ML20198A811

AEOD/S602, Overview of Nuclear Power Plant Operating Experience Feedback Programs

ML20198A811
Person / Time
Issue date:	05/09/1986
From:	Crooks J NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
Shared Package
ML20198A780	List: ML20198A776 ML20198A783 ML20198A800 ML20198A811
References
TASK-AE, TASK-S602 AEOD-S602, NUDOCS 8605210270
Download: ML20198A811 (48)

Text

.-

s c.

Report No. AE0D/S602 Date May 9,1986 An Overview of Nuclear Power Plant Operating Experience Feedback Programs Prepared by John L. Crooks (Lead)

Program Technology Branch Office for Analysis and Evaluation "

of Operational Data U.S. Nuclear Regulatory Commission This subject matter is under continuing review.

This report supports ongoing AE00 and NRC activi-ties and does not represent the position or re-quirements of either other NRC program offices or the Nuclear Regulatory Commission.

e 8605210270 860509 PDR ORG NEXD

t a Table of Contents Page No.

EXECUTIVE

SUMMARY

............................................... 1 ,

1.0 INTRODUCTION

............................................... 4 1.1 Background ............................................ 4 1.2 Study Methodology...................................... 4 2.0 DISCUSSION ................................................. 6 2.1 Concepts, Requirements and Previous Assessments........ 6 2.1.1 Operating Experience Feedback Concept .......... 6 2.1.2 Requi remen ts and Guidance . . . . . . . . . . . . . . . . . . . . . . 7 2.1.3 Previous Assessments ........................... 8 2.2 Licensee Operating Experience Feedback Activities...... 10 2.2.1 Reasonable and Practical Elements............... 10 2.2.2 Characteristics of Actual Performance. . . . . . . . . . . 13 2.2.3 Variability in Licensee Activities . . . . . . . . . . . . . 16 2.2.4 Shortcomings ................................... 17 2.3 Operating Experience Feedback Documents - Their Use and Related Problems .................................. 19 2.3.1 Operating Experience Feedback Documents ........ 19 2.3.2 Burden, Overlap, Conflicting Information and Timing ..................................... 23 3.0 FINDINGS.................................................... 28 3.1 Operating Experience Feedback Programs ................ 28 3.2 Feedback Documents .................................... 29

4.0 CONCLUSION

S ................................................

31 4.1 Operating Experience Feedback Programs................. 31 4.2 Feedback Documents..................................... 32 S.0 RECOMMENDATIONS ............................................ 33 APPENDICES APPENDIX A Part 1 of Enclosure 2 to af 7, 1980 letter from D. Eisenhut to Licensees on Addit-ional TMI-2 -

Related Requirements.............................. 35 APPENDIX B Gene'rfc Letter No. 82-04......................'.... . 37 APPENDIX C Summary Tables from Visits........................ 40 11

EXECUTIVE

SUMMARY

In 1980, as a consequence of the lessons learned from the Three Mile Island accident, the NRC issued requirements for nuclear power plant licensees to implement an operating experience (OE) assessment function. This action was to stress the importance of operating experience to improve overall reactor safety.

The primary purposes of this AE0D study were to (1) determine the character-istics of representative licensee OE feedback programs; (2) determine how feedback documents from various NRC offices, Institute of Nuclear Power Operations (INP0), vendor, and other organizations, including AE0D publica-tions, are used in such programs; and (3) to determine if there is a need for changes to the NRC's operational experience feedback program or NRC requirements governing licensee activities related to OE feedback. The study, which began in mid-1984, consisted of reviewing the regulatory requirements for licensees to have OE feedback programs; visiting seven licensees to discuss their activities; observing INP0's evaluation of OE review activities; discussing program issues with involved personnel; assessing the resulting information; and developing findings, conclusions and proposed further actions. Because of the variability that exists among licensee programs, the deficiencies or problems noted in the findings and conclusions may not apply to every licensee.

The principal findings and conclusions from the study are:

(1) Over the last five years, many significant and worthwhile initiatives have been implemented to understand the lessons of experience. As a

~

result, there has been increased attention paid to OE feedback. However, many licensee programs have not yet achieved effective performance levels in all areas. Most, plants are making moderate, not extensive, use o.'

their in-house operating experience, and in general are making less use of the large body of knowledge associated with events and concerns that originate elsewhere in the industry. Thus, many licensee OE review activities still do not provide high confidence that all of the important lessons of operating experience are being thoroughly explored and used.

To increase the effectiveness of OE reviews in order to correct past and potential operational problems (i.e., assuring safety of operations),

licensees and the industry or the NRC need to take further actions.

(2) The NRC, INP0 and other organizations within the industry are expending large amounts of resources on the screening and assessment of operational data and the preparation and dissemination of OE feedback documants.

However, extensive in-depth assessments of this information by licensees and the expected voluntary actions in terms of changes in plant hardware, procedures, and training programs based on this industry-wide feedback were not readily apparent in many cases. The resources and attention being devoted by licensees to understanding and implementing the lessons from experiences at other plants are usually considerably less than those directed to in-house feedback. As a result, the yield from thase activities iri terms of corrective actions and improved operator knowledge and capability and the associated reduction in component failures and plant events seems smaller than expected.

I 1

s s (3) TMI action Item I.C.5 intended that the lessons of experience be routinely provided to all operating personnel in an efficient and effective way. However, plant operators (licensed and unlicensed),

trades personnel, technicians, and other non-supervisory personnel are not typically and effectively provided much OE information beyond events that occur in their own plants.

(4) The NRC requirements and guidance in this area are very general (i.e.,

technical program goals and objectives, evaluation criteria, or expected activities are not fully specified). The industry, through INP0, has developed more detailed guidance on how OE review activities can be implemented, but licensee activities still are widely diverse and vary considerably in effectiveness.

(5) NRC routinely verifies (i.e., inspects) the effectiveness of only a portion of current operating experience feedback practices on a plant-by-plant basis, although the NRC has attached great importance to this activity. The NRC also lacks information industry-wide on the range of effectiveness of OE review activities and the degree to which plants are meeting the established NRC requirements. At this time however, the general NRC requirements do not provide a sufficiently definitive basis to permit a meaningful evaluation of the total process by the NRC.

Current industry efforts also do not provide a suitable basis for judging the effectiveness of licensee activities in this area.

(6) The large volume of OE feedback is degrading the effectiveness of the feedback programs. Duplication and overlap are div,erting resources available to effectively use OE feedback.

(7) Conflicting information provided by two or more sources pose major difficulties for licensees unless the conflicts are identified. Thus, when the initiators of feedback are aware of differences in proposed actions or event details, the differences need to be noted and discussed in order to aid the licensees' evaluations.

(8) AE00's three periodic publications [the bimonthly Power Reactor Events (PRE) the quarterly Report to Congress on Abnormal Occurrences (A0s),

and the monthly Licensee Event Report (LER) Compilationj are not as useful and effective to licensees as was indicated by previous surveys and discussions with users.

Based on this study, AE0D recommends that IE consider informing licensees of the conclusions drawn from this study and suggesting that each licensee review their OE activities for effectiveness and conformance with the intent and scope of the existing NRC requirements. No major regulatory actions are being recommended because AE00 is aware that industry initiatives in this area are still on-going. AE0D recommends that the effectiveness of OE activities continues to be monitored by staff through operating experience reviews and observations of OE review activities. If, after sufficient 2

time (two years or so), the effectiveness of OE review activities has not improved, further regulatory actions would be appropriate. In this regard, AEOD recommends as a prudent measure that NRC initiate the development of better guidance for licensees' OE activities, and an inspection module based on that guidance. In addition, AE00 recommends that the staff consolidate existing systems for communicating operating experience into a single "NRC Notice" system. This system would thus be the sole NRC system for the feedback of operating experience to licensees.

e e

3

}

l

1.0 INTRODUCTION

1.1 Background

In 1980, as a consequence of lessons learned from the Three Mile Island acci-dent, the NRC and the nuclear industry took many actions to improve overall reactor safety. One of the principal changes was to stress the importance and use of OE feedback to prevent serious events from occurring and to mitigate the consequences of events that do occur. Two of the specific NRC actions taken were: (1) a reorganization within the NRC, including the formation of the Office for Analysis and Evaluation of Operational Data (AE00), to improve the use and feedback of OE, and (2) the issuance of general requirements for nuclear power plant licensees to implement an OE assessment function. The latter action was to ensure that the lessors learned from significant safety-related operating experiences both internal and external to the plant (s) were acted upon, communicated to the plant operators, and factored into various aspects of operations; (e.g., maintenance, testing, modification, and train-ing) at all applicable plants.

The industry-wide OE feedback activities are extensive because the U.S. nuclear industry is large and complex. Some lessons from operating experiences are very plant specific, while other lessons can apply to groups or types of reactors or to all plants. Thus, because plants often differ in details of plant layout, equipment or operational parameters, detailed assessment is frequently required to determine whether operating experiences are plant specific or applicable to other units.

The primary purposes for this study were to (1) determine the characteristics of representative licensee OE assessment activities; (2) determine how the feedback documents from various NRC offices, INP0, vendor and other organiza-tions, including AE0D publications, are used in such programs; and (3) deter-mine if there is a need for changes to the NRC's OE feedback activities or NRC requirements governing licensees' activities related to OE feedback. The study was started in mid-1984 and a preliminary report was issued for peer review in late December 1985.

1.2 Study Methodology The study consisted of reviewing the regulatory requirements for licensees to have an OE assessment function; visiting seven licensees to discuss their programs; discussing program issues with involved personnel; observing one INP0 plant evaluation; assessing the resulting information; and developing findings, conclusions and proposed further actions.

A list of questions was developed for informal discussions with licensee personnel. These questions dealt with the structure of the licensee's OE program, including which OE feedback documents were being used, how they were being used, how much material licensees were currently receiving, and what problems were being experienced.

t 4

During the visits with licensees, the discussions were oriented to prngrammatic issues and fact finding, and did not address details from an inspectinn and enforcement perspective. The scope and content of the discussions became more focused as the study and the sampling progressed. There was also some variety in the personnel contacted and in the content of the discussions, since differ-ent licensees assign similar responsibilities to different organizational ele-ments. The discussion section of this report includes the facts gathered and the observations made during the plant visits. The findings, conclusions and recommendations sections address the major issues identified during the study.

Because of tne variability that exists among licensee activities, the defi-ciencies or problems noted in the findings and conclusions vary, and specific findings may not apply to every licensee.

e o

e I

5

• 7

~

2.0 DISCUSSION -

2.1 Concepts, Requirements, and Previous Assess [nents

' " ^

2.1.1 Operating Experience Feedback Concept, ,

The concept of OE feedback, simply put, is to analyze unusual, abnorma'r, undesired or. unintended operational events; to determine:their cause, sionifi-cance and implications; and to develop and implement the specific corrective action to prevent recurrence or'to. mitigate consequences.

