ML20134A654
Text
__
s l
l QUALITY ASSURANCE CASE STUDY WORKING PAPER CASE D m
l Prepared for Nuclear Regulatory Connaission Washington, DC 20555 i
February 29, 1984 Fo/A44493
$DR hoffAG50703 LEICHTOB4-293 pop i.
2.
TABLE OF CONTENTS
- fage, I.
SUMMARY
l A.
Introduction 1
B.
Background
2 C.
Summary of Root Causes 5
II. ROOT CAUSES OF THE PROJECT'S PROBLEMS WITH QUALITY IN CONSTRUCTION 8
A.
Inexperience of Project Team in Nuclear Plant Design and Construction 9
B.
Inadequate Management Support for Quality 12 C.
Shortcomings in NRC's Licensing and Inspection Practices 13 D.
Inability of Project Team to Adjust to Changing Nuclear Power Environment 14 III. REMEDIAL ACTIONS TAKEN TO CORRECT CONSTRUCTION QUALITY PROBLEMS 15 IV. GENERIC IMPLICATIONS 16 A.-
Understanding of Nuclear Project and its Implications 16 B.
Nuclear Experience Vital Today 17 C.
Management Awareness and Involvement Necessary 17 D.
Need Qualified, Capable People 18 E.
Ultimate Responsibility Retained by Licensee 18 V.
UTPLICATIONS OF CASE STUDY D FOR NRC QA INITIATIVES 18 A.
Measures for Near-Term Operating Licenses (NTOL) 19 B.
Industry Initiatives 19 C.
NRC Construction Inspection Program 20 D.
Designated Representatives 21 E.
Management Initiatives 21 VI. IMPLICATIONS OF CASE STUDY D FOR THE FORD AMENDMENT ALTERNATIVES 22 A.
More Prescriptive Architectural and Engineering Criteria 23 B.
Conditioning the Construction Permit on the Applicant's Demonstration of His Ability to Manage an Effective Quality Assurance Program 23 C.
Audits, Inspections, or Evaluations by, Associations of Professionals Having Expertise in Approptfate Areas 23 D.
Improvement of NRC's QA Program 24 E.
Conditioning the Construction Permit on the Applicant's Commitments to Submit to Third-Party Audits of Its 24 Quality Assurance Program
\\
4 Pcgs 1 of 25 QUALITY ASSURANCE CASE STUDY WORKING PAPER CASE D I.
SUMMARY
A.
Introduction The Nuclear Regulatory Commission (NRC) has undertaken a study of selected nuclear reactor construction projects to determine the important factors, or root causes, that underlie effective and ineffective assurance of quality programs.
Several nuclear projects that have experienced major quality-related problems and several that have not were selected for the study. Data and findings from these site-specific studies will be used by the NRC in formulating generic policies and programs related to the assurance of quality during design and construction of nuclear reactor projects and in responding to Section B of NRC's FY 1982-83 authorization bill (the Ford Amendment). This working paper summarizes the findings from the fourth case study (Case D).
The Case D licensee was selected
- for the case studies because it had experienced problems in design activities and in quality of construction that resulted in the issuance of an NRC Show Cause Order.
The case study team was comprised of four subteams of two personnel each.
One subteam concentrated on project engineering / design, one subteam concentrated on construction, one subteam concentrated on quality assurance / quality control, and one subteam concentrated on project management. Prior to visiting the Case D licensee, the case study team members reviewed NRC inspection reports, investigation results, hearing transcripts and other documentation pertaining to the project. The NRC project team leader and one team member visited the licensee's corporate offices for two days the week before the Case D site visit to interview top level corporate personnel responsible for nuclear projects.
This advance team also visited the NRC regional office for two days the week before the Case D site visit to interview cognizant regional personnel and to review pertinent NRC records.
The case study team spent five days conducting interviews at the licensee's corporate and engineering offices and at the plant site.
About 60 individuals involved in the project were interviewed, including personnel of the licensee, the architect engineer /
construction manager (AE/CM), the constructor (C), and cognizant NRC regional personnel. Licensee personnel interviewed ranged from the President and Chief Executive Officer to site QC inspectors and included each level in between.
Cognizant NRC regional s
)
4 Pcga 2 of 25 personnel participated in some subteam interviews and in the exit conference. The licensee changed AE/CM and C organizations during the project.
Interviews during the case study were with the current AE/CM and C personnel. No interviews with the former AE/CM and C personnel were held.
In addition to conducting interviews, the case study team spent one-half day touring the construction site. The case study culminated in a briefing for licensee management and project staff members, in which the preliminary results of the case study were presented and the licensee was given an opportunity to respond to the team's preliminary results. The case study team did not evaluate the adequacy of design or quality of construction.
B.
Background
Early in the 1970s, the Case D licensee decided to construct nuclear generating plants. A possible natural gas shortage, the favorable economics of nuclear power, and public acceptance of nuclear power were reasons the nuclear option was deemed by the licensee to be a logical choice.
Two projects were initiated, one in which the licensee would be sole owner (and which was later canceled) and the other a joint partnership with the licensee as project manager for all aspects of engineering design, construction, and operation.
This latter project comprised two large (greater than 1000 MWe) units. The first-unit operation was projected for the 1981-1982 timeframe, with second-unit operation to follow about two years late r.
Both units have been delayed.
The licensee had no prior nuclear experience, but this was not seen as an insurmountable obstacle.
Many other utilities were (or had been) in the same position, and the leaders in the industry were viewed as not having that much more experience.
In selecting an architect-engineer / construction manager / constructor (AE/CM/C), the licensee had compiled a candidate list that included the firm selected. Because many nuclear plants had been on order in the late 1960s and early 1970s, most AE firms were committed and the licensee realized there would not be an opportunity to select from a large number of firms.
It selected a large engineering and con-struction firm that had a national and international reputation of doing things within time and cost, had been a successful contractor in chemical plants and oil field and pipe line construction and had committed itself to growth in the nuclear industry and to getting top people by hiring personnel from other architect engineer organizations having nuclear experience.
Its primary forte up to the early 1970s, however, had been in other than nuclear work.
It did not have as extensive nuclear experience as many other AE or constructor firms. Although it had previous experience as constructor at two nuclear plants, this would be its first major nuclear engineering and design project and the first nuclear project for which it was construction manager.
~
Peg 2 3 cf 25 When the. licensee applied for a construction permit in the mid-1970s, it was received about 6-8 months earlier than either the licensee or its AE/CH/C expected. This may have been the result of a national emphasis to streamline the licensing process (a few years previously, the oil embargo had taken place and there was national concern over energy independence). The licensee maintained (during the site visit) that rapid licensing resulted in construction being started before an adequate amount of design and engineering (estimated by the licensee at less than 25%) had been completed.
