ML20217B084

Forwards AEOD Diagnostic Evaluation Team Rept for Plant for Info.Rept Documents Team Findings & Conclusions & Includes Discussions of Licensee Strengths & Weaknesses

ML20217B084
Person / Time
Site:	Oyster Creek
Issue date:	03/05/1991
From:	Taylor J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
To:	Carr, Curtiss, Rogers NRC COMMISSION (OCM)
References
NUDOCS 9103080298
Download: ML20217B084 (2)

Text

. - . . . _ - .- - . - _ _ _ _ - ._ - . _. -.- . -

( s MAR 051901 MEMORANDUM TOR: Chairman Carr Commissioner Rogers Commissioner Curtiss Commissioner Remick FROM: James M. Taylor Executive Director for Operations

SUBJECT:

OYSTER CREEK NUCLEAR GENERATING STATION DIAGNOSTIC EVALVATION TEAM REPORT Enclosed for your information is a copy of the AE00 diagnostic evaluation team report for the Oyster Creek Nuclear Generating Station. This evaluation was initiated following discussions during the NRC senior managers meeting in June 1990. The need for a diagnostic evaluation was based on questions related to inconsistencies in the licensee's performance and protracted periods to implement improvements at Oyster Creek. The report documents the team's findings and conclusions, and includes discussions of the licensee's strengths and weaknesses, follow u) staff actions resulting from the diagnostic evaluation have aircady seen forwarded to Region 1 and NRR.

I would be pleased to provide any clarification or further information that you may require.

Original Signed By:

kmes M. Taylor James M. Taylor Executive Director for Operations

Enclosure:

Diagnostic Evaluation Team Report l for the Oyster Creek Nuclear Generating Station l cc w/ encl:

SECY OGC l ACRS l

Contact:

Edward L. Jordan, AEOD 49-24848 p Distribution (w/o enc 1)

See page 2

• See Previous Concurrence f T[ -

DEllB:00A DElIB:00A D0A AE0D E Eb

• SRubin *CGrimes RLSpeYsard O ass /Eldordan J. zek JM" lor Ol/2S/91 01/25/91 02/,)//91 (2/ /9 03k ./91 03 ' /91

// /91 f

File Name: G:COMM0DET.SD L

\ ggyg h EsR318ss!Za88881, 1l P PDR

( i 2

Qistrit'ution w/o enclosure:

DCS NRC PDR

' at al PDR EM R/F AEOD R/F DOA R/F DEllB R/F JTaylor JSniezek JBlaha EJordan Dross RSpessard CGrimes SRubin RLloyd TMurley FMiraglia TMartin, R1 SEbneter, Ri!

ADavis, Rlli RMartin, RIV JMartin, RV WBateman, OEDO JLieberman, OE HDenton, GPA JFouchard, PA WHehl, RI EWenzinger, R1 WRuland, RI Ecollins, R1 JPartlow SVarga EGreenman JStolz ADromerick

1

< i

• 1. y:,6 as:

k a UNiltD sTAf ts

[ " NUCLEAR REGULATORY COMMISSION

$ W A SHING T ON, D. C, 20666 s

% ..... TE9141991 Docket No: 50-219 General Public Utilities Nuclear Corporation ATTN: Philip R. Clark, President and Chief Executive Officer

.One Upper Pond Road Parsippany, New Jersey 07054

Dear Mr. Clark:

This letter forwards the Diagnostic Evaluation Team Report for Oyster Creek Nuclear Generating Station.

The team assessed the effectiveness of licensed activities performed by GPUN in achieving the safe operation of Oyster Creek and determined the causes of identified performance deficiencies. The team of 15 evaluators, led by an NRC manager, was onsite at the Oyster Creek station during November 5_through 16 and December.3 through 7. Team members also conducted reviews at the corporate offices in Parsippany, New Jersey during these periods. The team's findings were discussed with you and your staff during an exit meeting held at the plant site on January 18, 1991.

To achieve an independent perspective, the team was staffed with individuals

-who did not have responsibility for the regulation of Oyster Creek. Safety ,

performance was evaluated in the areas of operations, maintenance, and testing, including the influence of management and engineering support.

The team concluded that, at the time of the evaluation, C .er Creek was in a .

state of transition. _ GPUN was still in the process of recovering the material condition of the plant and implementinq a variety of improvement initiatives to enhance future performance. Manyoftheperformancestrengthsobservedat Oyster Creek were primarily attributable to the experience and skills of individual personnel and managers. As a result of the effective conduct of plant operations and the improvements observed in the plant's performance indicators, continued orogress was anticipated; however, the pace of .

improvement was being hampered by r,everal- weaknesses involving work. practices which were identified by the team.

The' team concluded that some of the weaknesses in performance could be attributed to the root causes described on Page V of the Executive Summary "

-and in Section 3 of_ the report. These postulated root causes should be of- '

special interest to senior nuclear managers as they direct your Plan for Excellence in Performance to further improve Oyster Creek.

b/9'2#27~TWN l

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

]

i General Public Utilities Nuclear .

i The slow rate of impimentation of the varied and numerous improvement initiatives indicated a need for increased management attention and support at all levels in the company, to address root causes and to produce a more ,

timely resolution of plant deficiencies. GPUN had previously identified many of the deficiencies found by the team. This fact reflects positively upon your self-assessment capability and_ your acknowledgement of the need for i further improvements.

I urge CPUN management to carefully review the enclosed report, with special emphasis on the areas requiring additional management attention. I request that, within 60 days of the date of this letter, you describe the actions you intend to take to address the root causes for the deficiencies in work practices. In particular, you should describe any changes to the Plan for Excellence in Performance which could enhance accountability and the timeliness of your future improvement initiatives, in accordance with 10 CFR 2.790(a), a copy of this letter and the enclosure will be placed in the NRC Public Document Ronm. Should you have any questions concerning this evaluation, we would be pleased to discuss them with you.

7 Sincerely,

/

esM.ksTa r ecutive D rector <

for Operations

Enclosure:

Diagnostic Evaluation Team Report '

for Oyster Creek Nuclear Generating Station i

cc w/enci' See Page 3 1

1

_.a._.._______._._ _ _ _ _____-._.._._ . _ _ ,

i , t ,,

General Public Utilities Nuclear .

3 cc w/encli  ;

J. J. Barton,_ Director j

Oyster Creek Nuclear Generating Station -

E. L. Blake, Jr. i Shaw Pittman, Potts and Trowbride J. B. Liberman, Esquire Bishop, Liberman, Cook et al.

BWR Licensing Manager GPU Nuclear Corporation Mayor, Lacey Township Licensing Manager Oyster Creek Nuclear Generating Station ,

Commissioner New Jersey Department of Energy Kent Tosch New Jersey Department of Environment Protection 1

1 1

t T

't i.

e en y t=-c-e*, 'tia---*Aetm? ev, pew- 3,*n vwper-g ww ei F#w-- NTtvyt trww wS tuth. __m w m rw ep w a m. t- yt -- p M t- weP m*+e'Wi-* c tP-'("M

8 5

OFFICE FOR ANALYSIS AND EVALUATION OF OPERATIONAL DATA DIVISION OF OPERATIONAL ASSESSMENT Licensee GPU Nuclear Corporation Facilityt Oyster Creek Nuclear Generating Station <

Location: Ocean County, New Jersey Docket No 50-219 Evaluation Period: November 5 through December 7, 1990 Team Manager: Christopher I. Grimes Administrative Assistants Laurette Moorin Team Members: Patrick W. Baranowsky Bruce A. Breslau Anthony J. D'Angelo Dennis L. DuBois Robert G. Freeman Patrick L. Hiland Michael A. Junge John V. Kauffman Ronald L. Lloyd Subinoy Mazumdar Thomas R. Staker Hershell A. Walker Consultants: Sonja B. Haber Spyros A. Traitoros Herbert J. Worsham Submitted By: $

Christophur I. Grimes, Manager Date Oyster Creek Diagnostic Evaluation Teair Approved By: - '87 M f/

? /

'Tdward L. rdan, Director office f /Dht6 nalysis and Evaluation of ope lonal Data l

l 1

S fY l

t f(.O']/2fO{~(* l'

- - - - - - - - - - - - - - - - ~ ~~ ~~ " ~ ' ~ ~ ~ ~ ~

. - ~ . - - - - _ - - - __ -

EXECUTIVE SUMMI.RY A diagnostic evaluation was conducted by the Nuclear Regulatory Commission (NRC) to assess the performance of GPU-Nuclear in assuring the safe operation of the oyster Creek Nuclear Generating Station. The evaluation was requested by the NRC Executive Director for Operations to obtain additional information so that NRC management could make informed decisions about future regulatory activities at Oyster Creek because of concerns about inconsistent performance and protracted periods to implement improvements.

A team of 15 evaluators and an administrative assistant, led by an NRC manager, conducted onsite evaluations at Oyster Creek and the GPU-Nuclear offices in Parsippany, New Jersey, during November 5 through November 16 and December 3 through December 7, 1990. The areas covered during the evaluation included operations and training, maintenance and testing, engineering and technical support, and management effectiveness and corrective actions.

Conduct of plant operation was good, and training facility improvements were evident. The operation department had an effective plan to ensure adequate staffing. Plant operators had a positive attitude, were professional and attentive, were knowledgeable of the plant and plant conditions, and conducted effective shift turnovers. Communication between the operation and the maintenance departments was good, while communication with training and radiological controls departments showed signs of " growing pains" as the relationship between these departments continued to evolve. The material condition of the control room

. was good, operator performance was hampered by having to work l

around some equipment-problems. Operations improvement initiatives were having a positive effect on the conduct of plant

' operation, particularly the equipment labeling program, the operator concerns program, the professionalism program, and the expectations and standards program. Plant operating procedures were improving; however, a schedule had not been developed for L

rewriting all the operating procedures to the new format.

The maintenance and testing practices were adequate and improving., Current improvement initiatives were directed mostly towards recovery of the material condition of the plant, which was improving as evidenced by the condition of plant housekeeping. The scope of preventive maintenance was relatively 11

a .. ,~ . ,, .

I t g low, although preventive maintenance was increasing as reliability-contered maintenance techniques were applied to troubicsome equipment.

poorly controlled. Skipped preventive maintenance tasks were A new system was being implemented to improve the prioritization of maintenance tasks. Efforts to make substantive programmatic improvements to maintenance, including implementation of life of system maintenance plans and equipment failure trending, were impeded because these initiatives lacked formalization with a clear definition of programmatic goals and objectives, implementation schedules, and dedicated resources.