Although the concept of OE feedb$ck has been an element of the nuclear power industry since its inception,'its importance at times has not been fully appreciated. As a result, corrective. action to preclude repetition of past ,

problems has not always been implemented as effectively or promptly as desired.

Consequently, events sdbh as the Browns, Ferry fire, the Three Mile Island accident, and the Salem anticipated transiegt without scram (ATWS) have occurred even though precursor events were known th,at indicated the potential for such serious events. 'y OE feedback activities are to include:'. operational experiences pertinent to nuclear safety from both in-house and external sources; the collection and storage of that information; the analpis of the information for significance and applicabilit/ to a li'censee's plant; the Aissemination of the analysis results for action; and the eventual incorpuration of ch procedures, and training to make plant Operation safer.The '3nges in hardware, objectives of /

these activities as viewed by AE0D are to (1) reduce the f equency of ' -

undesired safety-related events, and (2) prevent the occurrence of, serious nuclear incidents by identifying safety significant precursor events and taking appropriate preventive or mitigative actions. ~

The effectiveness of the activities should be mesmrab7e in terms of fewer repetitive events; improved plant performance factors (i.e., cap sity factors -

and availability); improved availability and performance of safety ecuipment; fewer repetitive human errors and more knowledgeable, better performing operat-ing staffs. More specifically, effectively implemented programs should reduce '

the frequency of events such as reactor trips due to. improperly performed -

surveillance or maintena'nce; safety systems being inadvertently valved'out and thus unavailable for extended pe';iods of time; and safety problems such as uncontrolled personnel entries into a very high radiation field (31000 R/hr) in the reactor cavity area with the instrumentation thimbles retracted. At present, these types of incidents seem to frequently recur. For example, the ,

eleventh and twelfth recurrence of personnel entering high radiatic? fields in the reactor cavity area occurred in late 1983 and early 1984 after repeated .

notices to licensecs regarding significant overexposures during such events.

The frequent recurrence of problems with motor-operated, safety-related valves malfunctioning due to improperly set torque switches and of problems with steam ,

binding in PWR auxiliary feedwater systems ate additional examples where considerable operating experience has been fed back, but corrective actions have not been effectively implemented in some instances.

Two aspects of OE feedba'c k about which we have become particularly concerned are, first, assuring that plant operations personnel (e.g., control room 6

9 - - - .

e D operators, auxiliary systems operators, trades and maintenance personnel) are knowledgeable about the problems being experienced in the day-to-day operations of their plant and other similar plants; and, second, insuring that such pro-grams draw from the broadest base of experience and perspective and all pertinent and significant lessons are acted upon.

2.1.2 Requirements and Guidance ,

One of the actions that NRC took subsequent to the Three Mile Island Unit 2 accident was to require that each nuclear power plant licensee formalize an OE assessment function to (1) incorporate the lessons learned from operating experience into plant hardware and procedures, (2) ensure that the lessons learn d are communicated to plant operators and other personnel, and (3) incor-porate OE information into training and retraining programs.

The NRC requirements consist of general guidance to licensees [e.g., the Three Mile Island (TMI) Action Plan, NUREG-0737, Item I.C.5], which was implemented via a May 7, 1980 letter sent from NRR to all licensees. The guidance states that licensees are to carry out an OE assessment function and to have pro-cedures covering their operational assessment and feedback activities. An appendix to the May 7, 1980 letter (see Appendix A) specified seven items the plant procedures were to include, provided general guidance in the discussion section, and stated that procedures governing feedback of operating experience should be in place by January 1,1981 for operating plants. This guidance, however, does not discuss the performance objectives or goals of the procedures with respect to safety nor does it discuss any measures for evaluating the effectiveness or acceptability of the activities. Therefore, the requirenents implicitly made the licensee responsible for structuring his activities, determining the goals and objectives for his activities, and implementing the procedures to meet the regulatory intent.

In support of industry efforts to reduce duplication of efforts in reviewing operating experience, to centralize screening of the data, and to ensure that plant resources can focus on evaluation and implementation of the lessons learned, NRC issued Generic Letter No. 82-04 (see Appendix B) on March 10, 1982, to endorse the use of INP0's Significant Event Evaluation and Information Network (SEE-IN) program as an acceptable method for meeting part of NRC's requirements. SEE-IN is a mechanism for the centralized collection and screen-ing of events reported by both U.S. and foreign nuclear plants. Full partici-pation by licensees enhances their capability.to meet the intent of TMI Action Plan Item I.C.5, by permitting them to focus their resources on evaluating and responding to OE information which has already been screened, rather than having to review the entire body of material available.

Subsequent to the issuance of Generic Letter 82-04, the NRC issued Generic Letter No. 83-28 on July 8, 1983, which required licensees to formal.ize their activities for post-trip reviews and for controlling. vendor product information, particularly that covering operating experience that required  ;

corrective actions (e.g., procedure changes, equipment modification, parts 1 replacement). Document control programs were also to be established. These activities are considered to be a part of the licensees' OE feedback functions. . .

7

O

~

Further guidance on implementing OE feedback review activities has been sepa-

< rately developed by the industry through INP0. INP0 has developed " good practice" guidance which is intended to promote excellence in the operation of its members' nuclear power plants. The INPO guidance addresses programs for (1) the review of industry and in-house operating experience, (2) the incorporation of operating experience in training programs, and (3) the use of SEE-IR and the Nuclear Plant Reliability Data System (NPRDS). General criteria for evaluating OE review activities at operating facilities have also been developed.

2.1.3 Previous Assessments NRC Reviews A combination of factors have led to limited NRC review and assessment of licensee OE feedback activities. These factors include '1) the general nature of the NRC requirements and the lack of supplemental guidance, (2) lack of NRC inspection guidance (i.e., the total scope of the OE review activity is not covered in an inspection module), (3) limited resources, and (4) an awareness that INP0 evaluates utility activities in this area.

Based on discussions with the NRC staff and with licensees, the NRC reviews included post-implementation reviews conducted by NRC inspectors for compliance with NUREG-0737, Item I.C.5 and associated NRR, IE, or Regional Office direc-tives. This review confirmed that licensees had or would have procedures meeting the NRC requirements. However, since the NRC requirements are very general, and in essence only require the existence of an assessment function and formal procedures, this compliance check did not review or confirm the adequacy of the program. Subsequently, NRC resident inspections have focused on such specific activities as licensee responses to IE bulletins and a sampling of IE information notices and NRR generic letters, LER reporting, and training programs for licensed reactor operators and other operations personnel. IE P.erformance Appraisal Team (PAT) inspections also have not reviewed the overall adequacy of licensee activities due to the lack of detailed program requirements, but have focused only on selected OE program elements. Occasionally, an enforcement action has been taken for a violation of NRC requirements for an event which might have been prevented had the licensee had an effective feedback program. For example, an NRC enforcement action was issued subsequent to problems associated with personnel performing maintenance at a PWR on instrumentation thimble seal table high pressure seals while the system was pressurized. In this case, the licensee failed to adequately communicate the lessons learned from two previous events that received significant emphasis in feedback documents.

NRC Incident Investigation Teams identified inadequate response to in-house and/or industry-wide operating experience as a contributing causal factor to three significant events in 1985 (i.e., the June 9th loss of main and auxiliary feedwater systems at the Davis-Besse plant; the November 21st loss of electrical power and water hammer event at San Onofre Unit 1; and the December 26th loss of integrated contrql system power and overcooling transient at Rancho Seco). The investigation of the Rancho Seco event specifically addressed the role of the utility's response to OE feedback as a major causal factor.

8

In late 1985, the NRC revised its inspection procedures to reemphasize its review of licensee responses to NRC's OE feedback. New inspection procedures werd developed and proposed for implementation in the future that incluc'e u '

sampling of licensee corrective actions in response to OE reviews. However, the NRC has conducted only limited reviews of the OE activities, and has not made periodic assessments of the range of effectiveness of licensees' OE review activities or of the degree to which licensees are in compliance with the intent and scope of the established NRC requirements.

In addition to the above, tne NRC periodically observes the INP0 plant evalu-ation activities described below and occasionally may review the programmatic aspects of INPO plant evaluation reports on this area. The NRC does nct routinely sample or assess the adequacy of ifcensee responses to specific 4 INP0 feedback documents. However, it has occasionally reviewed licensee responses to an INP0 SOER to assess whether additional regulatory actions were necessary. In this regard, there also have been discussions between the NRC and INP0 on some of the programmatic aspects of the industry's OE review activities. ,

INP0 Reviews INP0 evaluates a utility's OE review activities as an element in their plant evaluation program. INP0 evaluations are peer review in nature, rather than a regulatory compliance review, and are to promote excellence in the operation of its members' nuclear power plants. The utility's activities are compared to the performance objectives and evaluation criteria, that are usually subjec-tive, provided in related INP0 guidance documents, Concerns (i.e., weaknesses) .

are called to a utility's attention for corrective actions, and INP0 conducts followup evaluations to determine if the utility has taken actions to correct the noted performance concerns. Utility implementation of INP0 proposals for corrective action are voluntary, although INP0 indicates that pressures for improvements can be placed on a licensee via peer pressure and other means.

More specifically, approximately every 15 months, INP0 performs an assess-ment of each nuclear power plant's OE review activities. The assessment includes a review of operating performance since the previous plant evaluation, comparisons to industry-wide performance, and a two-week on-site evaluation of the OE review activities. Pre-visit reviews are conducted to identify trends in areas such as recurrent equipment or system failures, unplanned transients, significant events and human errors. The results of these reviews assist the INP0 evaluators by providing a partial assessment of effectiveness of the OE review activities and by identifying areas of potential strength or weakness for on-site review. The on-site evaluations (about seven work days of actual evaluation effort) cover the organization and management of OE reviews, in-house and industry OE reviews, and NPRDS activities.

The INP0 evaluations focus on the adequacy of (1) meeting 0E review per-formance objectives and supporting evaluation criteria developed by INP0 and reviewed by industry, and (2) the utility's response to INP0 feedback documents. INP0 identifies strengths and weaknesses (concerns) during the visit and communicates them to the utility verbally and in a written report.

The assessments are very thorough in some areas addressed by the evaluators; 9

however, the evaluators are not required to assess all the evaluation criteria in each evaluation. As a result, each INP0 evaluation is not necessarily a complete ~ effectiveness review nor does the evaluation state whether' the per-formance objectives have been met (i.e., only strengths, concerns and comments on the areas evaluated are documented). Further, INP0 does not routinely sample 6nd fully assess the effectiveness of the utility's response to NRC ~

feedback documents (i.e., INP0 evaluations focus on INP0 and industry feed-back activities). INP0 evaluation reports are not routinely distributed to the NRC or made publicly available.

