ML19347F622

From kanterella
Jump to navigation Jump to search
Proposed Findings of Fact & Conclusions of Law on Mgt issues.Short-term Recommendations Should Be Required Before Restart While long-term Recommendations Should Be Required as Soon as Practicable
ML19347F622
Person / Time
Site: Three Mile Island Constellation icon.png
Issue date: 05/15/1981
From: Blake E
METROPOLITAN EDISON CO., SHAW, PITTMAN, POTTS & TROWBRIDGE
To:
Shared Package
ML19347F620 List:
References
ISSUANCES-SP, NUDOCS 8105220129
Download: ML19347F622 (235)


Text

. , .

4 ...

t/.AY 131981 * ;-

~

3

.\

LIC 5/15/81 's -

y , ,y \ '

UNITED STATES OF AMERICA ,

NUCIEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of ) '

)

METROPOLITAN EDISON COMPANY ) Docket No. 50-289 SP

) (Restart)

(Three Mile Island Nuclear )

Station, Unit No. 1) )

LICENSEE'S PROPOSED FINDINGS OF FACT AND CONCLUSIONS OF LAW ON MANAGEMENT ISSUES i

1 SHAW, PITTMAN, POTTS & TROWBRIDGE George F. Trowbridge Ernest L. Blake, Jr.

Deborah B. Bauser Counsel for Licensee

{

l 8105220 v

- - - _- _ /.2 9-- '

o -

i LIC 5/15/81 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of )

)

METROPOLITAN EDISON COMPANY ) Docket No. 50-289 SP

) (Restart)

(Three Mile Island Nuclear )

Station, Unit No. 1) )

LICi'ISEE'S PROPOSED FINDINGS OF FACT

, AND CONCLUSIONS OF LAW ON MANAGEMENT ISSUES l

SHAW, PITTMAN, POTTS & TROWBRIDGE George F. Trowbridge Ernest L. Blake, Jr.

Deborah B. Bauser Counsel for Licerrae

.,.,,_,._.o.._,.__., .m-._,,,.,m_,, ,, _..,_,--,.,,.,-.-.-.y,, --,.v, _ , _ ,, .--

i

'~ .

e' l C 1 l

l TABLE OF CONTENTS l LICENSEE'S PROPOSED FINDINGS OF FACT

AND CONCLUSIONS OF LAW ON MANAGEMENT ISSUES I. Introduction...................................... 1 II. Licensee's Command and Administrative Structure......................................... 4 III. Quality Assurance................................. 22 IV. TMI-l Organization and Technical Resources........................................ 29
A. TMI-l On Site Organization.................. 30 E, -TMIA Contention 5............................ 56 C. Training.................................... 91 l

D. Aamod t Conten tion 2. . . . . . . . . . . . . . . . . . . . . . . . . 13 6 V. Views of NRC Inspectors.......................... 156 VI. Health Physics................................... 161 l VII. Financial / Technical Interface.................... 170 VIII. Safety Reviews and Operational Advice........................................... 176 IX. Validity of Comparing TMI-l Infraction, LER and Operating Exp7tience History with Ind us try-Wid e S ta tis tics . . . . . . . . . . . . . . . . . . . . . . . . . 18 7 X. Licensee's Management Response to the TMI-2 Accident................................... 203 XI. Technical Capability and Resources........................................ 214 XII. Othe r Specific Issue s . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 XIII. Conclusions of Law............................... 227 i

I i

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

LIC 5/15/81 l

UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION I

BEFORE THE ATOMIC SAFETY AND LICENSING BOARD 1

In the Matter of ) l

)

METROPOLITAN EDISON COMPANY ) Docket No. 50-289 SP

) (Restart) i (Three Mile Island Nuclear )

Station, Unit No. 1) )

LICENSEE'S PROPOSED FINDINGS OF FACT AND CONCLUSIONS OF LAW ON MANAGEMENT ISSUES I. INTRODUCTION

1. Licensee's management competence to operate TMI Unit 1 is an issue which the Board has evaluated very carefully and thoroughly, on the basis of an extensive record. The Board is particularly aware of the responsibility delegated to it by the Commission in its Order, CLI-80-5 (March 6, 1980), to de-velop a full record on this subject. The Board also recognizes the fundamental importance of the issue of Licensee's manage-ment competence, and the fact that assessing a company's capability to operate a nuclear power plant, from a technical, organizational and management standpoint, is a matter which is based not only on findings of fact, but also on the Board's subjective impressions of that company's management and

y . - .*-

C /

personnel. The Board is satisfied that during the course of the lengthy TMI-l restart hearing, it has had the opportunity to observe, first-hand, and direct questions to most if not all of the management and technical personnel who will have significant responsibilities in Licensee's organization. These findings of fact and conclusions of law therefore reflect both an objective understanding of the organizational structure and resources available to Licensee, and a subjective awareness of the people who will be making daily decisions at TMI-1 and within the larger GPU nuclear organization.

2. We have chosen to organize the vast amount of material on the subject, management capability, by following the issues identified by the Commission in CLI-80-5. In addition, in the body of our discussion on a particular order item, two management capability contentions, TMIA Contention 5 and Aamodt Contention 2, have been addressed directly. The Board has focused upon these two contentions because they were actively pursued by intervenors; consequently, they deserve our full attention and review. There are three other management-related contentions which have been considered by the Board, ANGRY Contention IV, CEA Contention 13 and Sholly Contention 14,1 even though these contentions were not pursued by their 1 ANGRY Contention IV states:

The Licensee lacks the management capability to operate a Nuclear Generating Station without endangering the public health and safety.

(continued next page)

A respective sponsors. Rather than address these three

~

(continued) ~

CEA Contention 13 states:

CEA contends that there is a specific need for the establishment of training for operators that addresses the problem of 'mindset' that denies information indicative of serious problems.

Sholly Contention 14 states:

The licensee's management capability, in terms of organizational, staffing, and technical capabilities, is not sufficient. Specifically, the following deficiencies in Licensee's management capability are contended:

(a) Licensee's administrative structure, both at the plant and corporate levels, is not appropriately organized so as to assure safe operation of TMI-1 while conducting cleanup operations at TMI-2.

(b) Licensee's operations and technical staffs are not sufficiently qualified to safely operate TMI-1.

(c) Licensee's Health Physics program is not appropriately organized, nor is it staffed with sufficiently qualified individuals to ensure safe operation of TMI-1.

(d) Licensee has not made sufficient provision for qualified individuals to provide safety review of and operational advice regarding TMI-1.

(e) Licensee's maintenance program is insufficiently staffed and inappropriately organized for the purpose of safel*1 operating TMI-1.

Intervenor Steven Sholly withdrew his Contention 14 in l December, 1980. Sholly Memorandum (December 23, 1980).

l Nevertheless, Licencee explicitly referenced the applicable portion of Sholly Contention 14 in its prefiled testimonies on management capability. Because Sholly Contention 14 l raised no subjects which were not explicitly identified by l the Commission in CLI-80-5, its formal withdrawal did not change the scope of the proceeding with respect to

Licensee's case on management competence. The Board (continued next page)

l contentions individually, however, the Board has considered and responded to the issue (s) raised by them in our discussion of the relevant CLI-80-5 issue item or, in the case of ANGRY Contention IV, a very general contention about Licensee's management capability, we have taken the contention into account throughout our findings.

3. Finally, as we noted in paragraph twelve of our Introductory Findings, the Commission has decided to take the issue of Licensee's financial qualifications out of the TMI-l restart hearing. CLI-81-3 at 9 (March 23, 1981). Conse-quently, the Board has made no findings of fact or conclusions of law with respect to issue item (12) in CLI-80-5, which questions whether Metropolitan Edison possesses the financial .

resources necessary to safely operate Unit 1 in addition to cleaning up Unit 2.

II. LICENSEE'S COMMAND AND ADMIl1ISTRATIVE STRUCTURE

4. CLI-80-5, Issue (1) identifies the following issue for the Board's consideration:

(1) whether Metropolitan Edison's command and administrative structure, at both

, the plant and corporate levels, is appropri-l ately organized to assure safe operation of

Unit 1.

(continued) is confident that the concerns raised by Mr. Sholly in Sholly Contention 14 have been taken into consideration by us in these findings.

Because of the reorganization of Licensee's nuclear-related activities, which we discuss in detail in the following paragraphs, the Board has examined this issue in the context of ~

the organizational entity called the GPU Nuclear Corporation, rather than simply in terms of Metropolitan Edison Company (Met Ed).

5. Virtually all of the witnesses who testified on the subject of the command and administrative structure used by Licensee to operate, technically support and manage its nuclear activities, including safe operation of TMI-1, explicitly stated or implied that the extensive modifications which have been made to Licensee's organization, including the changes in responsibilities of key management-level personnel and the influx of additional personnel, constitute a significant positive factor in their endorsement of Licensee's command and administrative structure at both the plant and corporate levels. Arnold, ff. Tr. 11,434; Tr. 11,528 (Arnold); Keimig, ff. Tr. 11,946, at 8; Tr. 11,988-98 (Crocker, A11enspach); Tr.

12012-15 (Crocker, Allenspach); Tr. 12024-25 (Haverkamp); Lee, ff. Tr. 13,251, at 11-12; Tr. 13,271, 13,274-75 (Lee); Wegner, ff. Tr. 13,274, at 8-12; and Tr. 13,309 (Miles). In order to understand Licensee's reorganization, we will briefly review the former organization in place at the time of the TMI-2 accident.

6. At the time of the TMI-2 accident, General Public Utilities (GPU), located in Parsippany, New Jersey, was the

4 .

I parent holding company of three operating subsidiary utilities and one company which previded common administrative and technical support to the operating utilities. Arnold, ff. Tr.

11,434, at 3.

The three operating utilities were Metropolitan Edison Company, Pennsylvania Electric Company, and Jersey Central Power & Light Company, which shared undivided ownership interests of 50%, 25% and 25%, respectively, in both TMI-l and TMI-2. Arnold, ff. Tr. 11,434, at 3. The company which provided common administrative and technical support was GPU Service Corporation. Id. at 4; Tr. 11,470-71 (Arnold). GPU Service Corporation's resources were available to provide technical support, management review and oversight, and audit functions. Arnold, ff. Tr. 11,434, at 4.

7. Met Ed was the exclusive operator of the TMI units with more than 500 of its employees stationed at Three Mile Island. In addition, Met I d's corporate staff in Reading, Pennsylvania provided administrative and technical support to l TMI and the other Met Ed generation and operating activities.

The chief operating officer of Met Ed was its President, who

- required certain status reports from the Vice President Generation with regard to the operation and maintenance of TMI-1, but who did not become involved with the day-to-day detailed activities at the facility. The Vice President Generation was in charge of the operation, maintenance, administration, quality assurance and related technical engineering support activities associated with all of Met Ed's

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

1 generating stations, including its nuclear plants at TMI.

Reporting to the Vice President Generation were six managers.

One of these managers, the Manager Generating Station-Nuclear, was responsible for operational control of TMI -- both units.

It was the job of the Manager Generating Station - Nuclear to assure that the station was operated and maintained in accor-dance with company policies, the requirements of the NRC license and the quality assurance program. .His accountability included responsibility for proper conduct of station adminis-tration, testing, repair, refueling, radiological controls, environment controls and emergency operations, as well as routine operation and maintenance. Reporting to the TMI '

Station Manager were four superintendents: Superintendent TMI-1, Superintendent TMI-2, Superintendent Maintenance and Su-perintendent Administration. Under the NRC licenses, the unit superintendents were assigned direct responsibility for safe operation of TMI Units 1 and 2. Arnold, ff. Tr. 11,434, at 3-6.

8. According to Mr. Robert C. Arnold, the man currently in charge of Licensee's nuclear activities,2 prior to l 2 At the time he testified on February 3,1981, Mr.

Arnold's titles were Senior Vice President of Met Ed and Jersey Central Power & Light Company, Vice President of GPU Service Corporation, head of the GPU Nuclear Group' and President of GPU Nuclear Corporation. Once the GPU Nuclear Corporation has been authorized under its various licenses to operate TMI-1, TMI-2 and Oyster Creek, Mr.

Arnold's only title will be President of that corporation.

Nevertheless, under both organizational scenarios, Mr.

Arnold's duties and responsibilities will be functionally identical. Tr. 11,483, 11,459-61, 11,125 (Arncid).

I I

the TMI-2 accident the management of GPU believed that GPU's nuclear activities, i.e., TMI-1, TMI-2, Oyster Creek, Saxton (decommissioned) and Forked River (a subsequencly cancelled project) would benefit from expansion of GPU's in-house technical capabilities, much greater involvement by the engineering groups active during plant design and construction with the technical functions necessary during plant operations, and a consolidation of the technical and management structures responsible for GPU's nuclear activities. GPU had embarked upon this program when the accident occurred at Three Mile Island in March of 1979. Tr. 11,537-38 (Arnold); Arnold, ff.

Tr. 11,434, at 6-7.

9. In his testimony, Mr. Arnold describes the genesis of the current organization of nuclear activities within the GPU system essentially as a combination of the idea to consolidate resources which had already been formulated within the organization, and as an outgrowth of the need for concentrated and integrated management and technical resources af ter the TMI-2 accident. Arnold, ff. Tr. 11,434, at 7-9. In order to maintain TMI as a priority, this method of doing nuclear business was formalized by GPU in July 1979 by the l

formation of the TMI Generation Group, under the direction of Mr. Arnold. The objectives of this new organization were to integrate the technical support capabilities available within the Met Ed and GPU Service Corporation Generation Divisions with the Met Ed operations and maintenance personnel for

~B-

support of day-to-day plant operations, to augment the manage-ment of non-operating functions and to apply additional technical and management skills to the activities at TMI. Id,.,

at 8-9. This group was nuclear-focused and essentially divorced from fossil and other generation or transmission and distribution utility engineering responsibilities. In addition to serving TMI, Units 1 and 2, the Generation Group wa s charged with continuit:s the GPU Service Corporation role of providing support to Jersey Central's Oyster Creek, the remaining operating nuclear plant in the GPU system. The Generation Group formally evolved into the GPU Nuclear Group with NRC's authorization of the amendment of the TMI 1 operating license in September, 1980; however, its basic purpose of consolidation and expansion of resources to support GPU's nuclear activities has not changed. Id.,at 9.

10. The head of the Nuclear Group is Mr. Arnold, who is assisted by a deputy, Mr. Philip Clark. Reporting to Mr.

Arnold and Mr. Clark are Directors in charge of operations at each of the three nuclear plants in the GPU Syst em, and six Directors in the areas of Technical Functions, Nuclear Assurance, Radiological and Environment Controls, Maintenance and Construction, Administration and Communications, Each of these individuals will be an officer in the GPU Nuclear Corporation, which was staffed by approximately 1,925 in 1980, and is authorized to be staffed by approximately 2,500 people by the end of 1981. Id.,at 9-10, Figure 1, and following

,e-, - - - - - , , , , , , - , - , , , - , - , - + , 4.,y.,,,,,-e-,..,, m. - , , - , , - , _,-.w,._en,_,-,-,-.,.,,,g- ,,,mw-,.,,.,.,e--w.-,,,-n.n-e-e-we,,,.--,----,---m-,, -~e,.na,,-s----e

o charts 5-7;3 Dieckamp, ff. Tr. 13,437, at Figure 5. Overseeing the activities of the GPU Nuclear Group is a management oversight committee composed of the presidents of Jersey Central, Met Ed, Pennsylvania Electric Company and GPU Service Corporation, plus the chief executive officer of GPU, GPU Service Corporation's vice president for corporate planning, Mr. Arnold and Mr. Clark. This committee meets monthly to review the operating performance of the GPU Nuclear Group, as well as the Group's budget. Tr. 11,441, 11,475 (Arnold).

Under the GPU Nuclear Corporation, the management oversight committee members will become GPU Nuclear's Board of Directors.

Tr. 11,475, 11,438 (Arnold).

11. It is the intention of GPU to move from the GPU
Nuclear Group, currently authorized by license to operate TMI-1, to the GPU Nuclear Corporation, a saparate corporation (rather than a group made up of segments of GPU's subsidiaries) functionally analogous in all respects to the GPU Nuclear Group. Arnold, ff. Tr. 11,434, at 10, 28; Tr. 11,435 (Arnold).

As of February 3, 1981 when Mr. Arnold testified, GPU Nuclear Corporation, a new GPC subsidiary, had been formed, although only one of the four principal regulatory approvals necessary for the GPU Nuclear Corporation to function in its intended role had been obtained. Arnold, ff. Tr. 11,434, at 28; Tr.

11,436 (Arnold). When functional, the Nuclear Corporation will 3 During the course of his direct testimony, Mr. Arnold l utilized nine slides, copies of which follow Mr. Arnold's prepared testimony following Tr.11,434. Tr. 11,477 (Smith).

l l

l assume responsibility for the operation and maintenance of all of GPU's nuclear plants as well as the management of the design and construction of modifications for those plants, i.e., it vill supersedo the GPU Nuclear Group as the organization responsible for operation of the plants. Arnold, ff. Tr.

11,434, at 10, 28. Ownership will not change; it will continue to vest with the individual utility operating companies that comprise the GPU system. Jd,., at 28.

12. GPU strongly endorses the concept of a separate

. company wholly dedicated to GPU's nuclear activities. Besides administrative convenience, this organization represents to GPU a means to emphasize the importance it placas on the safe and efficient ope ation of GPU's nuclear stations, and the impor-tance of centralizing man gement of GPU's nuclear activities.

Tr. 11,528 (Arnold); Arnold, ff. Tr. 11,434, at 29. Specifi-cally, during his direct testimony, Mr. Arnold explained what management identified as major elements of the organization in deciding upon GPU Nuclear and in transmitting to GPU personnel l what the new organization was all about:

First of ell, [] we wanted a full-time organization within the GPU system dedicated solely to nuclear generation. We believe that this is consistent with a number of recommendations that came out of the acci-dent.

We wanted increased onsite technir al and management resources. We wanted an organiza-tion that was characterized by a very strong central control of technical issues, and the technical integrity of the plants.

(

We wanted onsite, full-time management; that is, senior management for plant

- ---r,.4r-- - - ,, - - - - - - , . - , , - - - - , - , - . - - , . , , ._,.----.<-.,....--.-r,,,-,-.-,,w._ - - , , , ..r,a.,,-. , ----.------,-e - - - - - . , , - - - . - - - , - , -

o operations and maintenance, and that management onsite would be dedicated to operation and maintenance in the sense that it would have primary responsibility there, -

and we would provide other kinds of func-tional activities in a support role. . . . i e *

  • We wanted an independent nuclear assurance division and by that, we were attempting to gather into a division that has the same organizational status and of the same visibility, the same strength within the total organization as operations would have, and which would be responsible for those activities which are extremely important to the reliability and safety of operations, but which can relatively easily become kind of

[] secondary importance if they are lef t as sort of a collateral duty for people who have also assignment for line responsibilities.

We include in that training [,] quality assurance, a nuclear safety assessment department . . . and emergency preparedness.

We look for the advantage of the pooling of the resources that are necessary for support of the several generating units, including in that concept the advantages that come from the cross-fertilization of ideas and the cross-flow of information on operat-ing experience.

We felt it very important that we had personnel policies, procedures, practices and i

resources dedicated to those areas that would be able to address the unique aspects of those areas within the nuclear technology.

That is , that there are different requirements for our nuclear plants and the supporting personnel than there are for the traditional electric [ operations.4] +-

Tr. 11,438-40 (Arnold).

4 See Licensee's transcript corrections.

  • - r--.e -

r---- -.w.,.#, v-w.,,---.-----,,--, -.-,,me --.-,-r....- ,,-----.---,-,-,-v-.e-,.--.. .,e--,.--.----..---.-,,--,.,.-y---, - . , . - r .. - - - - -~>

Y. #

d I

13. Because of the functional equivalence of the GPU Nuclear Group and the GPU Nuclear Corporation, Arnold, ff. Tr.

11,434 at 28, and the Staff's willingness to endorse either organizational framework, Staff Ex. 14, at 22, the Board considers its findings with respect to management capability to apply to GPU Nuclear's organization, whether it exists in the group or corporate form. For purposes of convenience and clarity, however, we will use titles applicable to GPU Nuclear i

Corporation personnel.
14. The basic organizational structure which Licensee has adopted has been reviewed by the NRC Staff, as well as by other experts. The NRC Staff, both the Division of Nuclear Reactor Regulation (NRR), and the Division of Inspection and Enforcement (IE), are more than satisfied with the reorganization of nuclear activities into a single entity within the GPU system. Staff Ex. 4, 5 III.B.2, at 4-8; Tr.

11,961-64 (Crocker); Tr. 12,014 (Allenspach); Keimig, ff. Tr.

{ 11,946, at 8. It is the view of IE that Licensee, in making this change, has increased sub=tantially the total overall management and technical resources available for TMI-l restart.

Keinig , ff. Tr.11,946, at 8. Additionally, experienced managers and professionalc have been introduced at both the corporate and plant levels of the organization. IE expects this centralized, nuclear-oriented group to provide increased management capability to focus on nuclear operational matters, as well as personnel policies and procedures applicable to nuclear activities. Id,. The view expressed by NRR was that Licensee's reorganization paralleled the model developed by NRR af ter the TMI-2 accident to evaluate the adequacy of the organizational structure of applicants and licensees and, therefore, was totally acceptable. Tr. 12,014 (Allenspach, Crocker).

i 15. The endorsement of Licensee's organizational structure at both the corporate and plant level was reiterated l

by two well-qualified witnesses proffered by Licensee. Mr.

William S. Lee, President and Chief Operating Officer of Duke Power Company, and Chairman of the Board of Directors of the Institute of Nuclear Power Operations (INPO), testified that GPU's proposed organizational structure for consolidating nuclear activities into the new subsidiary GPU Nuclear Corporation is a strong management concept that can be effective in providing an integrated single-minded approach which will give additional assurances of safety. Lee, ff. Tr.

13,251, at 11-12; Tr. 13,273-76 (Lee). Mr. Lee is a graduate engineer whose experience with nuclear power began in the Navy nuclear program in the early 1950's, and continued with his extensive engineering and management involvement in the nuclear activities of Duke Power Company since 1955, as well as his l

numerous professional and honorary activities associated with the nuclear industry. Lee, ff. Tr. 13,251, at 2-3.

l

16. Mr. Lee's viewpoint was shared by Mr. William Wegner, a nuclear engineer who served as the Deputy to the

/

Director of the Naval Reactors program, Admiral Rickover, from 1964 until his retirement in 1979 when he formed a consulting

~

firm with three other individuals, each with some twenty-five years of experience in the Naval Reactors program. Wegner, ff.

Tr. 13,284, at 1, Attachment 1, at 2. Mr. Wegner and his associates conducted an extremely detailed review of Licensee's management structure, capability and technical resources,5 ,,

5 The assessment conducted by Basic Energy Technology Associates, Inc. (BETA), Mr. Wegner's firm, began prior to the development of the management criteria draf ted by NRC. In evaluating the issue of management capability and technical resources, because one,is dealing with people, with organizational structure, with attitudes and with many other at-tributes, none of which conform to any given laws, there is a lack of specific criteria which must be met in a very specific way; rather the overall capability of an organization must be judged by looking at the entire picture, not just one isolated I segment. This is what BETA attempted to do. After defining management capability and technical resources as that overall capability of a utility to own, operate, and be fully responsible for one or more nuclear power plants in such a way as to protect the health and safety of the worker and the public, BETA went on to develop a scope of inquiry. The following groups were assessed by BETA: (1) corporate headquarters; (2) both off-site and on-site organizations relating to (a) overall manage-ment; (b) operations; (c) engineering / technical; (d) licensing; (e) quality assurance; (f) nuclear safety assessment; (g) se-lection, training, and qualification; (h) radiological control; (i) emergency planning; (j) fiscal management; (k) personnel matters; (1) labor relations; (m) material management; (n) in-dustrial safety; (o) security; (p) facilities management; (q) public relations; (r) radioactive waste management; (s) fire, protection; (t) environment; (u) maintenance; (v) records control; and (w) water chemistry. Each of these groups was reviewed to determine if it was sufficient in the following areas: (1) detailed written procedures; (2) clear lines of responsibility and authority; (3) qualified personnel, number and qualifications; and (4) accountability for actions.

Particular attention was given to the working relationship be-tween comparable off-site and on-site functions and on the attitude of management. Wegner, ff. Tr. 13,284, at 1-5.

l A

well as the attitude of management, beginning in October 1979 and continuing intermittently into January 1981. Over 150 employees of GPU and its affiliated organi.9ations were inter-viewed, usually on a one-to-one basis for from one to as long as ten hours. BETA also worked with GPU on specific technical issues related to TMI-l and, consequently, had an opportunity I

to judge first-hand the technical and management capability of the GPU organization. Mr. Wegner ccncluded that the new organization and the management of the GPU nuclear plants through this single, unified structure is probably the most effective way a nuclear utility could be handled. Id., at 6,  !

