applicability of QA guidance provided to the industry and to ensure active, effective NRC partici-pation in standards development activities. To effectively implement this plan, NRC QA resources must be used in such a manner that the NRC QA staff actively participates in QA standards development activities.

The benefits which would result from accomplishing the above goals include the following:

clearer and easier-to-apply guidance which will facilitate QA program implementation and achievement of goals better definition of the scope of QA program applicability, resulting in better identification of those areas which need QA guidance, more germane guidance in areas needing guidance, more uniformity of programs and better focusing of resources more timely guidance and more timely NRC endorsement of consensus standards more active NRC commitment of resources to and participation in standards development activities greater use of standards activities to resolve differences between NRC.

staff and industry (as opposed to taking exceptions through regulatory positions).

The issues associated with this program element include the following:

development of useful guidance that helps industry achieve QA program goals and that contains an optimum balance between flexibility and prescriptiveness confusion which may result from changes to existing QA programs effects of backfit considerations and decisions which may result from changes to existing standards difficulty in developing guidance for performance-oriented QA programs the scope of QA program coverage This QA program element is addressed by five subelements in Sections 3.4.1 through 3.4.5 below.

3.4.1 Regulatory Guidance on QA P ograms and Activities: Endorsement of ANS-3.2, NQA-1 and NQA-2 Revisions to present national standards are an ongoing activity. As changes are made, the regulatory process incorporates the improved QA standards and 3-25

program guidance through en'orsement d of the revised standards. The NRC will evaluate and endorse newly developed standards, as appropriate. Additionally, guidance will be generated through the consensus standard development process for newly identified areas of QA program applicability (for example, computer software). Regulatory Guide 1.33, " Quality Assurance Program Requirements (Operation)," through endorsement of ANSI /ANS-3.2, is NRC's mechanism for describing an acceptable QA program for operating reactors. As ANSI /ANS-3.2 is revised and improved, it will be reviewed by NRC and, if acceptable, will be endorsed by Regulatory Guide 1.33. Possible exceptions would be made to cover Commission policy, recent agreement, or simply disagreement between NRC and the consensus body. Presently, Revision 3 to Regulatory Guide 1.33 is undergoing staff review. A major policy consideration is whether the regulatory guide, endorsing ANSI /ANS-3.2, should be applied to existing operating reactor QA programs. ANSI /ANS-3.2 contains a number of post-TMI improvements that most licensees have adopted but to which they were not committed under ANSI 18.7, the predecessor to ANSI /ANS-3.2. The current version of Regulatory Guide 1.33, Revision 2, reflects the nuclear industry QA standards of nearly a decade _ago.

Similarly, Regulatory Guide 1.28, " Quality Assurance Program Requirements (Design and Construction)," is NRC's mechanism for describing an acceptable QA program for reactors in design and under construction. The latest proposed revision to that guide endorses the design and construction applications of ANSI /ASME NQA-1 (1983 edition), which in turn consolidates a number of older N45.2-series programmatic QA requirements standards for nuclear facilities.

Under development is a third regulatory guide that will endorse ANSI /ASME NQA-2, which consolidates a number of additional, older N45.2-series technical QA requirements standards for nuclear power plants. ANSI /ASME NQA-1 and ANSI /ASME NQA-2 were each developed to consolidate, in one document, the bodies of QA program and technical requirements, respectively. Their endorsement by NRC will constitute major steps toward eliminating duplication and confusion resulting from the current multiplicity of standards for nuclear power plant QA programs.

General activities associated with this subelement have been. summarized above.

Specific activities to be conducted include the following:

Endorse ANSI /ANS-3.2 through Revision 3 to Regulatory Guide 1.33.

Establish the costs and benefits associated with applying the provisions of ANSI /ANS-3.2-1982 to currently operating plants.

Develop a new regulatory guide to endorse ANSI /ASME NQA-2-1983. This new regulatory guide will consolidate several older regulatory guides into one.

Develop appropriate guidance on QA program requirements for the pre-operational and start-up phases of nuclear power plants.

End Products The end products of these activities include the following:

issuance of Revision 3 to Regulatory Guide 1.28, endorsing ANSI /ASME NQA-1-1983, with minimal NRC exceptions to the standard (FY85) 3-26

amendment of Appendix B to 10 CFR Part 50 to state, with the exception of the regulatory positions identified in Revision 3 to Regulatory Guide 1.28, that commitment to NQA-1 is an acceptable way for licensees to meet Appendix B QA requirements (FY87) issuance of Revision 3 to Regulatory Guide 1.33, endorsing ANSI /ANS-3.2-1982, with minimal NRC exceptions to the standard (FY86) a decision on the issue of backfitting Revision 3 to Regulatory Guide 1.33 (ANSI /ANS-3.2-1982) (FY86) issuance of a new Regulatory Guide 1.XX endorsing ANSI /ASME NQA-2-1983, with minimal NRC exceptions to the standard (FY87) development of guidance on QA program requirements for pre-operational testing and start up phases (FY87).

3.4.2 Computer Software QA for Nuclear Applications Quality assurance requirements for developing and using computer software have been under consideration for several years. An analysis of events reported in NRC Regional Office Daily Reports, inspection reports, and Licensee Event Reports indicates that operational problems can be directly attributed to quality problems in software (purchased or self generated) written for application in nuclear power plant design, construction, and operation. In many cases, the software has been written by a vendor or supplier and sold to a higher-tier supplier or utility for use in plant activities having direct safety implications. The' goal of this subelement is to identify areas where controls for developing and using computer software for nuclear applications should be considered and to provide guidance on appropriate methods of control.

Initial efforts will be to examine the feasibility and benefits of adopting existing software quality assurance (SQA) standards that pertain to non-nuclear work for the nuclear industry and to establish a data base to help define the scope and extent of SQA problems in the nuclear industry. Based on l the results of these efforts, alternati.ve approaches that address the identified problems will be developed and evaluated.

The activities to be conducted under this subelement include the following:

Identify software applications in nuclear power plants, fuel facilities, and waste repositories, and identify where deficiencies in these applications have occurred.

Determine the existing regulatory base for achieving and assuring quality and safety in computer software for nuclear applications.

Determine the extent to which the existing regulatory base and existing standards are adequate to cover both the requirements for SQA and the identified problems or potential problems related to nuclear computer software.

3-27

End Products The end products of these activities include the following:

• endorsement of IEEE 730 or some other appropriate SQA standard, with

- minimal exceptions, for application to software used for nuclear applications (FY87) d'evelopment of inspection procedures fnr software QA inspections (FY87).

3.4.3 Improved Q-List A licensee or applicant must identify the safety-related structures, systems, and components to which it will apply a QA program that meets the requirements of Appendix B to 10 CFR Part 50. This list is called a Q-list. The QA program applied to the plant's safety-related items is often quite different from and more stringent than the QA program applied to other items. The anticipated transient without scram (ATWS) event at Salem in 1983 highlighted the signifi-cance of omitting safety-related items from the Q-list ana the resulting lack of quality, reliability, and maintenance attention paid an item not on that list.