2.2 Licensee Operating Experience Feedback Activities 2.2.1 Reasonable and Practical Elements Based on the NRC requirements and general good industrial practices in light of the objectives, one might expect that responsive OE review activities typically would include tha following:

(1) Management involvement, including direction, support and oversight:

Management defines the program, including establishing the goals and objectives that, as a minimum, meet NRC requirements. It creates the organizational structure and assigns clear responsibilities for the various functions. It fosters the positive attitude necessary for successful implementation. It also provides guidance to ensure that both in-house operating experience (i.e., from events at the subject plant) and applicable operating experience from external sources (i.e., from events at other plants) receive the appropriate action.

Management provides the necessary resources (i.e., personnel and funding) to meet the objectives. In addition, it approves the distribution of the resources and emphasis between in-house and industry-wide feedback.

Management maintains an oversight of the program performance through routine reviews of the activities and periodic independent assess-ments of the effectiveness in meeting the goals and objectives.

Effectiveness is measured both objectively and specifically. For example, measures are made of how well the program objectives are being met [i.e., is the recurrence of undesirable events decreasing; are operators aware of significant events that have occurred at their plant and others industry-wide; are potential problems (those occurring at other plants that can occur at this plant) resolved before they affect operations]. Measures are made of how well the individual elements of the program are working (i.e., are operators being continually informed of pertinent feedback without being deluged with information that det'racts from performance; are the screening and assessment functions conducted with in-depth multi-disciplined reviews; are root cause analyses conducted for in-plant events; and are reconrendations for actions implemented in a timely manner). ,

10 9 - ~ ,

o ,

(2) A system for collecting, controlling, screening, assessing and storing in-formation originating both within and outside the licensee's organization:

- The system provides for segregating the feedback by source, signifi-cance, priority, and applicability to the plant. The system is designed for ease in managing and auditing.

The screening function uses INP0's SEE-IN program for screening industry-wide experience. It subsequently involves a rather coarse screen for the SEE-IN and other feedback to delete what is clearly not applicable to the plant. The screening should be a multi-disciplined review. The screening function directs the applicable feedback to the various disciplines involved for further assessment and determines what groups are to receive information copies. The information routing should be such that pecple are kept well informed (i.e., given more than absolutely necessary) but are not overwhelmed with extraneous material. It assures a suitable balance between in-house and external feedback such that operators are well aware nf both in-plant and industry-wide experience.

- The screening function also provides suitable checks to assure that conflicting or contradictory information is resolved before it is conveyed to operators and other personnel.

- The assessment function involves utility personnel having collective competence in all areas important to plant safety. Assessments are made by personnel in job functions directly related to OE review responsibilities. Sound bases are provided for determining appli-cability, or non-applicability, of industry-wide feedback to the plant and for determining the need for taking, or not taking, action.

Root cattse analyses are performed, particularly on in-house operating experience, and corrective actions address the root causes. This function is conducted by a multi-disciplined group activity or a multi-level review activity (i.e., it is not dependent on only one or two reviewers).

- The minimum feedback sources are the NRC, INP0, the NSSS, other equipment vendors, and the licensee.

(3) Action tracking system:

- A system exists for initiating actions, tracking the status of actions, and closing-out completed actions. It also includes features for assuring actions are timely and that periodic reassessments of priority include timeliness considerations (i.e.,

overdue items are considered for upgrading of priority).

- It identifies the administrative and technical review steps necess'ary to translate recommendations by the operating experience assessment s,taff into actions (e.g., changes to procedures, plant modifications, revised training plans, etc.).

The system is designed for ease in auditing.

11

0 4 (4) Information dissemination systems:

- Information dissemination systems exist to assure that operating information pertinent to plant safety originating both within and outside the licensee's organization is continually supplied to plant operators, trades personnel, technicians and other non-supervisory personnei in addition to supervisory and technical support personnel, ,

and is incorporated into training and retraining programs, including simulator training.

- The recipients of various categories of feedback information are identified (e.g., STAS, plant operators, reactor operators, mainte-nance personnel, supervisory personnel, health physics technicians),

and means are provided through which such information can be readily related to the job functions of the recipients. The recipients are trained in how to deal with this information (i.e., they are familiar with the plant procedures and their responsibilities regarding feed-

, backinformation).

- The system has means to assure that the affected personnel are in-formed of and understand information of such importance that immediate action is appropriate. This type of information cannot wait for routine disseminatior, or emphasis through training or re-training programs. In such cases, prompt information dissemination features must exist such as special instruction logs, temporary changes to operating and emergency procedures, or special edition newsletters or memos. These features must include provisions to assure that operations management personnel are directly involved.

- Since the total volume of information handled is large, the system should assure that discrimination is used such that plant personnel do not routinely receive extraneous, unimportant information in such volume that it obscures priority information or otherwise detracts from overall job performance and proficiency.

(5) Internal Audits:

- Periodic internal audits are conducted at a reasonable frequency (e.g. quarterly or semi-annually) to assure that the feedback program functions effectively at all levels.

- Individual elements (i.e., action tracking, information dissemi-nation, screening, assessments, etc.) are audited.

- Audits assure that as a minimum the NRC requirements are being met (i.e., those stated in the May 7,1980 letter to licensees (Appendix A) and in Generic Letter 83-28).

(6) Documentation:

- Documentation exists that defines goals and objectives (those that are general as well as those that are plant specific are addressed); activities and the associated positions, authority, 12 1

responsibilities, accountabilities and approvals involved are identified; interfaces between various elements are defined; and measures for assessing effectiveness, of these activities are detailed.

- Procedures exist that incorporate the. items specified in the NPC requirements (TMI Action Item I.C.5 and NRR Generic Letter 83-28),

as a minimum.

(7) Effective staffing:

- The staff conducting the various functions are well-trained, conscientious, and aggressive in ccmpleting tasks.

- The assessment function, in particular, is staffed with knowledgeable personnel, preferably with multidisciplined backgrounds, including personnel with specific knowledge and training in nuclear plant operations (i.e., NRC licensed or the equivalent).

- The end-users (i.e., the reactor operators, mechanics, support staffs) are well versed in what they are expected to do with the feedback information.

2.2.2 Characteristics of Actual Performance Seven utilities were visited by AE0D staff and discussions were held with staff members at each plant site and several corporate offices to determine how actual programs were defined and working. Three utilities with one nuclear plant each; one utility with three operating units; one utility with two operating units and a third unit in the final stages of construction; one utility with four operating units; and one very large utility with eight -

operating units were visited. The sample represented a cross-section of nuclear utilities and OE review activities. (See Appendix C for summary tables based on the discussions with each utility.) AE0D also participated as an observer on one INP0 plant evaluation of a utility's operating experi-ence review activities.

From the plant visits and discussions, a typical utility's present 0E review activity can be characterized as follows:

(1) The licensee properly places primary importance on in-house operating experience. Events of an abnormal or unusual nature are usually identi-fied and reported by the plant operating staff. Assessments of signifi-cance are initially made by lower-level operations management with con-firmations of significance made by technical support personnel, Safety Review Committees, and upper management. The more significant an event

~is, the. more attention it receives (i.e., more organizational elements become involved) and more actions result (e.g., external reporting, procedure changes, staff training, plant modifications). Thus, in-house data are (1) collected; (2) assessed individually and collectively; (3) categorized as needing or not needing further actions, including reporting to the NRC or INP0, referral to others at the plant or 13

g _

performance of routine maintenance; and (4) stored for action tracking purposes and for future reference. In-house feedback constitutes 80% to 95% of the information fed back to plant personnel (operators, technicians, and maintenance personnel). ,

(2) Licensees are depending on the NRC, INP0, and the reactor vendors to screen industry-wide operating experience (i.e., from other than their units) and to feed back to them the experience that may pertain to their operations. Sources of industry-wide experience are indicated on Figure 1. The results of these screening activities are forwarded to the licensee corporate staffs, individual plant staffs, or both. This screened information is further screened and assessed by the corporate 1 and/or plant staffs for its relevance to their plant (s). In some utili-ties this screening is centralized with a small number of people review-ing the information to determine its significance, its relevance to the plant, and its priority. In some, this activity is highly selective, with only a limited amount of the material circulated outside the small group that does the initial reviews. As a result of these activities, plant operators, radiation protection personnel, trades personnel, and others directly see a relatively small portion of the industry-wide feedback available at the utility. These operators will notice changes to their

, facilities that are initiated as a result of operating experiences.

However, they often are not aware of the specifics of the experience or reason for the changes.

(3) Typically, the lower and mid-management levels at the plant (and some-times corporate offices) plus support staffs assess the OE to determine applicability and to determine the appropriate actions to be taken. The individuals reviewing the information assess it from their specific per- ,

spective (e.g. engineering, operations, trades, chemistry). Sometimes, independent assessment groups are utilized to broaden the perspective used in the process. In essence though, a few people review most of the material while most of the people only review a small quantity. Thus, the quality of the assessments is highly dependent on the few people assessing the data to identify what is pertinent and important to the licensee's operations.

(4) Typically,10% to 20% of the industry-wide feedback from the NRC, INP0 and the vendor is determined to be significant and applicable, and gener-ates some action at a specific plant. Such actions generally include (a) changing the operating or maintenance procedures; (b) changing the plant design; (c) incorporating the concepts of the problem into tra.ining programs; or (d) discussing the actual sequence of events, its conse-

i. quences, and lessons learned directly with the plant operating staff.

Item [(d)] is done less frequently than the former [(a)-(c)].

Since the OE review activities are dynamic, both positive and negative changes are occurring. For example, as experience is gained with an activity, changes may be made to improve it. Personnel turnover and increases or decreases in resources change the characteristics of the OE feedback programs. Some-times these changes improve the situation and other times they degrade the situation (e.g... personnel turnover can cause temporary vacancies and loss of institutional knowledge and thus delays are introduced in completing proaram activities). .

14

. a Figure 1 Sources of Operating Experience Feedback information m

INPO ,

NRC

• Site Evaluations (1/yr)

• NPRDS Reports

• Licensing Actions including

• NUCLEAR NETWORK Generic Letters (~25/yr)

• SEE - IN - Document:s, (100/yr)

• IE Bulletins and Information

• Utility Reports Notices, (~125/yr)

• Information Queries

• AEOD Periodic Reports (~20/yr) -

e NRC incident Response . NUREGs and Selected AEOD Center (50.72) Reports Technical Reports

• lNPO O &MR's Licensee's On-Site Experience Vendor's/A Es

• Deviation / Deficiency Reports (~100/yr)

. NSSS SILs, TILs, Technical

• LERs (~ 30/yr) if Bulletins, Event Notices, etc.