11-12.

4

17. In summary, there was a pervasive view expressed i

among the witnesses who testified on the subject of Licensee's command and administrative structure at the plant and corporate level that Licensee's reorganization is a very important factor contributing to their sense of confidence that Licensee's management capability and technical resources are adequate to l

2nable Licensee to safely operate TMI-1. The conclusion that this structure brings clear advantages to Licensee's nuclear activities appears to be very well-founded in light of the following observations, made by Mr. Wegner: (1) The establish-l ment of a single organization, reporting to a high corporate level and responsible for all aspects of nuclear plant opera-tion and sapport, is in agreement with many of the recom-mer.dations contained in post-TMI accident reports. (2) By l

l

combining the technical resources of the various GPU utilities, a larger pool of talent has been assembled which can be put at .

the disposal of the nuclear plants in order to resolve problems and to ensure a better flow of information between the plants.

(3) By having a larger base of technical and management talent the GPU Nuclear organization is less reactive to personnel i losses and can afford to move people to gain experience. (4)

GPU Nuclear can develop and use uniform policies between the plants on mat ters such as training, procurement and facilities.

(5) Because of its combined size and consolidated technical strength it can provide GPU corporate management with a much more professional assessment of matters which might affect reactor safety. (6) All the key technical positions within the GPU Nuclear Corporation are filled by nuclear-experienced personnel and their functions are not diluted with non-nuclear matters. (7) Th, person at the site responsible for the operation at TMI-l is a vice president of the GPU Nuclear Corporation and reports directly to the Office of the President of the corporation. He is not encumbered by organizational layers between himself and top management. (8) Those functions which need not be done at the site are performed off-site by personnel not reporting to the TMI-l unit Vice President. This provides the unit Vice President with more time which he can devote to matters directly related to the operation of the plant. (9) For all practical purposes, TMI-l and TMI-2 have been separated physically and organizationally. This is l

l important in that a separate group of capable people have been assigned to TMI-1, independent of TMI-2. (10) The new organization makes it very clear who is in overa11' charge of GPU nuclear matters. Wegner, ff. Tr. 13,284, at 10-11.
18. In addition to the organizational framework in which nuclear activities are conducted within GPU, there are i

two additional factors which bear upon our satisfactory resolution of CLI-80-5, Issue (1); namely, the qualifications of the team that will head up the GPU Nuclear Corporation (or Group), and the attitude displayed by key GPU management officials with respect to the importance they place on safely operating their nuclear facilities, including TMI Unit 1.

19. Mr. Robert C. Arnold, President of GPU Nuclear Corporation, received a Bachelor of Science degree in Science Engineering in 1959, and served for ten years in the U.S. Navy, six of which were spent working in the Navy Nuclear Power Program. Arnold, ff. Tr. 11,434, at 1. Mr. Arnold qualified as a reactor operator during that period of time, and held senior management positions aboard a nuclear powered ship. He left the Navy in 1969 and joined Met Ed. Since that time, Mr.

Arnold has assumed increasing levels of responsibility, including Vice President Generation at Met Ed and Vice President Generation at GPU Service Corporation. Id,. Thus, he h'as had extensive engineering and management experience in the commercial nuclear industry. Mr. Philip R. Clark received a B.C.E. degree in Civil Engineering in 1951. He also attended

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

g _

the Oak Ridge School of Reactor Technology from 1953 to 1954.

He was in the U.S. Navy's Nuclear Power Program from 1954 to 1964, at which time he became the Associato Director, Reactors, -

Naval Reactors Division, U.S. Department of Nnergy and Chief, Reactor Engineering Division, Nuclear Power Directorate, Naval Sea Systems Command, Department of the Navy. In these posi-tions, Mr. Clark was responsible for the direction of a major element of the U.S. Naval Nuclear Propulsion Program. He retired from the government in August of 1979, having spent his last 15 years of government service as an Associate Director in Admiral Rickover's organization, and joined GPU in January of 1980. Clark, ff. Tr. 6,225 (attached qualifications). In summary, Mr. Clark has had over 25 years of nuclear engineering and management experience in the Navy nuclear program.

20. The qualifications of Messrs. Arnold and Clark satisfy current NRC criteria. Tr. 11,995 (Crocker). Moreover, i

l in the words of one NRC staff witness, the folks that we see at  !

TMI -- Mr. Arnold, Mr. Clark, Mr. Hukill -- are fully as qualified as any we have found anywhere. Tr. 11,987 (Crocker).6 This fact weighed heavily in the testimony of Mr.

Lee of Duke Power, who felt that Mr. Arnold and Mr. Clark, 6 Mr. Hukill's qualifications are discussed in 1 36, infra.

Additionally, others of GPU Nuclear's tcp management whom the Board observed as witnesses and had an opportunity to question are discussed infra, including the heads of Technical l Functions (Mr. Wilson), Nuclear Assurance (Mr. Herbein), and Maintenance and Construction (Mr. Manganaro). i l

i

9 along with other management personnel including Mr. Herman Dieckamp, the current President of GPU Corporation who will serve as Chairman of the Board of the GPU Nuclear Corporation, possess technical qualifications, leadership qualities, and organizational skills which are outstanding. Lee, ff. Tr.

13,251, at 4, 12; Tr. 13,271 (Lee); Dieckamp, ff. Tr. 13,437, at 1. Mr. Wegner also commented on the caliber of upper management people. Tr. 13,303 (Wegner). Moreover, in Mr.

Lee's view, as well as Mr. Wegner's, the fact that Licensee's management demonstrated effective abilities to respond to a crisis environment with objectivity and calm is an added (positive) dimension to their qualifications. Lee, ff. Tr.

13,251, at 4; Tr. 13,273-74, 13,278 (Lee); Tr. 13,318 (Wegner).

The Board finds that the qualifications of Licensee's top management, particularly Messrs. Arnold and Clark, provide an added assurance that the command and administrative structure, as restructured into the GPU Nuclear Corporation (or Group),

l will function effectively.

21. Finally, the Board was particularly interested in the attitude of Licensee's management towards the responsi-l bility it assumes in operating a nuclear power plant such as TMI-1. While this factor is intangible, it is of crucial importance, especially if it is found to be lacking.
22. There were no negative comments received by the Board from the Staff, despite the Board's solicitation of such comments, with respect to the thoroughness, seriousness and

.. _. = - . _ - __

determination with which Licensee approached the restart of TMI-1. See Tr. 11,995-12,005 (Smith, Crocker, Keimig, Haverkamp). To the contrary, it was the view of IE, for example, that Licensee has made and continues to make a sincere effort to correct prior deficiencies in the operation of TMI-1, including corrective actions taken and those planned in response to Licensee and NRC reviews and investigations following the TMI-2 accident. Keimig, ff. 11,946, at 15. The presence of a responsible attitude within GPU management was confirmed by Mr. Wegner, who related his opinion of Licensee's attitude, af ter spending time talking with the top GPU manage-i ment.

t Part of our assessment was to spend time talking to the very highest levels within the corporation, which we did. In those discus-sions I was particularly interested in trying to find out if the commitment to doing things right was genuine or whether it was either forced because of the notoriety, and the pressures, and everything else -- you know,

{ it would be obvious at any stage of the game i

under these circumstances for the top people to say well, obviously we are going to do things right.

How genuine was the commitment? How deep was it? And [how] long would it last?

Weald it last a year from now, two years from now, and so forth.

l The conclusion that I came to af ter spending hours and -- many hours with these top people was that I could not detect that they were doing and they were reacting in the right way, because they, number one, were being forced to do it. I think that you have l a commitment in their own mind, and just as I you have indicated, it happened during their watch, and they are not going to walk away from it until it is fixed and it is working right.

Tr. 13,317-18 (Wegner).

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

. - . . - .= - . -

23. The Board's confidence in Licensee's reorganiza-tion stems in part from the testimony of Mr. Arnold, who assured us that first and foremost, GPU management has ar-ticulated and ccumunicated both directly and indirectly that I safety of operation at its nuclear power plants is an absolute requirement for effective operation. Arnold, ff. Tr. 11,434, I I

at 34. In addition, our sense of confidence 'is bolstered by the view shared by the Staff and Messrs. Lee and Wegner; namely, that tha Licensee has an appreciation of the commitment

necessary to operate TMI-l and consciously and deliberately made that commitment through the important step it took in making major changes in the manner in which its nuclear activities would be managed, including the command and adminis-trative structure responsible for those activities. On the basis of the strong endorsement of the new GPU Nuclear organi-zation by all or the witnesses as well as the Board's indepen-dent assessment of the reasons given for these endorsements, and the qualifications and attitude of GPU Nuclear's top l

managers, the Board finds that the command and administrative structure of GPU Nuclear Corporation (or GPU Nuclear Group), at both the plant and corporate levels, is appropriately organized to assure safe operation of Unit 1.

III. QUALITY ASSURANCE l

24. Although the issue of the sufficiency of Licensee's quality assurance (CA) program and organization is l

,yn,,_-_-v .....eu ,...~.,.,,,,,-,,..,y.,,.,,me,,,,.,e, -

r,-~_.,...,,....,_,,.,_r.,.._,., ,,__.,,,.,,,,,e-,o . . - ,

0 not separately identified by the Commission in its March 6, 1980 Order and there were no contentions on this specific

W subject, the issue of Licensee's operational quality assurance program was identified in the August 9,1979 Commission Order j and Notice of Hearing as a specific short-term action item to be resolved by Licensee prior to restart. The Board considers these matters to be subsidirry to the question of the adequacy of Licensee's command and administrative structure, at both the plant and corporate levels, to assure safe operation of Unit 1 (CLI-80-5, Issue (1)); consequently, we will address the issue of Licensee's QA program and organization at this juncture.
25. The rationale for the Board's association of QA i

with Licensee's command and administrative structure is based on the fundamental importance of quality assurance as the designated, institutionalized, internal check on the function-ing of the GPU Nuclear organization. Distinct from so-called safety reviews, quality assurance is that part of the formal structured organization which functions as a watch-dog for the system. Its purpose is to assure, on a day-to-day basis, that the system is working as designed, i e. , that the organizations .

which make up the plant and corporate structure are performing the functions for which they were intended and that this method of operation does in fact promote the safe operation of TMI-1.

Arnold, ff. Tr. 11,434, at 16, 33; Tr. 11,893-94 (Kazanas); Tr.

11,551-52 (Arnold).

26. The overall QA organization and staffing established for TMI-l has been restructured and improv ed since

,-ewwwwr-*--e---*= et----9--=cy w=--g-m -m , w w*---N-ws.m.-wme- r vww a w-e-w e gwm-w--N--+-ww--,.-ewem--ww-ww---mgsee----#pvg-7-e-y 9,- a-g c wwa sp w-a a 9m-wy--my--+w-**-m-+T-

'A , . - .*

c d the TMI-2 accident to include responsibility fu- implementation of the QA controls required in day-to-day activities and for _

involvement in the review and concurrence of procedures associated with these activities. The authority and respon-sibilities of the QA organization have been expanded to include all the important activities that occur during the operation of the plant. The QA staff has been increased in size and its qualifications relative to education and experience have been improved. Staff Ex. 1, at C6-8.

27. Licensee's Quality Assurance Department is one of four functional departments within GPU Nuclear's Assurance Division. Arnold, ff. Tr. 11,434, at 15. The three other departments are Training aUd Education, Nuclear Safety Assessment and Emergency Preparedness, each of which will be addressed by us in detail in later portions of these findings, i Id; Staff Ex. 4, at 5. The Director of the Nuclear Assurance Division is John Herbein. Arnold, ff. Tr. 11,434, at 15. Mr.

Herbein is a graduate of the Naval Academy with over twenty years of professional experience, fifteen years of which have been in nuclear power. Mr. Herbein's previous positions include TMI Unit 1 Engineering Supervisor, TMI Plant Superintendent, Manager of Nuclear Operations, and Vice l

President Generation at Met Ed. Because Mr. Herbein is intimately familiar with the TMI facility and individuals in the organization, he is particularly well qualified to head the Nuclear Assurance Division. Id.,at 15-16. In his testimony

before the Board, Mr. Herbein characterized the role of the Nuclear Assurance Division as a key one, particularly in light _

cf the lessons learned from the TMI-2 accident, such as the importance of training, quality assurance, and nuclear safety assessment. Tr. 11,907 (Herbein). It is Mr. Herbein's belief that the Nuclear Assurance Division, which provides technical capability in the home office as well as on-site at GPU's nuclear generating stations, can support the GPU Nuclear Corporation in these areas through its amassing within the Division of some seventy professionals with Bachelor of Science degrees in engineering, as well as a number of Masters and Ph.D. degrees. Id.; Staff Ex. 4, at 5.

28. Nuclear Assurance's Quality Assurance Department is under the direction of the Manager of Quality Assurance, Mr.

Nicholas Kazanas, who reports directly to Mr. Herbein. Arnold, ff. Tr. 11,434, at 16; Tr. 11,869 (Kazanas). However, in addition to this direct reporting chain, Mr. Kazanas has unencumbered access to Mr. Arnold or Mr. Clark, as well as to the Vice President, TMI-1, Mr. Henry Hukill. Arnold, ff. Tr.

11,434, at 16. Mr. Kazanas received an engineering degree in metallurgy in 1962, and has an M.S. in business administration.

He has been in commercial nuclear work since 1967. Mr. Kazanas joined GPU Service Corporation in 1978 as the Manager of QA.

In that capacity he worked primarily on construction activities at the Forked River project site until the accident at TMI-2.

l Tr. 11,871 (Kazanas).

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

. 1 o J l

29. It is the function of the Manager of Quality Assurance to evaluate the manner in which all activities important to safety, both on-site and off-site, are conducted with respect to quality, by means of review, audit, monitoring, and inspection. In order to facilitate this function s the Manager of QA, as well as the Director of the Nuclear Assurance Division, are independent of design, procurement, manufactur-ing, construction, operations and maintenance line responsi-bilities. Arnold, ff. Tr. 11,434, at 16. The Staff believes that the independence of QA from these line functions, combined with the reporting level of the Department, assures sufficient '

freedom from the pressures of cost and schedule to permit effective implementation of the QA program. Staff Ex. 1, at C6-8. The Manager of QA, through his staff, performs eval-uations on a planned and periodic basis to verify thst the QA Program is being effectively implemented. He identifies quality problems and initiates, recommends or provides solu-I tions through designated channels and verifies implementation of problem solutions. Id. The Manager of QA aas authority to issue stop work orders and to initiate through management unit j shutdown orders, in accordance with his assigned responsi-bilities and applicable QA procedures. Arnold, ff. Tr. 11,434, at 16-17.

l

30. The Quality Assurance Department is organized into six organizational units: Design and Procurement Assurance, including a Manufacturing Assurance sub-section, l

,.w-wev.,p-g-y- , , *-,-------yy,,.v w w+ ,ym,y.,g,y w ,_w g---,_,-gy y w._.,,e.c-rcwywac y , wee .yt we-- w w w w , r. , we ,,w-,g.w-w-mw,,.,+w _m-r .a

Modifications / Operations for TMI and a separate, parallel group i

for Oyster Creek in which Quality Control and Operational )

Quality Assurance work is conducted, Program Development and Audit, and Materials Technology. Id.,at 17-19, Figure 2; Tr.

11,881-84 (Kazanas); Staff Ex. 1, at C6-8 to C6-9. Th: total complement of QA personnel within GPU Nuclear, including Oyster Creek, is approximately 131. Tr. 11,884 (Kazanas). As of 4

February, 1981, approximately 65 to 70 QA personnel were assigned to TMI, 30 of whom were actively engaged in TMI-1 work. Licensee's 1981 budget provides for approximately 44 people working on the TMI-l QA program. Arnold, ff. Tr.

11,434, at 19. These figures contrast sharply with the 18 permanent QA personnel within Met Ed's total QA organization prior to the accident. According to Mr. Arnold, this increase can be attributed to the importance Licensee places upon the QA function, as evidenced by the fact that the scope of QA's responsibilities has been expanded to include systems and components having functions important to safety7 , ,, ,,11,,

the traditional QA scope of systems and components classified as safety related. Id,.,at 19-20; see also Staff Ex. 1, at C6-9, C6-12. In addition, the number of activities which have I

been classified as important to safety have been significantly l

7 It should be noted that Licensee's quality assurance program has been applied to systems and componenta beyond those considered "important to safety," as that term was used in the plant design and procedures phase of the hearing.

Conran, ff. Tr. 8372.

increased, and QA activities previously the responsibility of I l

other groups such as nondestructive examinations assaciated with inservice inspection, have now been added to the responsi- ~

bilities of the QA Department. Arnold, ff. Tr. 11,434, at 20.

In the Staff's opinion, the scope of the neM QA program represents a significant improvement relative to the acceptable scope in effect prior to the TMI-2 accident and to that of other operating reactors, and is acceptable for restart. Staff Ex. 1, at C6-7, C6-9, C6-12; Staff Ex. 14, at 23.

31. In general, it is the Staff'a view that Licensee's proposed QA program for TMI-l operations will satisfy the requirements of Appendix B to 10 CFR Pert 50 and provides QA controls improved significantly beyond those in use prior to the TMI-2 accident. The major areas where the QA controls have been improved, in addition to the expansion of the applicability of the QA program, are: greater involvement of the QA organization in the review and approval of quality-related aspects of procedures for operations, maintenance, inservice inspection, modifications and procurement; in the performance of inservice inspections, nondestructive examina-tions, routine inspections, verification, surveillance and audit activities; in the day-to-day operations and maintenance staff meetings to keep abreast of ongoing activities; in determining adequate close-out of corrective actions; and, better control of as-built drawings. Staff Ex. 1, at C6-7 to C6-8, C6-9 to C6-12.

l

32. The Board was able to discuss Licensee's QA program with the individuals having principal responsibility for QA activities associated with TMI-1, i.e., Mr. Herbein, Mr.

Kazanas, and Mr. Blaine Ballard, the Manager of TMI QA Modifications / Operations. Tr. 11,793-908 (Herbein, Kazanas, Ballard). On the basis of Mr. Arnold's detailed description of the QA organization and its areas of responsibility, the Staff's satisfaction with the organization, staffing and scope of QA activities, and the oral testimony by Messrs. Herbein, Kazanas and Ballard, the Board is satisfied that Licensee's QA organization and program will function as a watch-dog over the command and administrative structure at the plant and corporate level in that it will assure, or bring to the attention of top management in those cases where it cannot assure, that the organizations which make up the plant and corporate structure are performing the functions for which they were intended and that this method of operation does in fact promote the safe operation of TMI-1.

IV. TMI-l ORGANIZATION AND TECHNICAL RESOURCES

33. CLI-80-5, Issues (2) and (5) relate to the organization, technical resources, number of personnel and their qualifications available on site ta support the operation of TMI Unit 1. Issue (2) states:

Whether the operatigns and technical staff of Unit 1 is qualzfied to operate Unit 1 safely (the adequacy of the facility's maintenance program should be among the matters considered by the Board).

Issue (5) states:

Whether the Unit 1 Radiation Waste System is appropriately staffed with quali-  !

fied individuals to ensure the safe operation l of the facility. ,

i Under the general category of TMI-1 Organization and Technical Resources, the Board will consider ( A) the on-site organization dedicated to TMI-1, including its structure, staffing levels, technical resources, changes in the on-site organization since the TMI-2 accident and the relationship between on-site GPU Nuclear personnel and their technical counterparts located off-site; (B) TMIA Contention 5, which relates to the adequacy of the conduct of maintenance activities at TMI-1; (C) the GPU Nuclear Training and Education Department, with particular emphasis on the on-site TMI Training organization and its programs; and (D) Aamodt Contention 2, which challenges the i adequacy of the training given to TMI-l's licensed operators.

i j A. TMI-l On-Site Organization

34. Licensee's on-site organization, responsible for the safe operation and maintenance of TMI-l on a day-to-day basis, has been significantly reorganized and strengthened I

since the accident at TMI-2. In general, the major changes that have taken place in the on-site organization have been isolating the management and technical support of TMI-1 from TMI-2 decontamination activities; significantly reducing the

! responsibilities of lead TMI-1 management in order to allow

_ _..,,,._,,m_ . - - - - - " - ' - " ' " " " " ' * "~

these individuals to devote their full attention to Unit 1; and restructuring the TMI-l organization so that effective control over important unit activities and decisions is maintained by TMI-l on-site management while, at the same time, direct channels of communication exist between on and off-site technical and management personnel. Hukill et al., ff. Tr.

11,617, at 2-3. These basic changes have been endorsed by the Staff, as well as by Mr. Wegner and his associates in their management review of TMI-l. Staff Ex. 4, at 8, 10, 13; Staff Ex. 14, Table B-1, at 8; Keimig, ff. Tr.11,946, at 14; Tr.

11,995, 11,981 (Crocker); Wegner, ff. Tr. 13,284, at 12-19. A description of the TMI-l on-site organization, as currently constituted, follows.

35. The Vice-President of TMI-l is the senior member of management within the TMI-l unit organization. He is located full-time on Three Mile Island; however, he reports to the Office of the President of GPU Nuclear, located in Parsippany, New Jersey. Hukill et al., ff. Tr. 11,616, at 4.

The sole responsibility of the Vice-President of TMI-l is TMI-1. Staff Ex. 4, at 8. Moreover, he has been delegated absolutely minimum responsibilities not directly associated with the operation and maintenance of the unit in order that he may devote his full time and attention to ensuring that in all respects, TMI-l is being operated and maintained safely, in accordance with the plant's Technical Specifications and other applicable criteria. Hukill et al., ff. Tr. 11,617, at 4. In

1 order to meet this responsibility, the Vice-Preside',st of TMI-l utilizes the services of, and is the senior liason with the l engineering, design and analysis, nuclear assurance (which includes training and emergency preparedness), maintenance and l construction, radiological and environmental controls and administrative services available to TMI-l from the support divisions of the GPU Nuclear Corporation. Id.,at 5.

36. The Vice-President of TMI-l is Henry D. Bukill, j an individual with extensive involvement in power reactor operations. Tr. 11,522 (Arnold); Wegner, ff. Tr. 13,284, at
16. Mr. Hukill joined GPU as the Prospective Director, TMI-l

, in June 1980, and formally began serving as Director 0 on i

September 8, 1980. Bukill et al., ff. Tr. 11,617, at 5.

(During the interim period of time, Mr. Hukill worked full-time r.t TMI-l reviewing plant documents such as the B&W instruction i

books, and familiarizing himself with the plant's systems. Tr.

11,620 (Hukill).) Mr. Bukill received a Bachelor of Science degree from the U.S. Naval Academy in 1953, and served on i

active duty in the U.S. Navy for more than 22 years, working primarily in the construction, maintenance and operation of nuclear submarines. Mr. Hukill has qualified as a reactor operator, in addition to holding a number of senior level l

8 Operating as the GPU Nuclear Group, Mr. Bukill's title is Director; with implementation of GPU Nuclear Corporation, his title will be Vice President; however, his responsibilities will be the same. See 113, supra and footnote 9, infra.

I 4

-,-w, e .,.w---.-.w.,.y , . . . . - , -s-

positions aboard nuclear submarines, including commander.

During his last four years on active duty, Mr. Hukill was assigned as a special Assiatant and Senior Line Officar on the ~

Staff of the Director, Division of Naval Reactors. In this capacity, he was directly responsible for the selection and engineering training of all nuclear ship Commanding Officers,

( and was also directly involved in the establishment and enforcement of standards and procedures for the safe and proper operation of all naval nuclear propulsion plants. After leaving the Navy, Mr. Hukill served as the Project Operations Manager, Clinch River Breeder Reactor Plant Project for Burns and Roe (1976), and as a Senior Civilisn Special Assistant to the Commander, Naval Sea Systems Command (1977-1980). While serving in this latter capacity, Mr. Hukill was responsible for all matters related to the selection, education, qualification, training and professional performance of the Navy's more than 1,200 Engineering Duty Officers. Hukill et al., ff. Tr.

11,617, at 5-7. In summary, Mr. Hukill has had over 22 years l

of nuclear power plant experience, the majority of which has been in the Navy Nuclear Program in line operations and management jobs, and in senior level management positions. The Staf f is more than satisfied with Mr. Hukill's credentials, Tr.

11,987 (Crocker), as are we.