The licensee recognized that managing a nuclear plant construction project would require a greater involvement than that required for a fossil plant.
Early in the project, the licensee used a matrix-type organization to manage the project. The approach was recognized to be embryonic, but thought capable of doing the job.
Proj ect management rested on an organization that had responsibility for both nuclear and fossil projects. When starting the project, the licensee moved personnel from fossil into nuclear and hired several personnel with prior nuclear design and construction experience.
The licensee assigned personnel to its nuclear projects based upon the needs of e ch project.
In 1977, the licensee became concerned about meeting schedules and about disparities, such as the amount of concrete poured and the amount scheduled, and hired an independent organization to evaluate project management. The independent organization reported the project was not within schedule and cost and was not as complete as reported by the AI/CM/C.
In the mid-to-late 1970s, quality-related problems resulted in the licensee suspending concrete and welding activities. Allegations were made of harassment and intimidation of QC inspectors and inadequate support of inspection by supervision. The cognizant NRC regional office discussed low morale of quality assurance / quality control (QA/QC) personnel, QA/QC staffing below licensee specified levels, concrete placement problems, and weaknesses in QA/QC program implementation with licensee management.
In late-1978, the licensee initiated a study of whether the AE/CM/C should be replaced. Consultations with other AEs and constructors led the licensee to conclude that it would do best to support and improve the AE/CM/C organization and to become more involved in the design and construction activities. Thus, during the course of the project and up into the early 1980s, the licensee increased its involvement in the AE/CM/C activities.
In 1978, following a consultant report that there was a high likelihood of both cost and schedule overruns, the licensee acted to strengthen its project management.
It made the power plant engineering and construction manager the nuclear project manager and created a project management team reporting directly to him. About 30 experienced personnel were added from a consultant organization until the licensee could replace them with comparable personnel.
Pcg2 4 of 25 In 1979, following an independent management audit, the licensee expressed written concern about the AE/CM/C's performance and directed it to take several actions in the areas of construction supervision, planning, scheduling, control of construction work, labor productivity, and site housekeeping. The AE/CM/C agreed in large measure with the licensee's assessment and already had begun corrective measures to improve its performance. While some concerns were promptly resolved, others continued to require the attention of the licensee.
The NRC performed an acclerated mid-term inspection (middle of construction) that identified five noncompliances related to the QA program. As a result of the licensee being considered ineffective in correcting poor construction practices and continual allegations of harassment and intimidation of QA/QC personnel, the NRC performed a comprehensive team inspection / investigation, which resulted in a NRC Show Cause Order in early 1980. That order required the licensee to show cause why the construction permit should not be revoked or construction should not be suspended. The Show Cause Order indicat-ed that procedural and programmatic inadequacies in the licensee and AE/CM/C organizations resulted in a failure to systematically identify quality problems and to routinely correct and prevent recurrence of identified problems; that procedural, organizational, and personnel inadequacies resulted in a lack of adequate control of safety-related construction activities; and that lack of detailed knowledge and involvement hindered the licensee's ability to main-tain adequate control over the AE/CM/C.
Corrective action was started in each area of the Show Cause Order immediately after notification and before actual receipt of the Show Cause Order. The licensee reorganized the project, hired experi-enced personnel from other nuclear construction projects, added its own QC inspection presence at the site, required the AE/CM/C to change its QA/QC program and assisted the AE/CM/C in hiring and relocating personnel with nuclear design and construction experience.
In the early 1980s, the licensee hired an independent organization to review the design work for the project. The independent organization reported that design work was not sufficiently completed to support construction.
Subsequently, the licensee terminated the AE and CM parts of the AE/CM/C's contract and later the AE/CM/C terminated the construction part of the contract. The licensee replaced the AE/CM/C with two separate organizations having extensive nuclear construction experience. One has contractural responsibility for all design as the AE and functions as the construction manager and the other is the constructor. The licensee maintains overview responsibility of the daily activities of both organizations.
Safety-related work resumed in the fall of 1982.
Pcgo 5 of 25 The licensee has natured in managing and overseeing the project by:
hiring personnel with nuclear experience for key positions approving key AE/CM and C management personnel increasing training reviewing AE/CM and C quality assurance programs, training procedures and personnel qualification records establishing formal interface agreements, a joint records management system with the AE/CM, a stronger quality engineer-ins function and an engineering assurance group to perform independent design verification.
The licensee has become less directly involved in the daily activ-ities of the AE/CM and C and has concentrated on its overview role.
The licensee attends weekly meetings with AE/CM personnel and monitors planning, scheduling and cost.
Licensee site personnel monitor schedules, analyze monthly output records and review manpow-er loading.
The licensee also holds monthly project review meetings.
Constructor QC personnel have first line QC inspection responsibil-icy. AE/CM QC personnel reinspect selected work that the constructor QC has inspected. As an overcheck, licensee QC personnel reinspect selected work that the AE/CM QC has inspected and selected work that the constructor QC has inspected.
The licensee has over 500 personnel assigned to the project and consulting personnel hired earlier have been replaced. The licensee is developing its engineering capability for future work on the project by having 13 engineers work for the AE/CM.
In mid-1983, there were about 1,500 manual personnel at the con-struction site. This number was expected to increase to about 3,000 by the end of 1983.
C.
Summary of Root Causes The case study team believes the root causes that underlie ineffec-tive assurance of quality prior to the NRC Show Cause Order are:
1)
Inexperience of the Project Team While the licensee had extensive experience in constructing and operating fossil plants, it had not been involved with con-structing a nuclear plant.
It apparently failed to totally appreciate the difference in scope and complexity between the two, as reflected in the controls applied to the project. The licensee was organized by technical discipline into a matrixed fossil-nuclear organization.
Personnel were shuffled from i
fossil to nuclear and vice-versa as the need for a particular discipline arose. As a consequence, a requisite core of I
I
Ptg2 6 of 25 full-time professionals was slow in developing. The licensee did hire some staff with nuclear experience; however, they were not sufficient to provide the necessary core of competence.
The' licensee's lack of nuclear experience was further aggravat-ed by the lack of experience of key individuals involved with the construction project. This project was the first nuclear project for the licensee's project manager, project engineering manager, and the quality assurance manager.
Licensee inexperience resulted in four management levels between the site quality assurance organization and the execu-tive vice-president responsible for the project. The delay and filtration of information caused by this managerial superstruc-ture contributed to incomplete understanding at the executive level of the problems that were developing.