• Trip Reports (~6/yr) (~20/yrt -

• Routine Operating Reports

• Part 21 Reports

• Special Studies

• Equipment and Component Vendor Service Bulletins Licensee's Operating-Experience Feedback Program Owners Group u

• Reports Commercial Information Special Interest Groups Sources

• PUC Studies, State . Industry Publications & Data Banks Reports, Intervenors, etc.

• Special Subscription Services (AIF, Nuclear News, inside NRC, Nucleonics Week, etc.)

- EPRI

• Technical Studies ~

• NSAC Studies

{

15

6 2.2.3 Variability in Licensee Activities Each licensee v.isited had established an OE review activity that was intended to meet NRC requirements and to conform to existing guidance. However, be-cause of (1) the very general regulatory requirements and program guidance, and (2) the variability in the available resources, plant performance, and licensees' perceived needs for OE feedback, existing programs vary widely in their characteristics and effectiveness. This variety is not inherently good or bad; it is an issue or important only for those plants having an OE assess-ment function which is not fully effective or adequate. However, the variety of programs indicates that some additional guidance is needed to help assure that all programs reach and maintain an acceptable level of effectiveness.

Examples of the variability noted are:

Upper management involvement varied from very direct involvement to a peripheral involvement. Management's fostering of positive attitudes, allocation of resources including guidance on their application, and oversight varied from strong support to only limited support. In most cases, management had not developed or documented clear goals and objectives for the OE review activities. Only a general statement often existed that these activities are to assure safety and reliability.

- The systems for collecting, controlling, screening, assessing and storing information varied considerably. Some plants had very detailed procedures for these activities and some did not. At some plants all feedback docu-ments are sent to a centralized staff; in others the NRC documents go to one person, INP0 documents go to another and vendor feedback may go to a third; in others, all documents are sent to one individual. In the decentralized case, the identification of conflicting data is difficult.

In screening the feedback, some plants use one or two staff members acting individually, while at others the screening is done by a' committee of people with various backgrounds and interests. The assessment -function was also sometimes conducted by sending the information to separate organizational units for review from their perspective with limited coordination on interfaces and mutual concerns.

There is variability in the distribution of documents. Some licensees make multiple copies of each incoming document and send it to a standard distribution of management personnel for information, in addition to .

entering each document into their feedback review process. Others enter the document into the OE assessment process and circulate only screened material to specific distributions dependent on the topics'of interest.

The former may overwhelm people, while the latter may be too restrictive.

- The methods for information dissemination to plant operators and others vary considerably. In one plant, a daily newsletter, which ~ infrequently contains information on an operational event from a different plant, is posted on selected bulletin boards for operational staffs to read. In another, a few items, including occasionally industry-wide information, are issued once or twice a week as required reading. In others, sizab-le packets of information (often 100 to 200 pages) are assembled and issued

. monthly as required reading material to distributions that ranged from 16 e O

_. G . 4 small to large (over 150 persons). However, these large packets contained limited information on industry-wide operational. experiences. Only one of the seven plants prepared monthly a reasonable packet of information on industry-wide experience for each category of personnel (operators, chem-ists, mechanics, etc.) and issued it as required reading.

- The quantity of information from industry-wide experience fed back to plant operators via required reading ranges from little ("only what they need to know") to what they need to know plus some items cf general in-terest. AE0D observed that no operators seem overwhelmed with operation-al feed-back; in most cases they seem to receive too little. Licensees estimated that they directly feed back to operators from 5% to~ 15% of the industry-wide feedback that they receive from the NRC, INP0 and the vendor. Feedback via training and retraining also was in this range, with some overlapping quantity. Some licensees cover recent industry-wide operating experiences once or twice a year in a special half-day training session (1% to 2% of the train'.m time) and devote some time during the year to individually significant avents (another 2% or so). Others indicated that they devoted as mu a as 10% of the training time to industry-wide experiences.

The background and qualifications of the personnel involved in the feed-back program varied from experienced senior technical staff to newly trained inexperienced personnel. Thus, the quality of the screening and assessments can be expected to vary with the qualifications of the personnel.

- Attitudes regarding the program also varied. Most felt it was a worth-while program yielding benefits. But on the other. hand, some individuals felt that it was a "make work" program, or that the industry-wide feedback received was difficult to use effectively without additional significant effort on their part.

As noted previously, the variability among programs is not surprising nor an issue. There are many different ways to accomplish a task. However, not all programs were fully accomplishing the task of effectively understanding and applying the lessons of experience. As a result, this activity is not yield-ing the desired result.

2.2.4 Shortcomings There were several key areas where shortcomings were observed. One area is the lack of clear statements by the licensees of the goals and objectives for individual programs. These statements should help shape the programs, and allow measures to be defined for assessing effectiveness. The latter is also an area of shortcoming. Some reviews of OE activities (usually an annual review) do not focus on their effectiveness (e.g., are we still making the same mistakes, or the same mistakes that others have made). The reviews are typi-cally audits focused on the NRC's procedural requirements and the mechanics (i.e., " Paper Trail") of the action tracking systems. In short, these audits properly focus on assuring compliance with requirements; however, in addition, periodic management assessments need to be made to assure OE activities are conducted at a high level of effectiveness.

~

17

1 In many cases, it was also noted that only a small amount of industry-wide OE feedback was being disseminated to reactor operators, other operations staff, trades personnel and personnel responsible for procedure changes and plant modifications (i.e., the process is filtering too much information). For example, in some programs, reports of many significant operating events are not routinely (1) discussed in-depth in operator trainin readily available for operator information, and (3) practiced g sessions, (2) made on simulators.

In fact, significant operating events have not been widely incorporated into simulator training programs although this seems to be slowly improving now.

Feedback to maintenance and surveillance personnel is even more limited.

The screening in some facilities is conducted in an uncoordinated manner or by too few individuals to thoroughly and comprehensively review the majority of the OE feedback documents. The reason for this is believed to be the volume and varying degrees of perceived importance. Typically, the material is extensively screened by one or two individuals who distribute bits and pieces to others. The remainder is effectively unavailable to others in the organization.

Licensees have indicated that the estimated yield from reviewing industry-wide feedback is in the 10% to 20% range, i.e., one or two reports out of ten from the NRC, INP0 and the vender are fed back .to operations or maintenance personnel or result in some other action being taken at a plant. On the surface, this yield seems low in light of the extensive filtering, assessment and screening that has taken place by the NRC, INP0 and the vendor prior to receipt by the plant. For example, in AE0D's opinion, based on a review of the 94 Information Notices issued in 1984, about 50%, rather than the 10% to 20%

indicated, contained information that should have been provided to shift operations, maintenance and other plant per'sonnel.

The fact that the NRC does not routinely review the full scope of licensee OE review activities for adequacy or effectiveness is also considered a short-coming. The lack of a structured and routine NRC followup is due in part to (1) the existing NRC requirements being general in nature and not lending themselves to precise assessments of what is and is not acceptable, and (2) the emphasis of the regulatory guidance being placed primarily on the process rather than on the desired end-result. That is, the requirements state that licensees are to carry out an OE assessment function and identify specific items to be addressed in procedures (see Appendix A) that relate to the processing of feedback information, but addresses only implicitly that the thorough assessment of operational data should result in action to reduce or eliminate the frequency and severity of undesirable operation events. These aspects have undoubtedly contributed to the wide range of licensee approaches to implementing the required program. Without NRC followup, there is insuffi-cent publicly available evidence for assessing whether licensee activities are meeting the regulatory intent.

In summary, as a result of NRC requirements, all licensees have established OE feedback activities. Initial NRC reviews concluded that the licensee activi-ties complied with the existing requirements to establish an OE assessment function, including formal procedures. However, licensees have not had to routinely demonstrate to the NRC that their programs are truly effective, and that the lessons of experience are being understood and used in a systematic and disciplined way to change designs, components, training, and/or procedures 18

for safe operations of the plants. Thus, the NRC has limited information on the effectiveness of individual licensee OE review activities or on the range of variability among plants in meeting the established NRC requirements.

2.3 Operatina Experience Feedback Documents - Their Use and Related Problems 2.3.1 Operating Experience Feedback Documents Besides the acceptability and degree of effectiveness of licensee assessment programs, the adequacy of feedback documents is also important. Acain, feedback documents are not used or evaluated in a standard way, and the documents are screened and utilized as each licensee deems appropriate. In fact, different licensees put different emphasis on a particular feedback document, and in some cases use the document quite differently. The feedback document may (1) initiate action by the licensee such as a procedure change, training or a plant modification; (2) be circulated for information to others (including plant operators); or (3) simply be screened out as not applicable, or not requiring action, by the personnel assigned to the screening function.

The following discussion provides general comments on each source and its role in licensee programs as gleaned from the site visits.

Licensee Produced Feedback Documents Licensee produced feedback reports include:

(1) Deviation / deficiency /non-compliance reports, including post-trip reports; (2) Licensee event reports (LERs) submitted to the NRC per 10 CFR 50.73; (3) Routine operating reports (monthly or periodic compilations of operating data, including the history of significant operating events);

(4) Special studies (i.e., engineering evaluations, licensing reviews, bases for technical specification changes, etc.);

(5) NPRDS reports; and (6) Maintenance histories.

The licensee's plant-specific (in-house) OE feedback 'is a primary source of feedback into operations. Most changes to operating policies and procedures, plant design, training programs,' etc., originate from this source.

Licensed reactor operators are made aware of.the more significant in-house OE feedback through inclusion of this material in training programs or through routing of information on routine or required reading lists. This material often comprises the majority (about 80% to 95%) of whatever experience feed-back is provided to licensed operators.

Required reading lists generally contain procedure and technical specification changes, copies of major plant incident reports, design change descriptions, and special studies. The other in-house feedback documents noted above are available, but less often used by the plant operators.

19 n _ __ -- - -

. . c Unlicensed auxiliary plant operators and maintenance personnel also receive feedback on a limited amount of in-house operating experiences through periodic group meetings, training, and job instructions. The feedback to these persornel is, however, usually very minimal (i.e. , very job-specific and plant-specific).

• NRC Generated Feedback Documents Many NRC produced documents are specifically oriented to OE feedback, and are a primary part of the licensee OE feedback program. A variety of NRC documents are available and licensees choose which documents to use. The NRC offices which produce the principal feedback documents are IE, NRR, and AE0D. The documents produced by IE, NRR, and AE0D are discussed below.