A cost is associated with applying the Appendix B-level QA program to any specific component. The decision to include an item on a Q-list can be an expensive decision. Engineering judgments regarding whether a particular item should be included on the Q-list must be made by the licensee or its contractors because it is not always clear whether an item is safety-related. Currently, there is no uniform working-level guidance which provides a logical foundation for applying quality measures to items commensurate with their relative importance or for deciding whether an item should or should not be on the Q-list. The guidance provided in 10 CFR Part 100 and Regulatory Guide 1.26 is very general.

The goal of this subelement is to provide improved guidance for licensees on placing items on their Q-lists. A related goal is the development of a logical and cost-effective approach for establishing and applying quality requirements for nuclear power plant structures, systems, and components.

The activities to be conducted under this subelement include the following:

Conduct a pilot program of the NASA (NRC QA-sponsored research) methodology of system assurance analysis, which addresses reliability and safety analysis, and determine if it can be'successfully applied to a nuclear power plant. The NASA approach is predominantly a deterministic analysis.

Conduct a pilot program of the Sandia National Laboratory methodology (NRC QA-sponsored research) of using PRA to identify and provide a relative ranking of nuclear power plant equipment, components, and systems. The Sandia approach is a quantitative, numerical analysis.

• Based on the results of the pilot programs, determine if it is possible to develop a systematic, comprehensive method of identifying nuclear power plant equipment and activities which pose a potential safety risk and rank the equipment and activities according to that risk.

3-28

,. -,.<-.n-.,,,;

, ---y -,.,---:,,,.---

, , - ~ , , , ,- n ,_, , - , - , - , - - - - - -

If'the results indicate the possibility of cost-effective improvements over current methods, prepare guidance for NRC and licensee implementation on developing and preparing improved Q-lists.

End Products The end products of these activities include the following:

NUREGs or other reports describing -the pilot programs and their results, including how the pilot program results relate to the current NRC and industry decisional practices (FY86) a report or a guidance document describing approaches to Q-list classifi-cation of items and cost-effective application of quality measures to plant structures, systems and components (FY87).

3.4.4 Alter' native QA Requirements for Subtier Suppliers The demand for nuclear items (equipment, components) and services has changed considerably over the past several years. With-the decline in applications for new construction, as well as the cancellation and/or postponement of many other projects, the final customers for the many nuclear vendors / suppliers are limited to the approximately 85 nuclear plants currently operating and the approximately 30 plants wh~ere construction is actively under way. Recent NRC studies in support of the QA study have shown that "special" nuclear quality assurance requirements tend to cause items or services to be more costly than their commercial grade counterparts. Some vendors / suppliers view the require-ments as posing an increased paperwork burden while actually contributing little toward enhancing the quality of the items or services provided.

For most vendors / suppliers, nuclear-related work comprises a very low percent-age of their business. Complying with nuclear QA requirements is costly in their opinion, and maintaining the capability to do work for the nuclear industry when their percentage of nuclear business is very small can price them out of other markets and is often not perceived as being worth the investment.

Competition declines in the environment where fewer businesses do nuclear work, and licensees are impacted by losing leverage in negotiating for reasonable prices dnd/or lead time on orders. Such an environment has the potantial to result in poor quality work and to impact plant safety. .

As the large number of nuclear plant cancellations occurred, an increasing number of vendors found it too costly to maintain their ANSI /ASME N45.2 quality assurance programs and therefore are no longer able to supply, safety-related components. A few vendors retained their ANSI /ASME N45.2 programs. and because competition in this ' area is now limited, the vendors- demand a hign.p.remium for their nuclear grade products.

Currently, many vendors are supplying the military with high-reliability components under military standard MIL-Q-9858A. If components procured under MIL-Q-9858A quality program requirements were determined to be equivalent to these procured under ANSI /ASME N45.2 requirements, nuclear utilities would be able to' procure standard design components which satisfy nuclear quality requirements. The nuclear utilities could realize significant savings, particularly when needing large numbers of single procurement maintenance or repair items.

3-29

In addition to the quality program requirements, compliance with 10 CFR Part 21 and 10 CFR Part 50.55(e) is a problem for component vendors and their subvendors.

In many cases component vendors, and particularly subvendors, are not in a position to understand the safety significance of deficiencies or discrepancies which are discovered with the products they manufacture or purchase. This situation has also contributed to the problem of competent and reputable vendors abandoning the nuclear. marketplace.

~

The goals of this subelement are to assure that suppliers of. nuclear components have quality programs consistent with maintaining public health and safety, and to reduce incentives of suppliers / vendors to leave the nuclear business as a result of NRC regulatory requirements.

The activities to be conducted under this subelement include the-following:

Compare the requirements of ANSI /ASME N45.2, NQA-1 and non-nuclear quality or inspection program documents such as MIL-Q-9858A and MIL-I-45208A to identify differences in the standards and the resulting impact on product quality.

Determine a hierarchy which will provide for only that level of quality program actually necessary at any particular procurement level (e.g.,

owner, A/E, NSSS vendor, component vendor, subvendo', r etc.).

Evaluate the impact of 10 CFR Part 21 and 10 CFR Part 50.55(e) and define the best level in the procurement hierarchy for implementing the requirements of these regulations.

End Products The end products of these activities include the following:

a NUREG or other report on the results of the above activities (FY86) a regulatory position on the acceptability of non-Appendix B QA programs by certain suppliers or vendors (FY86) a regulatory position or legislative change, if necessary, on the minimum level in the owner-A/E-NSSS supplier-component vendor-subtier vendor chain at which Part 21 and 50.55(e) determinations should be made (FY86) a rule change, if appropriate, (to 10 CFR 50.XX) permitting the use of certain non-Appendix B QA programs by certain classes of vendors and suppliers, and modified Part 21 and 50.55(e) requirements for these classes, subject to certain audit safeguards (FY87) a guidance document for implementing the rule change (s) (FY87) revised vendor inspection procedures for non-Appendix B QA programs (FY87).

3-30

4 i .

j 3.4.5 NRC Participation in Consensus Standards Committee Activities The NRC will intensify.its involvement with the national consensus standards

committees that write quality or-QA-related guidance and will provide active i input to the developing standards. Active, cooperative participation in these activities by both the-NRC and industry is needed _ot give the standards development.

process the full benefit of NRC staff opinion. NRC's participation will-also help to minimize deviation among consensus standards and their NRC QA program-required counterparts (e.g., the regulatory guides which endorse them).

To improve the applicability and usefulness of QA standards and the clarity and timeliness of their NRC endorsement, the following activities will be performed:

• ' Commit NRC resources necessary to adequately support the QA standards-making activity. -

• Ensure active NRC QA staff participation in all QA and related national standards development committees.