Management only recently acknowledged the need for dedicated resources to implement maintenance improvement initiatives. A central computer program provided effective capability to generate maintenance work packages and equipment failure histories, but was hampered by the quality of the data base, data retrievability, and training weaknesses.

The quality of engineering and technical support varied as evidenced in several case studies performed by the team. Tasks perceived to be important were performed well. However, the performance on other tasks often lacked rigor, lacked inquisitiveness, or failed to clearly identify the failure mechanism. Further, there was no systematic means of determining the operability or functional capability of equipment with deficiencies necessary. to establish how soon corrective actions were Engineering roles and responsibilities appeared to be understood, and appropriate technical resources were available when the need was identified. Examples of good engineering work were identified, including the development of a comprehensive standard for thermal overload conditions and the design of modifications for isolation condenser piping. Examples of the lack of rigor and lack of understanding failure mechanisms incluoed line, One evaluacions of leakage in the feedwater minimum-flow lines, binding valves in the isolation condenser condensate return and thermal effects on the starting batteries for the diesels.

While the licensee had determined that the fuel oil tank needed replacement because of corrosion, no evaluation was performed to establish the structural capability of the tank in the intervenJng time or how soon the tank replacement was needed.

Site and corporate engineering departments began developing a new procedure for conducting graded root-cause evaluations based on their critique of the minimum-flow line analysis.

l 111 1

1 4 i l 9 1

Hanagement offectiveness was improving but still had weaknesses.

The plan for excellence in performance identified important goals and objectives that relate to most performance problems. Ample staffing and funding were availabin to support plant activitics, and relations with the union had improved substantially.

Considerable emphasis was placed on employee morale and development, communication horizontally and vertically in the organization was improving, performance indicatore were used effectively; however, at least one indicator, " rework," was misleading. Work tracking, which was performed at the department level, was not always prioritized or tracked well; many of the implementation schedules in the excellence plan had slipped.

There were a variety of means to ensure wat deficient conditions were identified. However, followup and closcout of some deficient conditions, including audit findings and material condition issues, were often untimely or superficial because of poor accountability, the complexity of corrective action systems, and weak independent verification of the resolution of problems.

Similarly, improvement initiatives were readily developed to address performance problems, but poor accountability and the complexity of work tracking hampered timely and effective impicmentation.

The team concluded that, at the time of the evaluation, Oyster creek was in a state of transition; the licensee was still in the process of recovering the material condition of the plant from the deterioration resulting from minimal maintenance during the early history of the plant and implementing a variety of improvement initiatives to enhance future performance. Many of the strengths observed in the licensee's performance were primarily attributable to the experience and skills of individus1 personnel and managers. As a result of the icproving performance trends and the effective conduct of plant operations, continued improvements were anticipated, although the pace of improvement would continue to be hampered by the weaknesses identified in work practices.

The weaknesses in the licensee's current performance generally applied to those tasks that were not perceived to be important to management and were attributable to the following root causes iv

l e.

s .

1 l

l (1) Weak administrative control of work because of the lack of an infrastructure for the systematic completion of work.

This weakness was reficcted in poor accountability, compicx and inconsistent methods for tracking and prioritizing work, complex corrective action systems, reactive management practicos, and the lack of a systematic means to determine when corrective action was necessary.

(2) Weak supervision and independent verification had caused some inadequate evaluations and poorly-reasoned decisions.

Management philosophy strived for decisions at the lowest level; however, there was no evidence that decisions were routinely challenged by supervisors or quality assurance.

Thus, the licensee relied heavily on the skill and experience of the personnel.

(3) There was a general lack of rigor and inquisitiveness in the licensee's work practices that hampered the ability to achieve consistently effective solutions to problems. The quality of work depended on the perceived importance of the task and there was a willingness to accept ready solutions to problems.

v

,e i e I

! TABLE OF CONTENTS I

run EXECUTIVE SUKKARY . . . . . . . . . . . . . . . . . . . . . . 11 1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Scope and Objectives . . . . . . . . . . . . . . . . . . 1 1.3 Methodology . . . . . . . . . . . . . . . . . . . . . . 2 1.4 Facility Description . . . . . . . . . . . . . . . . . . 3 1.5 Facility History . . . . . . . . . . . . . . . . . . . . 3 2 EVALUATION RESULTS . . . . . . . . . . . . . . . . . . . 5 2.1 Operations and operator Training . . . . . . . . . . . . 5 2.2 Maintenance and Testing . . . . . . . . . . . . . . . . 11 2.3 Engineering and Technical Support . . . . . . . . . . . 17 2.4 Hanagement Effectiveness and Corrective Action . . . . . 28 3 ROOT-CAUSE ANALYSIS OF PERFORMANCE PROBLEMS . . . . . . 36 3.1 Inadequate Administrative Control . . . . . . . . . . . 37 3.2 Weak Supervision and Independent Verification . . . . . 38 3.3 Lack of Rigor and Inquisitiveness in Work Practices . . 38 EXIT MEETING . . . . . . . . . . . . . . . . . . . . . . . . 39 APPENDIX A - Exit Presentation APPENDIX D - Exit Attendees vi i

l .. . e 1

1 INTRODUCTION 1.1 Daexaround Nuclear Regulatory Commission (NRC) management has been concerned about inconsistent performance at the oyster Creek plant, stemming primarily from frequent equipment problems often resulting in forced plant shutdowns and long periods to implement corrective actions. Despite substantial plant improvements and a demonstrated safety-conscious attitude, equipment probicas continued to challenge the plant operators.

The NRC's last Systematic Assessment of Licenseo Performance (SALP) identified significant weaknesses in the licensee's radiological controls program for a second time. As a result, the licensee implemented extensive programmatic improvements, similar to the improvenients impicmented to correct weaknesses in the operator training programs.

In addition to a variety of programs to improve specific weaknesses and correct past problems, the licensee developed a Plan for Excellence in Ferformance to lay out overall corporate objectives to achieve excellence in all phases of plant operations.

The performance of licensed activities at selected reactor facilities, including Oyster Creek, was discussed at a meeting of HRC senior managers in June 1990. Those discussions included the continuing challenges resulting from equipment failures at Oyster Creek and the accumulation of programmatic improvements. The Executive Director for Operations (EDO) subsequently directed the staff to obtain additional information conca-"4-- ** i" S r d '

perternance et Oyster Creek by conducting a dirgaoctic cycluation.

1.2 Egnpo and obiectives The EDO directed the staff to perform a broadly ctructured evaluation to assess overall plant operations and the effectiveness of the licensee's principal activities for supporting safe plant operation. The following goals were established for the diagnostic evaluation: (1) provide information to supplement SALP and other assessment data, (2) 1 l

evaluate actions of licensee management and staff with regard to safe plant operation, (3) evaluate the ef fectiveness of the licensco's improvement programs, and (4) determine the root causes of safety-related equipment and performance probicas.

1.3 Methodoloov A diagnostic evaluation team consisting of 17 members was organized into four subteams (each with a team 1cader) reporting to a team manager, with an administrative assistant providir.g support. Team preparation included document reviews and briefings by representatives from Region I, the Office of Huclear Reactor Regulation (NRR), and the Office for Analysis and Evaluation of Operating Data (.\EOD). On November 5, 1990, the team began a 2-week evaluation at the plant site, with visits to the corporate offices. The team returned for an additional week of evaluation on December 3, 1990.

The team conducted 134 interviews of licensee personnel, conducted tours of accessibic areas of the plant, observed various work activities, examined records and logs, and performed detailed studies of selected subsystems and components. During

. daily team meetings, the team discussed their findings to gain insights into the strengths and weaknesses of the licensee's performance in the areas of plant operation, maintenance, engineering, and management. The team leaders met with their licensee counterparts daily to discuss the team's activities and findings. The NRC resident inspectors attended the team meetings and provided technical advice to the team. Special emphasis was placed on identifying the causes of any observed performance veaknesses and the licent,ee's ability to identify and correct problems.

As pt f t of the evaluation of Oyster Creek, the team assessed the excellence plan, along with well over a hundred other improvement programs and initiatives, which the licensee identified as their means to provide the foundation for future safe and reliable lant operation. The team reviewed these programs and initiatives in conjunction with tha current plant conditions and operating practices to assess performance as well as the potential for lasting performance improvements in the future.

When the team arrived for the initial onsite evaluation, the 2

I

I licensco was preparing to repair a leaking valve in the minimum-flow line of the "A" feedwater train. plant operation was limited to 70 percent of rated power while the "A" feedwater train was isolated for repairs. The team's observations of the licensco's evaluation of the minimum-flow valve Icakage and plans for repair are discussed in following sections of this report.

In addition, the team reviewed the licensco's preparations for the 13R refueling outage scheduled to begin in February 1991, 1.4 Facility Descriotion The Oyster Creek Nuclear Generating Station, which began commercial operation in December 1969, is located in Ocean County, adjacent to Forked River, on the New Jersey coast. The plant is a boiling water reactor (BWR-2) designed by the General Electric Company, with a licensed power level of 1930 megawatts-thermal and a gross electrical rating of 670 megawatts-clectric.

The reactor is a single-cycle, forced-circulatiori (non-jet pump)

BWR producing steam for direct use in the steam turbine. The containment is a Mark I pressure-suppression system with a secondary containment reactor building.

GPU Nuclear Corporation and Jersey Central Power and Light Company (JCP&L) are co-licensees for the operation of Oyster Creek. Figure 1 illustrates the organizational structure for management and support of oyster Creek, including both site and corporate functions. The corporate offices are located in Parsipanny, New Jerse.y, about a 2-hour drive from the site.

1.5 Facility HistpIy oyster C ect was origina1Ly constructed as a " turn-key" project.

The General Electric Company directed the design and construction of the facility and, once operating, turned over the leadership role to the owner, Jersey Central Power and Light. For more than a decade of plant operation, little preventive maintenance was performed and the plant material condition generally declined.

GPU Nuc1 car became a co-owner of th? facility in 1982 to augment company resources available to address significant issues. These issues included requirements resulting from the accident at Three Mile Island Unit 2 (TMI-2), other new regulatory requirements, and reevaluation of the Mark I containment margins associated 3

Philip R. Clark President & CEO Vacant Executive Vice President l l 1 J. C. Devine R. L. Long VP & Director VP & Director Technical Corporate Functions Services I. R. Finfrock Eugene E. Fitzpatrickt P. B. Feldler VP VP & Director VP & Director Site Services Oyster Creek Nuclear Assurance John J. Barton Deputy Director ft Corporate Human Resources Engineering Licensing Planning Security Quality Assurance Construction Rad Controls Training i itti Robert J. Barrett D. K. Cronoberger operations outage Manager Radwaste Chemistry I

Arthur 11. Rone Lennis L. Lammers Plant Engineeting t;aintenance i

William V. Stewart Mark F. Budaj Safety Review Plans & Programs t In January 1991 J. Barton replaced E. Fitspatrick as Director.

tt Indicates indirect functional reporting.

ttt Reports to President, except 3 months before and during refueling outages.