I i 37. Reporting directly to the Vice P esident of i

TMI-l is the Manager of Administration, the Operations and l

Maintenance (O&M) Director and the Plant Engineering Director.

l l

_ . _ _ -_. __ _., ____ _._ _ - _ _ _ _ _ _ _ . - _ _ . _ _ . . _ . . . _ _ _ _ _ _ _ _ _ _ _ . _ , _ _ - n- -.. _ _ .

o Hukill et al., ff. Tr. 11,617, at 7, and chart following prepared testimony.9 The Manager of Administration, an individual with a civil engineering degree and some twenty years in the electric industry, manages the administrative activities of TMI-1, functions as a staff assistant to the Vice l President of TMI-1, and coordinates the administrative work and needs of TMI-l with the activities of GPU Nuclear's Division of Administration. Id., at 38-40. The OEM and Plant Engineering Directors have primary responsibility for Unit l's operational, maintenance and associated engineering activities, i.e., they i

manage the people with key functional responsibility for the safe operation of the unit. Id., at 7.

38. As a result of the orgaaizational changes instituted by Licensee since the TMI-2 accident, the responsi-

, bilities of the OEM Director have significantly decreased in i

scope. Id., at 9. This change is representative of Licensee's i

effort on a number of fronts to substantially increase the depth of understanding and involvement of GPU Nuclear managers in the technical issues with which individuals reporting to them are dealing. This goal is accomplished by. limiting the number of discrete areas of responsibility assigned to a 9 Because of recent title changes which Licensee adopted in moving from the GPU Nuclear Group to the GPU Nuclear Corpora-tion, it was necessary to clarify the prefiled testimony of Hukill et al., ff. Tr. 11,617. The new titles of key TMI-l employees are listed in a chart, provided as supplemental direct testimony, along with an organizational chart following the Hukill panel testimony. Tr. 11,616-17 (Hukill).

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

manager, and by eliminating functions for which his technical and managerial capabilities are not necessary, e.g., purely ,

administrative work. See, e.g., Tr. 11,452 (Arnold); Tr.

11,706-07 (Ross); Keinig, ff. Tr. 11,946, at 11; Hukill, ff.

Tr. 11,617, at 15. The OEM Director is no longer responsible for the unit's engineering, radiation protection, water chemistry, and administrative activities. Insteed, he can focus his attention on plant operations, and on the facility's preventive and corrective maintenance programs. Hukill et al.,

ff. Tr. 11,617, at 9. Essentially, it is the O&M Director's job to see to it that any and all operational problems which arise while the plant is operating or during an outage are properly diagnosed, so that appropriate action can be taken.

It is also his responsibility to effectively coordinate the activities of the Operations and Maintenance Departments, with a staff of approximately 260, and, in particular, to ensure that Maintenance personnel are responsive to the needs of Operations. Id., at 8. The current O&M Director, Mr. Ronald J. Toole, has diversifiad power plant experience, having worked extensively in both line engineering and management positions in nuclear power plants (including Shift Test Director at TMI-l during its low power physics and power escalation programs), and in coal facilities. Id., at 9-11; Wegne r , ff. Tr. 13,284, at 16. Reporting directly to the O&M Director are the Manager of Plant Operations and the Manager of Plant Maintenance. Hukill et al., ff. Tr. 11,617, at 11, 30.

9

- - - , , . . _ . . . , - , ,,...---,,.._-,,,n,----,,

...,,-e,, .,..,.. ,...n ._,..,w,.--. .,-n,--n,, ,_-,..,,...,,.--,----,,,--,n, . - , . -

o 4

39. The Manager of Plant Operations is responsible i

for the day-to-day operations of TMI-1; conversely, he has no

~

responsibilities that do not directly affect the daily opera-

, tic.: of TMI-1. Staff Ex. 4, at 8; Hukill et al., ff. Tr.

11,617, at 11. He has approximately 110 personnel assigned to him, 75 of whom are assigned shif t personnel. Staff Ex. 4, at 8; Tr. 11,937-38 (Crocker). On a daily basis, the Manager of Plant Operations reviews and schedules all routine and non-

routine operations; is in charge of requesting operations-related maintenance work from the Maintenance Department; reviews and writes operating procedures; and is available to -

consult with his staff with respect to p'lant operations.

Hukill et al., ff. Tr. 11,617, at 11. The current Manager of Plant Operations is Michael J. Ross. Prior to assuming this responsibility, Mr. Ross worked as a TMI-l shift supervisor and shift foreman, a member of the Operations staff and an operator i

! instructor at the Saxton Nuclear Experimental Corporation, and as a reactor operator and Engineering Officer in the U.S.

Nuclear Navy Program. Mr. Ross holds a current senior reactor operator license on TMI-1. Id., at 12; Tr. 11,673 (Toole).

40. The Operations staff, under the direction of the Unit 1 Manager of Plant Operations, is divided into three categories: the shif t operating staff, the radwaste group, and several operations engineers. In addition, working in conjunc-tion with the control room operating personnel are the Shif t Technical Advisors (STAS) who represent the input of and report

N, e

. . **~

o 1 1

i to the Technical Functions Division of GPU Nuclear. Hukill et al., ff. Tr. 11,611, at 12.

41. The shift operating staff control the reactor primary and secondary systems as well as associated plant systems and equipment during normal operations and plant shut downs, in response to reactor transients, and when emergencies unrelated to the operation of the facility are experienced on site, such as fires or personnel injuries. Hukill et al., ff.

Tr. 11,617, at 13. Licensee's licensed shift operating staff is composed of six shift supervisors, seven shift foremen (three of whom are candidate senior reactor operators (SROs),

i.e., they have not yet been licensri as SROs by the NRC), and about twenty control room operators (ten of whom have reactor operator (RO) licenses, although they will be retested along with the candidate ROs prior to restart of Unit 1), along with thirty-six (non-licensed) on shift auxiliary opc:ators. Staff Ex. 13, at 4; Tr.11,666-69 (Hukill); Hukill et al., ff. Tr.

11,617, at 14-24. The Staff considers this manning level to be sufficient for the long term. Staff Ex. 13, at 5. In addi-tion, although the Staff originally required Licensee to have two SROs and two ROs per shif t at the time of TMI-l's restart (consistent with the treatment rf.ven to a new plant applying for an operating license), in light of the Commission's belief "that Unit One should be grouped with reactors which have received operating licenses, rather than with units with pending operating license applications," CLI-81-3, at 7 (March i

l

. , , . . - _ - - - - - - - - - - - - - - - - - - - - - - - ~ ~ - - - - ~ ~ - - * - - - - - ~ ~ ~ ' ~ ~ " ~ ^ * ^ " ~ ' " ~ ~ ~ ~ ~ ~ ~ ~ ' ' ' ' ~ ~ ~ ~ ^ ' ~ ~ ~

l 23, 1981), the Staff has am9nded its earlier requirement.

Staff Ex. 14, at 22. Since the Staff finds nothing unique about TMI-l that requires a second SRO to be assigned on shif t i at restart, and the Staff feels assured that the capability of the TMI-l operating staff will be satisfactory with trained i STAS and highly experienced SROs and ROs, Licensee is required to have one licensed SRO and two licensed ROs per shif t at I

restart. Id., at 22-23. The Board agrees with the Staff's position, particularly in view of the fact that Licensee is required to meet the additional staffing requirement of a second SRO on shift on the same schedule as other operating plants, as set forth in NUREG-0737, i.e., July, 1982. Id.; Tr.

11,668 (Hukill).

42. In addition to considering the manning levels (per shif t and in total) of shift operating personnel and the professional backgrounds of these individuals, the Board has taken into account in its review of Licensee's operating staff for TMI-1 Unit 1 the training, both initial and requalifica-tion, which these individuals receive, see 11 123 to 151,
infra, and the method used by Licensee to assure that each l shif t properly turns over its control of the plant to the l

l incoming shift. Hukill et al., ff. Tr. 11,617, at 13-14; Staff Ex. 4, at 12. The Board concurs with the Staff's finding I

i that the shift relief turnover checklists and other management controls for plant operational activities which the Licensee has adopted since the TMI-2 accident provide improved manage-ment information on plant status and better control over plant operational activities. Staff Ex. 4, at 12.

}

l ( - -

J -

43. Another change and improvement in the quality of Licensee's shif t operating staff is the adcition of an STA, or shilt technical advisor, to each of the six rotating shif ts.

The position of STA was instituted at TMI-l in response to the accident at TMI-2, and was subsequently adopted as one of the recommendations of the NRC Lessons Learned Task Force, l

NUREG-0578 (1979). Hukill et al., ff. Tr. 11,617, at 28. The STA is a degreed cagineer who is assigned on shif t with the primary duty of assessing the impact which various plant operations may have on nuclear and environmental safety.

During accident or off-normal conditions, the STA's specific duties include recognizing and diagnosing unusual reactor and instrument responses. During normal operating conditions, the STA's duties include the review and evaluation of plant performance, and of the adequacy of procedures used to assess that performance. Thus, the STA monitors and provides direct technical input to the on-going activities in the TMI-l plant.

Because an STA must have a Bachelor of Science or Engineering degree, he provides additional analytical and technical capability to support the operator on an around-the-clock basis. The STA can, for example, analyze conditions in the core in the event of a transient. This analytical capability heretofore has not necessarily been present. At TMI-1, on an ongoing basis, the STA evaluates the need for and recommends corrective action on safety componcats and systems; advises the shift foreman or shift supervisor, as needed; and provides a

(

l

technical liason with the GPU Nuclear Technical Functions engineers. Id., at 28-29. While the Board expressed an initial skepticism about the willingness of the regular shif t crew to accept the advice of an individual who has a college ,

l degree, but who has not been licensed on the plant, these doubts were put to rest by Mr. Michael Ross, Manager of Plant Operations, TMI-1, who stated unequivocally that the STAS are accepted and, in fact, are looked to for advice by the rest of the shift crew. Tr. 11,645-46 (Ross); see also Tr. 11,642-45 (Hukill). Moreover, Messrs. Hukill, Toole and Ross, i.e., the Vice-President of TMI-1, the O&M Director and the Manager of Plant Operations, agree that the concept of the shif t technical advisor as an adjunct to the regular operating crew is a good idea. Tr. 11,646-47 (Hukill, Toole, Ross). Similarly, the Staf f concluded that STAS having the qualifications required by the Licensee will be able to provide an added measure of technical expertise on-site. Staff Ex. 14, at 46. Moreover, Licensee's STA program meets Staff requirements. Id,. The Staff's review also indicates that the initial incumbents of I

Licensee's STA program meet the qualification requirements as stated in Licensee's description of its program, and that Licensee is in full compliance with all NRC requirements regarding STAS as set forth in NUREG-0737. Id., at 46-47. The Board, too, is satisfied with Licensee's STA program, including the implementation thereof.
44. Before discussing the other personnel who work for the Manager of Plant Operations, it is noteworthy that a

I number of individuals in the operational chain of command have the independent authority and responsibility to shut down and cool down the reactor if it is necessary to do so in the interests of health and safety or if, in his judgment, such action is otherwise warranted. These individuals include the Vice-President of TMI-1, the OEM Director, the Manager of Plant operations, the shift supervisors, the shift foremen and the control room operators. Bukill et al., ff. Tr. 11,617, at 5, 8, 11, 15, 18, 22; Tr. 11,629, 11,635 (Hukill). This unila-teral authority is necessary in order for these individuals to properly carry out their duties. Tr. 11,630 (Hukill). Of course, guidelines do exist to direct the control room opera-tors. Id. (Ross). (Since an unlicensed operator cannot take any action in the control room that is not supervised by a licensed operator, this policy does not apply to STAS, ROs in training, or auxiliary operators. Tr. 11,632 (Ross).)

45. In addition to managing the shif t operating staff, the Manager of Plant Operations is in charge of the TMI-l Radwaste group, which carries out the daily radioactive waste activities at the facility. Hukill et al., ff. Tr.

11,617, at 24. It is the job of the Radwaste group to collect, decontaminate, package, prepare to ship or otherwise properly dispose of materials, liquid and solid, which exceed a specified level of radioactive material. The TMI-l Radwaste organization has changed significantly since March of 1979.

Prior to the TMI-2 accident, radwaste activities were a part of

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

the TMI health physics program; consequently, there was no staff specifically allocated with the responsibility of decontamination, packaging, preparation for shipping and

~

minimizing the quantity of radwaste at Unit 1. This is no longer the case. Not only are the individuals who work for the Supervisor of Ra(waste dedicated solely and on a full-time basis to Unit 1 activities, but they are also designated TMI-l Radwaste staff. Id., at 24-25. In addition to strengthening the resources dedicated to radwaste, this organizational change improves the control of radwaste because radiological control personnel review the radiological aspects of radwaste processing. Wegner, ff. Tr. 13,284, at 28.

46. The Supervisor of Radwaste, a highly qualified radwaste engineer with a B.S. degree in chemical engineering, directs the activities of the 24 individuals reporting to him.

{ Hukill et al., ff. Tr. 11,617, at 25, 26. One of the goals of the Supervisor of Radwaste is to develop, through experience, a rotating decomtamination system which will ensure that the protected and vital areas at TMI-1 are maintained in as clean and radioactively-free an environment as possible. The Supervisor of Radwaste meets several times a week with the Manager of Operations, coordinating the activities of radwaste personnel with the needs of the operating and maintenance staff. Id., at 25. The three Radwaste Foremen direct laborers, utility workers, and, at times, auxiliary operators in the performance of radioactive waste functions. Staff Ex.

l 4, at 25; Hukill et al., ff. Tr. 11,617, at 26. After packaging of waste for shipment, all Unit 1 and Unit 2 ship-1 ments are reviewed and approved by the Unit 2 radwaste process l t

support group. Staff Ex. 4, at 25. Packaging of Unit 1 radioactive waste material requiring a licensed container is the responsibility of the Unit 1 radwaste process support group. A qualified Unit 2 solid waste and disposal-supervisor is responsible for all radioactive waste being shipped off site in accordance with applicable requirements. Id. In addition to having foremen to direct radwaste activities, the Supervisor of Radwaste is assisted by a radwaste engineer, who writes procedures and trouble-shoots the system when problems arise which need immediate evaluation. Hukill et al.,.ff, Tr.

11,617, at 26 ; Wegne r , f f . Tr . 13,284, at 28. The routine day-to-day operation of the waste systems is performed by auxiliary operators who report to the Operations Shif t Foreman i

(on duty), who in turn reports to the Operations Shift Supervisor. Staff Ex. 4, at 25.

47. In its reviews of Licensee's radioactive waste program and organization, the Staff has concluded that (1) the Unit 1 organization is appropriately organized and staffed with qualified personnel in accordance with NRC guidelines (NUREG-0731, Regulatory Guide 1.8) and ANSI standards (ANSI /18.1-1971); Staff Ex. 4, at 26; (2) that the interface 1

between the Unit 1 and 2 radioactive waste organizations either has been or will be satisfactorily documented prior to restart;

Staff Ex. 13, at 8; and (3) that Licensee has submitted to the NRC an acceptable plan to expand existing low activity solid waste storage capability; NRC Staff Ex.14, at 20. In addi-tion, the available radwaste facilities have been improved by making modifications to the radwaste evaporator, and an extensive program has been started to reduce the amount of solid radioactive waste. Wegner, ff. Tr. 13,284, at 29. Based on these findings, the Board is satisfied with Licensee's radioactive waste program and organization. (We will discuss Licensee's technician training program later in these findings.

See 11 152 to 163, infra.)

48. In addition to the shif t operating staff and the radwaste group, the Manager of Plant Operations, TMI-1, is also in charge of the activities of several operating engineers from whom he can obtain immediate and short-term engineering work.

Bukill et al., ff. Tr. 11,617, at 26-27. These personnel assist the Manager of Plant Operations in writing operating procedures, reviewing these procedures for their effectiveness, and otherwise providing additional support for operations-related engineering problems. The availability of operating engineers within the Operations staff provides added depth to the station organization in that three levels of technical support -- operating engineers, TMI-l Plant Engineering Department and Technical Functions -- are now available to plant operations during normal plant operations, in addition to the numerous other sources of technical information available in the u+ent of a transient. Id., at 27; Tr. 11,648-50 (Hukill, Ross, Colitz).

49. The Maintenance Department is the other major station organization which reports directly to the OEM ,

Director. Hukill et al., ff. Tr. 11,617, at 30. The organiza-tion and practices of the TMI-l Maintenance Department have changed considerably since March of 1979. Shovlin et al., ff.

Tr. 13,533, at 1. In general, the scope of responsibility of key individuals has been narrowed to provide for a more intense focus on the various aspects of the maintenance within one nuclear power plant unit. Three major areas of responsibility have been identified as: Preventive Maintenance and Technical Specifica' tion Surveillances; Corrective Maintenance; and Shift Maintenance. The area of responsibility for the Manager of Plant Maintenance prior to March of 1979 was both TMI-1 and TMI-2. With the separation of the units, maintenance respons$-

bilities, too, have been separated; the Manager of Plant Maintenance at each unit has responsibility for maintenance for his unit only. Id., at 1-2; Wegner , ff. Tr. 13,284, at 16-18.

Finally, consistent with Licensee's policy of increasing and concentrating the technical resources and management strength being applied to its nuclear activities, Licensee has estab-lished a Maintenance and Construction Division of GPU Nuclear Corporation, headed by a Vice President, with the responsi-bility of (1) establishing and monitoring uniform policies, l practices and procedures for all maintenance, repair and construction activities at GPU's nuclear plants, and (2) carry-ing out assigned plant modifications, repair and construction

I I

activities and conducting major and specialized maintenance work, in accordance with corporate policies and all applicable laws, regulations, license and technical requirements.

Manganaro, ff. Tr. 13,643, at 1-2; Wegner, ff. Tr. 13,284, at  !

18-19.

50. The Manager of Plant Maintenance, in coordi-nation with the Manager of Plant Operations, is in charge of planning, organizing, integrating and directing the daily maintenance effort that takes place at Unit 1. Hukill et al.,

ff. Tr. 11,617, at 30; Shovlin et al., ff. Tr. 13,533, at 2.

It is the responsibility of the Manager of Plant Maintenance to coordinate preventive maintenance, and to direct the diagnosis and repair of all equipment that Operations has identified as in disrepair, requiring component replacement or in need of other corrective maintenance work. Hukill et al., ff. Tr.

11,617, at 30-31. In addition to the two major maintenance staffs, corrective and preventive, reporting to the Manager of Plant Maintenance there are 35 utility workers, who primarily perform a housekeeping function, a welding foreman, and a senior technical analyst responsible for maintenance work associated with the TMI-1 security and communications systems.

Id., at 38. In total, there are approximately 150 employees assigned to the Manager of Plant Maintenance. The Manager of Plant Maintenance obtains technical support from the TMI-1 Plant Engineering staff in carrying out his responsibilities.

In the event that the necessary work appears to require a great

- ~ - - - - . - - - , . - - . -,._e....n,c.--rm,.yw w ,c-y..$_-,m--- , - . . - . - - . . . - . -.-_--__..,,-.--.,.,,--...----.m.- . .

o deal of manhours or technical analysis, the Manager of Plant ,

1 Maintenance through the Plant Engineering Department calls upon l l

the technical resources available from the Technical Functions Division of GPU Nuclear Corporation. Major plant maintenance and construction activities are assigned to the Maintenance and Construction Division of GPU Nuclear. It is the responsibility of the Manager of Plant Maintenance to oversee all maintenance activity at TMI-1. Through this centralized organization, TMI-1 management coordinates and scrutinizes all on-site maintenance activities. Id., at 30-31.

51. The current Manager of Plant Maintenance, Mr.

Daniel M. Shoviin, is a Navy veteran of twenty-seven years, during which time he assumed major responsibility for main-tenance of several large combatant surface ships. Mr. Shovlin has worked at TMI since 1973 as Unit 1 Supervisor of Maintenance. Since that tim *, he has served as the Supervisor of Maintenance at TMI Unit 2, and then, in 1978, as the Superintendent of Maintenance responsible for all maintenance i

activities on Three Mile Island. He assumed his present responsibilities when the TMI-1 and TMI-2 organizations and units were formally separated in November,1979. Hukill et l al., ff. Tr. 11,617, at 32-33; Shovlin et al., ff. Tr. 13,533, at 3.

52. Maintenance at TMI-l is divided into two, entirely separate organizations: the preventive and the corrective maintenance groups. Hukill et al., ff. Tr. 11,617,

,= , 4 * . .* .l c 1 1

l at 33. The preventive maintenance (PM) group conducts the preventive maintenance program, which is a program of regular inspections and other preventive maintenance work on TMI-l systems, particularly those related to the facility's safety and reliability, in order to decrease the likelihood of equipment experiencing failure during operation. Shovlin et al., ff. Tr. 13,533, at 5-6; Hukill et al., ff. Tr. 11,617, at

34. Included in the PM group are representatives from the electrical, instrumentation and control, mechanical and utility disciplines. There are about 25 people assigned to preventive maintenance, with eight of these employees on rotating shif ts.

Through its preventive maintenance program and staff of 25 dedicated PM employees, it is the goal of Licensee to assure reliable performance of equipment and to reduce to an absolute minimum the amount of corrective . maintenance work required at TMI-1. Id.,at 4-5. It is Licensee's view that an established PM program such as the one now in place at TMI-l promotes safety while optimizing equipment availability and reliability.

Hukill et al., ff. Tr. 11,617, at 33. We agree.

53. In addition to the maintenance staff working full-time on preventive maintenance activities, Licensee's TMI-l Maintenance Department includes a group of about 94 workers under the direction of the Corrective Maintenance (CM)

Manager. On a daily basis, the CM Manager plans, organizes, and directs corrective maintenance work at Unit 1. Id., at 34.

In fulfilling this responsibility, he is assisted by a l

t_-_.-,_-._..-

Supervisor of Management Control and several other planners, who plan major maintenance tasks, develop schedules for the _

accomplishment of tasks within specified time frames, such as a refueling outage, and otherwise assist in the planning of the day-to-day corrective maintenance work at TMI-1. Shovlin et al., ff. Tr. 13,533, at 9-10. Reporting to the CM Manager are Lead Foremen in the disciplines of instrumentation and control-(I&C), mechanical and electrical. Hukill et al., ff. Tr.

11,617, at 35. Each of these Lead Foremen is responsible for the activities of the foremen and the 24-hour shif ts of maintenance workers in his respective disciplines. This i

responsibility encompasses all discipline activities related to the planning, organizing, and directing of day-to-day maintenance taking place at TMI-1. All work performed in the unit must be cleared with the Operations staff in order to ensure that it does not interfere with ongoing operational activities. Also, all corrective maintenance work must be performed in accordance with the Operational Quality Assurance Plan, and where necessary, with Radiological and Environmental Controls supervision. Id., at 35-36.

54. In addition to the preventive and corrective maintenance personnel employed on site during the day, TMI-1 has a shift maintenance work force composed of six rotating sections, each comprised of a minimum of two men from each of the following disciplines: electrical, I&C, mechanical and l utility. There are approximately ten men per shift section.

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

. - e?

o /

Shift maintenance works on corrective and preventive maintenance items that can be completed during an eight hour shift. Each section is headed by a maintenance foreman, who reports to the Lead Shif t Maintenance Supervisor. Shovlin et al., ff. Tr. 13,533, at 7.

55. Licensee has described in detail, and the Staff has approved, the qualifications of Licensee's maintenance managers including the Manager of Plant Maintenance, the CM Manager, the PM Manager , the Lead Shif t Maintenance Supervisor, and the Supervisor of Management Controls. Shovlin et al., ff.

Tr. 13,533 at 3, 5, 7, 8-9, 10; Hukill et al., ff. Tr. 11,617, j at 32-33, 35, 37; Staff Ex. 4, at 8-10, 13-14; Keimig, ff. Tr.

11,946, at 11-12, 13. These individuals have had either extensive Navy training in maintenance and operations, and/or extensive experience as maintenance workers and supervisors at TMI-1. Id.

56. The conduct of maintenance activities at TMI-l is discussed in detail in 1561 to 115, infra, in connection with TMIA Contention 5 on that subject. Briefly, Licensee utilizes GPU's central data processing equipment in order to record, plan, track, close out and maintain history with I

respect to all corrective and preventive maintenance work.

This computerized system has many advantages over the previous, entirely manual, system of conducting maintenance work. In particular, through this centralized computer bank of informa-tion, Maintenance now has the ability to easily prioritize t

l l

assig nments , to track and close out backlogged maintenance work, and to follow the current status of outstanding jobs.

Shovlin et al., ff. Tr. 13,533, at 12-21.