Actively participate on committees responsible for creating new QA-related standards to ensure that developed standards represent true industry and NRC consensus and that the standard can be endorsed with a minimal number of exceptions.

End Products The end products of these attivities include the following:

• the development of consensus QA standards requiring a minimal number of regulatory positions for endorsement (ongoing)

• NRC endorsement of consensus standards in a timely fashion (ongoing).

3. 5 PERSONNEL ISSUES IN ACHIEVING AND ASSURING QUALITY Inadequate qualifications of some personnel working at nuclear plants have been noted as a contributing factor to quality-related problems in many NRC investigations and inspections. Because the ultimate responsibility for performing and assessing high quality work rests with the licensee, ensuring that the qualifications both of personnel performing the work and of those inspecting it meet established standards is of great importance to the licensee.

Unless quality can be independently confirmed, work done or inspected by inadequately qualified personnel cannot be presumed to meet established

' acceptance criteria and may constitute grounds for failure to issue a license.

A goal of this program element is to characterize and focus the various levels of quality responsibility to ensure that personnel performing activities affecting quality have qualifications commensurate with the performance requirements of tiieir assigned jobs. A second goal is to try to facilitate management's instilling into their workers of a sense of personal responsibility for the quality of their work and establishing greater accountability for the work performed.

3-31

The benefits which would result from achieving these goals include the following:

• reduced risk to the health and safety of the public by assuring that quality-related activities are performed only by adequately qualified .

personnel  ;

increased NRC a 1 public confidence both.in the qualifications of personnel performing act h 8 ties affecting quality and in the quality of work performed -

reduction of the complexity of and confusion in interpretations of current requirements for maintaining the proficiency of personnel performing activities that affect quality improved selection and training programs for QA personnel increased quality awareness / consciousness of and clear responsibility for quality of work, which will result in high quality work and will enhance the chances of overall nuclear project success.

The issues associated with this program element include the following:

the impact on safety that improvements in this area will achieve versus maintaining the status quo the relationship of nuclear power plant practices and NRC QA requirements to personnel qualifications and quality accountability issues the role of licensees, vendors, and their suppliers and the NRC in personnel qualifications issues extent of NRC oversight and/or involvement in a licensee's personnel qualification issues and in those of the licensee's contractors and vendors.

This QA program element is addressed by two subelements - qualification /

certification of QA/QC personnel, and line organization responsibilities for achieving and assuring quality -- discussed in Sections 3.5.1 and 3.5.2 below.

3.5.1 Qualification / Certification of QA/QC Personnel In 1980, the NRC recognized the need for upgrading the qualifications and training of industry QA and QC personnel. In at least two nuclear projects (Marble Hill, South Texas), inadequate qualifications of QA/QC personnel contributed to quality problems. SECY-82-352, " Assurance of Quality," presented a number of staff initiatives designed to improve the assurance of quality in the design and construction of nuclear projects. One of the initiatives,

" Qualification and Certification of QA/QC Personnel," was initiated 1) to direct more attention to NRC enforcement of existing standards for the qualifications of QA/QC personnel, 2) to work with the industry in developing improved qualification standards, and 3) to consider the feasibility of formal certification of QA/QC personnel by the NRC or other appropriate independent parties. Staff efforts since SECY-82-352 was issued have identified the need 3-32

to clarify NRC requirements for maintaining the proficiency of personnel performing activities affecting quality (10 CFR Part 50 Appendix B, Criterion II). Staff efforts have also identified the need to better understand and characterize qualification issues for QA/QC personnel during operation as well as during design and construction of nuclear projects.

INPO has developed a number of performance-based guidelines for the training of personnel holding various positions at operating plants. .These guidelines a s currently available to member utilities -for self-evaluation of their training programs. In addition, INP0 uses these guidelines in their efforts to assess and accredit utility training programs.

Although this INP0-managed accreditation program presently excludes QA/QC personnel, it is possible that they will be included sometime in the future b(cause of the present interest in this issue. The NRC has endorsed the IM0-managed program as a way of meeting the proposed rulemaking for training and qualifications of personnel at operating nuclear power plants in response to Section 306 of the Nuclear Waste Policy Act (NWPA). INP0's accreditation program and implementation of NRC's "306 Package" will be closely followed during staff actions under this subelement to determine its applicability to the objectives of those studies.

The activities to be conducted under this subelement include the following:

Identify the extent and magnitude of deficiencies in the nuclear industry in meeting required qualifications for personnel, especially QA/QC personnel performing activities affecting quality, and determine root causes'of the problems.

• Evaluate industry programs (including INP0's performance-based guidelines for training and qualification) for maintaining the proficiency of personnel performing activities affecting quality; determine the adequacy of existing industry standards and/or NRC requirements which govern those programs; and develop and implement solutions where inadequacies are apparent.

(Implementation may occur in the form of revised industry standards and/or revised licensee programs for qualification and certification of QA/QC personnel.)

• Develop and implement methods for improving NRC inspection and enforcement of personnel qualifications requirements (new or revised inspection procedures, or changes to enforcement policy).

End Products The end products of these activities include the following:

• NUREG or other report describing the results of the analysis of the QA/QC qualification issue (FY86)

• a policy statement or regulatory position on training, qualification and certification of QA/QC personnel (FY87) 3-33

development / endorsement of new and revised consensus standards on qualifi-cations of QA/QC personnel, if appropriate (as such standards are developed).

Line Organization Responsibilities for Achieving and Assuring Quality (a)

~

j 3.5.2 SECY-82-352 included an initiative entitled " Craftsmanship," the objective of

! which was to explore the potential methods and incentives for enhancing the craft worker's role in achieving and assuring quality. Staff discussions with representatives of craft labor unions and information gained from site visits made as part of the QA study revealed the following points:

j.

• Crafts typically are not well informed of their role in the QA/QC process; 4 utilities and contractors often have not provided adequate training to crafts regarding quality. ,

* ' Repeated rework as a result of field changes has a demoralizing effect on a crafts and has an impact on the quality of the final work. Improved front-i' end engineering and procurement could reduce the amount of change and rework.

i Some utilities are not convinced that QA/QC is cost-effective; labor

, perceives that utilities regard QA/QC as a high-cost item rather than a cost-saving tool.

The NRC recog'nizes that craftsmanship plays an important role in achieving quality. Pride in one's work, combined with clear responsibility and direct accountability for performance, should provide the incentive necessary for high quality. The idea of directing responsibility and accountability back to the doer applies not only to craft workers but also to inspectors, engineers, i

, supervisors, and -- most importantly -- management. On this aspect, the case i

studies performed in support of the QA study concluded that, in the nuclear industry, management actions (involvement and commitment) have more effect on o craftsmanship and the quality of work than any other single factor. Quality is i

everyone's responsibility, but management must be involved and demonstrate visible support to the QA program, including staffing resources and personal involvement. Through this demonstration of support, craft workers know that top management expects work to be done "right the first time."