IE documents are individually screened by licensee technical support and operations staffs for site applicability. This individual attention is due in large measure because NRC inspections verify that there is a record of such screening. As a result, IE Information Notices, IE Bulletins, in-spection report open items, and enforcement items are generally tracked as action items and are closed out by completion of an action or by a memo to file documenting that no action is necessary. Licensees estimated that about 10% to 20% of the IE Information Notices result in some action (e.g.,

a procedure change, training, plant modification) at the plant. In AE0D's opinion, IE Information Notices do not generate the amount of voluntary licensee response that was expected, primarily because they are for informa-tion only and, as such, licensees do not routinely commit adequate resources to ensure that all needed actions are identified and implemented.

NRR issuances such as generic letters and licensing actions are screened for applicability to the plant and to the OE review function. Both generic letters and licensing actions are occasionally included in the required reading lists and in various training programs.

AE00, since its beginning in 1980, has issued three periodic reports (dis-cussed below) to licensees as part of the NRC OE feedback program, plus made technical reports available. Although they place no requirements on licensees, AEOD documents contain information that is intended to be useful in the licen-see's OE assessment function. Discussions with licensees indicated the following:

(1) Monthly LER Compilation Reports (NUREG/CR-2000)

The LER Compilation Report, which contains a computer listing of LER abstracts added during that monthly period plus indices, receives only limited use in licensee assessment and training programs.

Licensees indicated that the indices were helpful; that the. report was used to compare the types and numbers of events being reported by others, and that it was a good bibliographic reference; but con-tained too many events with too little detail on each to be of much greater use. They indicated that some summary data on particular trends or patterns in the data would make the document.more useful as feedback.

20 a

(2) Bimonthly Power Reactor Events (NUREG/BR-0051)

Power Reactor Events (PRE) is used at some facilities to communicate industry-wide experience to site personnel including licensed operators.

Licensees indicated that it is a source for training material because the event descriptions are comprehensive. Some items from PRE are occasionally included in the licensed operators' required reaaing lists at some plants.

Section 1 of the PRE report contains detailed event descriptions; these are the items most often used. Section 2 of the PRE report contain.s excerpts directly from LERs; this section however, does not appear to be receiving much use. Licensees indicated that Section 2 would be more helpful if trends or patterns of data are highlighted; for example, if a large number of events were highlighted that were previously thought to be of low frequency or a number of repetitive events industry-wide from the same cause (e.

surveillance test)g., Section a large number 3 of the PREof trips being report is acaused by a certain bibliographic source of recent NRC feedback issuances. Although not all plants used it, the PRE report was cited as the most useful of AE0D's periodic publi-cations. Although the material is usually 5 to 6 months delayed, this did not seem to present a major problem for training use.

(3) Quarterly Reports to Congress on Abnormal Occurrences (NUREG-0090)

The A0 report, which contains information on unscheduled incidents and events determined to be significant by the NRC and reportable per statute, is reviewed by some licensees and, as a result, events are occasionally included in the required reading lists and in training programs. Some licensees indicated that the update material was also of interest and use-ful. However, this document is fairly delayed in publication, and much of the information of interest is obtained from other sources. prior to receipt of the A0 report. Thus, this report is not screened or used in most licensee feedback activities.

(4) Specific technical reports (e.g., case studies, engineering evaluations, and technical reviews)

AE0D technical reports are not routinely distributed to licensees, but are available on request. As a result, some utilities have occasionally made use of AE0D technical reports in their training programs. However, most licensees do not use them in their OE feedback program unless the events described occurred at their plant, or unless AE00 or others (INP0, IE, NRR) call their attention to the report verbally or by reference in another document.

In addition to the specific feedback documents issued by IE, NRR, and AEOD, discussed above, NUREG documents are issued by various NRC offices. These documents cover a wide range of topics, and may discuss operational problems and suggest actions for improvements. Thus, some NUREGs are a source of OE .

feedback and are used as reference material in technical studies by the engi-nearing support staffs.

21

INP0 Feedback Documents The INP0 is a primary source of operating experience feedback information to licensees. All licensees are participating in INP0's SEE-IN program and are taking credit for INP0's screening of industry-wide reporting as part of their response to the regulatory requirements of TMI Action Plan Item I.C.5.

As a general rule, the INP0 SEE-IN documents receive very similar treatment to their corresponding NRC documents (i.e., SOERs and IE Bulletins, SERs and IE-ins). Licensees indicated that little use is currently being made of the NPRDS data, but the data processing equipment and files are currently being revised, updated and upgraded. When these activities are completed, it is expected by INP0 and AE00 that licensees will make more use of the information.

It currently is used for finding spare parts, component population information, failure cause identification, and similar limited uses.

NSSS and Component Vendor Feedback Documents Licensees have included vendor feedback documents in their programs as noted below. The NRC issued Generic Letter No. 83-28 instructing licensees to for-malize their programs for controlling these documents and to insure the vendor recommended actions are identified and tracked to completion. At the time of the site visits, some licensee activities in these areas were being revised.

(1) NSSS correspondence and reports The major Nuclear Steam System Suppliers (NSSSs) have OE feedback pro-grams that provide information to the utilities that bought their systems.

For example, General Electric issues service information letters (SIls),

technical information letters (TILs), and topical reports. Westinghouse issues similar documents but identifies them by different names; Babcock and Wilcox (B&W), and Combustion Engineering have similar feedback docu-ments. Licensees screen these documents for material to incorporate in their feedback programs. Selected items are put into the action tracking system and are incorporated in the required reading lists ard training programs.

(2) Component vendor reports Some component vendors provide feedback reports on problems with their equipment and on. corrective actions. When these are received, they are included in the feedback programs. In addition, 10 CFR Part 21 reports on significant deficiencies with safety implications received priority attention in the OE assessment activities by some of the licensees visited.

NSSS Owner's Group Feedback Documents Owner's groups have been formed by the utilities who operate plants designed by a specific NSSS. These groups meet and collectively respond to regula-tory and other matters. They can also deal with operational experience, For example,-the B&W Owner's Group has a program where B&W prepares detailed technical assessments of major transients experienced at B&W plants. The reports are called Transient Assessment Program (TAP) reports and are a key 22

source of feedback data for B&W plants. Other NSSS Owner's Groups are apparently considering similar programs.

Nuclear Industry Publications t

The following documents were mentioned in the discussion of industry feedback publications:

(1) Nuclear Power Experience, (2) Nuclear Safety Journal, (3) Inside NRC, (4) Nucleonics Week, and (5) EPRI and NSAC technical reports Licensees were aware of these publications. Most had them available within their organization and indicated that support staffs such as engineering used them as source or reference material. Several licensees indicated that one or more were sources of information in their OE feedback programs. For example, Nuclear Power Experience was used as an information source for feeding back industry-wide experience to operators, trades personnel, radiation and chemistry technicians, and others at one site; and was used at another utility by an offsite review group at a corporate office as a source for screening industry-wide experience. Inside NRC and Nucleonics Week were identified as good sources of information on current events. Nuclear Safety Journal was usually read casually for items of interest, but not incorporated into any structured feedback programs. Several licensees indicated that they included EPRI and NSAC technical reports in their feedback programs.

2.3.2 Burden, Overlap, Conflicting Information, and Timing Burden ,

Even though licensees use INP0's SEE-IN program and thus, each does not screen the large amount of OE feedback data contained in the approximately 3000 LERs reported annually, plus many foreign reports, the volume of OE feedback doc-uments is still very large and burdensome for the licensees to collect, assess, track and disposition. Figure 1 illustrates representative numbers of some items received by licensees.

The number of documents being processed adds to the difficulty of properly assessing and acting on the operational experience. Several licensee personnel indicated the number of documents that they have to deal with are tremendous.

The task is complex and difficult since the documented operating experience may come from many sources and may be applicable to only one plant, to a specific plant design, or to all plants. They indicated that some consolidation was needed and some order added to th6 feedback system.

~

For example, the number of feedback documents available on any one event is a function of its significance, or perceived significance, and the number of organizations that become involved. For a very significant event there is usually a large amount of documentation originated by the utility (e.g.,

NUCLEAR NETWORK entry, media- release, LER, special reports); the NRC (e.g.,

daily report, preliminary notification, Information Notice, Bulletin or generic 23 o - - -

I letter, .ibnormal Occurrence report, Power Reactor Events summary, inspection report); INP0 (e.g., 0&MR, SER, SOERI-special report); the reactor vendor; and many of the trade publications.

Most of the NRC feedback documents are for "infermation only" (i.e., informa-tion that could be used by licensees to improve the safety of plant operation but do not contain requirements). Documents such as AE0D's periodic publica-tions (Report to Congress on Abnormal Occurrences, PRE and LER Compilation),

NUREG reports, and IE Information Notices (ins) are in this category. Together they feed back lessons on about 120-150 events per year with some overlap.

Several licensees commented that many ins did not contain sufficient details on the event sequences for training use, while PRE did. The ins, however, are more timely than PRE.

Licensees would have to expend considerable resources to use some of the docu-ments being disseminated (e.g., documents containing raw data.such as the monthly LER Compilation Report). Most licensees do not have the resources or the inclination to analyze such data. As a result, in general, such documents are not frequently used by licensees in their evaluation and feedback programs.

Overlap Several of the licensees indicated that there was a fair amount of overlap in OE feedback. They indicated that this overlap presented some problems:

(1) different writeups on the same event sometimes contain different or conflicting information; (2) important points are sometimes missed by licensee personnel because less attention is devoted to an event description if the event has previously been assessed, and (3) resources are wasted reviewing multiple writeups on the same event. They indicated that occasionally signifi-cant overlap exists within NRC documents such as an IE Information Notice, an IE Bulletin, an Abnormal Occurrence report, and a Power Reactor Event summary of the same event. Also there ts a high degree of overlap between NRC, INP0 and reactor vendor issuances.

The problem has several facets: Some overlap is due to differing organiza-tional responsibilities; while other overlap results from different time frames such as when conveying new or more detailed information, revisions and corrections.

To help put the problem in perspective, the topics discussed in 11 IE Informa-tion Notices (ins), issued during November and December 1984, were compared

, to topics in IE Bulletins; NRR Generic Letters; AE00 A0 reports and PRE; INP0 SERs, SOERs and 0&MRs; and NSSS service letters (see Table 1). Six of the 11 ins were discussed in at least one other document. Two ins that addressed

, events determined to be significant to public health and safety (i.e., A0s) were discussed in at least two other documents.

! The NRC has in the past attempted to coordinate its efforts to keep redundancy to a minimum. For example, AE00 generally does not include items in PRE that have been covered in an IE Information Notice or Bultatin, unless it is con-sidered appropriate to disseminate significant additional details or to update s information.

o 24

There is also some coordination between the NRC and INP0 to minimize overlap in information feedback. Routine information exchanges provide prior know-ledge about whether the other organization is planning to issue materiel on a given subject. However, some overlap exist. due to the different responsi-bilities of the NRC and INP0, due to differing technical opinions and due to the availability of the information provided (i.e., INP0 releases have restricted availability while NRC's are publicly available).