Figure 1 GPU Nuclear organization for Oyster Creek 4

_. l l

with suppression pool dynamic loads. At that time, oyster creek

)

also was one of 10 older nuclear power plants being reevaluated j by the NRC against current regulatory requirements under the l Systematic Evaluation program.

As a result of these activities, substantial changes to the facility were identified and implemented throughout the 1980's. .

During the mid-1980's, the aging of plant equipment became evident as a significant cause of equipment failures. Although ,

preventive maintenance initiatives were under development, )

corrective maintenance took precedence while a backlog of safety improvements resulting from regulatory requirements (like those mentioned above), fire protection, and equipment qualification, were being implemented.

As this backlog declined, the licensee focused more attention on future needs. Some initiatives were delayed in the mid-1980's, like the purchase of a plant-specific simulator. Nevertheless, the licensee continued to make subrtantial investments to ensure the long-term viability of the plant. The licensee developed its plan for excellence in performance in 1988 to consolidate and communicate the overall goals and objecti', . of the company.

2 EVALUATION RESULTS The following sections describe the team's findings relative to particular performance attributes in plant operations, maintenance, engineering and technical support, and management effectiveness. For the evaluation of engineering and technical support, the team's findings are described for particular case studies that were considered representative of technical performance attributes.

2.A oneiations and operator Trainina overall, the conduct of plant operations at oyster Creek was good. Operators were professional and attentive. The control room material condition was good. Communication within and outside the department was good and improving. The staffing plan was working well although the training staff had a heavy workload. The team did not evaluate the training program, but found that training materials and equipment were very good.

The effective conduct of plant operations and overall improving 5

trends in plant performance were attributed to increased management attention and the positive influence of recent improvemont initiatives. However, current performance strengths appeared to result primarily from the experience and skill of individuals.

2.1.1 Conduct of Plant Operations operators were aware and knowledgeable of plant conditions and quick to respond to annunciators. Operator shift turnover was formal and included full panel walkdowns. Licensed operators used procedures appropriately during plant evolutions. Log entries of shift events were appropriate. Shift activities were planned and coordinated with other departments effectively by shift briefings and by use of action plans for special activities such as the repair of the feedwater minimum-flow line. Access control to the control room was adequate and the noise level was low. The plant passed 150 days of continuous operation during the diagnostic evaluation.

Operator performance was hampered by having to work around some equipment problems, like malfunctioning turbine building sump pump controllers and check valves, spurious fire alarms, actuation of the thermal overloads of some motor-operated valves, and the need to continually clean trash screens in the intake structure during inclement weather conditions. However, these equipment problems did not impede the operators ability to respond to events or maintain the plant in a safe condition. The office for both on-shift senior reactor operators (SRos) was out of visual range of the control room panels, and the SROs responded when they heard neveral alarms or if they were contacted by the reactor operators (Ros) when en 'bnorma.'

condition occurred. During the next refueling outage, the licensee planned to begin modifications to move the offices to the back of the control room so that the SROs can see the control panels.

The material condition of the control room was good. Adequate management attention to the control room was indicated by completed human factors modifications made to the control panels, a minimum number of annunciators in the alarm state, and the general cicanliness of the controls area. The number of control room deficiencies was relatively high (45), but all were clearly 6

1

identified, causes were known by the operators, and most (about

40) were less than 1 year old. The number of active instrumentation ftilures was known and frequently updated in the shift technical advisor's daily report to plant management.

2.1.2 Staffing The operation and treining departments had ample staffing.

Staffing for the operttions department consisted of six rotating shifts. Notably, thert were six Licensed operators (Ros and SRos) on each shift, as compared to the four required. Recent increases in staff allowed licensed personnel to be assigned other duties that provided opcrator experience and expertise to other dcpartments. A nuclear plant operator (NPO) program was instituted this year with a primary goal to license all operating crew personnel; thereby, providing a clear progression path and greater flexibility to rotate shift position assignments.

The training department had a dedicated and professional staff with a wide background of skills and experience. The workload in the operator training department required substantial overtime.

The workload included conducting training classes to meet expanding normal training needs and supporting the design and testing of the plant-specific simulator. An additional burden on the training staff resulted from poor past performance, identified in recent NRC inspection findings, in the areas of operations task analysis, requalification examination bank, and standardization of lesson plans and examination grading. The overtime condition appeared to be temporary until reactive work stabilizes; therefore, the licensee did not anticipate significant staff increases it, the training department.

2.1.3 Training Tccilities The training facilities contained extensive training aids, including typical plant components such as pumps, valves, valve operators, air compressors, small piping systems, nuclear instrumentation drawers, electric circuit breakers, and a recirculation pump seal area mockup. These aids provided hands-on training capability and were used to operationally check out components or systems before installing them in the plant and to validate maintenance and operating procedures. Other training aids included simulated contaminated areas, requiring appropriate 7

l I dress and adherance to radiological controls procedures while performing simulated work tasks. In addition, a " picture" control room simulator was used by all departments for training and familiarization of control room panel alarms, instrumentation, and controls layout. A partial task simulator vas used by operator training, but not observed by the team.

]l A plant-specific control room simulator had been ordered, but was not yet installed because it was undergoing construction and testing at the vendor's site. Continued simulator training was provided at Nine Mile Point simulator. The licensco's request for a waiver from NRC schedular requiroments for the simulator, because of expected delays in simulator certification, was pending.

2.1.4 Communication Communication was good and improving. Effective actions had been taken to improve interdepartmental communications, as demonstrated by the plan-of-the-day meeting, the daily "1:30" maintenance meeting, shift briefings between operations and the support organizations, and the newly created positions of operations outage coordinator and training coordinator. In particular, communication and interface between the operations and maintenance departments were notably strong, with operations having a direct input and ownership for planning and prioritizing daily work activities.

Communication between operations and the radiological controls department appeared to be experiencing " growing pains." For example, operators perceived radiological controls requirements to be excessive for relativelv minor jobn nn the hydraulic control un.its. Also, operators were concerned about work stoppages by radiological technicians without appropriate communication to the operations department and the lack of radiological support for some preplanned jobs. Conversely, the radiological controls department believed that needed actions were being taken to improve radiological practices.

Communication also was weak between operations and the training department primarily because of the operators' concern regarding the limited oyster Creek operuting experience in the training department; little effective action had been taken by either department to enhance the perceived credibility of the training 8

m___..___ _ _ _ _ _ _ _ . .

.. , s staff.

Intradepartmental communication was adequate and improving.

Examples were the operator concern program, effective shift turnovers, and use of night orders. The operations managers appeared to have close working relationships and their offices 1 were in close proximity. Operations managers were actively communicating standards and expectations through their plant tours, during which they documented discrepant conditions. ,

Managers attended simulator sessions to critique crew response in l relationship to their expectations and included excellence plan action items in employee performance appraisals. Operations management met weekly with the training crew to discuss plant '

status and provide a general, open discussion forum.

The onshift SRos expressed some concern that they had only limited input to the development of departmental goals and objectives. Operations managers indicated that they had made some effort to improve this communication by holding regularly scheduled meetings, but attendance at those meetings was minimal and the meetings were discontinued. There was some confusion and frustration with poorly defined workload and some changing or conflicting priorities among operations department technical 4 staff. This confusion was exacerbated by occasional contradictory signals from plant managers when they took people off high-priority tasks, did not authorize overtime for high-priority tasks, or did not allocate appropriate resources for a particular task.

2.1.5 operations Improvement Initiatives Noticable improvments in the overall enndort e# eterst4cno indicated that the improvement programs were Leing effectively implemented. Specifically, the operations department staffing plan allowed six shift rotation, filling of the training and l outage coordinator positions (SRos), and a progression path l within the operations department.

l The equipment labeling program was a visibic effort that improved the identification of equipment and components. In addition, this on-going process identified numerous deficiencies in plant drawings, procedures, and configuration control.

9

t .. . s i

The operator concern program had been expanded to include all ,

site personnel and was an effective means to ensure problems were documented. The professionalism program and-the expectations and 1 standards program were on-going efforts that also had the potential to enhance employee conduct. The recently developed operations department tracking system was a positive offort that captured open issues and work items and enhanced management attention to the total workload.

Improvements in the area of radiological controls had offectively lowered exposures. Figure 2 shows the licensee's performance indicator for radiological exposures. As a result the licensee had reduced the annual exposure goal (the licensee's exposure i " limit) from 450 rem to 395 rem in August, and to 330 rem in-November. While this goal reficcted one of the lowest annual exposures in the history of plant operation, even lower exposures are achievable.

• 1. M DEC$.st[ t IWIT 45W M5 400 -

Acm Excukst nt ora 1 l o

t _

}_ ~

tw

/

7 iM ,_ /

Y 1

/ '

/

, n . .

m ete w a on w a x, Ax up oct e as figure 2 1990 cumulative personnel exposures Two other programs that had potential for added improvements to

-the operation and training areas included the recently completed operations self-assessment and the procedure improvement program.

During the self-assessment, operations identified specific areas 10 l

l

.. . s 1 .

t that needed improvement. The planned actions in response to the licensee's self-assessment had not been established at the time of the diagnostic evaluation; however, if effective corrective actions are implemented, added improvements to the overall plant operation should be possible. The procedure improvement program had developed a writers' guide and draft pilot procedures.

Similarly, the potential for improving overall plant operation is possible if effective implementation o,f the procedure improvement program is pursued.

1 2.2 Maintenance and Testina ,

1 The maintenance program and surveillance testing were adequate  !

and improving. Improvement resulted primarily from recent reorganizations. However, continued improvement depends en the formalization and ef fective implementation of existing programs, including appropriate implementing procedures, completion schedules, and dedicated resources, rurther improvement in the plant material condition was planned to correct historical equipment problemn. This planned improvement is particularly important for components with multiple failure histories.

2.2.1 Maintenance organization The maintenance organization was in a state of transition. In 1989, the maintenance organization was moved from a corporate division to report directly to the Director of oyster Creek. In October 1990, the materials group, which consisted primarily of maintenance assessment engineers, was reorganized within the maintenance and plant engineering departments. These organizational changes enhanced the effectiveness of the maintenance activities.