57. Finally, in December, 1980 Licensee formally established its Maintenance and Construction Division, under the direction of Mr. Francis F. Manganaro as its Vice President. Manganaro, ff. Tr. 13,643, at 1, 6. Mr. Manganaro joined GPU in June 1980, af ter spending the previous thirty-three years in the U.S. Navy. During twenty one of those years, Mr. Manganaro was assigned to various positions involved in the design, construction, overhaul, refueling, conversion and maintenance of nuclear powered surface ships and subma-rines. Id.; Arnold, ff. Tr. 11,434, at 27; Tr. 11,524-26 (Arnold); Wegner, ff. Tr. 13,284, at 18. In general, it is intended that the Maintenance and Construction Division provide direction and support in the functional areas of maintenance, repair and construction to permit maxi;num concentration of attention and resources by plant management on safe and efficient operation and maintenance of the GPU nuclear generating stations. Manganaro, ff. Tr. 13,643, at 2; Staff Ex. 4, at 7; Wegner, ff. Tr. 13,284, at 19-20. The Manager of Plant Maintenance and his staff will remain responsible for assuring the material conditic: of the plant. Manganaro, ff.

Tr. 13,643, at 2. The on-site Maintenance organizations will I

continue to carry out some of this work themselves using their Maintenance Department, but they may also identify work they

, , ,- - y ,-an,--,---,-e-- ---w,,7,. v.v, --,---~-w,,-.-m-~n- .,,.w-,---r-, ,,_s-. ,,. --.,,-rw- -r--- --,e,-.-we----,m~,,--e, -,++%e --ee.w,ww----e=*-

cannot or do not wish to do, e.g., because of a lack of technical resources. Id,.; Tr. 13,645 (Manganaro). Like an outside consultant, the Maintenance and Construction Division will never perform maintenance work at TMI-l without the approval of the O&M Director. Tr. 13,648 (Manganaro). The Maintenance and Construction Division will have a resident Director as well as functional managers located at TMI and at Oyster Creek. Manganaro, ff. Tr. 13,643, at 2-3. These individuals will coordinate their activities with their functional managers, administrative support, and Vice President located at Licensee's corporate headquarters. Id., at 3-6.

58. In addition to the supervision of activities for which the Operations and Maintenance Director is responsible, the Vice President of TMI-l oversees the activities of a large TMI-l Plant Engineering Department, under the direction of the Director, Plant Er.gineering. Hukill et al., ff. Tr. 11,617, at 40; Keimig, ff. Tr. 11,946, at 11. The TMI-l Plant Enginaering group consists of 45 individuals who provide the on-site technical capability to support the day-to-day safe operation and maintenance of the generating facility. Hukill et al., ff.

Tr. 11,617, at 40; NRC Staff Ex. 4, at 8. This support covers l

the electrical, mechanical, nuclear and I&C engineering disciplines, plant chemistry, and fire protection. These rather diverse activities are managed by the Director of Plant Engineering, who works closely with the O&M Director in order to ensure that appropriate priorities are maintained in those

l areas where plant Operations or Maintenance require technical support from the Plant Engineering staff. In addition, the I

Director of Plant Engineering works with the O&M Director and i his staff in preparing operating and emergency procedures; ensuring that the Technical Specification requirements are met; providing engineering and other technical support to on-going preventive and corrective maintenance work; reviewing and evaluating changes in plant design or procedures; and support-ing refueling outage activities. The Director of Plant Engineering is the major TMI-l liason to the Technical Functions Division of GPU Nuclear Corporation. Bukill et al.,

ff. Tr. 11,617, at 40-41; Staff Ex. 4, at 8.

59. The present Director of Plant Engineering is Mr.

Joseph J. Colitz, an individual with a B.S. degree in mechanical engineering, extensive experience as a practicing engineer at both fossil and nuclear power plants, a former TMI-l Superintendent SRO-licensed on TMI Unit 1 and clearly, from the Board's first-hand experience, an individual with a detailed working knowledge of the technical details of the plant and its systems. Hukill et al., ff. Tr. 11,617, at 41-42; Tr. 3,115-16 (Jordan). The TMI-l Plant Engineering Staff is composed of lead engineers and supporting engineers in the mechanical, nuclear, electrical and I&C engineering disciplines. Hukill et al., ff. Tr. 11,617, at 43. All four of the lead engineers at TMI-l have Bachelor of Science or Engineering degrees or better, and from four to eleven years of l

l

___.____.__,___.__,_._m,_,_.-_,__,,,,_,,,,,,._.,,_._.,.-..-.,,,.___.,_.....__.,__,-,__,,,_,,,_..._.,...m.. _ . - _

experience in their respective disciplines. Id., at 43-48. In overseeing the activities of the TMI fire protection engineers, the Director of Plant Engineering is assisted by a senior TMI l fire protection engineer who is currently pursuing his i Associates degree in fire technology af ter having ' qualified as a U.S. Navy S3G prototype mechanical operator and engineering laboratory technician in the U.S. Navy Nuclear Power Program.

The fire protection engineers are responsible for the overall readiness of all fire service and fire protection systems at Three Mile Island. Id., at 48-49. The TMI-l Chemistry Department, under the direction of the Supervisor of Chemistry, also reports to the Director of Plant Engineering. This Department conducts all TMI-l water chemistry-related work, including sampling and laboratory analysis on the primary and secondary systems of the TMI-l reactor in order to ensure that the water chemistry meets plant Technical Specifications, manufacturer specifications, and discharge limits. It is also the responsibility of this group to provide technical super-vision and assistance in the operation of the water treatment, chemical addition, and waste treatment systems at TMI-1. Id.,

at 50; see 1 50, supra. The Supervisor of Chemistry, who has a B.S. degree and approximately 12 years of experience in chemistry and radiochemistry, directs the activities of the twelve chemistry technicians, who operate on a six shif t basis, I

with the assistance of a technical assistant and a chemical foreman. Id., at 50-51. Finally, TMI's Generation Maintenance i

System (GMS) coordinators report to the Director of Plant Engineering. The GMS analysts coordinate the scheduling of and provide the data to computer operators regarding preventive maintenance work at TMI-1. Id. They also review for complete-ness maintenance records, machinery history files and gen-j erally, provide the interface between the computer system and i

its on-site users, particularly, the Maintenance Department.

Id., at 51-52.

60. The Board joins the Staff in its confidence that Licensee has assigned sufficient management and technical capability to the TMI-1 plant site to ensure adequate planning for, and technical support and direction of day-to-day opera-tions. Staff Ex. 4, at 13-14; Tr. 12,028 (Haverkamp); Tr.

11,995, 11,964-5 (Crocker); Tr. 11,962-63 (Allenspach, Crocker). Our review indicates that the Vice President, TMI-l has the proper background and experience necessary for the job.

In addition, the second, third, and fourth level managers in the TMI-1 organization have the necessary experience and qualifications for their jobs. Licensee has dedicated a large staff of technical personnel to the plant. Moreover, formal lines of communication have been established between the on-site organization, and the technical support groups located on-site and at corporate headquarters. These additional resources appear to be readily accessible to and regularly used by the plant staff. Although Licensee has instituted pervasive changes in its organizational structure, as well as in the

manpower assigned to TMI-l and to GPU's nuclear activities, there appears to be a common understanding among the more senior level managers who testified before the Board of the re-allocation o'f responsibilities within the new organization.

See also, Wegner, ff. Tr. 13,284, at 13. In summary, in an overall sense, the Board finds that the TMI-l operations, l maintenance and technical staffs are qualified, or will be qualified before restart, to operate Unit 1 safely.

B. TMIA Contention 5

61. Intervenor Three Mile Island Alert, Inc. (TMIA) filed a contention, accepted by the Board in its First Special Prehearing Conference Order, on the subject of maintenance practices at TMI-1. 10 NRC 828, 841 (December 18, 1979). This contencion, TMIA Contention 5, subsequently was amended by TMIA. TMIA Motion for Revision of Contention 5, July 26, 1980; TMIA Correction of Errors in its Motion for Revision of Contencion 5, July 31, 1980. With several clarifying modifica-tions, revised TMIA Contention 5 was accepted by the Board.

Memorandum and Order of Prehearing Conference of August 12-12, 1980 at 3 (August 20, 1980); Tr. %138-53. Several additional modifications subsequently were made to the contention.

Tr. 4602 (Selkowitz); Tr. 12,290-91 (Blake, Smith).

62. TMIA Contention 5, as finally revised, states:

It is contended that Licensee has pursued a course of conduct that is in violation of 10 CFR 50.57, 10 CFR 50.40, 10 CFR 50.36, 10 CFR 50.71 and 10 CFR Appendix

B, thereby demonstrating that Licensee is not

" technically . . . qualified to" operate TMI Unit 1 "without endangering the health and _

safety of the public." This course of conduct includes:

a. I deferring safety-related maintenance and repair beyond the point established by its own procedures (see e.g. A.P. 1407);
b. disregarding the importance of safety -

related maintenance in safely operating a nuclear plant in that it:

1. [ deleted]
2. proposed a drastic cut in the i maintenance budget; -
3. (deleted]
4. fails to keep accurate and eniplete maintenance records relarnd safety items;
5. has inadequate and understaffed QA/QC programs related to maintenance;
6. extensively uses overtime in performing safety-related maintenance.
63. TMIA Contention 5 was not litigated by the parties in the usual evidentiary manner, that 4*, with Licensee first presenting its case on the subject, followed by the Staff and by any intervenors presenting direct evidence. Rather, because of "a failure by TMIA to respond fully to licensee's interrogatories on the contention" after approximately five months of discovery, motions to compel by Licensee, responses by TMIA, and Board orders granting Licensee's motions to compel, the Board directed TMIA to proceed first with its affirmative case on Contention 5 at the start of the

- _ . . - . -. - --- --..-m. ~. ,-,.r,,-..,_ --w, ---,__re.,my,,,m.,, w-.w.-e... - . . . - . . - . , , ,, ,,, _ -m my ,

o evidentiary hearing in October,1980. Memorandum and Order of Prehearing Conference of Aagust 12-13, 1980 (August 20, 1980), _

at 3-4. In our view, on the basis of TMIA's affirmative case, Licensee would be able to discover the specifics upon which TMIA relied in asserting its Contention 5, and therefore would be able to meet TMIA's affirmative ctee. I d,. , at 4, citing Tr. 2,106-28.

64. Beginning on October 15, 1980, TMIA put on its affirmative case on revised TMIA Contention 5. TMIA called fif teen witnesses to testify, thirteen of whom were Licensee employees subpoenaed by TMIA. In ridition, during the October, 1980 session on TMIA Contention 5, an additional Licensee employee testified on the subject of overtime practices in the Maintenance Department, at the recuest of the Board.

Tr. 4,017-19 (Smith).

65. Licensee's responsive case on the subjects included in TMIA Contention 5 (as revised) was conducted in February, 1981. Two pieces of direct testimony by a panel of five witnesses and an individual witness, respectively, were introduced into evidence. Shovlin et al. ff. Tr. 13,533; i Manganaro, ff. Tr. 13,643. In addition, a third piece of prefiled testimony entitled, " Licensee's Respor.se to Board Question Concerning Maintenance Practices in the Sample Year, 1978," prepared at the request of the Board, Tr. 3,352-58, was admitted as Licensee Exhibit 29, without any cross-examination of Licensee witnesses by TMIA, the NRC Staff, or the Common-wealth of Pennsylvania. Tr. 13,659-61 (Blake, Smith).

k, '

66. The NRC Staff prefiled two pieces of testimony, both by witnesses Keimig and Haverkamp, along with a supporting _

memorandum, directed at issues raised by TMIA and by the Board on the subject of previous and present maintenance practices at TMI-1. These documents also were received into evidence without any cross-exauination of the sponsoring witnesses.

Tr. 16,408-09; 16,411-12 (Smith, Keimig, Haverkamp).

67. TMIA Contention 1 asserts that Licensee is not technically qualified to operate TMI-l for a number of distinct reasons, each having to do with the sufficiency of Licensee's past and present maintenance practices, including mai tn enance record-keeping. Each of these five allegations constitutes a serious charge against both the technical capabilities of Licensee, and Licensee's management practices; consequently, we have evaluated each allegation separately.
68. Deferral of safety-related maintenance and repair beyond the point established by Licensee's own proce-dures. Historically, corrective maintenance has been performed i

at TMI-1 by use of a work request (now called a job ticket) system. Tr. 2,638 (Shovlin). In order to initiate a repair on any component or system associated with the plant, an individ-ual originates a work request, identifying the malfunction and I the cause, if known. Tr. 2,639 (Shovlin).

This identification triggers a chain of events within the Maintenance Department with which Operations is frequently intimately involved, Tr. 2,683-88 (Shovlin), whereby the identified repair is 59-

,re--,,-a,--,---r-- +-- -- w--r- - - ~ + - - - - - , , - +

o planned, scheduled for work, approved for work, and fixed.

Tr. 2,639-62 (Shovlin). Depending on the nature of thu work performed -- e.g., whether an engineering modification is required, rather than a repair in kind -- the work request itself will go through many hands other than those of individuals in the TMI-1 Maintenance Department. Af ter the repair has been completed, a similar chain of approvals and signatures is required, the number of which is also dependent on the nature of work performed, e.g. , whether QC must review the work package because the repair involved a component included in GP 1008, the Quality Assurance Systems List, Board Ex. 1. Tr. 2,646, 2,659-62 (Shovlin); Tr. 13,595-97 (Dyckman).

69. It is TMIA's contention that a review of a selected sample of work requests, the forms utilized by the Maintenance Department to parform corrective maintenance, along l with a review of the systems in question and the priorities
given to the work, per se prove that portion of their conten-tion which alleges that Licensee improperly, and at the expense of the public health and safety, deferred safety-related main-tenance. Tr. 3,032-35 (Selkowitz). In TMIA's view, some of
these work requests, in and of themselves, would support TMIA's allegation of grievous maintenance deferral practices at TMI-1; other work requests, when considered as a group, will provide ev'idence of a reprehensible pattern of deferral of safety-related maintenar.ce, a pattern which if true, would present a serious challenge to Licensee's present management capability i to operate TMI-1. Tr. 2,567, Tr. 3,341 (Smith). 1 1

l

',e .

o /

70. In order to consider the merits of TMIA's first allegation, two preliminary questions must be answered: (1)

What constitutes improper deferral? and (2) What is safety-related maintenance? ,

71. Improper Deferral. TMIA arbitrarily chose one year as a bench-mark indication of improper deferral of safety-related maintenance. Bonetti, ff. Tr. 3,310, at 1. Licensee, on the other hand, declined to formulate a bench-mark, and instead addressed in its testimony the propriety ~ in each particular instance of delays in the repair identified in the work requests admitted into evidence. Shovlin, et al. ff. Tr.

13,533, at 23-27, 52-55, 58-61, 63-68, 75-77.

72. The evidence presented makes clear that the time frame in which maintenance work is accomplishad at TMI-l is not formally or rigidly defined by Licensee's procedures, nor has it been so defined in the past. However, work requests are identified by their assigned priority. Tr. 2,673-82, 2,",01-03, l 3,061-73 (Shovlin). The manner in which priorities are assigned within the TMI-l Maintenance Department is radically different now with the promulgation of new definitions of priority work in March, 1980, than the system previously used at TMI-l to categorize the large numbers of repairs -- from cutting the grass to working on the fuel handling door seal --

with which Maintenance contends. Tr. 2,885-86 (Colitz);

3,071-72 (Shovlin). Prior to that time, a priority system was in effect at TMI-l which defined priority 1 maintenance work as l

urgent, priority 2 work as routine, and priority 3 work as a low priority. Although this system was time-related, as

~

obviously work defined as urgent in theory would be accom-plished before routine maintenance, in fact the system did not effectively operate to screen out the truly important maintenance, and thus was not used for that purpose.

Tr. 2,674, 3,068 (Shovlin). Rather, "real" priority maintenance work was scheuuled for work through the use of regular " plan of the day" meetings attended by Maintenance and Operations personnel, as well as individuals from other appropriate organizations such as QA. Tr. 2,702, 3,085 (Shovlin); Lic. Ex. 29, at 12-14. These "real" priorities were assessed on the basis of the description given in the work request itself of the malfunction needing repair, not on the basis of the priority assigned to the work. Tr. 3,100 (Shovlin). Generally, this system functioned effectively because "real" priority jobs usually were identified initially by Operations, Lic. Ex. 29 at 12, and, in any event, they were the maintenance items that were truly tracked by Operations through the plan of the day and the 1600 hour0.0185 days <br />0.444 hours <br />0.00265 weeks <br />6.088e-4 months <br /> meetings.

Tr. 3,085 (Shovlin).

73. The priority system described above was clearly unsatisfactory, a fact which Mr. Shovlin, the Manager of Plant Maintenance at TMI-l and the long-time senior member of the Maintenance organization at Three Mile Island, recognized. The numerical system utilized on work requests to identify priority I

Y work inaccurately reflected "real" priorities established at the plant, in part because of the broad definition given to ,

priority 1 work. Tr. 3,063-64 (Shovlin). In addition, because the priority was assigned F, the initiator of the work request, a tremendous element of subjectivity entered into the designa-tion of a priority. Tr. 3,071-72 (Shovlin). (As Mr. Shovlin explained, a janitor with a leaky valve that is spilling water all over his floor considers the valve repair to be urgent, and hence a priority 1 job. Tr. 2,677 (Shovlin).) Moreover, a very large volume of duplicative work requests existed because the system permitted a work request to be initiated by any individual; however, no administrative method existed to weed out jobs already identified on a work request, e.g., by a worker on a different shift. Consequently, Mr. Shovlin would

" purge" the system on occasion; that is, work requests which were no longer valid because the work had been completed through a different work request or work requests that dupli-cated other outstanding work requests were discarded.

l Tr. 2,679-80, 2,698-701 (Shovlin); Shovlin et al., ff. Tr.

13,533, at 30-31.

74. The priority system, as revised, completely changed the definitions of priorities, in addition to adding a fourth category of work, priority 4. Shovlin et al., ff. Tr.

13,533, at 39-45. The new definitions classify maintenance work according to the health and safety of the public and plant personnel, and the operability of the nuclear plant:

I Priority 1: Can only be classified by superintendents, department heads or shift supervisors; will cause a plant shutdown; reduce generation; has a time clock of very _

short duration; is an immediate industrial or nuclear safety hazard; compromises nuclear safety or security, reactor control or power conversion cycle control system in so far as to present a clear threat of initiation of a trip or severe transient; imposes or threatens increased personnel radiation exposure; constitutes one element of a j multievent failure which would result in L initiation of a trip or transient.

Priority 2: Could cause a plant shutdown if operation is continued too long; redundant

component and backup is no longer available; could cause a plant limitation in the near future; time clock on the component that will require it to be repaired in a timely fashion; items that should be repaired when plant conditions allow.

, Priority 3: Routine corrective maintenance that does not impact plant operation.

Priority 4: Corrective maintenance to clear minor problems that don't actually affect the operation of any components; all change modifications and any improvements that are not related to plant performances. As the paper work on change modifications become available a suitable priority will be assigned.

Lic. Ex. 2. While a priority assignment is recommended by the initiator of the work request, the actual priority assigned to a job is determined by the Supervisor of Maintenance (now called the Manager of Plant Maintenance) or his designee.

Tr. 3,097 (Shovlin).

75. In addition to the revised priority system, l

however, the Maintenance Department continues to make use of the plan of the day meetings, as well as daily (1600) sched-uling meetings, to review with Operations and obtain joint

concurrence on the appropriate schedule of work. Shovlin et al., ff. Tr. 13,533, at 45-47. In addition, Maintenance -

utilizes planners, who organize the paperwork, and a computer system with numerous printouts, to organize and track all TMI-1 corrective maintenance activities. Among other things, these computer summaries provide a means whereby currently outstanding work can be identified by priority. Id., at 36.

76. In reviewing the maintenance system described by TMIA's witnesses, as well as the testimony of Licensee and the Staff, the Board finds no basis in the record for TMIA's allegation that Licensee has in the past sub'stantially departed from a company standard in failing to perform maintenance in a timely fashion. The Board also finds no evidence of such a practice under the current maintenance system. These deter-minations are based in part on the fact that Licensee had and continues to have no firm standard defining the time within which work, of whatever importance, was and is required to be accomplished. Rather, through the use of regular meetings as I

well as currently utilizing computer printouts of outstanding job tickets, it has been Licensee's practice to track what its key Operations and Maintenance personnel perceive to be important maintenance work. Tr. 3,085 (Shovlin); Shovlin et al., ff. Tr. 13,533, at 45-47.

l 77. Although TMIA failed to show that Licensee l

ignored its own procedures in improperly deferring safety related maintenance, the Board considers it reasonable to l

l l  ;

evaluate on its own whether the alleged examples of improperly deferred safety-related maintenance presented by TMIA, either independently or as a group, indicate a lack of attention on the part of Licensee to significant maintenance work at TMI-1.

Similarly, during the proceeding, one of the topics on which the Board requested additional information was whether, prior to November, 1979, Licensee had in place a reliable method of identifying nuclear safety work requests which required maintenance. Tr. 3,352 (Smith).

78. Safety-related Maintenance. TMIA Contention 5 is limited to maintenance which is safety-related. A great deal of time was sp=_ , during the hearing attempting to clarify the Board's and the parties' understanding of this concept.

TMIA and Licensee agreed that safety-related is not equivalent to and should not be confused with safety-grade, or other terms 1

l of art frequently used in the industry. Rather, it was the consensus of the in't .udd parties that the term safety-related, as it pertains to TMIA Contention 5, should be interpreted by what would be a dictionary or ordinary defini-tion of the term; that is, an educated judgment by an individual as to whether or not a particular maintenance item and the systems involved in the maintenance item have some safety significance. Tr. 2,859-68 (Selkowitz, Blake).

l 79. The parties also agreed, at least initially, to rely upon the expert opinion of Mr. Joseph J. Colitz, the Manager of Plant Engineering at TMI-l and one of the witnesses '

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

subpoenaed by TMIA, as to whether the particular maintenance activity identified in each work request offered into evidence _

by TMIA would be safety-related, in light or the component (s) and system (s) involved. Tr. 2,573-79 (Blake, T. Adler); Tr.

1 2,861-67 (Blake, Selkowitz). During the course of Mr. Colitz's direct and cross examination, however, it became clear that TMIA disagreed sharply with Mr. Colitz's conclusions as to whether particular maintenance activities should be considered nuclear safety-related. Compare Tr. 3,487 (Selkowitz) with Tr. 2,847-50 and 3,134-35 (Colitz); compare Tr. 3,560-61 (Selkowitz) with Tr. 2,948-52 and 3,238-43 (Colitz).

80. Mr. Colitz, who has worked at the plant in a senior engineering capacity for approximately nine years and has been licensed as a senior reactor operator at TMI-1, exhibited an Lapressive familiarity with the TM' -1 facility.

Tr. 3,115-16 (Jordan); Tr. 2,994 (Colitz); Hukill et al., ff.

Tr. 11,617, at 40-42. Mr. Colitz testified that he determines whether a particular maintenance activity is nuclear safety-related by looking at the particular component or the problem associated with it and the consequences of doing that job. If the act of doing the repair does not affect the integrity of the reactor coolant system boundary, if the component or system being taken out of service to do the repair is not required for safe shutdown of the plant, and if the inoperable component or system is not required for any accident conditions or mitiga-tion of any consequences and releases to the public, Mr. Colitz

_ _ _ ._ . _ . _ _ ___._- _ __._--_ _.___..-. _ - _ _.-. _ ,_ _ _ . _..__ _ _ _ _ _ -~.. _ _ --

d l would maintain that the maintenance job is not safety-related.

Tr. 2,994-95 (Cclitz). Thus, although Maintenance personnel regularly work on safety-related components or systems, the job itself -- considering both the nature of the problem and the work required to fix the problem, e2 g., a packing leak -- may not necessarily be a safety-related repair. Tr. 2,995 (Colitz).

81. TMIA offered no alternative meanc of determining

! whether a particular maintenance activity was safety-related.

Tr. 3,030-39 (Selkowitz). Other than relying on priority '

designations, no consideration was given to whether the work request activity itself or the system or component which was the subject of the work request activity was safety-related.

Tr. 3,317 (Bonetti).

82. While the Board relied heavily upon the testi-mony of Mr. Colitz in assessing whether particular components or systems are nuclear safety-related for purposes of perform-ing maintenance on those components or systems, the Board also utilized its own expertise in evaluating the testimony, as well as its non-technical judgment, in light of the common usage of the term, safety-related , to which the parties agreed with respect to the evidence presented on TMIA Contention S. See 1 78, supra.
83. The Board admitted twelve exhibits (excluding subparts) into evidence on the topic of deferral of safety-related maintenance: TMIA Exs. 11, 12, 13, 15, 17(a-f), 18,

19, 20, 23, 28, 31, and 39. Each of these exhibits consists of one or more work requests for which the TMI-1 Maintenance organization was fully responsible. According to TMIA, eight of these work requests (TMIA Exs. 11, 13, 17(a-f), 18, 19, 20, 28 and 31) represent a pattern of deferral of safety-related maintenance activities from which one could infer generally that Licensee's management of its maintenance activities was inadequate. Four work requests (TMIA Exs. 12, 15, 23 and 39) in and of themselves support TMIA's allegation of grievous maintenance deferral practices at TMI-1. See 1 69, supra.