Historically, the licensee had depended upon its contractors to do the bulk of the planning and execution of fossil plant construction jobs. The licensee assumed that this same ap-proach would be appropriate for the nuclear project and, consequently, placed too much reliance on the prime contractor.
While not adequately involved at higher levels of management, in some respects the licensee became too involved at lower levels.
Licenne e personnel found themselves directly in the approval chain for AE/CM/C design approvals and other docu-ments. This had the effect of unduly restricting work flow.
Everyone in the chain had veto authority, and everyone had to agree. Toward the end of the AE/CM/C's tenure, the licensee assumed much of the contractor's responsibility in an intensive but vain effort to help the contractor's effectiveness.
In effect, the engineering work that was performed was the product of the AE/CM/C and the licensee instead of the product of the AE/CM/C with licensee overview.
The licensee failed to recognize and understand that the problems encountered were symptoms of larger problems in the licensee's control of the project.
Although the licensee was involved in providing direction from the beginning, their experience level rootricted the direction and resulted in insufficient and unclear management direction and involvement and the diffusion of responsibility to inappropriate levels of authority.
The AE/CM/C, like the licensee, had inadequate nuclear experi-ence. As a consequence, according to the licensee, the AE/CM/C did not understand the complexity of nuclear plant design and construction and did not bring to bear the necessary technical and management skills. These problems were aggravated by the earlier-than-expected approval of the construction permit and, therefore, the AE/CM/C did not have the pinnned time to come up 1
P g2 7 ef 25 to speed on design and personnel competence. The AE/CM/C's lack of nuclear experience was also further aggravated by the lack of experience of key individuals involved with the construction project. According to licensee personnel AE/CM/C first level management on the project had no nuclear experience nor did the principal AE staff.
The AE/CM/C inexperience resulted in construction being started before an adequate amount of design and engineering had been completed (estimated at less than 25% by the licensee), which resulted in an excessive amount of design evolution and rate of design change. Design work proceeded slowly and specifications and procedures were inadequate and formatted in complex ways.
There appears to have been insufficient engineering support for design and construction. The capabilities that the AE/CM/C did have were channeled into those areas in which it had experience, to the neglect of other equally important areas, according to the licensee.
Engineering efforts were scheduled based upon dictates from construction. This led to unrealistic demands on the engineering groups.
Licensee personnel stated that quality assurance and quality control were also dominated by construction. There were many conflicts between QA/QC and construction in which construction generally prevailed.
Licensee personnel also indicated that inexperience of the AE/CM/C resulted in insufficient and unclear management
~
direction and involvement. According to licensee personnel, project management did not have an adequate understanding of the interfaces and responsibilities for such functions as QA/QC, engineering, design, and construction. As a result, the constructor did not react in a timely, effective way to problems and did not employ proper management systems to reveal the causes of problems and to prevent them from recurring.
2)
Inadequate Manasement Support for Quality Neither the licensee nor the AE/CM/C appeared to have had a full understanding of quality and quality assurance concepts as they applied to nuclear plant construction. Although both made commitments to quality, these were not actualized in the construction process. The licensee was not appropriately involved in monitoring the total scope and details of activ-ities and did not know how to take effective corrective action to prevent recurrence of problems. The AE/CM/C did not suffi-ciently insulate QA/QC from cost and schedule demands, nor shield them from intimidation or harassmant. Consequently, construction supervisors dominated the QA/QC functions, both in the field and in the form of published policy, which emphasized minimizing cost and maintaining schedule. The long chain of command filtered information and introduced inefficiencies into the decision making and implementing processes. To further
Peso 8 ef 25 compound these problems, the licensee had none of its own QA inspectors at the site until 1980. This gave low visibility to management support of quality, which may have been interpreted as a lack of backing from top management for quality.
3)
Shortcominas in NRC's Licensina and Inspection Practices A recurrent these was that the NRC licensing process did not adequately address the ability and experience of the project management, nor was there adequate evaluation of whether the nuclear industry had over-extended itself at the time this plant was contracted. The inspection process also tended to ignore management issues. The irregular presence of NRC inspectors at the site early in the project was cited by the licensee as a contributing factor. The process used by NRC in identifying and dealing with problems was cumbersome and required excessive amounts of time.
In effect, the NRC ap-proach was one of allowing troublesome situations to progress to the point that a case could be built for taking the drastic action represented by a Show Cause Order.
Some of the problems involving the NRC required up to two years to resolve.
4)
Inability of Project Team to Adjust to Chantina Nuclear Power Environment The rapid proliferation of regulations during the mid-1970s was cited by the licensee as particularly troublesome, especially since the design of this particular plant was probably only about 25% complete when construction began in 1975 and pro-ceeded more slowly than it should have in relation to con-struction activities.
Regulatory changes from the TMI and Brown's Ferry incidents were also a severe blow to the project, according to the licensee.
Declining energy projections and increasing interest rates made funding plant construction more difficult.
Incidents within the industry, such as TMI and Brown's Ferry, reflected into changed design requirements. All of these changes coming in rapid succession further complicated the task for the relative-ly inexperienced nuclear staff of the licensee and its AE/CN/C.
II. ROOT CAUSES OF THE PROJECT'S PROBLEMS WITH QUALITY IN CONSTRUCTION Based upon review of NRC inspection reports, investigations, hearing transcripts and other documentation and interviews with current licensee, AE/CM, constructor and cognizant NRC regional personnel, the case study team believes that the following root causes were significant in contributing to the major quality and quality assurance problems experienced by this project before the NRC Show Cause Order.
Pcgo 9 cf 25 A.
Inexperience of the Project Team (Licensee and AE/CM/C) in Nuclear Plant Design and Construction.
1)
Inexperience of the Licensee This was the first nuclear plant project for the licensee. The licensee had successfully constructed fossil plants but did not change its approach adequately to adjust to the difference in complexity between fossil and nuclear plants.
In retrospect, the licensee exhibited inadequate understanding of what was involved in constructing a nuclear plant and didn't seem to recognize its inadequate understanding.
Licensee inexperience resulted in inadequate staffing for the project.
In addition to the lack of prior nuclear experience of the licensee as a corporation, the licensee appointed personnel without prior nuclear construction experience to key project positions.It was the first nuclear project for the licensee's Project Manager, Project Engineering Manager and Quality Assurance Manager.
Licensee personnel interviewed indicated it was company philosophy in the early 1970s to promote from within the company and to hire young people out of college. Outsiders were brought into the company only in capacities the company did not have, such as nuclear engineers.
Interviews with licensee personnel indicated that the licensee believed it was in the mainstream of their contemporaries in regard to staffing for the project. They had studied what other utilities were doing and found that in 1972 the leaders in the industry had about the same level of experience, which was an average of two to four years. Licensee personnel stated they had a bright, young, but inexperienced team, which, because of their inexperience, did not know how to expeditiously solve some of the problems that developed.