Tha maintenance planning group had been supplemented with contractors. However, as a result of a task force assessment of the status of maintenance initiatives in August 1990, the licensee had determined that the planning group needed to be supplemented with more permanent staffing. In addition, on the basis of the task force's findings, the licensee was planning increased staffing with qualified personnel for the maintenance j assessment group, which was responsible for most improvement l

initiatives.

l 1

11 l

l l

+ .

2.2.2 Plant Material Condition At the. time of the diagnostic evaluation, the licensee was still l working primarily on the recovery of the material condition of the plant. Up to about 1984, maintenance at Oyster Creek was minimal. In 1982 and 1987, studies of the plant material condition were performed and the results were intended to be used as the basis for improvements to both the material condition of the plant and overall maintenance practices.

Other initiatives to improve the fundamental maintenance program and equipment reliability issues, which began primarily in 1989, l included the valve improvement initiatives, isolation condenser l modifications, recirculation pump seal modifications, and the l control rod drive system improvement initiatives. Other specific improvements included modifications to the instrument air compressors and the recent installation of a new refueling bridge.

The licensco's initial efforts appeared appropriately direct (4 on the basis of the ma* ority of equipment failures having been attributed to valves, a history of plant shutdowns to repair recirculation pump seals, and the control rod drive system having had the most component failures of any system. Most of the past efforts were modifications to correct specific equipment problems, although some efforts were made to develop and validate practices to be used for substantive programmatic improvements.

Plant housekeeping was good and improving. The licensee's painting efforts were evident during plant tours. A decontamination program was on-going. Although the goals had not been met, Figure 3 illustrates continued reductions to the contaminated areas. These efforts enhanced tha overall material condition of the plant and the conduct of plant maintenance.

2.2.3 Maintenance Work Planning and Scheduling The licensee was effectively controlling the corrective maintenance work backlog. The backlog for nonoutage corrective maintenance had been reduced from 630 tasks in December 1989 to 410 tasks in November 1990. In maintenance work planning, scheduling, and prioritization, several improvement initiatives 12

.e .

• l SIM AG MCA V V 93 79,330 80,000 -

1500 GXL 75,000 -

55, t.01

- 70,000 -

YEA 4 TO DATE f 67,037

{es,000 -

8 60,000 -

- ~ - - ~

, 55,000 -

W 000 -

JAN FEB WAA APA HLY JW JJL A!E SEP OCf KN DCC Figure 3 1990 area contamination reductions were under way. Duriny the diagnostic evaluation, the licensee issued a work planners guide to provide guidance to maintenance planners. The licensee stated that a training program for planners was under development. An integrated work scheduling system for corrective maintenance, preventive maintenance, and surveillance testing during system outages was also under development, as well as a procedure for a new, simplified prioritization system.

The GMS2 computer system was a good information management system for planning maintenance work and equipment surveillance.

However, accessing information appeared difficult and the data base was incomplete. Instructions for loading data for each job l nrder or work request were not available, which appeared to be at least part of the reason for the variation in the quality of the data base. Some entrie s contained cross-referenced information such as deviation repo!.t numbers and other job order numbers, and other entries only inci.uded brief descriptions of the problem that existed. At timei, plant personnel required several attempts to obtain inf>rmation from the system. The licensee was planned to conduct tra.ning for maintenance personnel to improve 13

the utility of the system and the quality of the data.

2.2.4 , Failure Trending and Analysis Equipment failure trending analysis was not being used effectively. The failure data trend program was established in 1988 to identify repetitive equipment failures and initiate corrective actions by issuing a trend action notice (TAN).

However, the equipment failure reports and TANS were not issued in a timely manner; some reports were issued almost a year after 4

the trend occurred. Closure of most TANS were overdue. The number of TANS issued was low in proportion with the number of components with multiple failures, primarily because of the high u

threshold for issuing a TAN. The licensee expected to reduce the TAN threshold in the future when the maintenance improvement initiatives reduced equipment problems to a more manageable level.

Maintenance histories for several component types, in most cases, did not have an adequate root-cause analysis of the equipment failure by the maintenance assessment engineers. Some abnormal equipment conditions had been referred to engineering for further evaluation, but no followup action had been taken and the engineering evaluations were not performed. The deviation report appeared to be the licencee's principal mechanism for conducting root-cause determinations. Closure documentation for deviation reports was not readily accessible. In some cases, when the team requested equipment failure information, the licensee had to reconstruct closure documentation because of a lack of formality when performing and documenting corrective actions taken for equipment failures. In other cases, the licensee had to take supplemental actions because the initial correctiva action had been inadequate.

2.2.5 Preventive Maintenance The licensee's preventive maintenance program was developed in 1984 using recommendations of plant personnel. The licensee recently assembled complete and up-to-date vendor manuals for equipment. A systenatic evaluation of systems and components, using vendors recommendations and maintenance histories, had not been performed to determine required scope and performance intervals for preventive maintenance.

14 l

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

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

Although the licensee stated that some preventive maintenhnce had been performed on electrical components for most motor-operated valves, there were no electrical preventive maintenance tasks for motor-operated v.11ves in the program. However, the licensee provided a draft procedure that was being developed to perform preventive m e tenance tasks on motor-operated valve electrical components.

The scope of preventive maintenance was increasing, as shown in Figure 4. However, there were 134 requeste for additional preventive maintenance tasks waiting for review and development; most of these requests were overdue.

1 (WE ASURED N CANHOURS) 70 -

60 -

50 - -

40 - /

O i i i i i i i i i i i ,? i.!iiiiiii,ii

- /.

i JJA$0NDJFWAWJJA$0ND.FkAWJJA$0 l 88 I 89 l 90 l b WONTHLY + 1 REND Figure 4 Ratio of preventive to tor.ai maintenance effort i There were many examples of preventive maintenance tasks that were skipped up to eight or nine times for particular equipment, without adequate justifications. The licensee changed the procedure controlling skipped preventive maintenance in April 1990 to require justification for skipped preventive maintenance tasks. In response to the team's questions about the control of skipped preventive maintenance, the licensee stated that 15

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

=.. . . .

1 I

additional changes would be made to the procedure to increase

- managementEawareness by requiring supervisory review and monthly.

management reports.

2.2.6 Work Practices .

- Most maintenance work practices appeared to be good or acceptable although several examples of poor work practices were noted.

Examples of poor work practices included inconsistencies in the root-cause analyses, a failure to complete an evaluation on the effect of an over-thrust condition on a valve and valve operator,

' failure to maintain the motor-operated valve thrust window listings current, and the lack of documentation for skipped preventive maintenance tasks. An= attempt to incorporate two preventive maintenance tasks resulted in the inadvertent deletion of one of the tasks because of poor work practices. In addition, on-one occasion, test equipment that had exceeded its calibration due date was_used during post-maintenance testing that was

' performed without the use of the procedure or work package.

2.2.7 ' Maintenance Improvement Initiatives

, Although there were several maintenance improvement initiatives land programs,=most of these did not provide programmatic descriptions, implementing procedures, or completion schedules.

As a result, adequate resources had not been allocated for effective-implementation of these programs. Active initiatives

. with a potential-for improving maintenance activities included the~11censee's radiological performance committee formed early in L

=

1990 and the maintenance procedures upgrades scheduled for completion in January 1991.

m- x The licensee intended.tx) use the. life of systems maintenance

. plans (LOSMP)= program to-perform-a-thorough review-of systems by.

- applying the principles of_ reliability-centered maintenance to D establish the scope of equipment maintenance and maintenance-y .related needs on a system-by-system basis;-however, the process was not formalized. ;A draft procedure for LOSMP was in an early l stage of Edevelc) ment at the time of the-diagnostic evaluation, U and a schedule for implementation did not exist. At the-current rate of implementation, it would take-over 20 years to implement the process on all 61 designated systems.

16

. . .~, .- . -. .

- - _ . .. = _ - - - - . . - - - - . - - - . - - . - . _ _ - . _ -

The material condition issues (MCI) program was initiated in 1988 to address long-standing equipment deficiencies, including many of the. material condition issues from studies conducted in 1982 and 1987. Similarly, the MCI program was not formalized and no implementing procedure existed. The program was idle for over a year. The materials condition review board had not met in 1990 to add or delete any items to the material conditions issues list. A review of a material condition issues matrix indicated that most of the issues were being addressed; however, the resolution of some issues was insufficient or, where the resolution was sufficient, the issue was still being monitored to verify that the corrective action would be sufficient. Closcout of nany issues was overdue. The corrective actions for many issues were scheduled for the upcoming (13R) refueling outage.

A valve maintenance committee was formed in 1989 to address valve problems, which had constituted the majority of equipment failures up to that time. Although the valve committee had addressed many issues, a program description with goals, implementing procedures, and assigned responsibilities had not been developed. The intended scope of the valve improvement efforts were reviewed in conjunction with the licensee's actions in response to Generic Letter 89-10 (motor-operated valves) and the Institute for Nuclear Power Operation Significant Operating Event Report (SOL 2) 86-03 (check valves). Check valve performance monitoring in response to SOER 86-03 was not implemented in a timely manner. Check valve inspections were initiated during the 1989 refueling outage when 10 valves were inspected and 2 replaced. Seven valves were scheduled for inspection during the next outage. All of the industry recommendations were scheduled to be completed by December 1992.

2.3 EngineerJna and Technical Support The quality of engineering and technical support varied. depending on the perceived importance of a particular task and the inquisitiveness of the assigned personnel. Some tasks were performed well. Lesser tasks, however, often exhibited a lack of rigor and thoroughness, and the failure mechanism of the problem was not always clearly understood. Failure trending was not used offectively to identify recurring problems. Safety analyses were L not always performed or properly documented. There was no l

systematic means to ensure that, when deficient conditions were 17 l

l -

.._ . e t

l

-l identified,_' timely evaluation of the operability or functional capability of the affected equipment would be performed to determine how soon corrective action was necessary. '

2.3.1 Technical Support organizations There were two engineering and technical support groups at Oyster  !

Creek: Plant Engineering, which was located on site and reported directly to the Plant Director, and Technical Functions, which was a corporate group located in Parsippany with a satellite group on site. Both groups perform engineering activities, initiated design changes, and responded to assist the other departments 'in diagnosing unexpected equipment failures and plant' transients.

Engineering activities for both groups consisted of (1) reactive support for routine and nonroutine plant activities, including response to equipment failures and transient events, and (2) processing plant design changes, of which about 10 percent were engineered in-house. The licensee had a qualified staff to do quality engineering work. Many of the engineers had substantial experience in similar engineering work. The working environment appeared conducive to quality engineering work with good horizontal and vertical communication within both groups. The engineers' morale and company loyalty appeared high.