Moreover, TMIA would contend that there is no basis for finding any improvements in current management oversight of maintenance practices. Tr. 2,582-83 (T. Adler). (In addition, two of the exhibits reviewed by the Board in connection with this portion of TMIA Contention 5, TMIA Exhibits 20 and 31, also pertain to TMIA's allegations concerning the adequacy of Licensee's QA/QC programs related to maintenance. See,TMIA Contention 5(b)(5).

In these two iustances, the major deferral of work evident from the face of the work request is not in the performance of the necessary maintenance; rather, it is in the time it took QA to

{

review the paperwork associated with the completed job.

Tr. 2,662 (Shovlin); Tr. 3,586 (Jordan) Tr. 3,690-98 (Selkowitz, Smith). (The issue of Licensee's maintenance record-keeping practices is discussed in 11 98 through 103, infra.)

1

84. Mr. Colitz was of the view that only two of the twelve work requests at issue here, TMIA Exhibits 13 and 31, l  !

had potentir.1 safety-significance. Tr. 3,155-58, 3,252-55 (Colitz). (It should be noted that Mr. Colitz was not asked .

about the safety-significance of the work identified in TMIA Exhibits 12, 15, 23 and 39, so-called " surprise" exhibits.

Tr. 3,470, 3,474 (Smith). This was due to the fact that TMIA's counsel did not question Mr. Colitz about the safety-significance of particular maintenance work, and the Staf f's and Licensee's counsel were not aware of the possible admission into evidence of these documents at the time Mr. Colitz testified; hence, no questions were asked of Mr. Colitz along these lines.) In each of the other instances, it was Mr.

Colitz's view that the maintenance work identified on the face of the work request, including delay in conducting the maintenance, wculd not affect the safe operation of the facility, and thus was not safety-related. Tr. 3,118, 3,133-35, 3,239-43, 3,249-51 and 3,175-79 (Colitz).

85. The Board admitted into evidence the exhibits listed above, applying a broad perception of the potential association of the components or systems referenced in the i

j exhibits to the safe operation of the facility. See, e.g.,

Tr. 3,038 (Jordan). TMIA Exhibits 11, 12 and 13 reference valves in the feedwater system, the system which provides water to the atsam generators. Tr. 2,843-50 (Colitz); Tr. 3,491-92

( (Jordan). The subject of TMIA Exhibit 15 is hydraulic snub-bers, components installed as pipe supports and to protect l

l against pipe breaks under seismic conditions. Tr. 2,874 I

l

, 1 is

, - , . , - - _ , . . . --.,,.---,..n. -,.,n. . . . _ , , , -, . , -. ,, ,-..,.. ,, , , , _ . , - - - , . , . . . - - . , , , - - , , , , . - , , , ,

o (Colitz). TMIA Exhibits 17(a-f) and 18 concern the nucleac river water pumps which cool the four nuclear services closed ccoling water heat exchangers in order to keep these systems at their design conditions. Tr. 2,949 (Colitz). TMIA Exhibit 19 involves repairs to the manways and handholds associated with openings into the primary side of the once-through steam generators -- openings which constitute part of the primary coolant system boundary. Tr. 3,252-55 (Colitz). TMIA Exhibit 20 concerns the motor-operated coolant block valve, RCV-2, which functions to isolate leaks resulting from failure of the power-operated relief valve, or PORV, to close. Tr. 3,250, 2,953-55 (Colitz). TMIA Exhibit 23 involves repairs to the reactor building access hatch interlocks. These access hatch interlocks prevent direct contact of the atmosphere in the reactor building with that of the auxiliary building.

Tr. 2,936 (Colitz). T a subject of TMIA Exhibit 28 .s prefil-tecs located in the makeup and purification system, the system which normally functions, amota other things, to maintain the pressurizer IcVe1 within operating range and to provide makeup water to the reactor coolant system. Tr. 2,898-901 (Colitz).

TMIA Exhibit 31 concerns repairs to the fuel handling building door seal. Tr. 2,885-86 (Colitz). In TMIA Exhibit 39, incorrect information was being relayed to the computer in the control room from four control rod drive thermocouples.

Tr. 2,893-94, 3,286-90 (Colitz).

86. The time lag involved in these work recuests, from date of initiation to final sign-off, ranges from i

h, ) . * ** .

o /

approximately seven months in the case of TMIA Exhibit 31 to approximately thirty-one months in the case of TMIA Exhibit 12.

87. Although TMIA questioned numerous Licensee employees on the work request process, particularly focusing upon the meaning of the questions included and information provided on the work request forms, see, e.g., Tr. 3,477-87 (Good); Tr. 3,494-99 and 3,511-13 (McGarry), no evidence was adduced as to the teason(s), if known, for the delays identified in the work requests.
88. In its responsive testimony, Licensee explained in detail how the various repairs were conducted in accordance with Licensee's procedures in the case of TMIA Exhibits 11, 12, 13, 15, 19, 20, 23, 31 and 39. Shovlin et al., ff. Tr. 13,533, at 25-27, 52-55, 58-61, 75-77. With respect te TMIA Exhibits 17(a-f), 18 and 28, Licensee acknowledged that the docu-mentation of these repairs was improper; however, the work was performed properly, in a manner consistent with current maintenance practices. Id., at 63-68. Also, in the case of TMIA Exhibit 20, both Licensee and the Staff recognized but saw no safety-significance in the QC sign-off delay of nine months, id,. , at 75-76; Keimig and daverkamp - TMIA 5,10 ff. Tr. 16,412, Table B at 8, or the approximately two and a half year delay 10 "NRC Staff Testimony of Richard R. Keinig and Donald R.

Haverkamp In Response to TMIA Contention 5," dated March 17, 1981, will be referred to as Keimig and Haverkamp - TMIA 5.

11 Table B was received into evidence without pagination.

In order to precisely cite to the table, page numbers (1-20) have been provided.

l in completing the design chLnge initiated by TMIA Exhibit 12.

Shovlin et al., ff. Tr.13,533, at 23-24; Keimig and Haverkamp

- TMIA 5, ff. Tr. 16,412, Table B at 2.

89. In its independent review of the individual work i requests offered by TMIA to prove its allegation regarding improper deferral of safety-related maintenance, the Staff evaluated the impact of the delays associated with the work requests received as TMIA exhibits and concluded that, " [t] here appears to be no example of inappropriately deferred safety-re-lated maintenance work." Keimig and Haverkamp - TMIA 5, ff.

Tr. 16,412, at 3-4.

~

90. While the Board found that certain of the exhibits offered by TMIA into evidence were of potential safety significance, there is no evidence supporting TMIA's allegation that, in fact, deferral of these potentially safety significant jobs endangered the health and safety of the public. While the delays identified in the suLject work requests initially raised concerns about the adequacy of Licensee's maintenance practices, detailed and reasonable explanations have been provided, without challenge, as to why, in each individual instance, such delays occurred. On the basis of Licensee's explanation, with which the Staff concurs, the Board is satisfied that Licensee has not in the past and does not presently defer safety-related maintenance and repair beyond the point established by its own procedures, thereby endan-gering the health and safety of the public, as alleged in TMIA Contention 5(a).

~

C /

91. Proposal of a drastic cut in the maintenance budget. An across-the-board operations and maintenance (UEM) .,

l budget cut was proposed by GPU management for the year 1979.

Tr. 4,038-40 (Wise). (The proposal was instituted by GPU management at the beginning of 1979, af ter the original 1979 budget had been develoned. Tr. 4,042 (Wise).) It is TMIA's contention that this budget cut proposal was drastic, at least 3

insofar as its impact on the TMI-l Maintenance Department, and constituted a disregard by Licensee's management of tne importan e of safety-related maintenance in safely operating a nuclear plant. TMIA Contention 5(b)(2).

92. There is no evidence as to what portion of the 1979 budget cut, as proposed, was ever instituted, nor is it clear what portion of the budget cut items applicable to TMI-1 were costs previously budgeted to the Maintenance Department.

TMIA called two witnesses on this subject, Mr. Donald Wise, Assistant Comptroller for Met Ed, and Mr. John R. Knoll, Administrator, Budgets and Reports, Three Mile Island. Mr.

Wise was not sure whether or not any of the proposed budget cut items were in fact instituted in light of the proximity in time of the finalization of the budget reductions to the accident at TMI-2. Tr. 4,057 (Wise). Mr. Knoll was of the belief that some, but not all, of the proposed budget cut items were actually reduced. Tr. 4,086 (Knoll) ; see also, Keimig and Haverkamp - TMIA 5, ff. Tr. 16,412, at 8. The two witnesses did not agree on what percentage of the identified reduction

items were attributable to corrective maintenance, and not other OEM expenditures. Compare Tr. 4,089-91, 4,107 (Wise) i with Tr. 4,125 (Knoll).

93. TMIA does not allege nor does the evidence support an allegation that drastic cuts were instituted in the TMI-l maintenance budget. However, the methods by which the budget would have been cut, but for the TMI-2 accident, and therefore, the manner in which budget cuts might be instituted at TMI-l today were of interest to the Board in assessing the management competence of Licensee.
94. The 1979 proposed budget cut, applicable to all elements of the GPU O&M system, was divided into two phases.

Tr. 4,043 (Wise). That portion of, the budget cut applicable to Licensee, Met Ed (one of the GPU subsidiaries), was: $2.5 million cost reduction applicable to 1979 without regard to whether these costs would be deferred to subsequent years; and, a $2.8 million permanent (i.e., non-deferred) Met Ed expen-diture reduction. Of these targeted amounts, in the case of phase one, $346,000 was identified as proposed Met Ed budget cuts for TMI-1, out of a total Met Ed TMI-l budget of $8.6 million. (These dollar figures represent Met Ed's budget for TMI-1. Since there are two other operating companies which support TMI-1, with c ntributions together which equal the contributions from Met Ed, a like amount would be cut from the budgets of these two contributing companies. Thus, the total phase one budget cut applicable to TMI-l was $692,000 out of a 1

1

total TMI-1 budget of $17.2 million. Tr. 4,118-19, 4,043-44 (Wise).)

95. At TMI-1, as well as within other Met Ed organizational units, the items proposed to management as eligible for inclusion in the proposed budget cut list were initiated and proposed within " root departments," such as the TMI-l Maintenance Department. These proposed items were reviewed within each organization, re-examined-on a production division level at Met Ed, and finally approved by the President of Met Ed. Tr. 4,046, 4,049-50, 4,059-62 (Wise). At TMI-1, management of Met Ed's Generation Division (the production division in which the TMI-l budget was located) undertook a thorough review of the priority list for budget cuts estab-lished by plant personnel. At a series of meetings, these priorities were examined in detail. Tr. 4,059-60 (Wise);

4,096-98, 4,116 (Knoll). This method of identifying potential cuts to the 1979 budget as originally proposed made it pos-sible, at least theoretically, for a department within a division to avoid any cuts to its budget if the department

{

could identify no savings greater than the risks associated i with cutting costs. Tr. 4,062 (Wise).

96. The NRC Staff was made aware of the proposed O&M budget cut applicable to TMI-l through a routine, programmatic Inspection & Enforcement management meeting with Met Ed's Vice President Generation and others on February 9, 1979. Keimig and Haverkamp - TMIA 5, ff. Tr. 16,412, at 6. Because of their i

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

i early notification of the proposed budget cut, NRC staff were confident of their ability to monitor the impact, if any, of ,

the proposed budget reductions on the safety of plant opera-tions. On the basis of IE inspection reports for that period, Staff witnesses Keimig and Haverkamp conclude that there is no apparent basis to suspect that budget reductions were. having adverne affects on plant safety. Nevertheless, in order to fully respond to the allegation raised by TMIA, an IE Region I inspector conducted interviews with Met Ed personnel, including the Superintendent of (now, Manager of Plant) Maintenance, maintenance supervisors and foremen and reviewed documentation on this issue. The inspector found no discrepancies between the documented material and the verbal information given during interviews. Although some reduction in contracted preventive maintenance support in early 1979 was identified, the inspector

found no evidence that the maintenance budget reduction affected any safety-related corrective maintenance. On the
basis of the review undertaken by the Staff of its own docu-i mentation, as well as the documentation and recollections of Licensee personnel, the Staff witnesses concluded that there is no evidence to support TMIA's allegation that Licensee dis-regarded the importance of safety-related maintenance in safaly operating a nuclear plant by proposing cuts in the TMI-l maintenance budget. Id., at 7-10.

l 97. The Board is satisfied with the review under-taken by the Staff to assess the merits of TMIA's budget cut l

l l

I.- . . - _ _ _ - - - - . - . - - , - - - - . _ . , . - _ . . - . . . - . , . _ , - - - - - . . . - . - - - - - . - - - - - - _

allegation, and finds no record evidence in support of a contrary conclusion. While a budget cut was proposed in early 1979 which would affect TMI-l maintenance activities, there is ,

no evidence to support the conclusion that this cut was drastic, or would have been drastic if the TMI-2 accident had not occurred. Nor is there any reason to believe that the method used by Licensee to prioritize items for reduction did not satisf actorily identify items which could not be eliminated from the 1979 budget without affecting safe operation of the plant, and therefore protect these items from inclusion in any cost reduction plan.

98. Failure to keep accurate and complete main-tenance records related to safety. In its case-in-chief, TMIA questioned Licensee employees on the record-keeping system used by the TMI-l Maintenance Department in the past and in the present. See, e.g., Tr. 2,662-72 (Shovlin) (discussion of how work request information is put into the computer, review of differences between computer summary and manual Maintenance Log information, genesis of current procedures); Tr. 3,377-437 (Good) (discussion of Corrective Maintenance Component History

, Report (TMIA Ex. 9) and Corrective Maintenance Master Job t

Ticket Report (TMIA Ex. 10), including function of these documents); Tr. 3,839-88 (Dyckman) (discussion of purpose of manual Maintenance Log, explanation of Misplaced Job Ticket Reconciliation Form (TMIA Ex. 42), description of current document control. practices). In addition, TMIA offered into

l

! evidence a number of work requests in support of subsection (b)(4) of TMIA Contention 5, which alleges that Licensee's ,

record-keeping practices with respect to safety-related maintenance are inadequate and, as such, contribute to Licensee's disregard of the importance of safety-related maintenance in safely operating a nuclear plant. See TMIA Exa.

16, 21, 22, 33(a-m), 34(a-k), 40 and 43(a-b). These work requests present a variety of record-keeping problems. TMIA Exhibit 16 is a work request that may have been used twice to-perform the requested job. Tr. 3,524 (McGarry). TMIA Exhibits i

21, 22 and 40 are marked, " cancel, purged" or " cancel" without any explanation given on the face of the work request as to the reason for the cancellation. Tr. 3,592-96 (Shovlin); Tr. 3,797 (Selkowitz). TMIA Exhibit 33(a-m) consists of thirteen work requests identifying the same piece of work; however, whether the work involved is nuclear safety-related is not answered

identically in these work requests. Tr. 3,716 (R. Adler).

TMIA Exhibit 34(a-k) also involves a series of work requests which identify the same repair; in this group, some of the work requests identify the component involved as a QC component, per GP 1008, while others do not. Tr. 3,722 (Selkowitz). TMIA Exhibits 42, 43(a, b) and 44 all relate to the manner in which misplaced work requests or job tickets are handled.

Tr. 3,867-81 (Dyckman).

99. On the basis of the evidence presented to the Board during TMIA's case-in-chief, the Board was motivated to  ;

i l

inquire further into Licensee's past record-keeping system, that is, whether the Licensee had in place a reliable system of records which would identify and assure that the work was

)

either done oE made unnecessary for some other reason. ,

Tr. 3,352, 3,896 (Smith). By agreement of the parties and with the Board's concurrence, the year 1978 was chosen as a repre-sentative or sample time period for purposes of meeting this Board inquiry. Tr. 3,358-59 (Smith); Tr. 3,8?5-36 (Blake, Selkowitz).

100. Licensee and the C7ff responded to the Board's request for information, in addition to responding to TMIA's case-in-chief. Licensee, in several pieces of testimony, described in detail the maintenance record-keeping system in existence Et TMI-l now, as well as the system utilized in the past, including discussion of the Maintenance Department's interface with the QA Lepartment. Shovlin et al., ff. Tr.

13,533, at 14-23, 29-39, 47-51, 72-75, 77-79; Lic. Ex. 29. The Staff also responded to TMIA's presentation and the Board's inquiry, particularly focusing upon the auditability of maintenance work in the past ( the sample year ,1978), and in the present, as suggested by the Board and agreed upon by the interested parties. Tr. 13,662-67 (Smith, Little, Swanson, L.

Bradford, R. Adler and Blake). Based on information indepen-dently obtained by IE during and in response to the Management Appraisal Inspection 50-289/80-21, described in the management safety evaluation report, NUREG 0680, Supp. 1, App. B (Staff a-.--m----,,.,--,.-,..w,-m_.,w,-,,,-w w,.,,,---,-,.w.,.,y+,+, ,ym,--..,,-,--.-,em-~ , , - - - . -,--,--..w.- - , , , , v----r,--- , y

Ex. 4), as well as routine NRC inspections in those areas conducted during 1978, the Staf f concluded that Licensee's prepared testimony was an accurate representation of the Licensee's current and past (1978) maintenance and QA/QC programs and practices. Keimig and Haverkamp - TMIA 5, ff. Tr.

16,412, at 2; Keimig and Haverkamp - Sample Year 1978,12 gg, Tr. 16,412, at 2-3.

101. With respect to the particular work request exhibits received in evidence in support of TMIA Contention 5(b)(4), Licensee ackncwledged that TMIA Exhibit 16 was improperly utilized twice; however, current procedures provide for the use of so-called " blanket work requests" to perform repetitive jobs on a particular component. Shovlin et al., ff.

Tr. 13,533, at 61-63; Tr. 13,626 (Dyckman). This position was confirmed by the staff testimony which states: "Use of this work request twice was a poor practice, but did not affect safety because the work was otherwise properly performed and affected systems were properly controlled and returned to service." Keimig and Haverkamp - TMIA 5, ff. Tr. 16,412, at 5, Table B at 4. With respect to TMIA Exhibits 21, 22 and 40, the

" purged" or " cancelled" work requests, Licensee explaine6 how the operations-related work identified on TMIA Exhibit 21 was 12 "NRC Staff Testimony of Richard R, Keimig and Donald R.

Haverkamp In Response to the Board Questions Concerning Auditability of Maintenance Practices in the Sample Year, 1978, and Currently," dated March 17, 1981, will be referred to as Keimig and Haverkamp - Sample Year 1978.

I i

t a

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

o completed in short order, with other minor work deferred and eventually cancelled; how TMIA Exhibit 22 identified a modification initially desired by Engineering, but subsequently cancelled and replaced by a better . modification, and another work request, and how TMIA Exhibit 40 initially was classified as a high priority item but, upon a closer look, the problem was discovered to be of less importance than originally thought and required an engineering change modification with an associated work request to correct. Shovlin et al., ff. Tr.

13,533, at 55-68, 27-29. In each of these instances, the Staff concurred with the actions taken by Licensee, including the work request cancellations. Keimig and Haverkamp - TMIA 5, ff.

Tr. 16,412, at 5, Table B at 8-9, 18-19. According to Licensee, TMIA Exhibits 33(a-m) and 34(a-k) do not involve QC components or nuclear-safety related work, although some of the work requests included in these exhibits incorrectly identified the work to the contrary. Shovlin et al., ff. Tr. 13,533, at 68-69. In the Staff's view, while the ventilation system l installation is regarded as QC, the filters which are the subject of these work requests are not within QC scope, and the work involved was not nuclear safety-related. Keimig and Haverkamp - TMIA 5, ff. Tr. 16,412, Table B at 11-17. The absence of formal documentation therefore was not a noncompli-ance with NRC requirements, although this shortcoming did impede the timely Licensee and NRC reviews of these work requests. Keimig and Haverkamp - TMIA 5, ff. Tr. 16,412, at 5.

1 r-,--- -

-.c. r-- -..,,---ew-me.-%.,---, --.-,y -- - -----,.-,,,--.ww---e-y- -, --,wy .--w- r-,g m,e-,e,r-- --.--y-w-ww%-, ,.v - - - - - - . --w--.. - .--w-,

Y.

c ,

l l

Finally, TMIA Exhibits 42 and 43(a, b) were discussed in detail by Licensee, including the rationale used to generate the so-called Job Ticket Reconciliation Form in order to correct the absence of work request records on particular jobs. Shovlin et al., ff. Tr.13,533, at 30-33; see also, Keimig and Haverkamp -

TMIA 5, ff. Tr. 16,412, Table B at 19-20.

102. Throughout Licensee's explanation of the record-keeping discrepancies associated with the work requests described above, Licensee acknowledged problems where it perceived the existing records to b2 inadequate , and addressed the methods currently in use to prevent such errors. See, e.g., Shovlin et al., ff. Tr. 13,533, at 68-69 re preventive maintenance program's focus upon timely replacement of filters such as those identified in TMIA Exhibits 33(a-m) and 34(a-k).

In addition, t'ie Staff's extensive review of maintenance activities during 1978 found that record-keeping associated with the activities reviewed was complete and auditable. In particular, the work requests showed that all necessary reviews and approvals for work were obtained as required by applicable

\

Technical Specifications, regulatory guides and standards, and Licensee procedures. On the ba;is of this finding, the Staff concluded that Licensee had in place an acceptable and audit-able method of maintaining records which demonstrated that the work was properly identified and either performed or determined to be unnecessary for appropriate reasons. In addition, Licensee's maintenance records provided an acceptable method of

.-ev--e- ,e, .,-,---v,- - . - - - - - , - ,_.----,-~.-,..--,----,,em.,-,,,w,.,,wm,,,----,mn.--.,--,e, ,-.,-,we,-,--,,.-,ven,_--w,ns.,.-w,,- - - . -

o obtaining necessary work review, approval and reporting.

Keimig and Haverkamp - Sample Year 1978, ff. Tr. 16,412, at 10-11.

103. Although the Board initially had concerns about the record-keeping system in operation at TMI-1, both in the past and at the present time, on the basis of the investiga-tions conducted by the Staff, as well as the detailed informa-tion provided to the Board by the Licensee, the Board's concerns have been satisfied. In addition to having confidence in the current maintenance record-keeping system, see, e.g. ,

Tr. 3,888 (Dyckman), the Board is satisfied that past record-keeping practices did not impact adversely upon plant safety.

Moreover, the Board is convinced that the attitude of Licensee management displays a villingness to admit shortcomings in its past record-keeping practices, and to initiate new methods for I

dealing with record-keeping problems.

104. Inadequate and understaffed QA/QC programs related to maintenance. Closely related to TMIA's allegation concerning maintenance record-keeping practices and to the' I

Board question on that subject is T: IIA's allegation that Licensee has inadequate and understaffed QA/QC programs related to maintenance. TMIA Contention 5(b)(5).

105. Quality Assurance and Quality Control personnel l are involved on a daily basis, in a number of different ways, with the performance of maintenance at TMI-1. Tr. 11,875-77 (Ballard). The controlling document for purposes of QC I

'A , e' . - .' ,

G J involvement in the work request or job ticket process is GP 1008, the Quality Assurance Systems List. Board Ex. 1. QC's scope of review currently includes components and systems which are consideced "important to safety," in addition to those components and systems which are classified as safety-related.

Tr. 11,899-901 (Herbeln). Under the current computerized job ticket system as well as the previous work request system, the Maintenance Department is required to determine whether the repair in question involves a component referenced in GP 1008 (or its predecessor procedure). See Board Exs. 2a-j (sample work requests and job tickets from 1974-1979). If so , the wor k '

request / job ticket package must be reviewed and approved by QC prior to initiation of the work. Shovlin et al., ff. Tr.

13,533, at 20; Lic. Ex. 29, at 18-19. QC also witnesses certain maintenance jobs in the field while the work is being accomplished. Shovlin et al., ff. Tr. 13,533, at 49-50; Lic.

Ex. 29, at 18-19. Once the job is complete, including testing, realignment end release of the component for normal use, the work request / job ticket package is sent to GC for review of the paperwork. Shovlin et al., ff. Tr. 13,533, at 74-75; Lic. Ex.

29, at 19-20.