The licensee created the position of QA Manager in 1973 and filled the position with a nuclear engineer who had one year of quality assurance experience as a QA supervisor in industry before joining the licensee. There had not been a quality assurance position in the company, or a quality assurance progrca, before that time. The licensee established its quality assurance program because it was an NRC requirement.
Prior to embarking on its nuclear program, the licensee had not been subject to NRC requirements and had not established a QA program of its own as a management tool.
In 1977, the QA Manager became Manager of Construction for both fossil and nuclear plant construction and was replaced by a metallurgical engineer in the QA department, who had demonstrated management ability but who did not have nuclear QA experience.
From 1972 to 1975 there was a staff of about 15 in QA/QC working on the two nuclear projects the licensee was involved in.
In 1977, the Case D site had about neven licensee QA personnel.
i Pcg3 10 of 25 i
An unuous11y long chain of command, consisting of three management levels at corporate offices, existed between the licensee site QA Supervisor and the Vice President of Power j
Plant Construction and Technical Services, who reported to the i
Exe'cutive Vice President. Apparently this long chain of i
command between relatively inexperienced personnel resulted in upper management's failing to receive the type of information needed to make decisions.
In 1973, the licensee Project Manager functioned as a l
coordinator between the licensee and the AI/CM/C had a staff of three personnel, and tried to monitor engineering and l
procurement work. Construction work was the responsibility of the Construction Supervisor. Until 1977, the licensee received 7
only milestone schedules from the AE/CM/C.
In 1978, the licensee established a project team consisting of personnel from various departments matrized to a Project Manager.
l Licensee personnel interviewed indicated that neither the r
licensee nor the AE/CM/C.had effective project control systems to estimate project status and to control the project.
r i
Licensee inexperience resulted in over-reliance on contractors.
l Mistorically, the licensee had depended upon its contractors to construct fossil power plants. While this approach may be adequate with contractors experienced in nuclear plant design and construction, it is not adequate with inexperienced l
contractors. The licensee placed more reliance on the AE/CM/C than the complexities of nuclear construction would justify j
and, in retrospect, was ineffective in oversight and control of the project.
It took independent evaluations of the project for the licenses to understand the status of the project. The
+
licensee indicated that the NRC Show cause order made it fully j
understand the severity of their quality-related problems.
Lincensee inexperience resulted in failure to recognise and understand that the problems encountered were symptoms of larger problems in the licensee's control of the project.
Licensee personnel stated that the emphasis was on detail and l
no one was looking at the big picture. Review of NRC l
investigations revealed the licensee was responsive to specific l
WRC identified deficiencies and implemented ~orra.ctive action in a timely manner. However, insufficient licensee management i
involvement at the detail level in the total scope of I
construction activities coupled with corporate and individual l
I inexperience, was an apparent reason for procedural and programmatic inadequacies, which resulted in a failure to systematically identify problems and to routinely correct and
[
prevent recurrence of problems.
Licensee personnel stated that the licenses was involved in j
providing direction to the project from the beginning but their experience level restricted the direction. They indicated they had greater involvement in the design arce than in construc-tion.
P:g2 11 cf 25 Licensee inexperience resulted in insufficient and unclear management direction and involvement, as evidenced by the ineffective project management system, and a diffusion of responsibility in the project to inappropriate levels of authority.
For example, engineering changes were made by the AE/CM/C based upon QA/QC or licensee engineer's comments without approvals at appropriate levels of management.
While not suitably involved at higher levels, in some respects the licensee became too involved at lower levels. Licensee personnel became directly involved in the approval chain for design changes and other documents.
This had the effect of unduly restricting work flow--everyone in the chain had veto authority and everyone had to agree to everything. Toward the end of the tenure of the initial AE/CM/C, the licensee stated it had assumed much of the contractor's responsibility for engineering work in an effort to correct the situation.
In effect, the engineering work that was performed was the product of the AE/CM/C and the licensee instead of the product of the AE/CM/C with licensee overview.
2)
Inexperience of the AE/CM/C The AE/CM/C had previous experience as constructor at two nuclear plants but this was the first nuclear plant project in which the AE/CM/C had responsibility for design, construction managesent and construction. The licenses stated that the AE/CM/C also did not fully appreciate the difference in complexity between fossil and nuclear plants and lacked the necessary technical and management skills.
The AE/CM/C inexperience was reflected in their starting and expediting construction before an adequate amount of design and engineering had been completed (estimated at less than 25%).
The small amount of design and engineering completion resulted in an excessive amount of design evolution and rate of design change, according to licensee personnel. Licensee personnel indicated the AE/CM/C was accustomed to having total control of other projects and when the licensee attempted to impose their controls, the AE/CM/C was reluctant to accommodate them.
AE/CM/C inexperience as a corporation resulted in inadequate staffing for the project.
In addition to the limited prior
- nuclear experience of the AE/CM/C as a corporation, the AE/CM/C appointed personnel without prior nuclear construction experience to key project positions. The licensee indicated that first level AE/CM/C management on the project had no prior nuclear experience nor did the principal AE staff. The licensee personnel interviewed indicated that design work i
proceeded slowly and was cumbersone, and that specifications and procedures were inadequate and in an unnecessarily complex format. They said the AE/CM/C lacked systems design capability and provided insufficient support for engineering design work.
According to licensee personnel, QA/QC was accused of engineering the job and talent was not available in engineering l
P 32 12 of 25 to the depth required to turn QA/QC away. Licensee personnel also indicated there was insufficient engineering support for construction and insufficient communication between engineering and construction. They said capabilities that the AE/CN/C did have were channeled into those areas in which they had experience. The constructor had what was characterised as a
" bulldozer mentality," that is the project operation was run by construction activities. Licenses personnel said that engineering efforts were scheduled based upon construction dictates, which led to unrealistic demands on the engineering groups, such as completing an engineering cycle of 18 months in two months.
The AE/CM/C inexperience resulted in insufficient and unclear management direction and involvement. Licenses personnel said that the former AE/CM//C's project management did not adequately define and provide for the interfaces and responsibilities for such functions as QA/QC, engineering design, and construction. As a result, the AE/CN/C did not react timely and effectively to the problems and did not have at their command the management systems to effectively establish the causes of problems and to prevent them from recurring. Licensee personnel indicated quality assurance and quality control were also dominated by construction. Many conflicts occurred between QA/QC and construction and construction generally won. The licensee indicated that as an overall complicating factor, it became clear in the mid-to-late 1970s that nuclear plant construction work in the U.S. was not going to increase as had previouriy been anticipated and it appeared that nuclear construction became a less desirable market area for the AE/CN/C. As a consequence, according to the licensee, the AE/CM/C reduced their cosuittaent to this project, and many of their most capable personnel lef t this project for other nuclear projects or for positions in other industries.