In summary, the overlap is still a problem that ~tends to decrease the effi-ciency of the feedback process. The overlap in NRC feedback may be reduced or eliminated through further coordination of activities and consolidation of various feedback documents. While some overlap between the NRC and INP0 may be reduced through increased coordination of activities, some overlap seems unavoidable.

Conflicting Information In the discussions, licensees indicated that the various feedback sources sometimes provide conflicting information. A careful reading and a comparison of information from the different sources is required to recognize the con-flicting information. Licensees generally have procedures to resolve the differences, once the conflict is identified. One example given was the Salem ATWS event where the stated cause varied ard where the NRC recommended certain corrective actions and INP0 and the reactor vendor recommended others.

Licensees indicated that it would assist them if the differences, when known, were identified and discussed briefly in the feedback document.

Timing Timing of the feedback of operating experience is an important factor. In general, the more timely the information the more useful it is. Detailed and accurate feedback, however, will always be delayed, because of the need to conduct assessments to evaluate significance. Although important or special interest events are usually fed back in days or weeks, the routine feedback of in-depth analysis of an event or situation usually takes from a few menths t.o a year or more. It is not uncommon for material to appear in the required l reading lists and in training programs three to 'nine months after the event

! date.

Several licensees indicated that the development by the NRC of a computerized information exchange system comparable to the INP0 NUCLEAR NETWORK would l improve the timeliness of their receipt of feedback information and also l

establish a centralized receiving ooint for this type of NRC material. Other attractive features would be simultaneous on-line transfer of information to all plants; the ability of licensees to ask questions and to query responses to l and from others; the ability to retrieve previous information entries; and l

consolidation of all NRC information on an event in one place.

l 25 A

a Table 1 Feedback Comparison - Redundancy NRR Generic IE Information Letter / AE0D Report INP0 Vendor LER Notice (Issue Date) IE Bulletin A0 PRE SER 0&MR (NSS) (Docket /no.) Remarks 84-79 Failure to Properly Vol. 6 93-84 271/84-21 install steam separator No. 5 (12/84) at Vermont Yankee (3/85)

(11/5/84) 84-80 Plant Transients Vol. 6 44-84 312/84-15 2 incidents-Induced by failure No. 2 (6/84) 302/84-10 Rancho Seco and of Non-Nuclear (9/84) Crystal River 3 Instrumentation Power Vol. 6 (11/8/84) No. 3 (11/84) 9?

84-81 Inadvertent Reduction 49-84 231 GE SIL 324/84-11 5 incidents at 3 in Primary Coolant (7/84) (3/85) No. 388 373/83-105 plants-WNP-2, Inventory in Boiling .

(2/83) LaSalle 1 and Water Reactors During Brunswick 2 Shutdown and Startup (11/16/84) 84-83 Various Battery None Vendor Manuals Problems (11/19/84) caution on problems. Battery problems are men-tioned in various other feedback documents.

84-84 Deficiencies in Ferro- W Technical None Resonant Transformers Hulletins (11/27/84) NSD-TB-84-08 (9/84) and NSID-TP-84-11 (11/84)

i Table 1 (Continued)

Feedback Comparison - Redundancy NRR Generic IE Information Letter / AE0D Report INP0 Vendor LER Notice (Issue Date) IE Bulletin A0 PRE SER O&MR (NSSS) (Docket /no.) Remarks 84-86 Isolation Between None Waterford con-Signalstof the Reactor struction de-Protection System and ficiency report Non-Safety-Related Equipment (11/30/84) 84-87 Piping Thermal Deflec- 69-84 W RADAR None SER covered five tion Induced by Strat- (9/84) ho. R-02 events.

i ified Flow (12/3/84) 84-88 Standby Gas Treat.uent Many LERs Summarizes AEOD

$ System Problems referenced report E410.

(12/3/84) in report (5/84) Also ex-pands IEN 83-25.

84-89 Stress Corrosion NRR Generic Update Vol. 6 277/84-16 Cracking in Non- Letter 84-11 Item No. 6 Sensitized 316 Stain- (1985) (4/85) less Steel (12/7/84) 84-92 Cracking of Flywheels None on Cummins Fire Pump Diesel Engines (12/17/84) 84-93 Potential for Loss of IEB 84-03 84-11 Vol. 6 72-84 GE SIL 388 213/84-13 Also IEN 84-81 Water From the Refuel- (8/84) (4/85) No. 4 (10/84) (2/83) 316/84-60 (11/84); NSAC-52 ing Cavity (12/17/84) (2/85) 92-84 (I/83)

(12/84) .

! o 3.0 FINDINGS The operating experience program review and discussions with licensees provided the basis for the following findings:

3.1 Operating Experience Feedback Programs (1) Over the last five years considerable resources and attention have been expended on operational experience feedback and major improvements in

. the quality and availability of OE information are evident. Similar significant improvements in the use of OE information are not readily apparent in many instances because existing licensee programs are highly diverse in their characteristics and effectiveness. The programs are also routinely changing as personnel, funding and other factors change.

(2) Most licensee programs tend to focus attention primarily on the operating experience of their facility and industry-wide feedback that demands action on their part (e.g., IE Bulletins, INPO SOERs, etc.). Other industry-wide feedback that communicates lessons learned that are appli-cable to the licensee receives some attention, but is often given very little emphasis.

(a) In some programs, industry-wide experience is highly filtered and little of the information is fed back to reactor operators or to individuals responsible for plant procedures, systems operations, radiation monitoring, and maintenance.

(b) Due to the extensive screening and limited distribution, much infore mation is unused and unavailable to those in the organization who are not part of the screening effort.

(.3) The NRC requirements for the establishment of an OE assessment function and the procedures for administration of the feedback program are very general in nature. The industry's program guidance is more specific and detailed in many areas. Neither, however, address all the appropriate program elements. For example, the need for program goals and objectives, criteria for the expected treatment of in-house versus industry experi-ence, and a complete listing of expected program activities are not fully addressed in available guidance documents. Important objectives such as increasing the availability of safety systems, reducing the frequency of undesirable safety-related events, preventing serious accidents, and ensuring that plant operating staffs are more knowledgeable are unstated.

The need for periodic assessments by the licensee of overall program effectiveness is also not fully or adequately addressed.

(4) Regarding reviews of the OE assessment function, the NRC, INP0 and licensees conduct periodic reviews of the OE activities. Each organization generally conducts reviews as follows:

28

+ a (a) NRC currently through its resident inspectors reviews licensee responses to NRC feedback requiring actions (IE bulletins) and a sampling of responses to other feedback (IE information notices, NRC generic letters, and some vendor and NSSS notices). In addi-tion, procedures have been proposed for reviews of licensee corrective actions by IE for future implementation. No periodic, focused review of the effectiveness of the OE review program at each plant is conducted.

The NRC reviews have been limited due to (1) the general nature of the NRC requirements and lack of supplemental guidance, (2) lack of NRC inspection guidance, (3) limited resources, and (4) an aware-ness that INP0 evaluates utility activities in this area. Because reviews have been limited, the NRC has not periodically assessed the range of effectiveness of their OE review activities or the degree to which plants are meeting the established NRC require-ments.

(b) INP0, approximately on a 15-month cycle, evaluates most of an indi-vidual utility's OE review activities as part of their plant evaluation program. INP0's evaluation includes a two-week on-site visit, is of a peer review nature, and assesses (1) Whether perfor-mance objectives developed by INP0 and reviewed by industry are met and (2) the adequacy of utility re:ponses to INP0 OE feedback, particularly SOER recommendations. All OE activities, however, need not be reviewed each evaluation and as a result incomplete assess-ments occur. Further, INP0 evaluation reports do not routinely state whether performance objectives have been met instead strengths, concerns and information on assessed criteria are documented.

(c) Licensees perform audits of the various OE review activities usually as part of their quality assurance efforts. Some licensees perform management reviews of effectiveness of the OE activities, although the majority of the plants visited in this study did not.

(5) Inadequate response to in-house and/or industry-wide operating experience was a contributing causal factor to three significant events in 1985 that were investigated by NRC Incident Investigation Teams.

3.2 Feedback Documents (1) The volume of OE feedback is large. Many documents are produced by many sources. Many of the documents contain different types of information of varying levels of significance and usually contain no indication of priority or significance. Others contain information that is similar to or overlays that from other sources. Licensees indicate that they must cope with vast quantities of paper, some redundant, and most of varying importance and usefulness. The review of a large number of documents, in addition to those issued by INP0, the NRC,'and the NSSS vendor, is diluting resources that could be devoted to more effectively utilize that feedback.

29

e .

(2) Overlap in OE feedback documents is often cited by licensees as wasteful of resources and a problem. This study independently con. firmed that overlap is present. The principal overlap is associated with NRC and INP0 attempts to bring significant events and generic concerns to the attention of licensees. Some overlap is also noted with other sources such as NSSS vendor communications, and in the documents issued by single organizations (e.g., various documents prepared by the NRC).

(3) The licensees believe that the timing of feedback is an important factor (i.e., the more timely the information the more useful it is), particu-larly when multiple reports of an event are received over a period of time. While the capabilities exist to provide prompt feedback, unless the event is very significant, it currently takes weeks or months for

, detailed event information to reach the licensees. This is due to t'.:e time required to collect, assess, and disseminate the information. Tnis results in the situation where a licensee screens a document on a certain event, assesses whether action is necessary or not, and prepares docu-mentation and action items. Then, sometimes months later, a more teetailed account (or accounts) of the same event is received. The license: is then faced with yet another review cycle. Thus, this untimely feedback some-times receives little or no attention even though it contains letsons learned which may be important and timeless.

(4) Licensees report that conflicting information is occasionally fed back by one or more sources. Thus, licensees have to carefully r(ad and compare information from all sources in order to identify d'fferences and conflicts. Licensees generally have procedures for re:alving the conflict, if identified, prior to acting on the informativa within their organization.

(5) In general, the focus of licensee activities for revirning industry-wide experience is on documents for which a record of rev'uw is perceived to be necessary (e.g., NRC Bulletins and Information N';tices, INP0 SOEPs and SERs,vendornotifications). As a result, other daeuments that contain relevant information, such as NRC's Power Reactor dvents, INP0's Opera-tions and Maintenance Reminders, and vendor serv be bulletins, often receive little attention.

(6) Most licensees stated that the best and most d^ sired feedback is prompt, detailed, and comprehensive descriptions of significant events or safety concerns. Trends and patterns analyses and o*her summary information that address the operati~ng experience collectively are also desired.