I The system engineer position within both Plant Engineering and Technical Functions was not an entry level position within the company. Most of the system engineers-interviewed had over 10

-years of industry experience. Although duties and l responsibilities of Plant Engineering and Technical Functions L system engineers were generally described by a procedure, their interface responsibilities for design changes were not well defined. Informal communications _between the two groups appeared to be good; however, Plant Engineering apperently could complete a design change without the concurrence of Technical Functions, although Technical Functions was-ultimately responsible for L

maintaining the design basis. Thus, there was a chance-for miscommunication and missed opportunities for feedback in the development of modification designs.

Several significant changes were made in recent years to improve j the support-that Technical Functions provided to the plant.

l' 18 I'

.. . e These changes included reorganization of the corporate engineering group, appointment of a new Director of Technical r Functions, assignment of corporate system engineers, selection of a principal engineering firm to provide outside engineering assistance, establishment of design bases documents to facilitate the reconstitution of the plant design bases, conduct of safety system functional inspections, and a continued effort to increase the technical ability to perform sophisticated in-house analysis work.

The licensee established a program to conduct safety systems functional inspections to identify system vulnerabilities.

However, the scope appeared to be limited. For example, the scope of electrical distribution system functional inspections did not include the mechanical aspects of the emergency diesels and their supporting systems.

2.3.2 Performance of Technical Support Organizations The findings from specific case studies involving examples of problem solving and the design of plant modifications are described below and illustrate the varied performance of the licensee's technical support groups.

(1) Feedwater Minimum Flow Line Leakage On November 3, 1990, the licensee discovered a through-body leak on the "A" feedwater minimum-flow control valve. The licensee reduced reactor power to permit the isolation and subsequent repair of the feedwater minimum-flow control valve in a safe manner. Corporate engineers were mobilized within several hours of the event to accict in the assessment of the valve body leak.

During subsequent investigation of the leak, engineering personnel focused on a theory that the control valve was obstructed from fully closing, thus producing a high-velocity jet that impinged on the valve body and eroded the body material beneath the valve seat. Engineering personnel believed that this theory of the failure mechanism was credible because the maintenance personnel who had disassembled the valve found a small piece of quarter-inch-diameter stainless steel tube in the valve. Engineering 19

personnel also believed the same high-velocity. flow was responsible for the wall thinning of the first elbow down stream of the valve, approximately two pipe diameters away.

After the through-body valve leak was found, the licensee performed ultrasonic wall thickness measurements of the line in the vicinity of the valve, which included one' pipe fitting on either side of the valve.

Engineering personnel had not tested their theory of obstructed valve closure by attempting to reassemble the valve with the tubing inside the valve body. At the team's request, the licensee conducted such a reassembly and the licensee was unable to demonstrate that valve closure could be obstructed by the tubing.

Upon further review of the records, the licensee determined that the minimum-flow line was not in the same configuration as the plant piping and instrumentation drawings becauce a flow restricting orifice had been removed from the line.

The orifice was removed early in the plant's life to increase the minimum flow rate so as to satisfy feedwater pump requirements. The effect of removing the orifice was to increase fluid velocity from approximately 10 feet per second to 28 feet per second. At the higher fluid velocity, the licensee calculated an expected wear rate (at a maximum temperature) of approximately 0.2 inch per year. Another failure mechanism could, therefore,-have been the higher fluid velocity resulting from the removal of the orifice.

This possibility was not investigated by the licensee until a second failure occurred on a section of pipe further downstream.

Because the team questioned the engineering support for the repair of the valve leakage in the minimum-flow line, the l engineering groups conducted a critique of their performance. They-concluded that, among other things, a simplified process was needed to ensure consistent and thorough root-cause evaluations. -In addition, engineering management concluded that the evaluation process needed criteria that would assign the level of engineering and b technical support for particular deficiencies and events, based on representative views of operation, maintenance, l

engineering, and technical functions departments. As a 20 l

l

result, the licensee established a plan to develop a new, graded procedure for root-cause evaluations.

(2) Emergency Diesel Generator Modifications Modifications were being implemented to replace the starting batteries for the emergency diesel generators, as a result of a low-cell voltage condition identified during surveillance testing. When the battery problem was initially identified, the licensee failed to analyze the early failure of batteries and recognize the reduced battery life. The emergency diesel batteries were expected to last for eight years, but the licensee simply planned to replace them after 5 years without conducting a detailed analysis of the reduced life.

The licensee was apparently aware of the environmental conditions that could affect battery life but had not conducted a detailed root-cause evaluation so that timely corrective actions could be taken. The emergency diesel batteries were located in an enclosed cabinet with high ambient temperatures. High ambient temperature can cause reduced buttery life, which the licensee had experienced in the past. The licensee planned to install a ventilation fan as part of the modification to replace the batteries; however, the analysis supporting the modification did not assess-the effect on battery life. Similarly, the licensee had not evaluated diesel cold-starting capacity. In 1988 the diesel vendor provided the licensee with cold-starting capacity information, but the licensee failed to conduct any followup evaluation and incorporate this information into the design of the modifications for the battery replocement.

Similarly, an analysis was not conducted to ensure the adequacy of emergency diesel starting batteries for their loading duty, calculations were not performed to establish the battery capacity for slow and fast start of the emergency diesels,-including voltage drop, before procurement of the replacement batteries.

Although the diesels and their associated switchgear were qualified for seismic conditions, seismic requirements were not specified for the replacement batteries. These 21

o batteries were used in diesel locomotives in which they were subjected to a much higher level of acceleration and vibration than earthquakes and the diesel vendor had furnished the licensee with test results that demonstrated this capability. In locomotive applications, the battery cells are clamped to make them withstand severe vibrations; however, the battery cells at Oyster Creek were not clamped because the licensee had not specified the seismic requirement .in its purchase specifications and the modification design for the battery frame.

The piping code of record for oyster Creek is ANSI B31.1, which recommends against the use of threaded piping near vibrating equipment. Contrary to this guidance, the licensee had allowed certain modifications to the fuel oil lines during the 12R outage that included threaded joints adjacent to vibrating equipment. Because of fuel oil leakage, the licensee planned to replace the threaded joints with welded joints during the 13R outage.

The licensee's assessment of weather effects on the emergency diesels was not comprehensive. For example, although the licensee had examined corrosion of the skids the diesel engines rest on, which resulted from rain I

accumulation on the floor, the team visually detected an area more degraded than the one chosen for ultrasonic testing. Similarly, the licensee had not evaluated water leakage into the diesel fuel oil tank room.

l (3) Diesel Fuel Oil Supply The assessment of the functional capability of the fuel oil tank was not well founded. External degradation of the diesel fuel oil tank had been detected and documented since 1985. Although the initial evaluations had recommended examination of the tank condition during the 12R refueling outage, no inspection had been performed. Such an inspection could not be readily made without draining the

tank. Therefore, engineering recommended replacement of the l tank during the 13R refueling outage. However, no analysis was performed to demonstrate the structural capability of the tank in the intervening period.

22

The records of fuel oil tests provided evidence of internal corrosion of the fuel oil tank because of water in the fuel, and possible microbiological 1y induced corrosion. There also was evidence of external corrosion at the base of the tank because of the accumulation of rainwater. N.either plant engineering nor plant chemistry were aware of the test results. While ultrasonic testing had been performed on the tank, the testing had not been conducted close to the bottom of the tank where fuel oil test results indicated greater corrosion would be expected. In addition, the licensee had calculations for the seismic design of tanks, but the analyzed condition of the fuel oil tank had different dimensions than the tank that was installed.

The uncertainty in the structural capability of the fuel oil tank, until replacement, was referred to Region I for appropriate follovup and resolution.

(4) Isolation Condenser Piping Replacement and Valve Failures Engineering analysis of isolation condenser functional operability, resulting from a " steaming" incident documented in a licensee event report (LER 88-21), showed strong technical capabilities and good cooperation between Technical Functions and Plant Engineering.

Initial assessments by Plant Engineering and a prudent decision to declare the isolation condenser inoperable showed good judgment. Subsequently, the licensee set up a task force to analyze the incident and develop corrective actions. The work performed was found to be comprehensive, technically well suonorted, and well documented. The interim corrective actions (operating procedure restrictions) were based on sound engineering analyses. The licensee used and validated sophisticated thermal-hydraulic computer analyses for this ovaluation.

A major system modification was planned for the isolation condenser during the next refueling outage (13R). This will involve replacement of all piping and valves outside of the containment. Numerous instances of intergranular stress corrosion cracking had occurred in the past requiring substantial in-service inspection (ISI) and wold repairs.

23

.. . o I

I l

In consideration of this, valve problems, and the inability '

to inspect portions of piping in the containment penetrations, the licensee decided to proceed with complete l replacement of the piping and valves. A piping I configuration modification was included that could preclude future " steaming" events. This work appeared to*be well coordinated, and the material and equipment changes should effectively improve the safety and reliability of the system.

Plant Engineering and Technical Functions did not, however, satisfactorily determine root cause of the isolation condenser condensate return valve failures over the past several years. Numerous condensate return valve failures had occurred since 1984, more than a half dozen apparently attributable to thermal binding. Systems operating procedures were modified several times over this period to alleviate the thermal binding of the condensate return valves. However, the problem persisted. As such, the plant operated a number of years with the isolation condenser in a potentially degraded condition without proper analysis of the system operability and overall effect on plant safety.

In addition, the licensee had to recreate a safety evaluation supporting the most current operating procedures, which specified cycling of the condensate return valves at intervals of 50 *F during plant heatups and cooldowns.

Plant Engineering.had repeatedly identified the need to determine the root cause and required corrective actions for the valve binding problem. At least initially, Technical Functions did an inadequate assessment of the problem. The change in the operating procedures, which required cycling of the valves to prevent them from binding, at least in part, was based on informal conversations and unsupported engineering judgment. The procedure was changed to require cycling of the valves at intervals of 50 *F, during plant heatup and cooldown, rather than the previous interval of 100 *F that had been used in an attempt to preclude binding of one of the valves. However, neither the fundamental cause of the valve binding nor the basis for valve cycling had been clearly established. On-several occasions, deviation reports were not written for isolation condenser condensate return valve malfunctions. Several deviation 24

reports had been closed without performing root-cause assessments. The closure referred to the planned modification to replace the valves during the next refueling outage (13R), which should alleviate the problem in the future.

The uncertainty in the operability of the isolation condenser condensate return valves, until replacement, was referred to Region I for appropriate followup and resolution.

(5) Instrument Air Systems The technical support provided and actions taken in response to Generic Letter 88-14, regarding problems with instrument air systems, appeared appropriate, timely, and sufficient to fully address the areas identified in the generic letter.