B.
Inadequate Manasement Support for Quality.
Neither the licensee nor the AE/CM/C appeared to have had an ade-quate understanding of quality and quality assurance concepts as they applied to nuclear plant design and construction. The licensee and the AE/CM/C were not use to functioning with quality assurance requirements. The licensee developed its quality assurance program in response to NRC requirements and the AE/CM/C had to rewrite its quality assurance program in order for the licensee to get their construction permit. Although both made commitments to quality, these were insufficiently supported through action. Neither the licensee nor the AE/CM/C staffed their key project positions with appropriately qualified and experienced personnel.
In addition, there was a high turnover rato in AE/CM/C personnel assigned key site positions.
Since 1977, there were six General Managers and seven Site Managers for the project. There was inadequate licensee and AE/CM/C project management to direct the project. The project lacked adequate QA, planning and scheduling and, according to
Pega 13 of 25 licensee personnel, executive understanding of interfaces and responsibilities. Neither organization implemented sound management systems to prevent problems from recurring. The licensee was not appropriately involved in monitoring the total scope and details of activities and did not take effective corrective action to prevent recurrence of problems. The licensee relied heavily on the AE/CM/C and until 1980, did not have its own QC at the site. Until then, all corrective action requests were handled and tracked by the AE/CM/C. According to licensee personnel, corrective action requests were issued and closed out and the problems would recur.
Licensee personnel stated they were aware of problems since 1973 and there was a continuous effort to get them corrected but no one was looking at the big picture.
Allegations were continually made regarding harassment and intimidation of site QA/QC personnel and lack of support by inspection supervision. Audits to provide feedback to management concerning the effectiveness of the QA Program were improperly implemented and at times not performed.
Audit reports were not issued beyond the level of the audited organiration. No effective program had been implemented to perform trend analysis of nonconformance reports. There was an overall lack of aggressive implementation of effective QA/QC programs.
The licensee indicated that the AE/CM/C did not follow the principle of QA/QC independence from cost and schedule and as a consequence, construction supervisors dominated the QA/QC functions in the field.
The AE/CM/C also did not take effective corrective action to prevent recurrence of problems.
Published AE/CM/C policy emphasized minimizing cost and maintaining schedule and stated that QC inspector's decisions were subject to question, challenge and reversal.
The long chain of command between onsite QA functions and top licensee management resulted in a filtering of information and introduced inefficiencies into the decision making and implementing processes.
Although licensee personnel stated that management was involved in the project from the start, their actions resulted in low visibility for management support for quality, which tended to be interpretated as a lack of backing for quality from top management.
C.
Shortcomings in NRC's Licensing and Inspection Practices Opinions expressed by both regional and headquarters NRC personnel, as well as licensee personnel, suggest that in some respects the NRC could have been more effective in preventing or taking action earlier on the problems that occurred at this project. A recurrent theme in the interviews was that the NRC licensing process does not do enough to address the ability and experience of the project team as it relates to managing a nuclear construction project. The inspection procens also tends to ignore management issues prospec-tively and tends not to address management issues until major technical, programmatic or quality problems have developed. The irregular presence of NRC inspectors at the job was cited as a
}
Pegs 14 cf 25 problem, along with the observation by members of the licensee staff that there is a great lack of consistency among the various inspec-tors as to their capabilities, their interests, and the depth to 4
which they pursue problems. The case study team discovered that the approach used by NRC in identifying and dealing with problems of the nature experienced by the Case D project was cumbersome and required excessive amounts of time.
In effect, NRC's approach was to allow troublesome situations in construction to progress to toa point at which a case could be built for taking the drastic action represented by a Show Cause Order. According to licensee personnel, some of the technical issues involving NRC headquarters required up to two years to resolve, and then the resolutions were too generic to be specifically useful.
Licensee personnel indicated that the rapid proliferation of regulations during the mid 1970s resulted in uncertainty in the regulatory process and a constantly changing target. Compliance with other (non-NRC) government requirements, such as not being able to refuse employment on the basis of drug or alcohol use or convictions of felonies, was also indicated by licensee personnel to have posed problems.
It should be noted that although the project team of Case D experienced difficulty in adjusting to regulatory changes, others in the nuclear industry were apparently able to adjust.
Some members of the licensee staff strongly expressed the opinion that NRC policy decisions were based more on political than on technical considerations. They indicated their belief that the NRC personnel involved in the special investigation (79-19) were con-cerned with expediency and avoiding controversy at the risk of doing an inadequate technical job.
The NRC identified problems before the special investigation that led to the Show Cause Order and the licensee had stopped construction in several areas of concern.
It was felt that more inspection and forceful action by the NRC earlier in construction would have resulted in earlier detection of problems and more effective corrective action, which may have avoided the need for a Show Cause Order.
Contrary to these views, several licensee management peraonnel expressed the view that the NRC had done the licensee a great favor in imposing the Show Cause Order because it made the licensee appreciate the significance of their problems and resulted in strong corrective actions.
Licensee personnel made a strong plea to depoliticize the NRC regulatory process and to replace the Commission with a single administrator. They expressed the view that, in the absence of such action coupled with a consolidation of the nuclear purview in Congress, further licensing and construction of nuclear plants would be virtually impossible.
I D.
Inability of Project Team to Adjust to Changing Nuclear Power Environment During the design and construction of this project, the environment surrounding nuclear power in the U.S. underwent drastic changes.
j 1
Psgs 15 of 25 One of the most obvious and significant of these changes was the proliferation of regulations imposed by the NRC. Other important changes occurred, however. One stemming in part from the increase in regulatory requirements has resulted in the necessity of a different role for the owner utility in nuclear plant design and construction; there has been a fundamental change in the level and degree of licensee management involvement in design and construction.
Before the early 1970s, nuclear plants were often constructed with minimal licensee involvement on a turn-key basis. There are factors other than changing requirements that dictate the need for greater utility involvement in nuclear construction projects. With the increased costs and complexity of nuclear plants, overall individual project costs have soared near the full capitalized value of some utilities. This alone dictates a more active concern by licensee management to all phases of nuclear plant construction. A licensee is also drawn into more active involvement because of the heightened political concerns surrounding nuclear power that developed during the 1970s. Projections of declining energy needs, increasing costs of nuclear plants, and increasing interest rates made funding plant construction far more difficult.