These are most useful for information, train'ng, and other purposes.

Descriptions which are brief and which conta:n no assessments of signifi-cance are not widely used. Documents that r1 quire extensive analysis are also not widely used.

(7) AE0D's periodic publications are not widely used as a source of OE information for plant operators or as a primary source of information

- which licensees use to initiate improvemen's in plant operations.

O e

30

o .

4.0 CONCLUSION

S 4.1 Operating Experience Feedback Programs (1) Over the last five years, many significant and worthwhile initiatives have been implemented to understand the lessons of experience. As a result, there has been increased attention paid to OE feedback. However, many licensee programs have not yet achieved effective performance levels in all areas. Most plants are making moderate, not extensive, use of their in-house operating experience, and in general are making less use of the large body of knowledge associated with events and concerns that originate elsewhere in the industry. Thus, many licensee programs still do not provide high confidence that all of the important lessons of operating experience are being thoroughly explored and used. To increase the effectiveness of OE reviews in order to correct past and potential operational problems (i.e., assuring safety of operations), licensees and the industry or the NRC need to take further actions.

(2) The NRC, INP0 and other organizations within the industry are expending large amounts of resources on screening and assessing operational data, and preparing and disseminating OE feedback documents. However, extensive in-depth assessments of this information by licensees, and the expected voluntary actions in terms of changes in plant hardware, procedures, and' training programs based on this industry-wide feedback, were not readily apparent in many cases. The resources and attention being devoted by licensees to understanding and implementing the lessons from industry-wide are usually considerably less than those directed to in-house experience feedback. .As a result, the yield from thdse activities in terms of cor-rective actions and improved operator knowledge and capability and the associated reduction in component failures and plant events, seems smaller than expected. .

4 (3) TMI Action Item I.C.5 intended th_at the lessons of experience be routinely provided to all operating personnel in an efficient and effective way.

However,, plant operators (licensed and unlicensed), trades personnel, technicians and other non-supervisory plant personnel are not typically and effectively provided much OE information beyond events that occur in their own plant.

(4) The NRC requirements and guidance in this area are very general (i.e.,

technical program goals and objectives, evaluation criteria, or expected activities are not fully specified). The industry through INP0 has developed more detailed guidance on how CE review activities can be imple-mented, but license activities are widely diverse and vary considerably in effectiveness.

(5) NRC routinely verifies (i.e., inspects) the effectiveness of only a portion of current operating experience feedback practices on a plant by plant basis, although the NRC has attached great importance to this W

31

l l

l activity. The NRC also lacks information industry-wide on the range of effectiveness of OE review activities and the degree to which plants are meeting the established NRC requirements. At this time however, the general NRC requirements do not provide a sufficiently definitive basis to permit a meaningful evaluation of the total process by the NRC.

Current industry efforts also do not provide a suitable basis for judging '

the effectiveness of licensee activities in this area.

4.2 Feedback Documents I

(1) The large volume of OE feedback is degrading the effectiveness of the feedback programs. Duplication and overlap are diverting the resources available to effectively use OE feedback, i (2) Conflicting information provided by two or more sources pose majcr difficulties for licensees unless the . conflicts are identified. Thus, when the initiators of feedback are aware of differences in proposed actions or event details, the differe.nces need to be noted and dis-1 cussed in order to aid the licensees' evaluations.

! (3) AE00's three periodic publications (PRE, the A0 report and the LER

! compilation report) are not as useful and effective to licensees as indicated by previous surveys and discussions with users.

1

\

I e

4 o

\

32

5.0 RECOMMENDATIONS Based upon this review and the 'resulting findings and conclusions, the following recommendations are made:

(1) IE consider issuing an Information Notice (IN) to licensees to communi-cate the conclusions from this AE0D study and to suggest that licensees review their OE activities for effectiveness and for the intent and scope of conformance with the existing NRC requirements. The IN should note that licensees and the industry have lead respcnsibility for ensuring that the OE review activities are effective and reemphasize that the NRC considers the OE activities important to the safety of operations. The a findings from the three IIT reports can be cited in this regard. Specifi-cally, the IN should suggest that licensees, consistent with existing NRC requirements:

(a) Emphasize root cause analysis and implementation of permanent corrective actions to prevent recurrence of safety-related in-house events, (b) Be diligent in the screening and assessment of industry feedback for applicability to their plant and in implementing corrective actions when appropriate, (c) Disseminate information to non-supervisory plant operators, trades personnel, technicians and other support staff both directly and through training and retraining to keep them aware of anomalous events and their actual or potential consequences, and (d) Implement self-monitoring of the effectiveness of their OE review activities through management reviews.

(2) AE0D have further discussions with the industry on additional initiatives, such as on means to increase the effectiveness of OE review activities, and on assessing the effectiveness of the OE review activities.

(3) The NRC continue to monitor operating experience and industry performance in OE activities for sometime (two years or so). If after this period of time, sufficient improvement in effectiveness has not been observed, then further regulatory action would'be appropriate. In this regard, AE0D believes the following actions should be initiated at this time:

(a) NRR, in conjunction with AE0D and IE, reappraise how the.

operating experience feedback activities might be improved through further guidance. The guidance could address the shortcomings noted in this s.tudy. The elements of an accept-able program could be defined in sufficient detail ~ to permit determinations of program adequacy. Further, the guidance G

33

-- m -

a .

e 4

could address that both in-house and industry-wide experience are to be thoroughly assessed; that the applicable lessons are to be fed back to the ,various non-supervisory operating staff through direct feedback information systems and training pro- i grams; a'nd that corrective actions are to be taken through procecure changes, training, and plant modifications. The guidunce developed should, hoyever, not be overly prescriptive, and should be consistent with present reouirements.

(b) Based upon the above guidance, IE develop a draft inspection module for reviewing the effectiveness of licensee practices for implementing operating experience feedback _ activities, including. verification by NRC on a plant by plant basis (at leastannually). <

(c) IE consider revising current OE feedback reviews to routinely include a sampling of indastry f,sedback documents.

.s -

(4) Regarding OE feedback, as soon as practical, NRR, IE and ACOD. consider consolidating _ existing systems (e.g., Generic Letters th'at deal with operational events, Information Notices that discuss operational events, Power Peac_ tor Events) for transmitting' operational experience into a single "NRC Notice" system. This " notice *'would constitute the sole system for the feedback of operating experience to licensees for appro-priate action and closeout. This system could specify the initiating office an,d the purpose (i.e., OE feedback), provide an indication of the significance and of the priority of< attention expected, and could be sent to a singl'.e standard distribution. list. In addition,ethe system could make sufficient information available for licensees to fully understand the concerns for assessment purposes and to conduct training when appro-priate. In light of several licensee suggestions, consideration should be given to transmitting these " notices" vis an electronic mail system. .

(5) In conjunction with the implementation of recommendation 4vabove, AE00:

(1) terminate the publication (of,, Power Reactor Events, ced'(2). terminate the publication of the monthly Licensee Event Report (LER) Compilation, unless, there are conflicting ~ statutory or legal requirements or suffi-cient subscriptions to offset the cost. ,

^

y W

e W

34 I

f Appendix A 1.

,','. 1.C.5 PROCEDURES FOR FEEDBACK OF OPERATING EXPERIENCE TO PLANT STAFF l.

POSITION In accordance with Task Action Plan !.C.S. Procedum, for Feedback of Operating Expertence to Plant Staff, each licensee snall naview its procedures and revise them as necessary to assure that operating infor-

[

mation pertinent to plant safety oriqinating both within and outside the uttitty organization is continua'ly supplied.to oprators and other personnel and is incorporated into training and retraining programs.

These procedures shall:

(1) Clearly identify organizational responsibillttes for review of operating expertence, the feedback of pertinent informa-tion to operators and other personnel and the incorporation of such information into training and retraining programs; l (2) !dentify the administrative and technical review steps necessary in translating recossendations by the operating ex-perterice assessment group into plant actions (e.g., changes to procedures operating orders);

-(3) Identify the rectptents of various categories of information from operating experience (e.g., Supervisory personnel STA's, operators maintenanca personnel. H. P. technicians) or' other-wise provide means through which such information can be readily related to the job functions of the recipients.

(4) Provide means to assure that affected personnel become aware ~

of and understand information of suffletent importance that should not wait for emphasis through routine training and re-training programs:

(5) Assure that plant personnel do not routinely receive extraneous and untm>ortant information on operating experience in such volume taat it would obscure priority information or otherwise ,

detract from overall job performance and prof tetency:

(6) Provide suitable checks to assure that conflicting or contradictory information is not conveyed to operators and other personnel untti resolution is reached; and, (7) Provide periodic internal audit to assure that the feedback program functions effectively at all levels.

_0ISCUSSION_

i Each utility is expected to carry out an operating experience assessment

^

function which will involve utility personnel having collective cogetence in all areas importar.t to plant safety. In connection with this assessment function it is important that procedures exist to assure that important Part 1 to Enclosure 2 to May 7,1980 Letter From D. Eisenhut to Licensees on Additional TMI-2 Related Requirements 35

Appendix A '

- 2-infonnation on operating experience originating both within and outside the organization is continually provided to operators and other personnel and that it is incorporated into plant operating procedures, and training and retraining programs. '

Those involved in the assessment of operating experience will review in-formation from a variety of sources. These include operating infonnation fromthelicensee'sownplant(s),publicationssuchasJEBulletins, Circulars and Notices, and pertinent NRC or industrial assessments of operating experience. In some cases, infennation may be of sufficient importance that it must be dealt with promptly (through instructions, changes to o precautions,perating and emergency procedures, issuance of special etc. and must be handled in such a manner to assure that operations managem)ent personnel wopld be directly involved in the pr In many other cases, however, imortant information will become available which should be brought to the attention of operators and other personnel for their general infonnation to assure continued safe plant operation.

Since the total volume of information handled by the assessment group may be large it is important that assurance be provided that high-priority matters are dealt with progtly and that discrimination is used in the feedback of other information so that personnel are not deluged with unigortant and extraneous infonnation to the detriment of their overall proficiency. It is important, also, that technical reviews be conducted to preclude premature dissasination of conflicting or contradictory infor-mattoa.

CLARIFICATION Review of and modifications to procedures governing feedback of operating experience to plant staff shall be completed and the procedures put into ef fact on or before Janupry 1,1981.