( 106. Licensee and the Staff recognized notable delays in CC final sign-offs with respect to certain work requests introduced into evidence by TMIA. See TMIA Exs. 20,

31. However, both parties assertad tha t this method of review is not in violation of Licensee's procedures, nor has it had

i

\ l i

any impact on plant safety. Lic. Ex. 29 at 20; Shovlin et al.,

ff. Tr. 13,533, at 75-77; Keimig and Haverkamp - TMIA 5, Table B at 8 and 11; Keinig and Haverkamp - Sample Year 1978, ff. Tr.

16,412, at 11.

107. Overall, the Board is satisfied with the extent of QA and QC involvement in the conduct of maintenance at l TMI-1. This conclusion is based in part on the maintenance j review procesa now in place at TMI-l for safety-related j repairs, in part on the Staff's conclusion that where past practices, such as quality review and documentation, have been found deficient or marginal, the Licensee has improved these problems through a revised maintenance organization structure, procedures, and management as well as through the institution of a computerized information control system, and in part on the Board's personal observations and discussions with Licensee's senior QA management. Shovlin et'a1., ff. Tr.

13,533, at 72; Keimig and Haverkamp - TMIA 5, ff. Tr. 16,412, at 5; Tr. 11,758-908 (Clark, Herbein, Kazanas, Ballard).

108. Witn respect to staffing, no evidence has been presented challenging the sufficiency of Licensee's staffing j levels, other than to suggest the inference that delays in CC l

l sign-offs could be ettributed to insufficient staffing. In view of the large number of personnel within the QA Departinent under the new GPU Nuclear organization, and the significant increase in number from the earlier QA organization, the Board finds that Licensee's QA/QC programs related to maintenance are

not understaffed. Tr. 3,181-84 (Kazanas); Lic. Ex. 29, at 7 and Attachment 2; Arnold, ff. Tr. 11,434, at 19-20.

109. Extensive use of overtime in performing safety-related maintenance. TMIA's final allegation charges that Licunsee has disregarded the importance of safety-related maintenance in safely operating a nuclear plant in that it extensively uses overtime in performing safety-related main-tenance. TMIA Contention 5(b)(6).

110. At issue here is the prudent use of overtime.

As the Staff has pointed out, it is important to note that the quality of work and, in some cases, the immediate safety of plant operations is enhanced by the use of overtime. This is particularly true for prompt corrective repairs of safety-related equipment. The maintenance work may be better if the same person or crew starts and completes the repair because specific techniques are sometimes learned during the trouble-shooting and disassembly of a component. Also, overtime may be particularly beneficial where certain specialized quali-fications or talents are limited to only a few individuals, such as would be the case for certain welding operations or complex calibrations. Keimig and Haverkamp - TMIA 5, ff. Tr.

16,412, at 12. On the other hand , if the use of overtime is abused, e.g., overtime is used without consideration of the physical and mental condition of the worker (s) involved so that maintenance personnel are exhausted while on the job and cannot properly perform the repairs assigned to them, it no longer

y serves a beneficial purpose and, in fact, may jeopardize plant safety.

111. The normal shift lengths for Maintenance personnel working at TMI-1 has varied since the plant went into commercial operation. Tr. 2,688-94 (Shovlin). At its peak, normal shifts during outages went for 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> a day, seven I

days a week. Tr. 2,688-89 (Shovlin); Tr. 4,166 (Reismiller);

Tr. 4,140 (McCuray); Tr. 3,991-92 (Eberle). Currently, Licensee uses daylight maintenance workers on eight hour shifts five days each week, plus six rotating shifts, eight hours each

shift. Shovlin et al., ff. Tr. 13,533, at 4. The use of overtime at TMI-1 has also varied. Under Licensee's old overtime policy, the unit superintendent was required to approve overtime for an individual after he had worked sixteen hours. Tr. 2,691 (Shovlin); Tr. 4,169 (Reismiller); Tr. 3,991 (Eberle). Under the current policy, which conforms to the 4

guidelines of NRC IE Circular No. 80-02, the maximum hours allowed, absent special circumstances, are 12 hours1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> a day, 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> each week. Shovlin et al., ff. Tr. 13,533, at 70 and Attachments 13 and 14.

112. The Board received conflicting evidence on the burden felt by Maintenance workers from the overtime required l of them. Mr. Norman Reismiller, a former Licensee employee, complained bitterly about the long worki.1g hours required of him. Tr. 4,178 (Reismiller). On the other hand, Mr. Lawrence i

E. Eberle, a current Licensee employee, appreciated the

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

opportunity to volunteer for overtime work. Tr. 3,986 1

(Eberle). Similarly, Mr. David McCurdy, a current Licensee '

employee,13 was of the opinion that the overtime available during refueling outages was optional, Tr. 4,150 (McCurdy), and that he was not forced to stay and work long hours if he did not feel up to it. Tr. 4,144 (McCurdy).

113. While it is clear that maintenance staff worked long hours during plant shutdowns, breaks were taken during the day, and a hot dinner break lasting for one to two hours was provided in the evcaing. Tr. 4,143, 4,140 (McCurdy); Shovlin et al., ff. Tr. 13,533, at 71-72 and Attachment 11; Tr. 3,992 (Eberle). Overtime was not mandatory; at the sare time, absence without prearrangement was not condoned. Shovlin et al., ff. Tr. 13,533, at 71 and Attachment 10; Tr. 4,011-12 (Eberle); Tr. 4,144 (McCurdy). Overtime assignments were made with an effort, as required by management's contract with its employees, to equalize the overtime work offered to personnel of the same classification, e.g., first class instrument men.

Shovlin et al., ff. Tr. 13,533, at 70; Tr. 4,006 (Eberle). In its review of Licensee's past and present overtime practices, l the Staff considered whether, during IE inspections, an 13 Mr. McCurdy testified on the subject of overtim9 as a result of his being randomly selected by the Board from a list of eleven former or present Licensee employees referenced in TMIA Exhibit 44(a-k). Tr. 4,138 (Smith). TMIA

, Exhibit 44(a-k) lists the overtime hours of these employees from October, 1977 through March, 1979.

f inspector had ever made note of the fact that an individual was excessively mentally or physically fatigued, or if Licensee's maintenance history Jords for TMI-1 revealed any abnormally repetitive maintenance which might indicate the existence of poor workmanship caused by fatigued workers. Keimig and Haverkamp - TMIA 5, ff. Tr. 16,412, at 11-13. However , the 1

Staff found no apparent basis on which to conclude that extensive overtime was used at TMI-1 in a manner which adversely affected the quality of safety-related maintenance work. Id., at 11.

1 114. In light of the Staff's conclusions af ter investigating Licensee's overtime practices, as well as the testimony by former and present Licensee employees, the Board is satisfied that Licensee has not in the past and does not now disregard the quality and importance of the safety-related maintenance being performed by improperly permitting or requiring employees to work overtime.

i 115. In summary, the Board finds Licensee has not (a) deferred safety-related maintenance and repair beyond the point established by its own procedures, or (b) disregarded the importance of safety-related maintenance in safely operating a nuclear plant by (1) proposing a drastic cut in the maintenance budget, (2) failing to keep accurate maintenance records related to safety items, (3) having inadequate and understaf fed QA/QC programs related to maintenance, and (4) extensively using overtime in performing safety-related maintenance;

consequently, Licensee's course of conduct with respect to these issues is not and has not been in violation of 10 CFR SS 50.57, 50.40, 50.36, 50.71 and 50 Appendix B, nor does it I

demonstrate that Licensee is not technically qualified to operate TMI Unit 1 without endangering the health and safety of the public.

C. Training 116. Training at TMI-1 has been of special interest to this Board. Training has been the subject of considerable attention and discussion since the accident at TMI-2 and was the subject of extensive testimony during the evidentiary hearing. See generally, Tr. 12,126-13,011, 13,108-227, 20,576-639, and 20,686-782 (Long, Knief, Newton, Ross, Gardner, Christensen, Kelly, Boger, Crocker, A11enspach, Aamodt) .

Licensee presented two panels of witnesses: the first composed i

of employees, namely Dr. Bebert Long who heads all of GPU Nuclear's training, Dr. Ronald Knief, who heads up training at TMI-1, Mr. Samuel Newton, who is in charge of Licensed Operator Training at TMI-1, and Mr. Michael Ross, Manager of Operations at TMI-1; and the second comprised of three Licensee consul-

[

tants, Dr. Eric Gardner, en educational psychologist with special expertise in educational and psychological measurement, psychometrics, test construction, and curriculum and program evaluetions, Dr. Julien Christensen, an engineering psycholo-gist and human factors specialist, and Mr. Frank Kelly, an i

1

independent consultant on training for nuclear power plant staffs. See Long, et al., ff. Tr. 12,140; Gardner, Christensen and Kelly, all ff. Tr. 12,409; see also Lic. Ex. 33.

117. The NRC Staff addressed training in its SER and Supplements (see, e.g., Staff Ex. 1, at Cl-16, C6-5 to C6-7; i

Staff Ex. 4, at 10-11, 21, 40-41; Staff Ex. 13, at 2-5, 7) and in testimony by Mr. Bruce Boger, a reactor engineer with NRC's Operator Licensing Branch, and Messrs. Lawrence Crocker and Frederick Allenspach, management engineers it NRC's Licensee Qualifications Branch (See Boger (Aamodt Contention #2), ff.

Tr. 12,770; Boger (CEA Contention No. 13), ff. Tr. 12,772; Crocher and Allenspach, ff. Tr. 12,653). One of the intervenors, Ms. Marjorie Aamodt, herself sponsored testimony l on Licensee's licensed operator preparedness. See Aamodt, ff.

Tr. 12,931. Our approach to the large volume of evidence on this subject will consist first of a review of Licensee's training program generally in this section and then a more detailed look at the specific concerns raised by Intervenor Aamodt in the next section. We will start in this section by reviewing Licensee's overall training organization and the principals involved in its management, each of whom we had an opportunity to view when they testified. Next we will describe the training programs which Licensee has developed and is implementing. Following the discussion of the programs generally, we will review the specific training programs for licensed operators and for unlicensed personnel (including

Shift Technical Advisors) which Licensee has instituted since the accident at TMI-2. And finally, we will discuss the NRC's views as well as those of Licennee's consultants who have reviewed Licensee's training programs.

118. The operation and maintenance of a nuclear facility such ra Three Mile Island Unit 1 must be supported by an extensive and diversified training program, including formal classroom instruction as well as on-the-job training ac-tivities. Since the accident at TMI-2, Licensee has embarked upon a major training and retraining effort for both licensed and nonlicensed TMI-l personnel. Training for operations staff is emphasized through continuous training, retraining, and testing programs, including increased simulator experience. In addition, Licensee has instituted comparable training programs for other staff, such as its maintenance, health physics and chemistry workers. With the exception of security personnel, all TMI-l shift workers are scheduled on a six shift work cycle with one of the aix shif ts dedicated to training. Thus, these personnel spend one in every six weeks in training. To accomplish this effort the TMI-1 Training Department has increased its staff from 7 to 45, is diversifying its curricu-lum as well as the educational opportunities it makes available to TMI-1 personnel, and is developing a much more sophisticated administrative capability in order to manage the TMI-1 training program effectively. Long, et al., ff. Tr. 12,140, at 3.

119. All training associated with GPU's nuclear facilities is coordinated by the Director-Training and i

1

Education of Nuclear Assurance located in Parsippany, New Jersey. The activities of the Training Departments located at each of GPU Nuclear Corporation's facilities are coordinated and reviewed by the Corporate Training Department. The Corporate Training Department is the organization which coordinates the development of common training needs of the three GPU Nuclear facilities, e.g. , radiological control training. In addition, the Corporate Training Department provides training to GPU Nuclear Corporation employees in specific areas where supplemental education is beneficial, e.g., nuclear engineering training for Technical Functions employees with B.S. degrees in electrical or mechanical engineering. The Corporate Training Department also coordi-nates the development of course plans in management and supervisory skills. These coursec are implemented at TMI-l by the TMI-1 Training Department. Id., at 4-5.

120. The current Director-Training and Education, Dr.

Robert L. Long, not only has had a broad range of experience in the nuclear industry, but also has been actively involved in university and industry nuclear engineering education programs for over fifteen years.14 Id., at 7. Dr. Long joined GPU in l

l 14 Dr. Long received a B.S. degree in electrical engineering from Bucknell University (1958), of utility power system data bases, and M.S. nd Ph.D. degrees in nuclear engineering from Purdue University (1959; 1962). From 1965 to 1978, Dr. Long was a member of the Chemical and Nuclear Engineering Department faculty at the University of New Mexico.

In addition to teaching at the University of New Mexico, Dr.

Long served as the Assistant Dean of the College of Engineering (continued next page)

. , . . . , ,. - - - - ---+ - ,. ,.. -,,.. , - - - , , - - - . . - , - - - , _ . . .- ,- . . . , , -.- --,. .-.

l o i 1978. In February,1980, Dr. Long became Director-Training and l Education. He is familiar with personnel involved in GPU's nuclear activities, and understands the nature of and basis for 1

the changes in the training organization since March of 1979.

Id., at 8-9.

121. The site Training Department is responsible for all classroom training of TMI-l personnel. It is subdivided into four sections: Operator Training, Technician Training, Training and Educational Development, and Administrative Support. Each section is administered by a Supervisor. The Manager of Training directs the activities of the sections and provides a liaison with the station organization and the Director of Training and Education. The overall training effort is coordinated b tween e the Training Department and the (continued)

(1972-1974), the Acting Chairman of the Chemical and Nuclear Engineering Department (1974-1975), and the Chairms.n of the Chemical and Nuclear Engineering Department (1975-1978). Dr.

Long has also actively participated in a large number and variety of research projects and industry consultancies while teaching. Long, et al., ff. Tr. 12,140, at 7.

i Dr. Long has over fourteen years of reactor operating experience including two years as an operator of the Argonne Thermal Source Reactor, two years as Facility Supervisor of the White Sands Fast Burst Reactor, one year as senior reactor physicist on the United Kingdom Pulsed Experimental Reactor, one year as Associate Reactor Engineer at the Con Edison Indian Point Unit 1 Power Reactor, and eleven years as an AEC/NRC licensed SRO on the UNM AGN-201 Training Reactor.

In addition he spent one year with the Electric Power Research Institute as project engineer with responsibility for managing projects in availability engineering and development of utility power system data bases. Id., at 8.

l

- m - --------e e- .-..---ra,.c-.----e w e,-w------<-----2 ,9.--.----.c--g-s- pcw wme- ,,,,--v++. . - - , , - - , . - , . . ,e-e,.9y e,,, , , .y.,,g p- ,ew-,.-i-.-.,,,-o,e-- w---,-r-i,---

slll y '

j functional operating organizations through lesson plan review, and approval by and frequent meetings with the cognizant supervisory personnel. For example, the Manager of Training and the Supervisors of Operator Training, Technician Training, and Training and Educational Development meet once every six weeks with the TMI-l Operations sad Maintenance Director, the Managers of Plant Operations and of Plant Maintenance, the Director of TMI-l Plant Engineering, and other designated individuals to discuss and coordinate the training to be conducted in the next six-week cycle. Additional, more frequent meetings are conducted for specific working-level considerations of operator, maintenance, radcon, chemistry, and other training programs. Id., at 9-13.

122. The current Manager of Training at TMI-l is Dr.

Ronald A. Knief, who joined the organization in June of 1980.

Dr. Knief's educational and technical credentials provide to the TMI-1 Training Department important leadership in which university-type educational expectations are combined with necessary practical experience and understanding.15 His 15 In 1967, Dr. Knief graduated with highest honors from Albion College in Michigan with a Bachelor of Arts in Physics, Mathematics and Economics. In 1972, he received his Ph.D. in nuclear engineering from the University of Illinois at Urbana-Champaign. While studying for his Ph.D., Dr. Knief was a Woodrow Wilson Kational Fellow (1967-1968), and a United States Atomic Energy Commission Special Fellow (1968-1971).

From 1972 to 1974, Dr. Knief was employed as a senior physicist for Combustion Engineering, Incorporated, doing reactcr physics

, and computational analysis. Dr. Kniaf began formal university teaching in 1972 when he worked as an adjunct faculty member in the Physics Department at the University of Hartford, In 1974, (continued next page)

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

extensive educational background includes teaching graduate-level courses in general nuclear engineering, reactor i

theory, radiation detection and measurement laboratory, reactor design, and reactor safety and safeguards -- all areas with applicability to power reactor training programs. He was also an NRC-licensed senior reactor operator on the University ot New Mexico's AGN-201M Training Reactor and served as its Chief Reactor Supervisor for four years. Dr. Knief also has exten-sive experience in non-university educational activities directed to a wide range of audiences.16 Id., at 13-14.

123. Under the direction of the Supervisor of Operater Training, the TMI-1 reactor operator training program (continued) he became an assistant professor at the University of New Mexico, in the Department of Chemical and Nuclear Engineering, and in 1977, associate professor. Lcng , et al. , ff. Tr. 12,140, at 13-14.

16 Dr. nief prepared and taught a video-taped introductory nuclear engineering course at Sandia Laboratories for heterogeneous audiences that included technicians, scientists and managers with educational backgrounds ranging from associates degree to Ph.D. He has served on the faculty for three International Training Courses on physical protection and material accountability for nuclear facilities. Dr. Knief was director and lead instructor for seven short courses on Nuclear criticality Safety. As a consulting fuel-facility inspector for the U.S. Nuclear Regulatory Commission's Region III Office, in summer 1977, Dr. Knief conducted a training program on l general criticality safety for headquarters staff and a course on computer calculations for fuel-facility inspectors. He also has extensive experience in public-education activities as director of three U.S. Department of Energy sponsored l

programs--two Summer Energy Workshops for high school science teachers; and the " Citizens Workshops on Energy and Environment" for New Mexico (including over 200 presentations to high school classes and various civic organizations). Long, et al., ff. Tr. 12,140, at 14-15.

l r

V

. . .t c /

is being conducted and continuously upgraded toward higher company standards, which also reflect changes in industry standards. The Operator Training Section is divided into cwo groups for the licensed and nonlicensed training curricula, respectively. It is the goal of Licensee to ensure that prior to restart, all operators have received a thorough education which includes the lessons Iaarned from the TMI-2 accident and subsequent analyses. Moreover, Licensee intends its current and future operator training curricula to alleviate the l concerns raised by the Kemeny Commission, the Rogovin Inquiry, and other investigations with respect to necessary training and education-based criteria prerequisite to qualifying individuals l

to operate a commercial nuclear facility. This is clearly not a fixed target. Rather, Licensee's operator training program will provide to operators all of the fundamentals necessary to operate the plant ander normal and transient conditions. In order to accomplish this task, the program must have sufficient flexibility to incorporate new material on short notice. At the same time, a basic but thorough curriculum must be estab-lished and maintained. Id., at 16.

124. The Supervisor of Operator Training at TMI-1 is Samuel L. Newton, who joined Licensee in April of 1980. Mr.

Newton has a B.S. from the United States Naval Academy and an M.S. from the Naval Post Graduate School. He has twelve years of nuclear experience in the Navy's program, where he qualified in positions roughly comparable to that of an SRO or a super-

{ visor of operations. It is Mr. Newton's responsibility to i

i develop and institute effective auxiliary operator, CRO, and SRO training programs at TMI-1. Mr. Newton reviews the course outlines developed by the thirteen instructors under his suoervision, and in general coordinates the operator-related training activities at TMI-1. In order to accomplish this task, Mr. Newton and/or his Group Supervisors meet regularly l with the Manager of Plant Operations and the shif t supervisors.

Id., at 17-18.

125. The training of licensed control room operators and senior reactor operators (the latter including shift foremen and shift supervisors) as well as the non-licensed auxiliary operators, is continuous; that is, initial training is always followed by requalification training or by training for new positions within the shif t operating staff. This is in contrast to qualification programs established for other professions where it is possible to study once, receive a degree cace, and/or pass tests once to permanently qualify for the job. The operator's education continues as long as he I

holds a position on the shif t operating staff. This is true for both licensed and nonlicensed (i.e., auxiliary) operators.

Id., at 18.

126. Auxiliary operators are the most junior members of the operating staff. There are approximately 42 auxiliary operators employed at TMI-1. Thcse individuals are divided into three levels, C, B, and A, based upon senicrity, associated training, and level of responsibility. In order to

be admitted into the auxiliary operator training program, an individual must have graduated from high school, with a course in algebra, or hold an equivalency certificate. Prospective f

auxiliary operators are interviewed by the Manager of Plant operations, who considers their maturity and their potential for advancement through the shif t operating chain of responsi-bility. During his first 90 days as an auxiliary operator, an individual is on " probation" and can be removed from the program by the Manager of Plant Operations for unsatisfactory performance. Id., at 19.

127. Licensee's auxiliary operator training program is two years in duration and begins with approximately one year of classroom instruction followed by approximately one year of on shif t on-the-job training and experience. The trainee must satisfactorily complete the entire two yuar program prior to classification as a fully qualified Auxiliary Operator "A".

During the initial (classroom) phase of the program the Auxiliary Operator "C" attends lectures, receives assignments, has specific study periods and is periodically tested for eight hours each day, five days each week, in a large number of subject areas: nuclear power orientation, basic math, basic nuclear concepts, reactor physics, fundamentals of heat transfer and fluid dynamics, mechanical equipment construc-tion / operation, radiation protection, chemistry and water / waste treatment, electrical fundamentals, instrumentation and operational analysis, procedures, fire brigade training, plant l

-100-

safety, and the primary and secondary systems of the reactor.

Examinations are given to these operators at the end of each topical section. Individuals who fail (less than 70%) are reexamined within a week. Those individuals who fail two final examinations on the same topical section are dropped from the program and returned to their prior job. In addition, auxil-1ary operators in training take a comprehensive exam at the end of the classroom training program. The minimum passing score for this exam is also 70%. A retest can be taken up to two weeks later. If an individual fails to pass the exam a second time, he is interviewed by the Supervisor of Operator Training for purposes of determining the appropriate course of action.

Id., at 21-22.

128. Auxiliary operators are trained on a particular reactor, i.e., TMI-1. At the end of the extensive classroom training program, auxiliary operators work on shif t as trainees. In this capacity individuals receive on-the-job training (0JT) in the plant systems of approximately 12 months duration. Because on-the-job training by definition occurs within the facility, it is administered by the TMI-1 Operations Department; however, the shift supervisors and shift foremen overseeing this program utilize check-off/ sign-off sheets prepared by the Operator Training staff. These check-off  ;

assignments exercise and test the auxiliary operator in systems, procedures and practical factors associated with the plant. Id., at 22-23.

-101-1

129. During the final four wecks of OJT, oral exams are ccnducted by Operations personnel and written exams are prepared and administered by the Training Department. The oral {

t exams must be satisfactorily passed on the basis of materials prepared by Training. If an auxiliary operator receives an

" unsatisfactory," he can retake the exam once. Failure to I

i receive a " satisfactory" on the second attempt results in dismissal from the auxiliary operator ranks. The written exam covers the plant's systems and procedures, as well as reex-amination in fundamentals of heat transfer and fluid dynamics, mechanical equipment construction / operation, radiation protec-tion, and chemistry anc water / waste treatment. The passing grade for the written exam is 70%. Like the oral exam, the written exam can be retaken once; if an auxiliary operator fails a second time, he is dropped from the program. Id., at 23.

130. Once auxiliary operators complete their initial j training, they then participate in one week of retraining every six weeks for the duration of their tenure as auxiliary operators. The retraining program consists of reviews of important material taught in the initial program as well as lectures in subject matter recently introduced in the auxiliary l

operator training program and new developments in the industry l with which auxiliary operators should be familiar. Id., at 24.

131. To become a reactor operator, an individual must l meet certain educational requirements, participate in an

-102-

extensive training program, and pass the NRC-administered reactor operator license examinations designed for the TMI-1 facility. Licensee's reactor operator training program has changed significantly since the accident at TMI-2, in response to new corporate policy, lassons learned from the accident, anticipated NRC requirements and INPO guidelines. Elements of the program include the initial or replacement reactor operator training program; the Operator Accelerated Retraining Program (OARP)17, a one-time program designed and implemented for the unique circumstances of the long shut down of the TMI-1 unit; the reactor operator requalification program; the senior reactor operator training program; and the senior reactor operator requalification program. Id., at 25.

132. Under Licensee's licensed operator program, a candidate CRO participates in a nine month training program consisting of four phases. Phase one will be six weeks of classroom training mainly in primary, secondary and support systems. During this phase weekly written topical exams will be given, with a passing grade of 70% required. For all failures, reexams will be given within two weeks. Failure of l

1 the second written examination will require the Manager of Plant Operations and the Supervisor of Operator Training to evaluate the student's performance and decide on the corrective action to be taken. Id., at 27.