An increase in public skepticism and more active involvement of intervenors also occurred during the 1970s.
Incidents within the industry, such as Three-Mile-Island and Brown's Ferry, have increased public concerned and resulted in significant regulatory change and design modifications.
All of these changes, coming in rapid succession and imposed upon an inexperienced licensee-AE/CM/C coalition created a situation in which the licensee and the AE/CM /C were not effective in fully recognizing the significance of the changes as they occurred, in keeping up with changes, and in meeting the requirements of nuclear plant construction as they evolved during the 1970s.
Several personnel having prior nuclear design and construction experience and who have been assigned to the project af ter quality-related problems developed, expressed the following opinions:
1.
There was a lack of clear management direction both on the part of the licensee and the engineer-contractor.
2.
There was an incapacitating fusion of responsibility between and within each company.
3.
Neither company had sufficient confidence in the other.
4.
The document control systems of both concerns were inadequate.
5.
Design evolution and the rate of design change were excessive.
6.
Neither company had sufficient prior experience.
III. REMEDIAL ACTIONS TAKEN TO CORRECT CONSTRUCTION QUALITY PROBLEMS Following the Show Cause Order and analysis of the project by an indepen-dent company, the main actions taken by the licensee to remedy the prob-lems that occurred were to replace the AE/CM/C with two firms, one having architect engineer and construction management responsibility and the second having construction responsibility, and to increase the nuclear experience of its own staff through hireing.
Both of these latter two firms have great depth of experience in nuclear plant design and con-struction. The system now being implemented is founded upon three general elements that characterize a good quality program. These are:
Pegs 16 of 25 1.
Considerable effort and thought devoted to planning.
2.
An experienced project management team which includes balanced representation from the licensee, AE/CM and construc*.or.
3.
Easily understandable and comprehensive procedures.
Significant features of the new system are as follows:
1.
The constructor has cognizance for QA/QC but its QA reports to corporate headquarters offsite. At the time of the case study, the constructor QC organization consisted of approximately 140 people.
2.
The AE/CM maintains an audit function that reviews all QA/QC done by the constructor. The AE/CM maintains a staff of 60 QA/QC personnel.
3.
The licensee is very much involved but in an overview and monitoring role. They maintain a staff of about 30 QA/QC personnel, which is twice as many as were involved before 1981. This group monitors all QA/QC (design, construction, purchasing, document control, and records management) performed by the constructor and audited by the AE/CM.
Internally, the licensee has made several changes. These include, in addition to shortening the chain of command, a highly increased and visible emphasis on quality. The licensee hired several well-qualified, nuclear design and construction experienced personnel, added their own quality control function at the construction site, personally reviewed and approved the hiring of many of the AE/CM and constructor personnel, reviewed and required modifications to the AE/CM and constructor QA/QC programs, and provided considerable assistance to both the AE/CM and constructor in hiring and relocating experienced personnel. Under the new system, the licensee Executive Vice President has been given full-time responsibility for the project and has removed intervening layers of management that previously separated him, as the responsible corporate officer, from the site QA function. The Executive Vice President meets with site QA personnel for detailed discussions at least monthly. The new system appears to have many redundancies, with the resultant diffusion of responsibilities. The licensee and present AE/CM and constructor management personnel have indicated that the new system, although still evolving and not fully tried, appears to be workable.
IV. GENERIC IMPLICATIONS Based on the information reviewed and analyzed by the case study team, several possible generic implications, or lessons, emerge. These are highlighted in each of the case studies to provide input and to help form overall conclusions concerning factors that constitute important elements in nuclear plant construction quality.
From the information considered by the case study team, several possible such implications emerged from this study:
A.
Understanding of Nuclear Project and its Implications It is essential that licensees and contractors understand the differences in complexity between construction of fossil and nuclear
Pags 17 of 25 l
power plants and the implications associated with the difference.
Nuclear plant construction is more complex and demanding than fossil plant construction and licenseca and contractors must provide a strong, experienced organization possessing the necessary management skills to effectively sanage the project. Although experienced contractors for design and construction activities may be involved, the licensee is fundamentally responsible for the safe construction of the plant. Given the complexity of these projects and the major implications of failures, the licensee must, in his own self de-fense, implement effective review, surveillance and audit of all phases of activity. Stated commitments to quality must be supported by positive actions.
B.
Nuclear Experience Vital Today The complexity, cost, regulation and safety requirements of nuclear plant construction makes it vital that the project team have prior nuclear design and construction experience before embarking on a new proj ect. Not all corporate members of the project team must have prior nuclear experience, but collectively, as organizations and individually, they must have it.
This experience must be manifest at all working and management levels.
In the absence of this kind of competence and experience, a licensee will be unable to recognize when problems are occurring, accurately assess their importance and devise approaches to solve them effectively.
In the opinion of licensee personnel interviewed in this case study, the NRC would be doing the utility that lacked this capability a great favor by denying them a construction permit.
C.
Management Awareness and Involvement Necessary The need for management involvement and support at the highest levels in nuclear plant construction is well illustrated by this case. Top-level involvement is important whether the licensee and its contractors have nuclear experience or not.
High-level involve-ment is required from the licensee, the AE, the CM, the constructor, and other contractors involved in the project to assure timely recognition and effective resolution of problems. A well-defined and effective decision making process followed by everyone involved eliminates the reality as well as the perception of indecision and provides needed direction. The requisite commitment to project objectives must be incorporated into all working guidelines as clearly stated approaches that are easily understood at all manage-ment and working levels.
Licensee management must interface actively with contractor manage-ment to assure that the licensee becomes aware of problems and their impact in a timely way so that their solutions can be effectively initiated. Management support for quality must conspicuously emanate from the tops of the licensee and contractor organizations and permeate all phases of activity.
If top-level management fails to visibly support quality, the perceived importance of quality diminishes and introduces tendencies to emphasize cost and schedule goals to the detriment of quality.
Management support must be
i P g2 18 of 25 implemented as well as verbalized, preferably through active and frequent involvement of top management with project management functions, including quality.
D.
Need Qualified Capable People Qualified people with indepth nuclear design and construction experience are required at all levels for the construction project to be successful.
Qualifications extend to education, training, and experience acquired on previous projects and should be supplemented by training, by the licensee or his contractor.
Personnel quali-fication requirements extend to the licensee as well as to the contractors. Without suitable qualifications, licensee employees will not be able to adequately assess the contractors' work, or the contractors their own work.
E.