This was a tracked licensing action item in an area in which system improvements were well under way to correct recurring operational problems. Modifications had been made to air compressors, air dryers, filters, and piping to improve system reliability and air quality. In addition, new check valves had been installed and tested to ensure availability of stored air supplies in accumulators for safety-related valves, in the event of a loss of the air system.

The licensee's review of recovery procedures for loss of air was poorly documented; thus, it was not possible to determine the sufficiency of that review although the team's review did not identify any notable deficiencies. There was also a concern regarding the adequacy of a single vacuum breaker check valve to fulfill containment trolation requirements. This configuration was vulnerable to a single failure and the licensee was unable to readily explain the design basis for the valve. The licensee was awaiting the NRC staff's review of the current design before taking any action. The uncertainty in this design configuration and completion of the NRC's review of this matter was referred to the Office of Nuclear Reactor Regulation for appropriate followup and resolution.

Continuing efforts on the air supply system included completing the filter modifications and initiating a 25

n. . e surveillance program for safety-related air operated valves planned for the next refueling outage, which-should complete

.the identified' actions in response'to the generic--letter.

(6)- Rosemont Transmitter Problems The licensee's actions in response to NRC Bulletin 90-01 regarding Rosemont transmitter problems were timely and;

. appropriate. - The- tracking of actions taken and analysis performed was well documented andLappeared well done. In this case, the assigned individual-appeared to have devoted-extra? effort to the assignment and continued close scrutiny _

oof Rosemont transmitter problems.

4 (7) Valve 5 Packing Leaks e The licensee had an effective program to switch to Chesterton valve packing and revised preventive maintenance to better = control valve packing leaks,- which had been a continuing problem'at Oyster Creek. As a result, the licensee assigned an individual the. responsibility for

, selecting the: appropriate' packing for long-term. resolution and prioritizing packing replacements. The. licensee'had not 4- decided whether _ to. ' initiate wide creplacement 'of: the' valve packings during the ' next refueling . outage or to just repair-the few known packing leaks.

_ (8) Electrical' Overload-Protection [

u L 'The' licensee'siengineering_ standards for overload protection

? 'for electrical motors-(ES-024)"and-selecti'on andL settings of-h protective devices'(ES-025) were well written, Lthorough,J and covered state-of-the-art practices. Similarly,:the calculations for thermal overload set points for! safety--

L related motor-operated; valves (TDR 519,-Rev. 2) were L _ excellent examples _of good. quality. engineering work on a difficult: subject.

~ The licensee had -conducted a -thorough investigation of- past L failures of 5-kV cables and established a1 reasonably thorough cable testing program. The licensee effectively resolved the cable failure problem with a well-developed surveillance program.

26

• • mt- 'w7.

- - m - - - 4  %+- --w - -- _ __m- +u _ - A * - 4-+-en -

.Je+ -ew-wwa-w---- 6 w em e er W

.=, .

The electrical bus transfer scheme used at oyster Creek is an automatic, sequential transfer in which the connected motors, switchgear, and the transformers might be damaged because the design does not include protective devices to prevent out-of-phase switching transients. The licensee had not evaluated related industry experience that described the vulnerabilities of automatic bus transfers.

(9) Geomagnetic Disturbance Studies In 1989 a geomagnetic disturbance event caused the failure of two main-power transformers at the Salem Nuclear Plant; extensive _ equipment failures in Pennsylvania, New Jersey, and Maryland; and a total blackout in the Hydro-Quebec system in Canada. In association with the neighboring utilities, the licensee took the initiative to conduct an extensive study on the vulnerability of its system to geomagnetic disturbances and effectively implemented the necessary preventive measures.

(10) Design-Basis Reconstitution Although the licensee instituted a design-basis reconstitution program through the development of design-basis documents (DBDs), the schedule only provided for completion of DBDs for two to four systems each year. There were more than 30 systems remaining to be completed. The licensee had established priorities for completion of the DBDs, but the basis for the priorities was not clear.

In several instances, der.ign-basis calculations did not exist, including calculations for ventilation in the emergency diesel room and flooding calculations for the emergency diesel and fuel oil tank rooms. Similarly, there did not appear to be a systematic means to update l engineering calculations when the opportunity arose during.

the design of plant modifications and the evaluation of new information. Many of the licensee's design calculations had not been updated in last 4 years or more. Some of these l calculations became outdated because of plant modifications or improved calculation methodology. As a result, about a year ago, the licensee had established a computer program to 27

update all electrical calculations. However, the licensee did not always take advantage of opportunities to update

~~

calculations. For example, the replacement of the diesel batteries could have addressed the cold-starting capacity information from the vendor and deficiencies in the diesel starting voltage drop transient analysis (effects of the energency diesel voltage regulator, contribution from connected motors, and an appropriate emergency diesel impedance model).

2.4 MAnag.ement Effectiveness and Corrective Action Although performance improvements were evident, management attention and oversight of Oyster Creek was considered weak in some areas because plant and corporate attention was driven primarily by the visibility of issues and initiatives. Highly visibic projects were adequately planned and executed. However, routine work activities often lacked adequate tracking, prioritization, resources, and management oversight.

2.4.1 Management Oversight Management oversight was accomplished by several means, including daily communication, performance indicator trend reports, audits, assessments, project evaluations, various management meetings, and management plant tours.

The prevailing upper-level management philosophy was that problems should be solved at the lowest possible level and that the site was the " customer" for corporate functions. While a variety of audits and assessments were performed to identify problems, it was not apparent that quality assurance (QA) was being used as a management tool to ensure the effectiveness of problem solutions. Some items on the QA monitoring report checklist were marked "not observed," quality deviation reports (QDRs) were generally not used, and QA did not review deviation reports (DRs).

This approach placed an additional burden on mid-level managers and supervisors to ensure effective and timely completion of work assignments. While there were several initiatives under way to enhance management practices, including management development training down through the first-line supervisors and the teamwork 28

.- , i and leadership program, there was little evidence that the supervisors were held accountable for this burden.  :

Upper-level management appeared to be aware of most performance problems at oyster Creek, including many of the weaknesses identified by the team, and plans had been or are being developed to resolve them. Recent self-assessments conducted in the operation and maintenance areas were found to be thorough and identified significant performance problems. Although upper-level management was aware of the findings identified in the self-assessments, they had not completed their evaluation of those findings at the time of the diagnostic evaluation.

2.4.2 Staffing and Resources In general, ample resources appeared to be provided to support plant activities, including funding and personnel staffing. For highly visible or large-cost items (primarily long-term projects), corporate attention appeared to result in adequate planning and scheduling. Examples included improvements in the radiological controls and in outage planning and scheduling efforts for 13R. Further, capital improvements (corporate projects) were effectively evaluated and prioritized. The licensee had made significant, long-term improvements during outages 11R and 12R and had many additional work items planned during the upcoming 13R outage.

The plant had an ample staff and the general experience level of Lite and corporate personnel was higher than typical at other plants. There was considerable management attention on addressing employee morale, work practice, and labor-relations issues, becanas of the low employee esteem that had developed from past plant problems.

Management-union relationships had greatly improved in recent years. Employee grievances dropped significantly in the last year. Many union issues appeared to have been resolved over the last 2 years, with the most notable regarding the selection process and pay grades for control room operators. Substantial improvements also were made in the area of industrial safety through the implementation of several programs, including a safety committee, professionalism, employee recognition, and prohealth. The union actively participated in these programs.

29

2.4.3 Plant Monitoring Most of the objectives in the excellence plan had an associated performance indicator, site goals were established, and plant performance was monitored in monthly trend reports. 'The plant monitoring reports indicated that an improving trend in Oyster Creek performance for the first 9 months in 1990 was evident.

Areas showing improvement in plant performance included (1) reduction in the forced outage rate, (2) reduction in the number of unplanned automatic scrams, (3) increase in unit equivalent availability, (4) reduction in the corrective maintenance backlog, (5) increase in the ratio of preventive to total '

maintenance performed, and (6) reduction in the overall plant radiation exposure. In addition, improvements in plant housekeeping were evident. i However, maintenance rework as used at Oyster Creek was too narrowly defined. The maintenance rework performance indicator reflected "zero" rework for 1990, through October. The equipment failure trending program had not been effectively addressing equipment failure problems in a timely manner because of the high threshold for trend action notices. Thus, the rework indicator did not accurately reflect actual plant performance. Actions to address weaknesses in the trending program were being initiated and a revision of the definition of rework was being prepared at the time of the diagnostic evaluation. Additional performance indicators were being considered for the monthly trend reports to monitor more of the objectives contained in the excellence plan.

2.4.4 Work Planning and Scheduling The plan-of-the-day meeting and rolling schedules were effective management tools to communicate plans, to modify plans, and to allocate appropriate resources for operating cycle work activities. The licensee was making a substantially greater effort in integrated outage planning for the 13R refueling outage scheduled to begin in February 1991. These efforts included the use of an operations outage coordinator and substantial completion of engineering work before the start of the outage.

In addition, input to the outage plans from radiological controls and other departments also had improved. Planning and scheduling of lower-level issues, however, was not as effective as those 30

_ _ _ _ - _ _ _ _ _ _ - -_ - - - - - - - - - - --- ~-~~

.. . s perceived as "important" by management. Progress on some initiatives was limited. Most notable among these was the life

~~ of syster maintenance plans, which initially lacked a clear definition of goals and objectives, procedures, completion schedules, and dedicated resources. Most of the initiatives that did not rise to the level requiring corporate planning'did not have good estimates of the resources needed, well-developed implementation plans, or well-defined completion schedules.

These activities appeared to be accomplished on an ad-hoc basis, usually by assigning them to particularly competent individuals whos workload were already heavy. This was a prime stated reason for delays in lower-level programs and initiatives.

Work tracking and prioritization was the responsibility of the individual departments, in accordance with plant management practices. Consequently, the quality assurance group used or accessed 42 tracking systems to conduct audits of pending actions.

Operations had an excellent tracking system for all identified work, however, the responsibility for prioritizing and delegating the work was not well-defined or clearly understood. Priority assignment to operations personnel was not evident for some overdue issues. For example, several pending operator concern itens were open 12 months past their initial due date with no increased attention focused to resolve the issues. While the operations managers were confident in their supervisory skills to prioritize staff workloads, continual shifting of priorities occurred because of reactive needs. One example of shifting priorities was the effort to improve the plant's configuration control procedure that had been on-going for over a year.

In the other departments, similar weaknesses in either the prioritization or tracking of work assignments were identified.

As a result, when the excellence plan was initially developed, the completion schedules were not realistic; delays encountered during the implementation of the action plans would likely extend the completion schedules beyond the expiration of the plant license. Overall, there did not appear to be a consistent, systematic means of prioritizing and tracking work that would promote a transition from reactive to proactive management practices.