17 The OARP is discussed in more detail at 11 146 to 151, infra.

l

-103-l i

I l

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

133. Phase two is 12 weeks of on-the-job training which consists of mandatory completions by task sheets, oral checkouts by three levels of Operations Department personnel, I

and spot checking of student progress through oral questioning by Training Department licensed instructors. Failure of the oral checkout at the third level requires the student's shif t supervisor to review his performance and recommend corrective action to the Manager of Plant Operations. Using the same criteria as the classroom examinations, written examinations are administered every four weeks with a comprehensive written examination at the end of this phase covering all areas from phases one and two. Id., at 28.

134. Phase three of the licensed operator program is six more weeks of classroom instructicn focusing on reactor theory, heat transfer, fluid flow, and thermodynamics, inte-grated control system, transient analysis, safety analysis, mitigation of consequences of accidents resulting in core i

damage, normal and emergency procedures, and technical specifi-cations. Weekly written examinations are administered as described in phase one. Id.

135. Phase four is further on-the-job training 1

administered as in phase two and concentrating on emphasis of those areas presented during the second classroom training period. This period also includes a minimum of three weeks of simulator training during which the startup certification examination would be administered. Id.

-104-l l

136. At the end of the nine month training period, the candidate CROs take a written and oral examination adminis-tered by Licensee. These examinations are modeled after the NRC-administered exams, and are intended to prepare the candidates for and to verify that the candidate is ready to take the NRC license exams. Successful completion of the mock exams requires an overall pass rate of 80% and no less than 70%

on each individual topic (i.e. , the same standards as for the NRC exams). Upon completion of the mock exams, the student's training files are sent to the O&M Director of TMI-l to approve NRC exams for those who passed and to decide on corrective action for those who did not pass. The records of those approved are then transmitted to the Vice President of TMI-l for final approval and transmission of an examination request to the NRC. Upon successful completion of the NRC exam, the candidates receive their licenses and are assigned to shif ts as control room operators. Id., at 31-32.

137. In accordance with corporate policies and newly issued criteria used by the NRC staff in evaluating reactor operator training and licensing, Licensee has added specific technical material to its candidate CRO and requalification training programs. Candidate CROs and licensed operators receive extensive training in heat transfer, fluid flow, thermodynamics, the use of installed plant systems to control or mitigate an accident in which the core is severely damaged, and particular topical training in reactor and plant trans-ients. As one step toward accomplishing the latter two

-105-

. . . - - - - . . - - , - .,,-,-,,,.-.v.-._.-.r__-

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

..,y.,

o training o'jectives, o Licensee has developed, through its Technical Functions Division, a transient analysis method which plots primary and secondary system pressures and temperatures and compares the multiple routes for various normal and abnormal conditions in the reactor. These computer plots aid the operators in identifying significant transient events by observing the values and trends of key parameters. This method is useful in discussing how transients affect key parameters, and how key parameters should respond to automatic or operator-initiated corrective actions. Licensee has under development a computer-assisted instructional program using this method for analyzing transients which will become part of the overall simulation training program. In addition, Licensee has contracted with B&W for B&W's Abnormal Transient Operator Guidelines manual, a fault tree diagnostic tool which is an effective aid in training operators in accident conditions and analysis. Id., at 32-33.

138. In addition to the extensive initial candidate CRO qualification training program, Licensee's TMI-l Training organization conducts an ongoing CRO requalification program.

There are six shifts of CROs, as well as other shift operating staff, with three shifts required to cover each twenty-four hour period. On any given day there are three shifts that are not manning the control room. These three shifts are divided into one off-duty shift, one relief duty shift, and one shift participating in the retraining program. In this manner, each

-106-

$. E shift spends one out of every six weeks in training.18 Id.,at 33-34.

139. The CRO requalification program is designed to keep the operators up-to-date on developments in nuclear technology that have occurred since they were originally trained; to review important subject matter in order to ensure that the operators ' level of knowledge is maintained; and to meet specific requirements established by the NRC for requalification and license renewal of reactor operators, including those in 10 CFR Part 55, Appendix A, "Requalification Programs for Licensed Operators of Production and Utilization Facilities." Id., at 34.

t I

1 140. The CRO requalification program consists of classroom instruction; twenty-seven specified control manipu-l lations, six of which are annual and twenty-one of which must 1

l be performed once every two years; and annual Licensee-administered written and oral examinations, with minimum pass rates af 80% overall, and at least 70% on each section of the exams. Although NRC regulations and Licensee Administrative Procedure 1006 specify a minimum of 60 hours6.944444e-4 days <br />0.0167 hours <br />9.920635e-5 weeks <br />2.283e-5 months <br /> of classroom requalification training ee.h year, the TMI-l one-in-six-week shift cycle actually provides in excess of 200 classroom training hours per year. Id.

18 Six shifts is not an NRC requirement; the NRC requires that the plant be adequately staffed. Tr. 20,773 (Crocker).

A07-l

, - - - - - . . - - - --,.,.,-,.,..--,-,_,,,,,,,,,-.,--.,,,-..-r.. . . , , - , . , - , , , , . ,_,.,n_,,,.--,,,,-,--.n. _,,.- - . ,- ,,,.- , ,

l 1

I 141. The classroom instruction portion of the CRO requalification program is administered by the Training i Department. Numerous topica are covered in the classroom training.19 Although there are certain constant features of l

l the requalification program, such as reviewing reactor theory and normal, abnormal and emergency operating procedures, the requalification program is also the mechanism whereby new material is thoroughly reviewed with the operators and inte-grated with informat ion previously taught, and whereby the operators have an opportunity to discuss and raise questions concerning such information. In this manner, Licensee ensures that developments in the industry, particularly those important to safety, are fully incorporated by the operators into their i

understanding of the plant. Id., at 34-35.

142. Licensee verifies the effectiveness of its requalification lectures and the extent to which operators are absorbing the material reviewed by administering periodic Subject matter which is not satisfactorily evaluation quizzes.

understood by individual operators is reviewed with them.

19 These topics include principles of reactor operations, principles of thermodynamics, heat transfer, and fluid mechanics; features of facility design and unit modifications; general operating characteristics of TMI-1; instrumentation and control; safety and emergency systems; normal, abnormal and emergency operating procedures review; radiation control and safety; technical specifications and safety analysis; major I

operational evolutions such as refueling; review of regulatory requirements and procedures; mitigation of accidents involving a degraded core; and, industrial experience reviews. Long et al., ff. Tr. 12,140, at 34-35.

-108-

e

h. **

d /

These makeup sessions are concluded only when an oral or written evaluation is sacisfactorily completed, i.e., a grade of 80% or greater is received by tho operator. Id., at 35-36.

143. The reactivity control manipulations which operators must perform are undertaken while on shif t, e.g., the startup manipulations, or during their annual week at the B&W l simulator, e.g., various loss of coolant accidents. On-the-job manipulations are conducted under the supervision of a senior member of the shif t, either the shif t foreman or shif t super-visor. At the simulator, the abnormal or emergency control manipulations are observed and evaluated by a member of the B&W Training staff. Id., at 36.

144. The annual examinations administered to licensed CROs include a written and oral evaluation. The results of the exams are used to identify specific areas in which retraining is necessary to upgrade an individual's knowledge in a particu-lar area. An individual who receives less than 80% on any section of the exam attends a requalification lecture and another exam on that subject matter. If an individual receives a grade of less than 80% overall or less than 70% on any single section of the annual exam, he is relieved of his licensed duties and is placed in an accelerated requalification program.

Upon successfully passing a second written and oral exam and receiving certification of a satisfactory rating by the Vice i

President of TMI-1, the individual is returned to his licensed duties. Failure to pass the exams a second time prevents the j individual from working as a licensed operator. Id., at 36.

l l

1

-109-

l l

l 145. The TMI-l Manager of Plant Operations, shift supervisors, and shif t foremen are required to hold a senior reactor operator (SRO) license which, like the CRO license, is awarded by the NRC upon successful completion of SRO examina-tions. SROs must also be requalified every two years. In order to qualify to become an SRO, an individual must have served as a TMI-l CRO for a minimum of one year. Licensed SROs presently participate in the regular operator requalification program, but are required to have a higher level of compre-hension as demonstrated on the examinations. Id., at 37.

146. In addition to these requalification programs, in 1979 Licensee designed and implemented a special five month licensed operater training program, called the Operator Accelerated Retraining Program (OARP). The extensive technical l subject matter covered in the OARP included a review of TMI-l l

l plant systems, plant operations, radioactive materials control, and plant transient analysis. OARP specific objectives were:

to improve operator performance during small break loss of coolant accidents; to assure that operators are able to recognize and respond to situations involving inadequate core cooling; generally to improve operator performance during '

transients, including events that are exacerbated by initially inappropriate operator action; to give operators an in-depth understanding of the TMI-2 accident and " lessons learned;" to assure that operators are knowledgeable of operating procedures and actions necessary upon initiation of the engineering

-110-

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

safeguards features; to assure that operators understand the manometer effects of water levels in the reactor coolant system _

under different prersure and temperature conditions; to assure understanding of the significance of simultaneous blocking of both auxiliary feedwater trains; to assure understanding of the NRC prompt notification requirements; to provide operators with an in-depth understanding of the methods required to establish and maintain natural circulation; to assure operators are knowledgeable of both short and long-term plant system modifications; to provide operators with a review of major plant systems; to provide specialized training on operations and procedural guidance requirements; to fully requalify operators through the administration of Licensee and NRC-administered written and oral examinations; to review with operators major administrative, normal, abnormal, and emergency procedures; and to provide to all licensed Unit 1 operators extensive experience on the B&W simulator, educating them on transients which occurred during the TMI-2 accident, as well as other abnormal reactor conditions. Id.; at 38-39; see also Staff Ex. 1, at Cl-16; C6-5 to C6-7; Kelly, ff. Tr. 12,409; Gardner, ff. Tr. 12,409; Christensen, ff. Tr. 12,409; Lic. Ex.

27.

147. The OARP was instituted by Licensee on its own initiative, to emphasize the lessons learned from the TMI-2 accident and the recognized need for an extensive and thorough training program for licensed reactor operators. The OARP was

-111-

presented from August, 1979 through March 1980 to all TMI-1 licensed CROs and SROs, and the STAS in training. The program ,

\

consisted of approximately 60 individual lessons or practice sessions. Teaching techniques included classroom presenta-1 tions, TMI-l control room training sessions, and simulator training sessions at the B&W simulator in Lynchburg, Virginia.

Shif ts participated as a group; consequently, Licensee was able to focus upon both the activities of the operators, such as an individual manipulating the reactor controls, and the aspects .

of operations which involve team effort and coordination. The program was divided into seven subject area modules including one week at the simulator. Each module consisted of four to five days of training, eight hours each day. Subject matter included the traditional areas of review, such as plant systems and radiation monitoring. However, particular emphasis was placed upon accident and safety analysis. Id.

148. At the end of the program, OARP participants took written and oral examinations designed and administered by PQS Corporation, an independent consulting firm which provides reactor training program reviews and audits. Individuals who did not score above 70% on any section of the exams, or who failed to achieve an overall score of 80% received remedial requalification training. Id.

149. An additional one week training session on Decision Analysis was given to all SROs and STAS. Decision Analysis trains individuals to handle complex situations for

-112-l

which written procecures do not exist; to develop a technique 1 to cope with uncertainty, stress, and conf.licting information, I and to make decisions in the face of such circumstances; and to ~

make " good" decisions, i.e., to consider fully and understand the significance of alternatives, and to factor in the most

-important considerations. Decision Analysis training develops in control room supervisory personnel--i.e., SROs and STAS--the tools and sensitivity to make the right decisions under highly adverse circumstances, and to do so in a systematic and thoughtful manner. Long, et al., ff. Tr. 12,140; Gardner, ff.

Tr. 12,409, at 13; Christensen, ff. Tr. 12,409, at 11; Lic. Ex.

27, at 70-75.

150. In order to assess the effectiveness of the OARP, Licensee sought the assistance of five highly qualified individuals,20 expert in relevant fields, to conduct a review analogous to accreditation reviews carried out by professional organizations, such as the Accreditation Board for Engineering 20 These individuals were Dr. Julien M. Christensen, Director of the Human Factor Division, Stevens, Scheidler, Stevens and Vossler, Inc., Dayton, Ohio, representing human factors engineering; Dr. O.ic F. Gardner, Professor of Psychology and Education of Syracuse University, Syracuse, New York, representing educational psychology; Dr. William R. Kimel, Dean l

of the College of Engineering at the University of Missouri, I

Columbia, Missouri, representing nuclear engineering education; Mr. Richard J. Marzec, Manager of Technical Training for Duke Power Company, Charlotte, North Carolina, representing nuclear power plant operator training; and Dr. Robert E. Uhrig, Vice President, Advanced Systems & Technology for Florida Power &

l Light Company, Miami, Florida, representing nuclear power generation. Long, et al. ff. Tr. 12,140, at 41-42.

-113-

r-and Technology (ABET) which accredits university engineering degree programs. These five individuals, who comprised the l

OARP Review Committee, familiarized themselves with the basic l philosophy of accreditation, including the quality required of I

an engineering program in order for it to become accredited; l i

attended OARP classes; evaluated the proper role of simulators in an operator training program; and evaluated the OARP in light of NRC requirements and " lessons learned" from the TMI-2 accident. Long, et al., ff. Tr. 12,140, at 41-42; Lic. Ex. 27.

151. Numerous comments and suggestions were made by the OARP Review Committee in its Report. See Lic. Ex. 27, at 135-49. Upcn completing its review, however, the Committee stated:

The conclusion of the Committee was that the Operator Accelerated Retraining Program carried out by Metropolitan Edison was a high-quality, well-executed program, having many features which should be incorporated into the regular Operator Retraining Program.

The Committee further believes that personnel who demonstrate satisfactory performance in l

the OARP should perform well on the NRC Reactor Operator and Senior Reactor Operator Examinations.

Lic. Ex. 27, at 3.

152. We turn now to Licensee's training programs for those other than operations personnel.

153. Consistent with its commitment to an ongoing formal training program at TMI-l for employees delegated with safety-related responsibility for the unit, Licensee has i

expanded its Training organization to include, under a

-114-I

. . = . , , --...w --,,e--.n. w~ ,-------r-, s-w xe~<v-*-~,-*'~~r*=v~**~- cv'*"'-* ' ' ~ * * ' ' ' " * * ~ * * ' " ' ' ' ' ' ' ' ' ' ' ' ~ ' ' ' ~ ' ' " ~ ~ ~ " ~ ' ' ' ~ ' * ' " ' ' ~ ~ ~ ~ ~

Supervisor, a Technician Training Section, and has embarked upon a number of new formal programs for training and re-training TMI-1 technicians in a variety of fields.' In addition to the Supervisor of Technical Training, eleven instructors and four group supervisors are employed by Licensee in Tecnnician Training, each with a minimum of three years experience in his or her field of inaccuction or a closely related field. Under the overall direction of the Supervisor of Training and the Technician Training Supervisor, the instructors within the Technician Training organization teach courses in the areas of maintenance, radiological controls, chemistry, security, emergency plan implementation, and general employee training.

Long, et al., fi. Tr. 12,140, at 43-44.

154. The Maintenance Technician Training Program is designed around a six shif t schedule and provides for up to 320 hours0.0037 days <br />0.0889 hours <br />5.291005e-4 weeks <br />1.2176e-4 months <br /> of training and retraining each year. The program is designed to provide a progressive update for the technician by review of basic concepts and basic skills and to provide instruction in plant systems and advanced specific systems, l components and concepts. During the first section or basic Phase of the Maintenance Training Program, all maintenance personnel are given an appropriate self-study course consisting of fundamental lessons on maintenance tools as well as the major principles involved in mechanical, electrical, I&C, and general maintenance work. Quizzes are given in each area of the subject matter as it is completed by the trainee with 1

I

-115-

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

provision for repeating the area if weak or unsatisfactory performance is exhibited. The second part of the first year of the Maintenance Training Program is subdivided into systems training, during which Training gives to Maintenance personnel a functional explanation of selected primary and secondary systems; and constituent theory, during which Maintenance personnel also receive training in their respective discip-lines, e.g., electrical maintenance. Phase three of the Maintenance Training Program consists of a review of basic skills, and the use of relevant instruments, e.g. , those needed for setting breakers. Finally, during subsequent years of the program, maintenance personnel receive advanced instruction in systems, components and concepts relating to their discipline and level of accomplishment. Id., at 44-45.

155. The Radiological Control Training Program directed by the Technician Training Section consists of two training programs: the Initial Technician Training Program, and the Technician Training / Retraining Program, called the Cyclic j r:ogram. At a minimum, Technician Training ensures that the i

radiological control training programs for TMI-1 meet the l

requirements "pecified in F.J1 N18.1 55.5 (1971). Id.; see also Heward, et al., ff. Tr. 16,292, at 6-7, and Attachment 1, at 10.0 to 11.0; Wegner, ff. Tr. 13,284, at 24; Tr. 16,317-18 (Potts).

l 156. The Initial Technician Training Program consists of approximately eight weeks of training which every new

-116-1

C radiological control technician must undergo prior to assuming responsibility for radiation control work at TMI-1. Basics of l

radiological control are taught in classroom sessions. A '

comprehensive examination is given at the end of the program.

Once a radiological control technician is assigned on shift at TMI-1, he or she participates in the Cyclic Program, which is l conducted continuously over a minimum of a forty-two week period, with each shif t rotating through the program for one week every six weeks. The initial Cyclic Program consists of three weeks of review of basic health physics principles and plant instrumentation. For the next two weeks, radiological controls technicians review the TMI-l emergency plan, and their role in the event of an emergency. A week is spent studying plant systems. Finally, problem areas in radiological controls are reviewed. During subsequent yearly training cycles, Licensee intends to continue to provide progressive training to j

assure that the technicians maintain their technical and professional abilities. Long, et al., ff. Tr. 12,140, at l

45-46. See also Heward, et al., ff. Tr. 16,292, at 6-7, and i

Attachment 1, at 10.0 to 11.0; Wegner, ff. Tr. 13,284, at 24.

157. The Chemistry Training Program taught by the Technician Program Training Section also consists of an initial Technician Program and a cyclic training / retraining program.

The newly hired technician receives nine weeks of classroom

( lectures followed by three weeks in the laboratory. This program must be completed prior to receiving a shift assign-ment. Included in the initial chemistry tr i ia n ng program is a

-117-S

- c-e,--w.1---*ym--------mee,----ev e n , y , .,-.-,ri.,,,,.-,--eqv.-,-.mw-.-,.., ..,--e- or--s,,,e-.--or--.m..e, w -

review of basic chemistry principles, e.g., titration; of laboratory procedures, e.g., the use of a multi-channel ~

l analyzer; and of TMI-l water chemistry specifications, includ- )

ing an explanation of why certain chemicals are necessary while i

others are not allowed into the reactor primary and secondary systems. Like the other cyclic programs, the chemistry retraining program consists of a review of important facts and principles, with emphasis on changes in relevant plant proce-dures. A progressive training program will be provided to maintain the technical and professional capabilities of the chemistry technicians. Long, et al., ff. Tr. 12,140, at 46-47.

158. The Security Training Program, formerly taught by the Security Department at TMI, is now under the direction of the Technician Training organization. In accordance with the requirements of Reg. Guide 5.20 (1974) and 10 CFR Part 73.55, Appendix B, Licensee reviews with its new security personnel the TMI-l security plan, legal issues of import, how I

security equipment works, methods to use in the event of a i

security alert, and other pertinent information. Id., at 47.

159. Basic General Employee Training (GET) consists of a four hour course for individuals working on Three Mile Island. During this sessian, employees as well as contractors working at TMI-1 are lectured on procedures in the event of a site emergency, basics of health physics, and basics of radiation, as specified in ANSI /ANS 18.1 (1971). Every person taking GET must receive at least a seventy percent on the

-118-I

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

o written exam administered by Licensee in order to be badged to work at TMI-1. Id., at 47-48; see also Heward, et al., ff. Tr.

16,292, Attachment 1, at 10.0.

160. In addition to basic GET, individuals who require access into radiological controlled areas receive an additional eight hours of training. This supplemental GET consists of a much more detailed review of radiation protection principles and procedures; four hours of practical facters training, e.g., procedures to follow in donning and removing the radiation protection clothing required in certain areas; and methods for minimizing radwaste. Written and practical exams are also given on this portion of the GET Prcgram. An individual is not allowed to work in a protected area until he or she receives at least seventy percent on this exam. Long, et al., ff. Tr. 12,140 at 48; see also Heward, et al., ff. Tr.

16,292, Attachment 1, at 10.0.

161. The annual retraining GET Program consists of a review of emergency-related procedures, followed by a written i

test or tests similar to those originally taken. If an individual does not receive the required seventy percent score

, on these exams, the entire applicable portion of the GET l Program must be repeated. Id.

162. In conjunction with GET, Technician Training conducts special sessions on fire brigade procedures and duties for personnel with fire brigade responsibilities. This instructicn is given by a certified Pennsylvania fire

, instructor. Long, et al., ff. Tr. 12,140, at 48.

119-y- .- --+,% - -- .,,.--,..c4 ,-,,,,we-,,,w,, e, p- . ,m-e-w-e--ry. -$-,.cw+ , , + - =,.,,----w,-w,,,-,-y3-+-,--+,- -,-. -r--,-. - - - -ee - -

o 1 l

163. Radwaste personnel receive additional training, )

taught by the Technician Training section, on shipping and receiving radwaste in accordance with the criteria set forth in I&E Bulletin 79-19. Id., at 48-49.

164. The third branch of the TMI-l Training Organization is the Training and Educational Development Section. Id., at 49. This is an administrative and service group which provides TMI-1 personnel resources fer additional technical and management training, is responsible for Shift 4

Technical Advisor (STA) training, and ensures that sound educational principles are used throughout the Training Department both in material development and lesson presenta-

tion. Id., at 49-50. A major focus of Training and Educa-tional Development is to actively encourage TMI-l professionals such as site engineers to enhance their capabilities, both technically and as potential management personnel. Id., at 50.

2 In order to promote degrees among on-site professionals (regardless of whether it is adopted by the NRC as a recom-i mendation or a requirement), the TMI-l Training and Educational Development Section is actively seeking college level educa-tional options for TMI-l employees. Id. Already, Training and Educational Development has been successful in contracting for a basic nuclear engineering course and a graduate level reactor shielding course at the Capitol campus, attended respectively by thirty-five and fif teen TMI staf f members including opera-tors (AOs, CROs and SROs). Id., at 50-51. Evening courses,

-1?0-

e'

    • 4*\@9 fg f A>Y

/f k[+>+y,,

/g,T,/e/

T,ST TARG,T (MT-3) s's l.0 En BM jl"NM i.i siis na .

- ! "' j l.8 1.25 1.4 l.6 I

i

/ = ._ s- =

4% 4 4 '4 Nf,,,kf 3  %,4,,(

W Mp 4 A.

  1. g,d ...eEe_ T.

%,z,*4 TEST TARGET (MT-3) l.0 M EM ifM y 'y lil11 I.l%,imRh8

- j .g 1.25 1.6

]i.4 1

l/ < 6" >

l l

  • il/*iy +hsp
  • kV,,,h %d.;W

_ m

too, have been instituted at TMI, taught by faculty members from Jarrisburg Area Community College, e.g., a course in the principles of management. Id., at 51.

165. The second area for which Training and Educational Development has major responsibility is in the development and Laplementation of the STA Training Program.

Id. Part One of the STA Training Program consists of a thorough exposure to TMI-l reactor specific material. Id., at

52. Like the training programc developed for the augiliary and the licensed operators, detailed attention is paid to individual systems on both the primary and secondary sides of the plant. Id., at 52-53. Also, lectures and study are devoted to auxiliary systems, engineered safeguards, instrumen-tation and control, electrical systems, procedures, docu-mentation (including Technical Specifications, the QA Plan and the Emergency Plan), accident analysis, unusual event recogni-tion and responses, B&W simulator training, nuclear power fundamentals, and health physics. Id., at 53. In addition, STAS participated with the operating shif ts in modules two through six of OARP (they reviewed by videotape OARP module l one), and in Decision Analysis and related command training.

Id.

166. STA courses are taught by an instructor i

dedicated solely to the STA Training Program, as well as by individuals with designated expertise, e.g., health physics, emergency planning implementation, simulator training. Also,

-121-

v o /

self-study is used on certain portions of the program, e.g.,

systems, procedures. Id.

167. By January 1, 1982, TMI-l STAS Mill have completed Part Two of the STA Training Program, which consists of college level courses in subject matter of particular importance. The areas to be covered are math, chemistry, physics and reactor physics, thermodynamics and heat transfer, control systems and electrical fundamentals fluids, communica-tions and materials. It is Licensee's intention to conduct a thorough ongoing requalification program after the STAS complete the initial two years of training. Id.