Ultimate Responsibility Retained by Licensee The licensee is ultimately responsible for a nuclear plant's con-struction and safe operation. Although the licensee may contract design, construction, and other activities to other organizations and may impose responsibilities on these organizations for complying with regulatory requirements, construction permits and operating licenses are issued to utilities and not to their contractors. The licensee has ultimate responsibility for successful construction and safe plant operation.
In its self defense, it must implement adequate controls to verify that its contractors comply with all requirements.
V.
IMPLICATIONS OF CASE STUDY D FOR NRC QA INITIATIVES NRC has under way several initiatives which are designed to establish additional confidence in the quality of design and construction activities, to improve the management control of quality, and/or to improve NRC's capability to evaluate the implementation of licensee programs. The initiatives are described in the NRC staff paper SECT 82-352, " Assurance of Quality," and subsequent correspondence between the Commission and the NRC staff. One of the purposes of this case study is to provide feedback on the relevance of the various initiatives to the Case D licensee's nuclear construction project. The following paragraphs discuss whether each initiative, had it been an ongoing activity at the time of CP issuance and up to NRC confirmation of the magnitude and nature of this projects design and construction problems, would have made a dif ference, that is, would the initiative have prevented or at least mitigated the design and construction problems that were discussed earlier. A more complete discussion of the scope and details of the various NRC-QA initiatives may be found in SECY 82-352 and SECY 83-32 "First quarterly report on Implementation of the Quality Assurance Initiatives." Most of these initiatives were discussed with the senior management of the licensee, his current AE/CM, and his current constructor.
No discussions were held with personnel from the original AE/CM/C.
Pcg2 19 cf 25 A.
Measures for Near-Term Operating Licenses (NTOL) 1.
Licensee self-evaluation - not applicable This initiative applies to actions that would occur when the licensee is in the process of receiving his operating license.
It requires that the licensee examine selected
~~~
portions of the engineering design or construction.
Licensee self-evaluation permits an evaluation of the project from beginning to end and would permit the Chief Executive Officer to state that the plant had been built according to its commitments.
In the Case D situation, construction had not proceeded to the point where a self-evaluation would have been appropriate. Therefore, this initiative would have had no effect.
2.
Regional evaluation - no The licensee regional evaluation is an action that would occur when the licensee is in the process of receiving its operating licensee.
For Case D, the operating license phase was well beyond the point in time where the problems discussed previously occurred.
As a result, this initia-tive would have had no effect in Case D.
3.
Independent Design Verification Program (1DVP) - no The licensee IDVP is an action that occurs when the licensee is in the process of receiving its operating license.
Construction had not proceeded to this point and, as a consequence, this initiative would have had no effect. Had such a verification been performed for this project before serious problems were discovered, it may have been effective in revealing the inadequate state of design completion and may have identified other engineering deficiencies. The licensee indicated that a continuous review of engineering work should be performed--not just a review at the end of construction and before the operating license is granted.
B.
Industry Initiatives 1.
INPO " Construction" audits - yes l
l INPO construction assessments potentially would have I
identified procedural and programmatic inadequacies that were not evident to the inexperienced licensee or his contractor.
l t
e Pcg3 20 of 25 2.
Utility Self-Initiated Evaluation Using INPO Method - maybe Applying the INPO criteria would likely have identified inadequacies; however, because of the nuclear inexperience of the licensee and his contractors, they may not have correctly characterized the extent, nature and seriousness of them, and probably would not have identified adequate corrective action to be taken in response to these re-vealed deficiencies.
C.
NRC Construction Inspection Program 1.
Revised procedures and increased resources - yes Revision of NRC inspection procedures to shift emphasis from review of records to work observation and inspection and an increase in NRC resources allocated to construction inspection may have detected the problems at an earlier stage of construction.
In the exit conference, the licensee suggested that a resident inspector should be present onsite for each discipline and that the NRC needs more resident inspectors or roving inspection teams to support all disciplines.
Licensee personnel also indicated that the competence of NRC staff must be upgraded to be equal to that of the utilities and the architect engineers. The licensee felt that the resident inspectorte presence onsite was very important on a day-by-day basis.
2.
Construction Appraisal Team (CAT) Inspection - (emphasis on management) yes A CAT inspection conducted before the Show Cause Order would likely have detected programmatic and procedural inadequacies in the project, pointing to shortcomings in project management. Because of the intensity of these special team inspections, they yield a comprehensive overview of a project sooner than the NRC inspection program. A CAT would have been particularly valuable in 1
determining the construction inadequacies characteristic of the Case D project. Because CAT inspections are performed only at about four plants under construction per year, this or any particular plant may not have been selected. Present CAT inspections focus on the quality of hardware, and on this basis, inferences may be drawn about the quality of management.
A modified CAT that addressed the management issues more directly would have been even more beneficial.
P:ga 21 of 25 3.
Integrated Design Inspection - yes The integrated design inspection is an activity that would normally occur while the licensee is receiving its operat-ing license.
It could be done earlier, however.
Such a design inspection would likely have revealed inadequacies in design control and the lack of systems engineering in Case D.
It also likely would have showed that insuffi-cient engineering support was available for construction and that engineering design was not sufficiently far along to permit construction to proceed.
4.
Evaluation of Reported Information - maybe This initiative would computerize 10 CFR 50-55c and Part 21 reports, facilitating trend and other analyses of these event reports.
Such an analysis would have provided an additional cross-check on the quality-related operations at the construction site.
Evaluation of reported inforra-tion from all licensees could have been effective in alerting the NRC or licensee upper management in Case D to impending problems.
D.
Designated Representatives - probably not The FAA designated representative program utilizes contractor or licensee individuals to represent the regulatory agency in
'. the field. This activity is generally considered to apply to the production process but could be construed to apply to design or other manufacturing activities as well.
If NRC were to adopt such a program, the designated representative would be an individual employed by the licensee, architect engineer, constructor or other firm external to the NRC.
In Case D, the inexperience of the licensee and their original AE/CM and the inability of the Licensee to effectively correct procedural and programmatic deficiencies suggest that a designated representa-tive chosen from their ranks would have had little effect in preventing the quality problems that occurred.
The licensee's view of designated representatives'as that they would erode w
responsibility for quality at the working level and may result in NRC assuming an inordinate share of the responsibility for quality.
E.
Management Initiatives 1.
Seminars - yes Seminars such as those conducted by the NRC or INPO in the past as well as presentations by utility executives who have had construction-related problems would have been helpful in increasing the licensees' management awareness of the importance of project management capability, prior nuclear experience, recognition of symptoms and quality control measures in construction.
Such presentations l
)
Pcas 22 ef 25 would have been particularly useful for the Case D project because of the inexperience of both the licensee and the original AE/CM/C and could have resulted both in more active and effective involvement of the licensee and an improved approach on the part of the AE/CM/C towards quality control and quality assurance.