31 ,

l 2.4.5 problem Identification A variety of means existed to identify deficiencies. Initiatives to improve identification and reporting of problems at the plant, such as the operator concern program and programmatic changes in the deviation reporting system, appeared to be having a positive effect in bringing problems to the attention of licensee management. The operator concern program was originally intended as an informal process to improve communications within the operations department. However, the program gained wide acceptance; plant management was encouraging personnel to identify concerns through that process. Consequently, although the operator concern program was successful in identifying problems, that process also had become a surrogate for other, formal deficiency reporting procedures; the issues then had to be translated into the appropriate corrective action system. In addition, a large number of the operator concerns have rewained open for an extended time period.

The deviation reporting system had been increased in scope to allow identification of equipment, program, and personnel performance deficiencies. In the past, the deviation reporting process was used almost exclusively for events that were reportable to the NRC. Based on the increase in deviation reports over the past year, the licensee's efforts to expand the use of the process appeared to be successful. The established method to review significant plant transients appeared to be effective. The plant transient review group (PTRG) and the independent transient review group (ITRG) provided a systematic method to ensure that transient events were understood, including the response of affected plant equipment.

2.4.6 Problem Resolution Problem resolution for highly visible concerns generally received adequate and effective management at .ntion and oversight. In these areas, resolution had specified both short- and long-term actions, and related plant performance trends had shown improvement. Areas that had demonstrable improvement included radiological control, work backlog reduction in maintenance and engineering, and plant housekeeping.

In areas with less visibility to plant management, problem 32

_-. - - - - ..-.. - . - . . -. - . - -- _ = . _ - - .

1 l

resolut$on'and decision making often occurred at the lowest levels of the organization, without ensuring adequate management oversight, support, or independent verification. The complexity of tra'cking systems used to ensure resolution had increased the ambiguity of accountability and ownership of the issues and, at times, appeared to have resulted in delays or ineffective resolution of problems.

Plant transient reviews usually developed effective short- and long-term corrective actions. Further, the licensee's use of a critique process effectively identified root causes for events that resulted in unexpected component or system responses.

However, some critiques were not performed in accordance with the established procedure, which led to a narrow root-cause determination. In addition, critique assignments were not performed in several instances, resulting in no root-cause determination. For example, the plant labeling program identified numerous errors in plant drawings. These errors were believed to be the result of drafting mistakes or poor past practices in controlling plant modifications. While the specific errors were corrected through the field change notice (FCN) process, a root-cause determination was not made either for individual discrepancies or on a generic basis. The FCN process was the proper means to determine acceptability of the as-found plant configuration, however, an assessment of the broader implications of these discrepancies was not performed.

The recent changes to the deviation reporting process did not appear to have improved the thoroughness of the root-cause evaluations or the timeliness of the subsequent corrective actions, as was found in some of the issues reviewed in the engineering and maintenance areas. In addition, some of the-deviation reports hed been closed without completing the required root-cause determination.

The effectiveness of efforts to improve the resolution of problems appeared hampered by weak administrative controls for tracking and closure and poor training on the process. The licensee indicated that training for the deviation reporting process was planned for the future.- Further, licensee personnel were addressing some deficiencies but had not formally documented the related deviations, such as inoperable sump pumps in the emergency diesel fuel oil tank room. This failure to document 33

.i . 4 some deficient conditions appeared to be symptomatic of a more general aversion to formalized practices. The licensee was in the process of reviewing 10 formal corrective action processes, with a goal of reducing the number of processes.

Problem resolution was further impeded by examples of' poor conduct or lack of a root-cause analysis. The quality of root-cause analyses appeared to depend principally on the perceived importance of the issue. In cases where the visibility was low, there was often a lack of documented root-cause analysis, erratic performance in the conduct of the engineering evaluations, and occurrences of multiple equipment failures without adequate resolution of the underlying problem. In addition, independent verification or some challenging of the root-cause analyses appeared limited. While the Safety Review Manager raviewed the closure of all of the deviation reports, that appeared to be an overwhelming task for one individual, which limited the effectiveness of the review.

2.4.7 Work Practices There was a lack of attention to detail in the practices for documenting work, which resulted in important information being overlooked. Examples included the lack of documentation for maintenance findings and material condition deficiencies. There was a general lack of inquisitiveness during problem analysis, as demonstrated during the evaluation of the valve leakage in the feedwater minimum-flow line and with the history of problems with the isolation condenser condensate return valves. This lach of a questioning attitude also was evident in some deviation reports, which had been closed out without the root cause or corrective action having been specified. There appeared to be a wilu ngness to accept ready solutions to problems without challenging the understanding of the failure mechanisms or the broader implications.

Actions to resolve deficiencies or make plant improvements often took a long time to implement. Some long-term initiatives seemed to be scheduled over extremely long periods, such as the maintenance initiatives, development of design basis documents, and the operations procedure upgrade program. Further, some action items resulting frem findings reported in self-assessments, audits, and critiques were often not implemented in 34

.- .~.

a timely manner. As previously noted, there appeared to be a heavy reliance on the skills of particular individuals to resolve problems, which at times led to inefficiency and untimeliness depending on the workload of those individuals. This practice also exacerbated the productivity of the entire workforce.

2.4.8 Improvement Initiatives The licensee did not have effective administrative control of the development and implementation of improvement initiatives. The licensee established over 130 improvement initiatives within the last 3 years, which included both formal and informal programs.

The number of formal programs was actually much smaller. The excellence plan consisted of a collection of improvement initiatives and performance goals. The most significant of these were labeled as " blue chip" items, which consisted of the radiologicai improvement program, the expectations and standards program, and the 13R refueling outage. The apparent proliferation of improvement initiatives appeared to be symptomatic of reactive management because of the tendency to initiate a new program whenever an issue was identified.

The team found that the excellence plan clearly laid out important goals and objectives and addressed most problems identified by the team or by the licensee's audits or assessments. However, not all of the action plans in the excellence plan appeared to be effectively controlled. The licensee indicated that schedular delays in implementing the excellence plan resulted because the initial schedules were overly optimistic and because reactive work (plant needs and NRC L requirements) made it necessary to slip schedules. Further, plant managers were not completely aware of tha total number and scope of improvement initiatives until they began preparations for the diagnostic evaluation.

t plant management recently initiated a number of programs to l

enhance employee work practices and productivity, including the

_ professionalism program, expectations and standards program, teamwork and leadership program, and the self-checking program.

In addition, model spaces had been used to communicate expectations for plant cleanliness, including decontamination, l lighting improvements, and painting.

35

s. . . - ,. -. . --. -. . , __- . _ _ _ _ - _ - _ _ _ _ _ - ___-

Ownership of the improvement initiatives appeared to vary. Some programs were closely monitored by licensee management, were highly visible, and received adequate support, resulting in a

~

positive influence on performance, such as a reduction in the overall plant radiation exposure. Other than the " blue chip" items of the excellence plan, however, a number of initiatives did not receive sufficient management attention and resources.

Consequently, there were some instances of poor coordination of initial assignments and general slippage of the action plan schedules. poor prioritization and tracking of work at the department level also appeared to have contributed to delays.

The licensee viewed many of the numerous improvement initiatives as lower priority work that should not conflict with more immediate needs. However, plant management indicated a desire to change this practice in the future and elevate the priority of improvement initiatives above " bulk work."

3 ROOT-CAUSE ANALYSIS OF PERFORMANCE PROBLEMS For more than a decade of Oyster Creck's early operation, little maintenance was performed and the plant material condition generally deteriorated. Following the TMI-2 accident, new regulatory requirements and evolving equipment problems resulted in a large backlog of reactive work. Only in the last few years has this backlog of work declined to a point where proactive initiatives were pursued.

The licensee was working primarily to recover the material condition of the plant that had deteriorated because of minimal maintenance during the early history of the plant. Although material condition improvements were evident, as demonstrated b'/

the striking differences between those areas that were improved and those areas awaiting improvement, substantial additicnal work was planned for at least the next three refueling outages.

Recent improvement efforts over the past ftw years had been largely directed towards employee morale and development. Many other improvement initiatives were only in early stages of development and implementation at the time of the diagnostic evaluation. 5 The licensee's accomplishments thus far were attributed to a 36

.,4  :,-_ t skilled and experienced staff and management attention to high-priority _ issues. Inadequate management attention to_the less-visible tasks led to the performance weaknesses discussed below.

3.1 Inadecuate Administrative control ,

Weak administrative controls hampered productivity. There was no apparent.i nf rast ruct ure in-the licensee's organization to ensure the systematic completion of work. The complexity of the work tracking systems, continued reactive management practices, and the lack of a consistent basis to assign priority for specific

. tasks hampered the timeliness of corrective actions and the

' licensee's productivity.

Although the excellence plan laid out important goals and objectives to promote performance improvements. - Many improvement-initiatives were not effectively controlled and there was little demonstrable accountability for completing assigned tasks.

Moreover, the licensee's tendency to develop improvement initiatives to solve problems compounded the administrative control of its work activities. Only reactive tasks, highly visible projects, and those influencing performance indicators were well defined and regularly monitored.

.A consistent, overall method for assigning priorities and s controlling schedules was_not evident for many overdue actions.

The less-visible tasks were prioritized and tracked at the department level,.such.that level of effort and oversight were driven.primarily by_ department agendas. Continual shifting of priorities, resulting from reactive management practices, '

frustrated the plant staff and tended-to overload particularly

competent individuals.

The-number and complexity of corrective action systems, while providing a variety of means to ensure that deficient conditions were identified, appeared to hamper the timeliness and effectiveness of corrective-actions. The licensee readily identified problems through self-assessments and reporting systems, and readily developed programs and initiatives to address the problems,-but appeared reluctant to formalize these programs.and initiatives with appropriate objectives, implementing. procedures, schedules, and assigned responsibilities.

37

-- s,ww- c- ,,~< , , e- e m v -

.. . /

l l

I Where deficient conditions were identified, there was no apparent systematic means to determine how soon corrective actions were necessary. Thus, even though effective corrective actions may have been developed, the basis for the timeliness of the action was ambiguous and the basis for the functional capability of the affected equipment in the intervening period was not clearly specified.

3.2 Weak Supervision and Independent Verification Weak r,upervisory oversight and independent verification caused some inadequate evaluations and poorly-reasoned decisions.

Weaknesses in accountability and independent verification of the bases for conclusions and decisions undermined the effectiveness of the work efforts and employee ownership of assigned responsibilities.