The QA studies of the quality programs of other agencies, industries, and forei.gn countries also recommended that the NRC re-examine its posture on QA to emphasize integrating QA into the overall management process of nuclear

projects. These studies suggested the need for NRC policies and programs to foster the belief that quality is a line responsibility rather than a responsi-bility that can or should~be delegated to the QA/QC organization at nuclear i projects. Quality cannot be " inspected in" at a project.

! The following activities will address the identification of responsibilities for

! quality and methods for directing these responsibilities toward greater assurance of quality:  ;

I (a) Because of budget considerations, no work is planned in this area.

l 1

i i 3-34 4

i

>r . - ~- - - - - - - .,n--n---r,. -,mm e n--, - vn,, , - - , , - -.--,g-,, -,-,,.-n -ew,. ,w,--,..nn., ~_-,,,a,, ,-.,,,n,, - ee,-,,,-,,,-r-,

. i Reexamine current NRC QA program guidance which has had the effect of isolating the QA/QC organization from line management and isolating line management from QA/QC concerns. In particular, the need for and effect of NRC-mandated (or the licensee's perception of NRC-mandated) organizational independence of the QA and QC organizations.

Propose changes and develop and analyze other alternatives to the NRC's QA philosophy, policy and practices to provide incentives for utility management to adopt policies and practices that programmatically implement the philosophies of " quality is everyone's responsibility" and "do it right the first time."

End Products The end products of these activities include the following:

a policy decision concerning changes to Chapter 17 of the SRP dealing with organizational independence (FY87) integration of the results of this work into the management and performance-oriented initiatives oeing pursued under other program elements (FY85, FY86).

3-35

4.0 REVISED QUALITY ASSURANCE PROGRAM FOR FUTURE PLANTS (a)

Most of th'e lessons learned from the QA study have been directed toward revising and improving NRC programs and requirements which would have current 2

applicability to plants in late stages of construction and in operation.

, This part of the implementation plan takes a long-range look at revising NRC

! programs for future nuclear power plants and for future issues affecting the quality of existing plants.

In addition to the program changes identified in Chapter 3.0, which would have

!. applicability to future plants (for example, readiness reviews, performance objectives, trend analysis, and configuration management), two additional QA efforts are planned to enhance the ability of the NRC to evaluate the

capabilities of a construction permit (CP) applicant before a CP is issued.

The first effort will be.to develop a NUREG or another suitable document to l assist the staff and the applicant in identifying project team management i characteristics which are fundamental to achieving an acceptable level of quality in design and construction. The second effort is to explore the possibility of the NRC establishing an advisory panel of experts, independent 4

of the NRC and the CP applicant, to prospectively review and assess the capabilities of the applicant's project team to manage the design and con-l struction of the proposed nuclear power plant. Both of these efforts respond i to one of the fundamental findings of the QA study: significant quality-

! related problems in design and construction were characterized by the inability i or failure of some utility managements to effectively implement a management system that ensured adequate control over all aspects of the project. A more

detailed discussion of the two QA efforts to address this finding are included I in Sections 4.1 and 4.2. The third element in this chapter addresses the need 1 to anticipate the quality problems of the nuclear industry in the future, i especially for operating plants that must purchase repair parts from a

, dwindling group of suppliers. That element may form the basis for long range

! planning capability for issues broader than QA.

4.1 CRITERIA FOR PROJECT TEAM QUALIFICATIONS, EXPERIENCE, AND CAPABILITIES A major finding in the QA report was the importance of prior nuclear construc-j tion experience for key members of the project team, including the owner's d

senior project management personnel. Some utilities experiencing significant i quality problems in design or construction operated under the assumption that j nuclear plant construction was not sufficiently different from fossil plant

! construction to warrant significant changes in their managerial approach. For

nuclear projects, some utilities believed incorrectly that with only minor

, modifications they could rely upon their existing organizational structures, staffing, contracting methods, and construction methods used previously for fossil construction.

(*) Note that due to the budget priority of operating plants, planned work that applies to future plants only is not being undertaken. In particular, the projects described in Sections 4.1 and 4.2 are being dropped.

j 4-1

The goal of this program element is to develop and disseminate guidance that will be useful to future CP applicants, the public, and the NRC in helping to avoid repetition of past design and construction quality problems. Based on experience and lessons learned, as discussed in the QA report, the guidance will include characteristics of successful project managsment arrangements and th'e qualifications, experience, and capabilities of successful project teams.

The project team includes licensee management (owner) and the management of the architect engineer, construction manager, and constructor. The guidance will allow the NRC, prospective plant owners,the public and the financiers of a nuclear power project to be more informed than in the past concerning the attributes of a proposed nuclear construction project team, and therefore, the likelihood of project success. Recently, the NRC re-instated the financial qualifications rule for CP applicants. This review of the applicant's financial resources should be an important element in NRC's judgment on the project's ability to be completed with acceptable quality.

The benefits which would result from accomplishing this goal are as follows:

increased assurance that future nuclear construction projects, once granted a CP, will be successfully completed and granted an operating license reduced likelihood that significant quality-related problems will occur in future nuclear construction projects as a result of project management team weaknesses and inabilities or financial limitations. .

The issues in this program element are as follows:

the difficulty in prospectively assessing the capability of a proposed project team, especially a new team assembled specifically for the project the difficulty in arriving at a meaningful set of guidelines that adequately accommodates differences in corporate organizations, philosophies, experience, and management approaches.

The following activity will be conducted under this program element:

Based on the lessons learned from the QA study, prepare a report describing the characteristics observed for successful nuclear power plant project teams (i.e., qualifications, experience and capabilities).

End Product The end product of these activities will be:

a NUREG or other report containing guidance on project team qualifications, experience, management capabilities and financial stamina to assist in the review of CP applications (FY86).

4.2 ADVISORY BOARD TO REVIEW CP APPLICANTS The QA report recommended that the NRC establish an advisory board to review prospective applicants for a CP and to recommend to the Commission the board's assessment of the management capabilities of the applicants. Advisory board 4-2

members would have the background, position, and experience appropriate to review the management qualifications, organization, experience, and capabilities of CP applicants. The board's review would supplement a prospective review by the NRC staff of the applicant's capabilities to manage the proposed nuclear power plant's design, construction, and subsequent operation.

The goal of this program element is to decide whether such a board should be established for future CP appi,1 cants. Before a decision is made, a cost / safety-benefit analysis of the advisory panel concept and alternatives to that concept will be performed. Public comments on the QA report were generally negative on establishing such a panel, and commenters suggested that existing bodies might be able to perform the same function. Action on this item is deferred pending the results of potential industry initiatives in this area.