2.
Qualifications / Certifications of QA/QC personnel -
probably not The working level QA/QC personnel involved with con-struction were basically qualified. They did identify problems but were not effective in correcting these because of the procedural and programmatic shortcomings of the licensee and the AE/CM/C. A greater degree of quali-fication or certification of QA/QC personnel may have resulted in earlier and more indepth awareness of the problems, but it is doubtful whether this awareness would have been effective in solving the problems.
i 3.
Craftsmenship - probably not The licensee indicated the AE/CM/C was a non-union employer with a reputation as a training ground for craftsmen. They paid relatively low wages but provided individuals the opportunity to gain the experience they needed to move into unionized organizations where wages were higher.
However, it does not appear that this was an l
important factor in the quality problems. The inability of the licensee and the AE/CM/C to effectively manage the project, the inadequacy of specifications and procedures and design completion not being adequately ahead of construction were much larger contributing factors.
VI. IMPLICATIONS OF CASE STUDY D FOR THE FORD AMENDMENT ALTERNATIVES Section 13 to the NRC's FY 1982-1983 Authorisation Bill requires NRC to
(
study existing and alternative programs for improving quality assurance and quality control at nuclear power plants under construction. This section, called the Ford Amendment, requires NRC to look in particular at I
the feasibility and ef ficacy of five specific alternative program con-capts. As part of this analysis, each alternative concept was evaluated as to whether it would have made a difference in the Case D project had it been in place at the time of the issuance of the construction permit up to issuance of the Show Cause Order.
Each alternative was discussed with senior licensee, AE/CM, and constructor management personnel. No discussions were held with personnel from the original AE/CM/C. The results of the discussions are given below.
Pcg3 23 of 25 A.
More Prescriptive, Architectural and Engineering Criteria - perhaps The Authorization Act requires NRC to evaluate the following alter-native:
13(b)1 - Adopting a more prescriptive approach to defining principal AE criteria for constructing commercial nuclear power plants that would be a basis for quality assurance and quality control, inspection, and enforcement actions.
The design and construction problems were not affected by the degree of prescriptiveness of the AE criteria. More prescriptive AE criteria would not have corrected the procedural, programmatic and management deficiencies noted in case D.
However, being prescrip-tive in the degree of engineering completion prior to the start of construction may have eliminated insufficient engineering support for construction and the scheduling of engineering efforts based upon dictates from construction. Licensee personnel indicated more prescriptive criteria would be good provided the methods of comply-ing with the criteria were left to the licensee, AE/CM, and constructor. To make decisions, the requirements must be clearly known. A need for clearly described standard construction method-ologies that would prevent known problems from recurring at other sites was also expressed.
B.
Conditioning the Construction Permit on the Applicants' Demon-stration of His Ability to Manage an Effective Quality Aasurance Program - yes 1
The Authorization Act requires NRC to evaluate the following alter-native:
l 13(b)2 - Requiring as a condition of the issuance of construction permits for commercial nuclear plants that the licensee demonstrate the capability of independently managing the effective performance of all quality assurance and quality control responsibilities for the plant.
Imposing this requirement would almost certainly have identified the lack of nuclear experience, lack of management understanding, and the programmatic and procedural inadequacies that affected the project, before initiation of construction.
It should have resulted in shortening the chain of command, in identifying the need for more experienced personnel at all levels, and in demonstrating the inherent lack of prior nuclear experience.
C.
Audits, inspections or Evaluations by Associations of Professionals Having Expertise in Appropriate Areas - Manatement Audits - yes The Authorization Act requires NRC to evaluate the following alter-natives 13(b)3 - Encouraging and obtaining effective evaluations, in-spections, or audits of consorcial nuclear power plant construction by independent industry or institutional organizations based on best experience and practices.
s
=
Page 24 of 25 Overviews of this nature should have been effective in identifying procedural and programmatic inadequacies and should have resulted in earlier management awareness of the existence and magnitude of problems.
It should be noted that the licensee did have evaluations performed during the early phases of the project to determine construction status. Later evaluations were performed to determine quality-related information and design adequacy. The licensee took action based upon these evaluations and likely would have taken action based upon inadequacies and deficiencies discovered by associations of professionals.
In mid-1983 the licensee had an l
audit of design and construction performed, using INPO criteria, by a team of 18 individuals representing five different organizations.
D.
Improvement of NRC's QA Program - yes The Authorization Act requires NRC to evaluate the following activ-ities:
1 13(b)4 - Re-examining the Commissions' organization and method for l
quality assurance development, review, and inspection with the objective of deriving improvements in the agency's programs.
Requiring that the licensee demonstrate his ability to manage a nuclear construction project and assessing the demonstration of implementation of licensee and AE/CM/C QA/QC programs would have
)
been helpful in earlier identification of the shortcomings experi-enced by the Case D project. The approval of a licensees' program
.. description in a PSAR is not sufficient evidence that he has fulfilled his commitments. The fact that a program complying with requirements has been written and documented in the PSAR says little about the subsequent successful implementation of that program.
Earlier appointment of a resident inspector (a full-time resident inspector arrived at the Case D site in 1979--four years after construction began) would have been helpful.
In particular, involv-ing an inspection team with expertise in each required discipline
)
earlier in the project would have been beneficial.
]
The Licensee indicated proliferation of regulations as a factor in causing the case D problems. They indicated that altering the NRC program to stabilize the regulatory base would be beneficial in avoiding such problems in the future.
E.
Conditioning the Construction Permit on the Applicant's Commitments to Submit to Third Party Audits of its Quality Assurance Program - yes The Authorization Act requires NRC to evaluate the following alter-native:
13(b)5 - Requiring as a condition of the issuance of construction permits for commercial nuclear power plants that the applicant enter into contracts or make other arrangements with an independent inspector for auditing quality assurance responsibilities for the purposes of verifying quality assurance performance.
An independent inspector is a third party and has no responsibilities for the design and construction of the plant.
\\
\\
Page 25 of 25 This alternative as it applies to Case D is closely related to Alternative 13(b)3 discussed above.
If this initiative had been in place, problems may have been detected earlier and the errors that occurred may not have developed into a project breakdown. Audits of this type, had they been in place at the time, should have iden-tified programmatic and certain procedural inadequacies and should have resulted in earlier upper management awareness of the problems.
The licensee did respond favorably to self-initiated evaluations and i
could be expected to respond similarly to inadequacies and defi-ciencies reported by an independent auditor. A comprehensive audit by a qualified independent inspection team should have identified the incompatibilities that arose between design, construction and quality control.
i
~~
-.e.,
,