ACTION: (Class!) _

(1) Licensee to implement actions and submit documentation of the methcd for staff review by scheduled dates.

l l

36

- - - - - -g-- * - - - - - - , - - , , e e - ,- ,-G- -w---*---- --w.- --~- ---w -me - - - - - - - - - - w

fh p5stuq% ,

UNITED STATES Apoendix B

! n

_ y, g e, - U j

e NUCLEAR REGULATORY COMMISSION wA HincTon, c. c. 2csss Parch 10, 1982

___..-. e=

/

w.- . *

.- . - . . . . ?. C '

TO ALL LICENSEES OF OPERATING PLANTS, APPLICANTS FOR OPERATING LICENSES AND HOLDERS OF CONSTRUCTION PERMITS Gentlemen:

SUBJECT:

USE OF INPO SEE-IN PROGRAM (GENERIC LETTER NO. 82-04 )-

Item I.C.5 cf the TMI Action Plan, requires. licensees to develop procedures toassure that important operating experiente originating both within and outside the utility organization is continually provided to operators and -

other per.sonnel, and is incorporated into training and retraining programs. '

Accomplishing this task requires that a utility assure that all events that occur throughout the industry are screen'ed to determin; whether or not they are applicable to the utility's plant, and that those e. vents which could be ,

significant are evaluated for applicability to the utility's plant. -

The initial screening process is a laroe undertaking involving several '

thousand event reports each year from the U. S. nuclear plants alone.

Such an undertaking by any one utility is both expensive and demanding of resources.

In an effort to provide a all events from both U. S., and mechanism foreignfor centralplants, nuclear collection theand screening Institute of of ,

Nuclear Power Operations (INPO) has established a Significant Event Evaluation and Information Network (SEE-IN). The latest description of the SEE-IN

_ program is provided in a Ja'riiiary 1982 program description from INPO entitled *

,. "Significant Event Evaluation and Information Network (SEE-IN)."

The NRC has now completed its review.of the SEE-IN program and a recent letter to INP0 (copy enclosed) ind6rses utility use of the program. As

' stated in the letter, use of SEE-IH will relieve individual nuclear plant operators and constructors of the necessity of setting up large staffs to obtain and screen the large volume of raw data pertaining to operational experience throughout the industry. The NRC believes that full partici-pation in SEE-IN will enhance y6ur ability to meet the intent of the procedures a'pproved under TMI Act. ion Plan Item I.C.S.

Participation in SEE ,IN does not relieve a utility from taking those actions specific to the utility's nuclear unit.which result from an evaluation of operational experiences. Each utility is .still required to have an internal -

procedure for handling operational experierice informatica, including the

' procedures necessary to a:.sure that appropriate individuals are provided the results of evaluations and that recommendations for corrective action identified as a result of evaluation are translated into actions. The interface between SEE-IN and, the functions and responsibilities of the utility necessary to satisfy- the requirements of Item I.C.5 are discussed in Section 3 of the SEE-IH program description. .

37 8203110001

, .  ;~ ,

j Appendix B The NRC believes that utilization of the SEE-IN program will result in the eliminat. ion.of:: duplication which will occur if each utility attempts

-a separate evaluation of significant plant operation events. The central-ization of this initial screening effort at INPO should result in a more efficient evaluation and will allow you to concentrate your efforts on evaluating the events that occur in your plant and those that are ident-ified through SEE-IN as being applicable or potentially applicable to your pl. ant.

The full potential of the SEE-IN program can be relized only if all utilities participate actively, both in furnishing event information to INPO and in taking corrective actions as necessary when potential problems have been identified as a result of INP0 efforts. The SEE-IN program is acceptable to the staff with no additional review required. ' Your partici-pation in the SEE-IN program is strongly encouraged.

,. Sincerely, .

~ .

r.

l b '

Darrell G. Eisenhut, Director Division of Licensing

, Office,of Nuclear Reactor Regulation

.c -

..~.

8 O .

i .

e e

e

. .~ ,

38 O

• . +

Appendix B g ese g .

4 jo, UNITED STATES

[ ]#-("((4;w E$ NUCLEAR @EGULATORY COMMISSION

~

~A / WASHINGTotJ. D. C. 20555

,.- Y,kY ,

u_. . . __ . ' . ._ .

, . . . January 11,1982 Mr. E. P. Wilkinson, President Institute of liuclear Power Operations 1E20 Water Place Atlanta, Georgia 30339

Dear Mr. Wilkinson:

We have reviewed the description of your Significant Event Evaluation and Information lietwork (SEE-Itt) program provided by your, letter of August 6,19El.

.The purpose and objectives of the program and the mechanics for its imple-

mentation, as described in yo'ur report, appear to adequately support the collection, evaluation and dissemination of operating experience. Your progra::t appears to effectively consolidate operating experience events, such as LERs,

• into one data assessment program.

We believe that the SEE-Iti program will make a significant contribution to reactor safety. At the same time, it will relieve the individual nuclear plant operators of the necessity of setting up large staffs to obtain and screen the large volume of raw da.ta pertaining to operational experiences throuihout the industry. Thus, we view the.SEE-Iri program as satisfying a large fraction of the licensee's obligations under TMI Action Item I.C.5.

The liRC/AE00 has a.MemorandOS of Agreement with liiPO u'nder which you are

.. regularly transmitting the results of your analysis effort. Our review of these output documents arid our knowledge of the events described allows us to ovarview your program's adequacy and timeliness. ,

We look forward to seeing a copy of the revised SEE-Iti program which consolidates the whole program at Ilip0.

, 4, sincerely, i

4;w/0 Harold R. Dentori, Director x Office of tiuclear Reactor Regult. tion .

39

c -

Appendix C Summary Tables from Visits to Seven Sites e

40

~

Ap' Appendix C Introduction The following summary ' tables were developed from discussions with various licensee personnel and from reviews of records such as samples of training plans, required reading packets, action tracking documentation and procedures. Tables C-1 thru C-4 illustrate conceptually how the operating experience information flows and how much is communicated to various personnel. The quantity indicated is a gross estimate of the percent of information available from the plant, INPO, NRC and the NSSS vendor. With regard to the legend for these tables, a

" fair amount" of operating experience feedback is over 30% of the plant's inci-dent or deviation reports plus most LERs or of the industry-wide information available from the NRC, INP0 and the NSSS vendor; a " limited quantity" is 10% to 30%; and "little" is less than 10%. Table C-5 illustrates how AE0D's three periodic publications are being used by licensees.

O

• 5 e e

41

.c .

ap. .

Table C-1 Operating Experience Feedback to NRC Licensed Operators (P0s and SR0s)

Feedback Plant Sites A B C D E F G

'1) In-house Experiences a) General Reading Circulation C C C C C C B b) Required Reading A A A A A A B c) Shift logs and other prompt A A A A A A A notifications d) Training and retraining B B B B B B B e) Simulator exercises C C B C C C B f) Discussions (Between"indi- B B B B B B B viduals, within shift crews or with support or opera-tions staff)

2) Industry-wide Experiences a) General Reading Circulation B B B B B B B b) Required Resding C B B B B B B c) Prompt Notifications C C C C C C C

. d) Training and retraining C B B B B C B

! e) Simulator exercises C B B B B C B f) Discussions (Between indi- C C C C C C C viduals, within shift crews

'or with support or opera-tions staff)

Legend:

A - Fair Amount B - Limited Quantity C - Little 42

N' ;

• Table C-2 Operating Experience Feedback to Other Plant Operators Feedback Plant Sites A B C D E F G

1) In-house Experiences a) General Reading Circulation C C C C C C C b) Required Reading N/A N/A N/A N/A N/A N/A B c) Shift logs and other prompt B B B B B B B notifications d) Training and retraining B B B B B B B e) Simulator exercises N/A N/A N/A N/A N/A N/A N/A f) Discussions (Between indi- B B B B B B B viduals, within shift crews or with support or opera-tionsstaff) -

2) Industry-wide Experiences a) General Reading Circulation C C C C C B B b) Required Reading C B B C C C C c) Prompt Notifications C C C C C C C d) Training and retraining C B C C C C C e) Simulator exercises N/A N/A N/A N/A N/A N/A N/A f) Discussions (Between indi- C C C C C C. C viduals, within shift crews or with support or opera-tions staff)

Legend:

A - Fair Amount B - Limited Quantity C - Little N/A - Not Applicable 43 l

q;- ,.

Table C-3 Operating Experience Feedbac,k to Maintenance, Trades, I&C, Radiation Protection, etc.

Feedback Pla'nt Sites A B C D E F G

1) In-house Experiences a) General Reading Circulation C C C C C C C b) Required Reading N/A N/A N/A N/A N/A N/A N/A c) Shift logs and other prompt C C C C C C C notifications d) Training and retraining C B C C C C C e) Simulator exercises N/A N/A N/A N/A N/A N/A N/A f) Discussions (Between indi- C C C C C C C viduals, with shift crews or with support or opera-tions staff)

2) Industry-wide Experiences ,

a) General Reading Circulation C B C C C C C b) Required Reading N/A B N/A N/A N/A .i/A N/A c) Prompt Notifications C C C C C C C d) Training and retraining C C C C C C C e) Simulator exercises N/A N/A N/A N/A N/A N/A N/A f) Discussions (Between indi- C C C C C C C viduals, with shift crews or with support or opera-tionsstaff)

Legend:

~ '

A - Fair Amount

. B - Limited Quantity .

C - Little N/A - Not applicable -

44

43a Table C-4 Operating Experience Feedback to Plant Management and Technical Support Staffs Feedback Plant Sites A B C D E F G

1) In-house Experiences a) General Reading Circulation A A A A A A A b) Feedback Program Activities A A A A A A A c) Shift logs and other prompt C C C C C C C notifications d) Training and retraining C B C C B C C e) Simulator exercises N/A N/A N/A N/A N/A N/A N/A f) Discussions (Between indi- A A A A A A A viduals, with shift crews or with support or opera-tions staff)

2) Industry-wide Experiences a) General Reading Circulation A A A A A A A b) Feedback Program Activities A A A A A A A c) Prompt Notifications C C C C C C C d) Training and retraining C C C C C C C e) Simulator exercises N/A N/A N/A N/A N/A N/A N/A f) Discussions (Between indi- A A A A A A A viduals, with shift crews or with support or opera-tions staff)

Legend:

A - Fair Amount B - Limited Quantity C - Little N/A - Not applicable -

45

a, r- .

Table C Licensee Use of Routine AE00 Publications Uses Plant Sites A B C D E F G

1. Circulate for general all all all all all all Selected information to a standard items distribution from I and II

2. Incorporate in Required Reading a) Selected Items none none all I,II none I,II I,II b) Entire Document none none none none none none none.

3. Use in training and retraining programs ,

a) Routinely none none I,II I I,II none none b) Occasionally I,II none all II all I,II I,II -

4. Use infregunetly to initiate all all all all all I,II I,II actions (procedure changes, plantmodifications,etc.)

Legend:

I - Power Reactor Events -

II - Abnormal Occurrence Report ,

III - LER Compilation Report l 1

All - I, II and III None - None are used 46

.