4-3

1 4.3 QUALITY IMPLICAIIONS OF THE CHANGING UTILITY-CONTRACTOR-VENDOR-SUPPLIER INFRASTRUCTURE'"'

The QA study.provided insight into achieving quality objectives for purchased l items and services. Attaining quality through the procurement process depends heavily on careful evaluation of bidders, attention to detail, and well-documented requirements. Active, informed involvement of licensee management

in procurement activities (just as in design and construction activities) is i fundamental to achieving quality in procured items and services. Careful evaluation of the capabilities of bidders is essential before a contract is placed. Effective follow-up is necessary to assure the contractor's quality i performance.

i The QA study also indicated the need for further analysis of the quality and

, safety effects of the current decline in nationwide nuclear business. That

, decline is causing changes in the utility-contractor vendor-supplier infra-structure. The long-term implications of those changes and their impact on the

! procurement and quality of nuclear items and services are indeterminate at this time but may have significant implications for quality in the nuclear industry j in 2, 5 or 10 years. NP.C's policy decisions on quality issues should be based i on an understanding of the changing nature of the traditional U.S. nuclear industry.. Long-range quality program planning is necessary, and it should be i i' based on info'rmed judgments of the likely nature of the industry's future infrastructure. .

i i

Although the NRC is concerned that individual supplier-related quality problems affecting safety be identified and corrected, significant improvements in the

vendor / supplier area depend upon assuring that,the root causes of adverse quality trends related to procurement of nuclear items and services are identified, corrected, and prevented from recurring. As pointed out-in the QA i report, some of the quality problems experienced by the nuclear industry in the l late 1970s and early 1980s were due in part to the surge of reactor orders and r CP applications that occurred in the early and mid-1970s. This surge made it i difficult for some utlities to assemble project teams with key members that had

! the requisite prior nuclear design and construction experience. At this same l time, extraordinary demands were also placed on component suppliers and sub-l contractors as well as the NRC staff. The QA report recommended that further l analysis be performed to identify 1) the rapid-expansion-related problems that (a) Because this topic has application to existing plants as well as future plants, some work is planned in this area. However, work that applies only to a future generation of plants, e.g., constraints on resumption of orders, will not be undertaken.

I l

I k

I L

i 4-4 I w -+ee-w-w--=-_m

previously occurred, and 2) the constraints that should be placed on resump-tions of orders to prevent similar problems in the future, should there be a resumption of new orders. This analysis will be conducted as part of this program element.

The goal of this program element is to provide decision-makers in the nuclear industry with a carefully researched, dynamic information base on the trends (and their underlying causes) in the relationships between utilities and their suppliers and contractors to improve calicy decisions affecting quality in the nuclear industry in the future.

The benefit which will result from achieving this goal is as follows:

better informed decisions by the NRC and the industry on policy matters affecting quality and quality assurance.

The issues in this program element are as follows:

economic considerations outside the purview of the NRC contributing to the shrinkage in the nuclear vendor / supplier marketplace (numbers and types of enterprises and volume of business) the degree to which the current institutional, statutory, and regulatory structure permits the NRC to take effective action in this area the point at which more aggressive vendor / supplier oversight becomes counterproductive, driving vendors / suppliers out of the nuclear business, reducing competition, increasing order lead time, or otherwise impairing the availability of quality items and services roles of industry and owners groups in assuring the quality of items and services.

The activities to be conducted under this program element include the following:

Develop an understanding of the changing environment in the nuclear industry and its near-term and long-term effects on the quality of procured items and. services.

Understand the regulator's role in this changing environment.

Identify alternative approaches to offset any negative impact of the changing environment on the safety and quality of nuclear facilities.

Gain a better understanding of the causal factors leading to utility-contractor-vendor-supplier QA problems.

Understand how various modified institutional and legislative frameworks postulated for the future will affect quality in the nuclear industry.

Periodically publish reports summarizing the latest results of research into the areas above, trends in the industry, their causal factors, and implications for quality in the future.

4-5

End Products The end products of these activities include the following:

NUREGs or other reports that summarize the results of research into the changing infrastructure of the nuclear industry, trends, and causal factors (FY86 and later).

O e

e 4-6

i 5.0. . QUALITY ASSURANCE FOR NON-REACTOR ACTIVITIES Chapters 3.0 and 4.0 of the QA program implementation plan focused on staff actions to improve the quality of the design, construction, and operation of nuclear power plants. This chapter addresses quality and QA programs for the balance of the nuclear fuel cycle -- the non-reactor activities. The non-reactor activities discussed in the following elements of the QA program implementation plan include mills, fuel facilities, transportation, high-level waste (HLW) repositories, monitored retrievable storage (MRS) facilities, and independent spent fuel storage installations (ISFSIs).

The Office of Nuclear Material Safety and Safeguards (NMSS) has the responsi-bility for licensing these non-reactor activities, including the development of policy, programs, and requirements for licensing. The Office of Inspection and Enforcement's (IE's) role in the non-reactor area is to develop inspection programs (including QA) for non-reactor areas and to provide support and QA expertise to NMSS in implementing agency programs. The program elements and IE activities presented in this chapter are directed both to support NMSS as requested (and as resources allow) and to review quality and QA policy, programs, and requirements in these non-reactor areas to ensure consistency and uniformity in the NRC's overall program for assurance of quality.

HMSS has specifically requested IE support in licensing review and in developing policy, programs, and req'uirements for quality programs for MRS facilities, ISFSIs, and HLW repositories. QA policies and programs for these three types of activities are in the early stages of development. In addition, the IE and NMSS staffs have developed an interface agreement on the respective roles of HMSS and IE for HLW repositories to address the uniqus features of that program.

In support of HMSS for mills, fuel facilities, and transportation, IE plans to review NRC's application of quality programs and requirements in those areas.

The objective is to assure that NRC's policy for quality in non-reactor areas is consistent and uniform with the policy for reactors and that quality programs applied to non-reactor activities are commensurate with both their relative safety significance and the requirements for protecting public health and safety.

5.1 MONITORED RETRIEVABLE STORAGE FACILITIES AND INDEPENDENT SPENT FUEL STORAGE INSTALLATIONS Pending the establishment of permanent HLW repositories for the disposal or storage of high-level radioactive wastes, measures are being taken to provide interim storage installations and capabilities (MRS facilities and ISFSIs).

10 CFR Part 72 contains the basic NRC requirements for independent storage installations. Subpart G of Part 72 contains the quality and QA requirements for ISFSIs. An amendment to Part 72 is presently being prepared to expand the applicability of Part 72 to include MRS installations. The Nuclear Waste Policy Act (NWPA) of 1982 requires the Department of Energy (DOE) to assess the need for and feasibility of one or more MRS installations. The MRS installations will provide for a longer-term storage of HLW than originally planned for ISFSIs.

5-1

The interim storage installations (ISFSIs and MRS facilities) will be licensed by the NRC/NMSS. NMSS has asked IE to provide support in developing QA policy, programs, and requirements. This IE support will also include IE licensing review of an applicant's quality program for both ISFSIs and MRS facilities.