The management philosophy that problems should be solved at the lowest possible IcVel and that the site was a " customer" for engineering and support organizations led to some poorly founded decisions and erroneous conclusions, because those decisions and conclusions were not routinely challenged or tested by supervisors or verified by the quality programs. Testing conclusions and decisions is usually intrinsic to the work practices of the staff and supervisors and/or routinely performed by quality assurance as a means of independent verification.

Neither testing or independent verification of decisions and conclusions seemed to occur for work that was perceived to be unimpertant to upper-management. As a result, the quality of work depended primarily on the skills of the personnel involved and their collective perception of the importance of the task.

3.3 Lack of Ricor and Inauisitiveness in Work Practices Work practices often lack rigor and there was a general lack of an inquisitive attitude. While many good or acceptable work practices were observed, several examples of poor work practices also were observed. The extent of rigor in a particular task appeared to depend primarily on the perceived importance. This inconsistency resulted from a general lack of rigor in the licensee's work practices, which hampered the ability of 38

5 i

personnel-to consistently achieve effective solutions to problems.

Moreov'er, several of the poor work practices appeared to result from the lack of an inquisitive attitude. Licensee personnel and management seemed willing to accept ready solutions to problems without questioning their understanding of the problem or broader implications of the problem. The licensee did not always clearly understand failure mechanisms.

4 EXIT MEETING On January 18, 1991, the Dir ector of the Of fice for Analysis and Evaluation of Operational Dana, the Regional Administrator for Region I, the Acting Assistant Director for Region I Reactors in the Of fice of Nuclear Reactor Regulation, the Team Manager of the Oyster Creek diagnostic evaluation team, and other NRC staff members met with the President of GpU Nuclear and several senior managers responsible for the plant, at the Oyster Creek site. At the meeting, the NRC presented the results of the Oyster creek diagnostic evaluation. Briefing notes summarizing the teams findings and conclusions are provided as Appendix A. The list of attendees is provided as Appendix B.

Following introductions, E. L. Jordan, Director of AEOD, began the meeting with an overview of the diagnostic evaluation process ,

and commented on the protracted time for improvements at Oyster 1 Creek. C. I. Grimes, the Team Manager, presented the findings and conclusions of the team in the functional areas of operations and training, maintenance and testing, engineering and technical cupport, and management effectiveness and corrective actions. The probable root causes were then discussed. Comnents and questions were then solicited and received frcm the licensee. Mr. Jordan then emphasized that the team's findings should be used in a way that improves safety performance.

E. G. Greenman, Acting Assistant Director for Region I Reactors, commented on the ambitious programs to improve performance and the well-skilled and dedicated staf f being the key to sustaining improvements.

39 l

l

e .,

P. R. Clark, President of GPU Nuclear, stated that the company will carefully consider the team's findings and integrate actions with those already underway. Mr. Clark stated that they have a ten-year commitment to make Oyster Creek an excellent asset and improvement efforts will continue.

J. J. Barton, Director of Oyster Creek, stated that he appreciated the efforts of the team and the efforts to minimize the burden that could impact the upcoming refueling outage. Mr.

Barton also stated that they were in transition and gave examples of improvement efforts. He then stated that they will continue those efforts and seek new opportunities for improvement.

40

f. ' ~ J*

APPENDIX A F

OYSTER CREEK DIAGNOSTIC EVALUATION TEAM RESULTS

SUMMARY

-Evaluation Period:

November 5 to December 7,1990 l

l A-1

.I - . _ - . - _

1 l

(

SUMMARY

OF FINDINGS

• Substantial improvements had been accomplished.

• Substantial work remained to be completed.

• Some lessons were learned from the DET.

• Performance weaknesses were identified.

A-2

I I

OPERATIONS & TRAINING 1

• Overall conduct of operations was good, e Operators were professional and attentive, o Control room material condition was good, o Communication was good and improving.

• Staffing plan was working well, but the training staff was strained, o Except for delayed completion of the simulator, training materials and equipment were very good.

e. Improvement initiatives were having a positive effect.

A-3 I

e e i

MAINTENANCE & TESTING .

• Organizational changes enhanced maintenance efforts.

• Material recovery efforts were still under way.

• Work planning and prioritization were improving.

• Failure trending was not used effectively and failure analysis was poor.

• Preventive maintenance scope was not systematically-defined.

• Work practices were generally adequate, with some exceptions.

• Improvement initiativas lack " program" attributes:

clear definition of goals and objectives, implementing procedures, completion schedules, and dedicated resources.

l L A-4 l

ENGINEERING & TECHNICAL SUPPORT E

o Recent organizational improvements.

• The quality of engineering and technical support varied.

. Some tasks performed well.

. Some tasks lacked rigor, thoroughness and inquisitiveness.

No systematic means to determine the existing adequacy of equipment with deficient conditions, or how soon corrective action would be necessary, e

A-5

MANAGMENT EFFECTIVENESS & CORRECTIVE ACTIO e

Adequate management oversight.

Performance improvements evident.

Lower-level planning varied and usually lacked clear definition of objectives and implementation procedures.

) e improvements in the identification and reporting of problems.

Efforts to improve employee inorale, employee work practices, and labor relations were evident, o

Effective resolution of " visible" problems, but poor resolution of less-visible problems.

Proliferation of improvement initiatives symptomatic of reactive management practices.

A-6

.. .. . r-ROOT CAUSE EVALUATION Oyster Creek was in transition, still recovering from minimal maintenance during early history and

-morale problems.

• Tasks that were not-highly visible often lacked adequate management attention and support.

There was weak administrative control of work which hampered the systematic completion of work.

Weak supervision and independent verification had caused some inadequate evaluations and poorly-reasoned decisions.

Work practices exhibited lack of rigor and lack of an inquisitive attitude.

l l

l l

A-7 RcV. I 1/24/91 L

r "1.<

APPENDIX B EXIT ATTENDEES JANUARY 18, 1991 Name Oraanization E. L. Jordan NRC, Director AEOD T. T. Martin NRC, Region 7 Administrator C. I. Grimes NRC, Team Manager E. G. Greenman NRC, Acting Assistant Director DRP I J. F. Stolz NRC, Project Director PD I-4 A. W. Dromerick NRC, Oyster Creek Project Manager C. W. Hehl NRC, Region I Director, DRP W. H. Ruland NRC, Section Chief, DRP E. E. Collins NRC, Senior Resident Inspector M. Banerjee NRC, Resident Inspector J. A. Nakoski NRC, Resident Inspector P. R. Clark GPUN, President J. J. Barton GPUN, Director of Oyster Crook J. C. Devine GPUN, Director of Technical Functions E. E. Fitzpatrick GPUN, Vice President I. R. Finfrock GPUN, Director of Site Services R. J. Barrett GPUN, Plant Operations Director L. L. Lammers GPUN, Plant Maintenance Director A. H. Rone GPUN, Plant Engineering Director J. E. Frew GPUN, Site Services Director M. J. Slobodich GPUN, Radiological Controls Director W. H. Behrle GPUN, Technical Functions Site Director M. F. Dudaj GPUN, Manager of Plans and Programs D. T. Barnes GPUN, Plant Engineering Supervisor, I&C C. Lefler GPUN, Technical Functions Site Manager R. W. Kuaten GPUN, Director of Performance Improvement R. F. Fonti GPUN, Site QA Hanager M. W. Laggart GPUN, Corporate Licensing Manager S. Polow GPUN, Site Communications Manager R. T. Ewart GPUN, Senior Security Supervisor K. R. Neddenien GPUN, Media Relations Manager R. J. Hillman GPUN, Plant chemistry Manager f C. A. Scarpenato GPUN, Labor Relations Administrator l

M. Bradley GPUN, I&C Superintendent B. T. Moroney GPUN, Nuclear Safety & Compliance Committee E. P. O'Donnell CPUN, Director of Corporate Planning E. F. O' Conner GPUN, Engineering Services Director C. R. Tracy GPUN, Engineering Project Director D. C. Smith GPUN, Corporate Assessor, Rad Con & Safety J. E. Hildebrand GPUN, Radiological & Environmental Controls l

l B-1 I

1 j

I .

. s, s ',

i l

I Name Oraanizatipa D. E. Tuttle, Sr. GPUN, Radiological Controls Deputy Director  !

R. L. Sullivan GPUN, Emergency Preparedness Manager I W. Muehleisen GPUN, Maintenanca Support Superintendent l J. K. Gulati GPUN, Oyster Creek Projects Manger l R. J. Blouch GPUN, Planning & Support Manager R. E. Weltman Consulting Engineer R. E. Brown GPUN, Radiological WP.ste Manager 1 P. Hayes GPUN, Operator Training J. Solakiewicz GPUN, Quality Assurance R. L. Long GPUN, Director of Coporate Services & THI-2 P. B. Piedler GPUN, Director of Nuclear Assurance C. Clawson GPUN, Director of Communications C. A. Mascari GPUN, Director of Quality Assurance G. W. Busch GPUN, Oyster Creek Licensing Manager P. F. Scallon CPUN, Plant Operat. ions Manager W. J. Quinlan GPUN, Station Services Manager H. Chrissotimos GPUN, Outage Management W. V. Stewart GPUN, Safety Review Manager J. D. Kowalski GPUN, Plant Training Manager D. MacFarlano GPUN, Site Audit Manger P. Thompson GPUN, Lead QA Auditor D. K. Cronoberger GPUN, Outage Management Director R. P. Coe GPUN, Training & Education Director G. Miller GPUN, Administrative Support Manager C. Brookbank GPUN, SSD Engineering Supervisor C. A. Pollard GPUN, Rad con Field Operations Manager E. Reilly GPUN, Expenditure Analysis Supervisor J. Halsey GPUN, Administrative Support Supervisor L. Schreiber GPUN, Startup & Test Manager J. A. Camire GPUN, Plant Analysic Manager W. Pelenski GPUN, Plant Process Computer Manager D. G. Weeden Vice President, Local 1289 D. Opdyle Local 1289 K. E. Dundon Local 1289 R. Cook GPUN, Human Resources Manager J. Derby ._ GPUN, Radiological Engineering J. M. Doblacio '" GPUN, Plant Engineering Manager C. L. Elliott, Jr. GPUN, Hochanical Maintentance Superintendent R. Renzi GPUN, Information Management Manager R. Martin GPUN, Mail / Administrative Support Supervisor D. L. Pysher GPUN, Facilities Site Services Manager R. Schwarz GPUN, Labor Management Director D. Myers GPUN, Director of Administration & Finanaco E. Roessler GPUN, Manager of Nuclear Safety J. L. Sullivan, Jr. GPUN, Director of Independent Safety Review A. L. Herfel GPUN, Director of Material Management B-2