Presently, one application for an ISFSI is under review; two more are expected in the near future. The scope and pace of this licensing activity will be accelerating. Associated with the license applications for ISFSIs is the submittal of dry storage cask QA topical reports for NRC review and approval.

Currently, a German firm has submitted such a report for NRC review and approval. Firms from Germany, France, Japan, Canada, and possibly other countries are expected to be interested in cask manufacturing.

In addition to these license application and QA topical report reviews, the NWPA requires 00E to submit to NRC for review in 1985, a proposed program for MRS installations. Appropriate quality and QA policies, programs, requirements, and guidance for ISFSIs and MRS facilities must be developed and implemented.

The goal of this program element is to provide QA support in assuring that ISFSIs and MRS facilities are designed properly, constructed as designed, and operated in a manner that does not pose undue risk to public health and safety. To attain this goal, appropriate QA program guidance and NRC insoection programs must be established.

The benefits which would result from accomplishing this goal are as follows: .

• assurance that MRS installations and ISFSIs will be properly designed, constructed as designed, and operated safely increased NRC, licensee, and public confidence in the quality and safety of HLW interim storage installations consistency in NRC QA policy, programs, and implementation.

The following issues are associated with this program element:

the potential differences in application and policy for quality and QA between two government agencies (NRC and 00E) the inherent differences in operation and in relative .isks (versus reactors) to the public in establishing quality program requirements for ISFSIs and MRS facilities.

the lack of directly relatable precedents or models the logistics and potential problems associated with the role of foreign vendors in the manufacture of dry storage casks the use of foreign government inspection organizations (for example, the German TUV) to inspect and assure quality and conformance to U.S.

regulations.

5-2

The activities to be conducted under this program element include the following:

For quality assurance, develop guidance for establishing and implementing

a program that meets NRC requirements and achieves program performance goals and objectives.

Develop an NRC position on handling the quality aspects associated with j; the activities of foreign vendors involved with storage casks.

For MRS installations and'ISFSIs, develop an NRC QA inspection program i

that provides adequate guidance to inspectors for evaluating licensee and

DOE implementation of ISFSI and MRS facility management control systems.

l Train NRC inspectors for ISFSI and MRS installation inspections.

End Products j The end products of these activities include the following:

interim guidance documents on QA for ISFSIs and MRS installations to coincide with the publication of revised 10 CFR Part 72 (FY85) a regulatory position on the use of foreign-inspection agencies to perform QA inspections of cask manufacturing in place of U.S. inspection agencies (FY85) .

I preliminary IE inspection procedures for the QA inspections of ISFSIs (FY85) preliminary IE inspection procedures for the QA inspections of MRS

facilities (FY86) l a regulatory guide or other guidance document on QA for ISFSIs (FY87)

) a regulatory guide or other guidance document on QA for MRS facilities 1

(FY88).

i' 5.2 HIGH-LEVEL WASTE REPOSITORIES The Nuclear Waste Policy Act of 1982 mandates certain activities and schedules for permanent HLW repositories. The stipulations of the law are in a sense

! " performance requirements" imposed on DOE, NRC, and the Environmental Protection l Agency (EPA). NRC's regulations for HLW repositories, including requirements for quality and QA, are found in 10 CFR Part 60. The QA requirements in Part 60 are based on the criteria of Appendix B to 10 CFR Part 50 "as applicable, and appropriately supplemented" by additional criteria specific to HLW reposi-l tories. 00E's nuclear QA policy is structured on applying the principles in the standard ANSI /ASME NQA-1. Although the QA programs for both reactors and HLW repositories have the same underlying principles, they cannot be uniformly

applied because of two conceptual differences in technical requirements, engineering considerations and missions. For example, there are differences in the magnitude of the effect of an operating reactor accident vs. an operating repository accident on the public's health and safety, and in the 30 ycar life j or mission of an operating reactor facility vs. the 10,000 year mission for a 1

HLW repository.

! 5-3 .

The key safety issue for HLW repositories is stated in the DOE Mission Plan (April 1984): "Will the geologic repository, consisting of multiple nature and engineered barriers, isolate the radioactive waste from the accessible environment after closure [ emphasis added], in accordance with the requirements set forth in 10 CFR Part 60, and the proposed Environmental Protection Agency rule to be codified in 40 CFR Part 191?" To answer this with reasonable assurance (i.e., adequate confidence that the requirements are met) demands effective implementation of a QA program tailored to meet the unique challenges presented by long-term storage of radioactive waste.

The licensing responsibility for HLW repositories resides with NMSS. IE will support the licensing activities of NMSS with the QA staff and expertise developed from the nuclear power plant licensing program. Agreement on the QA interface for HLW repositories has recently been reached between the two offices. The QA program implementation plan activities for the repository program will be developed jointly by the two offices in the near future.

5.3 FUEL FACILITIES (a)

The NRC program for licensing mills and fuel facilities, other than plutonium processing and fuel fabrication plants, does not contain explicit requirements for quality or QA. (According to 10 CFR 70.22 (f), each-application for a license to possess and use special nuclear material in a plutonium processing and fuel fabrication plant is required to contain a QA program based on Appendix B to 10 CFR Part S0.) The NRC assurance of quality and safety for these activities and facilities rests predominately on the requirements and controls imposed by NRC for criticality control, environmental protection, radiation protection (public and worker), nuclear material control, accountability, and security (10 CFR Parts 20, 70, and 73).

Many of the features of the existing NRC requirements for safety may be viewed as elements of an informal QA program for mills and fuel facilities. However, there is no definition of a comprehensive, integrated QA program which covers all activities (e.g. , radiation protection, criticality control, measurements, etc.).related to the design, construction, and operation of the facilities.

The goal of this element is to deter,mine if it is necessary to define, develop, and implement an effective, integrated program of quality and management controls that will provide additional assurance that mills and fuel facilities are designed properly, constructed according to design requirements and commit-ments, and operated in a manner consistent with public and plant worker health and safety.

The benefits which would result from accomplishing this goal are as follows:

• consistency in defining and implementing an NRC policy for assurance of.

quality as applied to both reactor facilities and, as applicable, to non-reactor facilities (a) Because of the budget priority of operating plants, no work is planned in this area.

5-4

assurance that plants which are part of the nuclear fuel cycle are subject to QA program requirements consistent with the requirements applied to the design, construction, and operation of nuclear power plants, but commensurate with the risk to public and plant worker health and safety posed by those facilities.

The issues associated with this program element include the following:

the development and application of 4A requirements for activities that have been conducted for years without any formal QA program requirements the necessity of an integrated, comprehensive QA program for mills and i fuel facilities in light of the relative risk (vs. reactors) the facilities pose to the health and safety of the public and plant workers.

The activities to be conducted under this element include the following:

Determine presently existing NRC requirements and programs which are or may be related as elements of a QA program, examine the completeness of this collection of elements of a QA program, and identify missing elements.

Determine the set of quality and performance activities that presently must be met by mills or fuel facilities and contrast that set to the attributes of a comprehensive, integrated program.

Prepare a report on the results of the QA program and requirements review for mills and fuel facilities. The report will contain recommendations on developing a QA program and performance requirements for those facilities.

End Products The end products of these activities include the following:

a NUREG or other report on the review of QA program requirements for mills and fuel facilities (FY86) a decision on whether existing informal QA measures applied to Part 20, 70, and 73 licenses are sufficient (FY86) depending on the results of the above products, development of rule changes, regulatory guides, guidance documents or other appropriate actions (FY87 and later).

5.4 TRANSPORTATION (a)

Licensees who transport or deliver to a carrier for transport licensed materials as defined in 10 CFR Part 71 are required to have an NRC-approved QA program satisfying the applicable provisions of 10 CFR Part 71 (Subpart H, " Quality (a) Because of the budget priority of operating plants, no work is planned in this area.

5-5

) Assurance"). NMSS has the QA licensing responsibility for all applicants subject to the requirements of 10 CFR Part 71. As part of this element, IE will perform a generic review of all QA regulations, policies, and programs related to transporting licensed material.

In this review, QA applied to licensees under 10 CFR Part 71 will be assessed.

This review will be similar to the one planned for mills and fuel facilities as discussed in Section 5.3; however, the transportation review will focus on the adequacy, consistency, and uniformity of existing Part 71 quality and QA requirements and programs as compared to those of reactor and other non-reactor quality applications.

As part of this element, IE also will consider the development of an inspection procedure to address the inspection of the suppliers of NRC-certified reusable radioactive transport packages. Present NRC inspection programs for NRC-approved QA programs do not address the adequacy of implementation of quality programs by the suppliers of reusable radioactive material transport packages. New spent fuel casks are an exception to the proble'm in inspection coverage because 10 CFR Part 71 requires spent fuel cask manufacturers to notify the NRC at least 45 days prior to cask fabrication. The notification is a trigger to the Vendor Program Branch of IE to perform inspections of new casks.

In general, pursuant to 10 CFR Part 71, QA programs are approved by the NRC for two broad categories of applicants: the user who usually obtains the transport package on lease or purchase, and the supplier. In some cases, the supplier also furnishes transportation and/or waste burial services. In any case, the supplier must perform all of the activities relating to design, procurement, fabrication, and initial testing of the transport package, as well as any periodic maintenance, repair or modification. In some instances, the supplier is also the user. It should be noted that the supplier will not in every case be an NRC licensee, although the supplier will have an NRC-approved QA program.

l The benefits which would result from accomplishing this program element are as follows:

1 consistency in defining and implementing an NRC policy for assurance of quality as applied to reactor facilities and as applicable to transporta-tion, commensurate with its importance to safety

'

• increased assurance to the NRC, licensees, and the public that the trans-portation of licensed materials conforms to regulations and assures adequate protection of public health and safety.

The following issue is associated with this program element:

the number and diversity of 10 CFR Part 71 licensees and the limited NRC resources available to inspect against QA requirements for

~

transportation.

The activities to be conducted under this program element include the following:

1 Following the review of transportation quality and QA programs, prepare a report on the results of the review and provide recommendations based on the review.

5-6

l Develop a new inspection procedure for inspecting suppliers of NRC-certified reusable radioactive transport packages. This will result in a more complete NRC licensee and vendor inspection program.

End Products The end products of these activities include the following:

a NUREG or oth- report on the review of QA for transportation (FY86) a decision on whether existing QA programs and inspection procedures for transportation are sufficient (FY86) depending on the results of the above products, development of rule changes, regulatory guides, guidance documents or other appropriate actions (FY87 and later).

5-7

o h O APPENDIX A INTEROFFICE COMMENTS ON THE QA PROGRAM IMPLEMENTATION PLAN A draft of the quality assurance (QA) program implementation plan was provided to other NRC program offices and the NRC regions for review and comment. As a result of the comments received, a number of changes were made. Statements noted as unclear or incorrect have been clarified or corrected, using where possible suggested wording provided by the commenter. Other changes that have been made in response to the comments include the following:

• The " Introduction" has been revised to explain the reasons for the level of specificity in the plan.

• The intent and need for regional office participation in some program development activities (e.g., performance trending) have been clarified.

• Based on an assessment of the resources needed to complete the activities described in the plan, a discussion on the resource needs compared with budgeted levels has been added to the " Introduction."

• In Section 3.1, the discussion of performance-oriented QA programs has been clarified to indicate that the existing QA regulatory base, which is necessary for' compliance inspections and enforcement actions, would remain intact. That is, no significant changes to the present QA regulatory base, which is built upon the requirements of Appendix B to 10 CFR Part 50, are planned. Additionally, there are no plans to elim.inate the Standard Review Plan submittals that obtain licensee QA commitments.

• Several sections in the plan have been revised to clarify the plan's position on the issue of performance vs. prescriptiveness.

Certain comments could not be addressed adequately. For example, one commenter noted that the plan's emphasis on NRC activities seems to diverge from the tenets of the QA program, which emphasize licensee responsibility for QA and the importance of licensee initiatives. The plan is intended to be NRC's plan for improving quality, and it necessarily focuses (as required by the Ford Amendment) on NRC actions as it tries to develop key program and policy changes aimed toward improved licensee actions. Each of the following implementation plan activities are oriented toward these goals: performance-oriented QA programs; recognition and support of industry efforts; readiness reviews (performed by the licensee instead of third party audits as recommended by the QA report); NRC-utility management meetings; reduction of QA inspection burden on licensees and vendors; alternative QA requirements for subtier vendors; and responsibilities of line organization management and personnel for achieving and assuring quality.

In the comments received, the most significant objection to the draft implemen-tation plan was on the issue of performance vs. prescriptiveness. In empha-sizing the importance of performance, the draft plan did not adequately address the importance of specific (prescriptive) commitments. Changes have been made to the plan to try to clarify the plan's position on the issue of performance vs. prescriptiveness. The plan provides for certain QA program modifications A-1

_ _ _ - _ - .- 1

that are more prescriptive than present practice. For example, acceptance criteria developed as part of a program for incremental NRC approval of com-pleted work would provide a better-defined and more prescriptive basis for acceptance of completed work than is presently the case (Section 3.2.2).

The plan endorses more prescriptiveness where it is necessary, e.g., for acceptance criteria or enforcement action based on chronic substandard per-formance. However, one emphasis of the plan is improving the NRC's and the licensees' ability to identify and correct substandard performance earlier than has been the case in the past. The plan envisions an inspection program that strikes a better balance between compliance-oriented and performance-oriented inspections. The underlying philosophy for this balanced approach is discussed below.

NRC inspectors inspect against license commitments or regulatory requirements.

Inspectors cannot arbitrarily cite a licensee; citations must be based on objective evidence of violations. The more specific (prescriptive) the requirement,is, the more objective and defensible are the inspector's findings.

The balanced approach would limit the flexibility of individual inspectors to make judgmental decisions, bringing a degree of uniformity to the NRC's inspection program. In the words of one Regional Administrator, however, "we

[NRC] wind up inspecting requirements for requirements sake" because the inspection process does not constantly challenge the inspector to ask the questions "is the requirement meaningful" or "is the licensee's implementation effective." With this " checklist" approach, inspections and inspection reports tend to emphasize individual, nonaggregated findings and do not provide adequate feedback on either of the above questions. This results in situations like those explored in the case studies of the QA study where "bandaids" are put on specific instances.of programmatic or generic problems. In such cases the root causes of the problems are not diagnosed and corrected, nor are the significance of the findings on the project's overall status put in perspective.

Consequently, the NRC has sometimes focused so much on detail (prescribed requirements and commitments) that both the NRC and the licensee has lost sight of the broader objectives of the NRC QA program. As a result, the NRC loses sight of whether a licensee's program as a whole is accomplishing its objectives.

The plan's emphasis on " performance" is not meant to eliminate objective, prescriptive commitments at the detail level. Those prescriptive commitments are necessary to provide legally defensible assurance that safety-related work has been performed correctly and to provide the basis for any required enforcement action. An intention of the plan is to improve the ability of NRC and licensees to see the " big picture" and to ensure that NRC inspection l of " requirements for requirements sake" is a rare exception to normal inspection practices. The shift in emphasis to bring performance up to a level comparable with compliance is also necessary so that NRC inspectors and managers can better understand how the inspection results fit into the overall picture of whether the licensee's program is accomplishing what it should. To do this, the NRC needs to define the program's performance objectives and needs to have tools (trend analysis and performance indices) that can measure whether these i

objectives are met (Sections 3.1.1 and 3.1.2). To do this most effectively, j the licensee needs to view performance orientation to quality assurance as something that it should do because it makes good sense, regardless of regula-tory considerations (i.e., not a requirement that the licensee does because it is required -- see the discussion of the QA program's basic tenets in the

" Introduction").

A-2 l

Another aspect of the performance vs. prescriptiveness issue is that the plan adopts the philosophy that as a rule, NRC should not tell a licensee how to do something. If the NRC tells a licensee how to do something, the NRC loses a degree of objectivity because the NRC now has a vested interest in the project's success. The regulatory base should define what should be achieved (performance) and allow the licensee, with the assistance of consensus guidance developed by professionals with appropriate expertise, to decide how to do it. The licensee then may describe how that performance will be achieved in specific, perhaps prescriptive, terms against which NRC inspects. A licensee's commitments can be phrased in prescriptive terms, but', except in special cases (such as chronic substandard performance), the NRC should be cautious of going beyond establish-ing requirements for what is to be achieved by defining how it should be done.

The following comments from one Regional Administrator specifically capture the essence of the above discussion:

"There is no question that we need to change the perception that NRC is more interested in the " form" of the QA program rather than the

" substance." While we support the need to focus more of our inspec-tion effort on the bottomline effectiveness of QA programs, we should not abandon our reviews of a good prescriptive compliance-oriented QA program. Program reviews of this type do provide the opportunity to identify and correct weaknesses or deficiencies before they result in hardware problems."

"We feel that effort is needed to more clearly define quality require-ments (i.e., be more prescriptive) to ensure that the program addresses only those attributes which are of importance. Many times, we wind up inspecting requirements for requirements sake. This is another reason why prescriptive requirements which are well conceived are of paramount importance to success. The success of a performance-oriented QA program will depend on the public's and the NRC's confidence that licensees are identifying most safety-related problems."

The first of the QA advisory groups (see " Introduction" on p. viii) of regional and headquarters staff to help implement this plan will specifically address the performance vs. prescriptiveness issues and will try to solidify the near consensus that appears to exist on this issue.

1 A-3  !

I i

r_

,A..

APPENDIX B SECY MEMORANDUM OF OCTOBER 19, 1984 i

l l

l l

l 1

)

I B-1 i - - - - - _

IN RESPONSE, PLEASE 4,

g UNITED STATES y } , ,, g Nh REFER TO:

NUCLEAR REGULATORY COMMISSION M841[16B g W ASHIN GTON, D.C. 20555 D,sg*=*/j/

9i '

ACTION - DeYoung Cys: cks

• October 19, 1984 P. g.,

CFFICE OF THE l['*; 7. /[ Q[ " { Rehm SECRETARY *vy .

$tg))g

1. is GCunningham g g '//Jl Denton Partlow, IE William J. Dirck's, Executive Director Malloy, IE MEMORANDUM FOR: Barry for Operations FROM: Samuel J. Chilk, Secretan f[\ ~q a f2

SUBJECT:

ION OF QA REPORT co./n STAFF TO REQUIREMENTS CONGRESS, 2:00 P.M., - DIS UES US{ DAY, OCTOBER 16, 1984, COMMISSIONERS' CON ERENCE ROOM, D.C.

OFFICE (OPEN TO PUBLIC ATTENDANCE)

' b The staff discussed with the Commission the Quality Assurance Report to Congress. The staff advised the Commission that it would send the quality assurance program plan to the Commissioners in early December 1984.

The Commission agreed with the staff recommendation to defer the three proposed rulemaking initiatives on: periodic third-party audits of plants under construction, management audits of organizations building future nuclear power plants, and establishment of an advisory board to review management and QA capabilities of CP applicants. Commissioner Bernthal requested that the staff continue to perform a careful long-term analysis of the concept of third-party audits that could be applied should ever construction of a new generation of nuclear power plants begin.

(IE)

The Commission, by a vote of 3-2 (Commissioners Bernthal and Asselstine approving) disapproved an initiative to have the staff investigate with one or two utilities support for a voluntary pilot designated representative program that would operate within existing NRC statutory authority.

There was strong Commission consensus, however, for more resident inspectors to be added to the current program.

The Commission requested that the staff review the concepts of the Integrated Design Inspection and the Georgia Power Company's readiness review program to see if they can be effectively integrated.

(IE) (SECY Suspense: 10/16/85) 1 Rec'd 011. gno Date. v  ?- 7 ~ 8 ",

B-3 Time .'.".. . .'(/.7. .W.

,,", ,,.d.,'

CS . ,e Chairman Palladino requested that the Commissioners' technical assistants meet to identify in Chapter II of the QA Report to Congress (SECY-84-124) items where consensus does exist and where additional information is needed.

,(SECY) (SECY Suspense: 10/23/84) cc: Chairman Palladino

• Commissioner Roberts Commissioner Asselstine Commissioner Bernthal Commissioner Zech Commission Staff offices PDR - Advance DCS - 016 Phillips B-4