168. The third area for which Training and Educa-tional Development has major responsibility is in ensuring that sound educational principles are used in the development and presentation of TMI-l training material. All permanent TMI-l instructors are required to attend a week-long instructor development program. The primary goal of this program is to furnish the instructor with an expanded foundation in training design, presentation, and evaluation to enable him/her to conduct effective employee training consistent with program goals and objectives, This program is conducted in a workshop l atmosphere with extensive student participation. A modified version of this course is planned for annual retraining. Id.,

at 54.

169. Three independent consultants who have been involved in reviewing Licensee's training programs testified on

-122-i l

the adequacy of Licensee training. See Gardner, Christensen and Kelly, all ff. Tr.12,409. Their involvement with TMI-1 and their testimony were directed at licensed operator training. Id. We now turn to discuss their views of Licensee's programs.

170. Mr. Frank Melly, President of PQS Corporation, is a consultant on power plant staffing and training, with over twenty-six years in the nuclear industry, engaged in the star tup and operation of nuclear power plants, the administra-tion of nuclear power plant training programs and the eval-uation of nuclear power plant operator proficiency.21 Kelly, ff. Tr. 12,409, Appendix A. Mr. Kelly's initial involvement with Licensee came in April of 1979, when PQS was requested by Licensee to review the operator training and requalification programs at Three Mile Island. Based upon this review, PQS concluded that Met Ed's operator training and requalification programs were representative of the industry, but could be improved in certain respects. These included (1) an expansion 21 Mr. Kelly graduated from the University of Vermont in 1954 with a Bachelor of Science degree in Electrical Engineering and completed graduate school course credits in Nuclear Engineering at the University of Pittsburgh in 1959. His industrial career spans the period 1954 through the present, with key assignments being Chief of the USNRC Operator Licensing Branch (then USAEC) from 1967 to 1969 and later, manager of all Westinghouse Electric Corporation nuclear training programs prior to founding the PQS Corporation in 1979. He is a registered professional nuclear engineer and a member and past Secretary of ANS-3, the American Nuclear Society's main standards writing committee on the conduct of operations at nuclear power plants.

Kelly, ff. Tr. 12,409, Appendix A.

l 1

-123-1

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

1 1

of the training organization, (2) certain modifications to the requalification program, (3) increased usage of the B&W simulator, (4) modification of the scope of all of the Operator training programs, and (5) additional evaluation examinations prior to NRC licensing examinations. All of these recom-mendations have been incorporated into the overall TMI training programs. Id., at 3.

171. Following initiation of the OARP, PQS periodi-cally monitored the program's progress. Id., at 4. PQS audited the scope, content and schedule of the program and the program's self-evaluation process, observed classes, interviewed individual trainees and reviewed records. Id.

PQS's view was that the OARP was a comprehensive, well adminis-tered program, that operators displayed great pride in their positions as qualified operators and that, based on training and self study, these individuals would successfully complete NRC's examinations and be well qualified operators. Id., at 6.

172. In December 1979, PQS was assigned by Licensee the task of preparing and administering complete written and oral examinations to the licensed reactor operators (ROs) and senior reactor operators (SROs) at TMI Unit 1. Id. The l purpose of those examinations was to exercise the operators with an examination of the type given by NRC to operator candidates in order to prepare the operators for such an exam and to provide Licensee with a comprehensive examination of its licensed operators and candidates by an independent testing

-124-

entity. Id. In April 1980, PQS administered written and oral audit RO and SRO examinations. Id., at 6-7. The examinations were developed based on PQS's knowledge and experience with NRC _

operator examinations, were formatted similar to NRC's teca-niques, and spanned all subject areas which NRC examinations of operators would 'e expected to encompass. Id., at 7. They were broad in scope but required detailed knowledge in each of the subject areas. Id., at 7, and Attachments B and C. The PQS examination for ROs and SROs also included an oral portion similar to exams administered by NRC operator license examiners--one-to-one orals, four to five hours in duration.

Id., at 9. During the examination, each candidate was required to tour the TMI-l plant and control room, to exhibit familiar-ity with the equipment and explain how and why he would conduct (or in the case of the SRO, direct) routine, abnormal and emergency operations. Id., at 9-10.

173. PQS administered in addition to the RO and SRO examinations a separate written test to all RO and SRO can-didates convering the TMI-2 event and principles of the small l

break LOCA concept (so-called Category T examinations). Id.,

at 9. For this examination,22 a 90% grade was required to i 22 This special examination required by NRC of all B&W

! operators following the TMI-2 accident covers lessons learned from that event. It is also a subject of the August 9, 1979 Commission Order in Item 1.e. The Staff will require, prior to granting licenses to individual operators, that each of them pass this examination. Mr. Newton testified for Licensee that, excepting four, the 36 individuals who recently took the NRC licensee written examinations had already passed this exam, (continued next page)

-125-i

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

i pass. Id. PQS recommended that those examinees who failed to obtain a 90% grade be given formal classroom retraining in the .

areas of required knowledge and then be administered a written 8

examination by the TMI Training Department. Based upon the results of the PQS audit exarinations,23 Mr. Kelly testified that overall, the TMI-l licensed ROs and SROs demonstrate a high degree of knowledge of how to safely and effectively operate the unit. Id., at 10. He attributed this in large part to the scope and depth of the OARP program and to the manner in which it was conducted by instructors who profes-sionally and conscientiously taught trainees who sincerely and i

enthusiastically participated. Id. Based on his knowledge of, l

t l' (continued) administered by PQS or by Licensee's training personnel; he further stated that the Licensee understood that the remaining four, too, must pass this exam prior to receiving NRC licenses.

Tr. 20,597 (Newton); see Newton, ff. Tr. 20,580. See also Tr. 20,688-91, 20,751-54 (Crocker).

23 The statistical results, i.e., numbers of examinees and their passes or failures, on the PQS-administered examinations, as well as reexamination criteria and the statistical results of further examinations conducted by Licensee personnel and by a second outside consultant in April of 1981, consumed hours of hearing time. See generally Tr. 12,159-69 (Long, et al.);

Tr. 12,607-16 (Kelly, Gardner, Long); Tr. 12,696-711 (Kelly);

Tr. 13,212-18 (Long); Tr. 20,576-639 (Newton); and Tr. 20,717-23 (Crocker). The testimony at times was confused and appeared inconsistent, although not materially '7. _C_cmpare, e.g., Tr.

13,213-18 (Long) with Newton, ff. Tr. 20,58D and Tr. 20,613 (Newton).

As a final step to provide an accurate record, Licensee presented Mr. Newton, who is in charge of TMI-l licet sad operator training, with detailed statistics. See Tr. 20,576-639. Mr.

Newton testified that 36 operators were certified to the NRC and took the NRC examination in April,1981; 20 were SROs (only 13 of whom would nornally stand watches on shift) and 16 were ROs. Tr. 20,581-86 (Newton).

-126-

and experience with, licensed operators throughout the industry, Mr. Kelly judges the TMI-l operators collectively to be retrained and evaluated to a greater extent than the industry norm. Id.

174. Licensee's second consultant witness was Dr.

Julien M. Christensen, a consultant in engineering psychology specializing in human factors design techniques, with some thirty-five years of experience in this field. Christensen, ff. Tr. 12,409, at 1-2.24 175. Dr. Christensen's involvement with Licensee has been as one of five members of a committee unich was con-stituted for the purpose of reviewing License.e's OARP, and as a 24 Dr. Christensen holds a Ph.D. in experimental psychology from The Ohio State University and has spent the period from 1945 to 1974 in the Human Engineering Division of the Air Force Aerospace Medical Research Laboratory at Wright-Patterson Air Force Base, Ohio. Christensen, ff. Tr. 12,409, at 1. During his last eighteen years there, he was the Director of the Human Engineering Division, during which time programs were developed that form a substantial part of the content of well-known and widely used human engineering source materials such as the Human Engineering Guide to Equipment Design and MIL-STD-1472 (Human Engineering Design Criteria for Military Systems, Equipment and Facilities). Id., at 1-2. The programs that he directed have also contributed directly to the safety and effectiveness of many non-Depar'.nent of Defense programs; the new NUREG-CR-1580, " Human Engirearing Guide to Control Room Evaluation," according to Dr. Jhristensen, relies heavily on l these materials. Id., at 2.

Fr:2 1974 to 1978, Dr. Christensen served as Professor and as Chairman in the Department of Industrial Engineering and Operations Research, College of Engineering at Wayne State University. He is currently an adjunct Clinical Professor of Psychology at Wright State University, as well as a consultant.

Id. He has received numerous awards for contributions to human f actors and is a Fellow in two national societies and a Diplomat in a third. Id.

-127-

1 human factors advisor. Id., at 4-5. The approach taken by the OARP Review Committee was described briefly by Dr. Christensen in his testimony:

The general mode followed in the evaluation by the Committee was based on that of the Accreditation Board for Engineering and Technology (formerly the Engineers Council for Professional Development (ECPD)).

This model requires that the agency or institution being evaluated conduct a rigorous self-examination, based on detailed requirements established by the Committee.

It is the procedure that the departments of engineering in colleges and universities are required to follow when they apply for accreditation. It is thorough; it is objective, and has been shown to yield reliable, valid evaluations of engineering I

training programs. In my view, it was highly successful in this application at TMI-l oecause it brought together and coordinated the efforts of the Met Ed educators and a committee of outside experts to conduct a critical salf-examination.

Id., at 5; see also Lic. Ex. 27, at 4-9.

176. As the individual on the OARP Review Committee l with a human factors background, Dr. Christensen took the lead in evaluating human factors considerations in training and the simulator training. Christensen, ff. Tr. 12,409, at 4-5; Lic.

Ex. 27, at 8. In his opinion, considerations of man-machine relationships reveals that human factors engineering and training both have significant contributions to make to assure that the man-machine interactions are safe and effective. Id.,

at 6. He further observed that at TMI-l the task has been attacked from both points of view. Id. On the training side, the OARP program was a positive response that, in the judgment I

-128-l

of the training experts of the OARP Review Committee, will add significantly to the capabilities of the operators. Id., at 6-7. He exami' n ed the Licensee's simulator program and ex-pressed the belief that it, too, will add substantially to the effectiveness of the training program. Id., at 7. He noted the simulator program has been broadened not only to facilitate initial learning but also to help assure that skills, once learned, do not deteriorate and that unusual circumstances are recognized and deal' with properly. Id. Further, he stated revision and additional development of the simulator part of the training curriculum should significantly enrich the program and serve as a source of positive motivation to the operators.

Id.

177. With respect to the other side of the coin (hu ~c.  ; engineering), Dr Christensen traced the activities of Licensee to evaluate the TMI-l control room from the operators' point of view, characterizing Licensee's approach as vigorous. Id., at 8. In his view, the result of this dual consideration of human factors and training should have a very positive effect on operator capability and should help assure an operation that is both safe and effective. Id.

178. The third consultant witness on training who appeared for Licensee was Dr. Eric Gardner, an educational psychologist with special expertise in educational and psycho-logical measurement, psychometrics, test construction, and curriculum and program evaluation. See Gardner, ff. Tr.

-129-

.-._m.- ~w -.--,.._---e-..,,---..-.----.,..,-.__-__,,-w_e.,~,, --

, , - , . . , _ , - . , ,._--.,_-e.,-,--_,

- - - - v.--,-.,. +-r--,

o 1

12,409, at 2. Dr. Gardner was Professor of Education and Psychology and later the Margaret O. Slocum Professor of Education and Psychology at Syracuse University from 1947 to 1978; he presently is Margaret O. Slocum Professor Emeritus at Syracuse and an educational psychological consultant.25 Id.,

at 2, 4.

25 During his career at Syracuse University, Dr. Gardner was Director of the Division of Advanced Studies in the School of Education from 1952 to 1961, Chairman of the Psychology Department of the College of Arts and Sciences from 1961 to 1976, and Director of the Psychological Services Center from 1961 to 1976. During these periods, he was called upon to participate as a member (and occasionally as chairman) of evaluation committees set up by the National Council for Accreditation of Teacher Education, Middle States, Southeastern

! States and other accrediting agencies to examine and/or accredit the educational programs in various colleges and universities. Among the institutions visited as a member of such committees were the University of Pennsylvania, the l College of William and Mary, the University of Buffalo, Cornell University and Indiana State College. Gardner, ff. Tr.

12,409, at 2-3.

I Dr. Gardner is the author or co-author of three books including a textbook on educational psychology, over twenty tests published by Harcourt Brace Jovanovich, ten chapters in technical books, and about fifty articles published in professional journals. He has been an active participant in the professional organizations in his field and has served as a member of the executive committees of the American Psychological Association Divisions of Educational Psychology and Evaluation and Measurement, and also as President of the Divisi-n of Evaluation and Measurement and President of the National Council on Measurement in Education. He is a Fellow l in the American Psychological Association and a Certified Psychologist in New York State. Id., at 3.

Dr. Gardner has served on a number of national, state and private institution committees. Included are memberships on the National Research Council Committee on Aviation Psychology (1951 to 1958), College Entrance Examination Board Committee on the Scholastic Aptitude Test (1956 to 1966), U.S. Office of Education Cooperative Research Committee (Chairman, 1956 to 1959), the New York State Committee on Proficiency Examinations and the New York Regents' External Degree Program Advisory Committee (1974 to the present) . Id.

-130-

S.

  • l
e. e =l 179. Dr. Gardner, in his testimony, provided his views on the training program established at TMI-l for NRC-licensed nuclear reactor operators, particularly in light of his participation in the TMI-1 Operator Accelerated  !

l Retraining Program (OARP) Review Committee. Id., at 4. He j explained that his duties as a member of the OARP Review Committee were to participate in all Committee discussions to insure that adequate attention was given to the educational and psychological aspects of the training program and to personally make the necessary observations of the various aspects of the  ;

program including those associated with teaching-learning cituations. Id., at 4-5. Further, he explained that all aspects of the final report were discussed by the Committee as a whole, but that his particular responsibility was to write

, Chapter Six, Evaluation of the Operator Accelerated Retraining I

l Program, and the section in Chapter Ten on educatio..al and training processes. Id., at 5-6; see ,also Lic. Ex. 27, Chs. 6,

10. Dr. Gardner was impressed both with the method Licensee utilized in planning the OARP anu the implementation of the Program. Id., at 5-6. Although, for example, he admitted an i

initial bias against the single oral examiner technique, he testified that he was convinced that the examinations given by the PQS team were effective. Id., at 6-7.

180. Dr. Gardner also provided his opinion on whether op rator mindset 26 would pose a problem to TMI-l licensed l

26 See n.1, supra.

l

-131-

? .

o operators. He explained that mindset is a term often used in general conversation to indicate a person's general feelings about a situation or issue. Since it is impossible to measure or determine directly the specific physiological activity taking place within the brain that causes observable behavior

, (either mental or physical), knowledge about mental functioning I

(ine'uding the measurement of intelligence) must be inferred from observaole responses related to specific stimuli presented to the person. Hence the term " response set" rather than "mindset" is a teru usually used by psychologists when the discussion involves training or educating. Response set may be related to a variety of stimuli and the resulting desirable behavior is the aim of either training or education. Id., at 9.

181. According to Dr. Gardner, the safe and effective operation of the plant by nuclear reactor operators requires them to respond to a variety of stimuli to which the appropri-ate response varies from an immediate S-R (Stimulus-Response) response to a delayed response based on higher level cognitive processes. A satisfactory training program requires a combina-tion of training in the narrow sense and education as prepara-tion for more complex situations. He pointed out that this issue was discussed in detail in the report of the OARP Review Committee for Three Mile Island (Lic. Ex. 27, at 115-17), and summarized the Committee's discussions. As he explained it, l training emphasizes mastery of scecific tasks through drill and i

-132-

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

+ . .- . - ** -

o 1 practice; education, on the . other hand, involves an open set of j operations, eventualities which cannot be entirely anticipated, and possibilities too numerous to be learned individually.

Emphasis is upon the transfer of knowledge to new situations through an understanding of concepts, and an acquiring of skills, attitudes and values. Id., at 9-10. Dr. Gardner then identified response sets that need to be established so that action by control room operators will ensure safety of opera-tion under all conditions, pointing out how the Committee considered each of them in passing favorably on Licensee's OARP. Id. at 10-13.

Dr. Gardner concluded:

. . . the OARP for retraining nuclear reactor operators for Three Mile Island-Unit One has been carefully developed and imple-mented to be consistent with effective educational and psychological principles. I agree with the other members of the Review Committee that the completed Operator Accelerated Retraining Program for TMI-l operational personnel and the addition to the shift operating staff of Shift Technical r

Advisors, who are degreed engineers, provide a blend of training and education that should result in the safe, reliable operation of TMI-1.

Id., at 14.

182. In its initial evaluation report of Licensee's

, readiness to restart Unit 1 (Staf f Ex. 1), the Staff reviewed 27 The NRC Staff's witness had a similar view of the steps taken by Licensee (training, revised procedures, emphasis of the shift supervisor's responsibilities, and on-shift STAS) .

l to prevent the development or presence of a mindset among TMI-l Operators. See Boger, ff. Tr. 12,772, at 10. There

[ was no contrary evidence.

-133-

I i

the requirements for training and Licensee's training activities. See, e.g., Staff Ex. 1, at C6-5 to C6-7 (Training of operating staff). The Staff pointed to Licensee's OARP, previous TMI-1 operator training programs and the audit and NRC examinations as factors that ensure enhanced operator perform-ance and the safe and effective operation of TMI-1. Id., at C6-7. Based on Licensee's operator retraining efforts covered by NRC 's rr quirements , on the improved capability of Licensee's training organization, on the additional retraining performed under the OARP, and on the requirement of satisfactory perform-ance on the NRC license sgaminatiens, the Staff concluded that training of the operations staff is adequate and that Licensee has complied with the requirements of the Commission's August 9, 1979 Order in this regard. Id.; see Tr. 20,751-54 (Crocker). The Staff pointed out, however, that it expected Licensee to provide revised training programs to meet then-developing criteria, which would be subject to later NRC review. Id.

183. In its later management supplement to the initial education report (Staff Ex. 4), the Staff observed that Licensee's Revised Licensed Operator Qualification and Requalification Training Program had been submitted by Licensee. Staff Ex. 4, at 21. The Staff stated, however, that review and approval of the revised program was not required prior to restart since all of the licensed personnel at TMI-l will be required to successfully pass an NRC examination prior to resuming their licensed duties at the operating plant. Id.

-134-l

184. In Supplements 2 and 3 to its evaluation report (Staff Exs. 13 and 14), the Staff addressed the training of unlicensed personnel which had not been complete at the time the initial evaluation report was issued. See Staff Ex. 1, at 10, 24, 38. The Staff reported that based on its review of the Training Department and the development of acceptable proce-dures for training unlicensed personnel, Licensee's unlicensed personnel training was acceptable and those persons (non-licensed) whose activities could have an impact on safe plant operation will be well qualified; the Staff also reported as acceptable Licensee's STA training program. Staff Ex. 13, at 4; Staff Ex. 14, at 46-47. The Staff also in Supplement 2 concluded that Licensee has made adequate plans for training of licensed operators for the long term, another item earlier identified by the Staff (see Staff Ex. 4, at 11 and 38) as requiring further review. Staff Ex. 13, at 5.

185. On the basis of the extensive record developed on training , the Board finds that Licensee has in place at TMI-l an ambitious and comprehensive training program. Since the accident, Licensee has substantially augmented its training i

department and headed it with professional educators who have solid backgrounds in nuclear training. Licensee's programs have been reviewed by NRC and by highly qualified independent l

consultarts. The ThI-l licensed operators have been trained, retrained, audited and reaudited by Licensee's training personnel and independent consultants. The operators have been l

-135-1

l exposed to training in the areas they should master before operating the plant. Nevertheless, prior to obtaining NRC licenses to operate the plant, these individuals all must pass NRC-administered examinations, both oral and written, with NRC's present grading criteria (70%/80%) and four individuals must pass as well the special Category T (TMI-2 lessons learned) exam with a 90% grade. The Board generally finds Licensee's training adequate and specifically finds Licensee has complied with the Commission's August 9,1979 and March 6, 1980 Orders insofar as they relate to training. See also 11 _, infra, in the Plant Design and Procedures section of this Recommended Decision as to specific. technical training issues.

D. Aamodt Contention 2

! 186. In the previous section of this Recommended l

l Decision, we have discussed the evidence on Licensee's training generally and licensed operator training in particular. We found that this Licensee has very competent personnel in charge of an ambitious and comprehensive training curriculum. We now turn to determine whether the program withstands scrutiny in the light of particular concerns advanced by intervenors.

l 187. In the area of training, only one contention was raised and pursued.28 The contention, raised by the Aamodts, states:

28 Other contentions which generally embrace training were advanced, namely, Sholly 14, ANGRY IV, and CEA 13, but not pur-(continued next page)

-136-

l l

It is contended that TMI-l should not open until the performance of licensee technicians and management can be demon-strated to be upgraded as certified by an independent engineering firm. This upgrading should include 100% test performance of job description with provision for retraining and retest, or discharge of those who cannot consistently and confidently master all necessary information for safe conduct of l their job descriptions under all anticipated critical situations as well as routine situations.

188. This centention was addressed in prepared testimony sponsored by Licensee witnesses, NRC Staff witnesses and one of the intervenors. Licensee's testimony on the contention consisted principally of that of Long, ec al., ff.

Tr. 12,140, Gardner, ff. Tr. 12,409, Christensen, ff. Tr.

12,409 and Kelly, ff. Tr. 12,409. NRC Staff testimony on this contention was that of Crocker and A11enspach, ff. Tr.12,653, and Boger, ff. Tr. 12,770. The Intervenor's testimony was by Ms. Aamodt herself, whose relevant background consists of an M.S. in experimental psychology followed by several years as an experimental psychologist with Bell Laboratories in the early 1950's and as a graduate assistant for about a year just prior to joining Bell Labs. Aamodt, ff. Tr. 12,931, at 10. Exten-sive cross-examination was conducted of each of these wit-nesses, spanning, together with related argument, some six total hearing days and more than a thousand transcript pages.

(continued) sued in the hearing through prepared testimony or cross-examina-l tion by intervenors. See 1 2, suora.

l l

-137-l L ------ -- -----

o 189. Aamodt Contention 2 on its face appears to be a rather straightforward broad attack on the adequacy of Licensee's training and testing program. Through the Aamodt prepared testimony and cross-examination of witnesses, however, i l

it became apparent to the Board that the contention had particular facets, focusing on licensed operators' preparedness and involving, for those operaters, elements of human factors engineering, simulator training, Licensee's training and testing methods, including reviews of Licensee's program by independent consultants, stress factors, operator attitudes and NRC's testing methods. Due in part to the evolving nature of Intervenor Aamodts' concerns, the record covering these various i areas is not all contained in the transcript of the almost five i

consecutive days at one stretch devoted exclusively to the contention. Additionally, the Board has taken into account l

l portions of the record containing Licensee's and the Staff's presentations on human factors engineering, NRC's review of Licensee training, and Licensee's overall management audit by an independent firm, BETA. See generally Walsh, et al., ff.

Tr. 10,234; Lic. Ex. 23; Tr. 10,235-355 (Walsh, et al.);

Ramirez and Price, ff. Tr. 10,452; Staff Ex. 2; Staff Ex. 15;

(, Tr. 10,460-590 (Ramirez, Price); Wegner, ff. Tr. 13,284; Tr.

13,285-327 (Wegner).

190. Intervenor Aamodts' first attack on the adequacy of Licensee's operator performance is based on the Aamodt view that human factors upgrading planned for the TMI-l

-138-1

, \

D. V * . .*

o

  • control room is insufficient, thereby negating the desired effects of Licensee's upgraded training. Aamodt, ff. Tr.

12,931 at 4. Specifically, Ms. Aamodt identified reduction of potential environmental stressors (such as glare from lighting) as needed items which were not getting proper attention. Id.,

at 2-3.

l 191. Licensee's case in chief on human factors upgrading of the control room was presented by a panel of five outstanding witnesses. See Walsh, et al., ff. Tr. 10,234; Lic.

Ex. 23; Tr. 10,412 (Smith). 2nfortunately, Ms. Aamodt chose not to attend that portion of the hearing when those witnesses testified, nor did she attend when the NRC Staff's witness on this subject appeared. See Tr. 10,412-13;. Tr. 10,460-590. We treat in some det*il the adequacy of the human factor. reviews of the TMI-1 control room in the Plant Design and Procedures portion of this Recommended Decision, infra. Suffice it to say, for purposes of Aamodt Contention 2, that the Board finds in that section that Licensee has conducted satisf actory reviews of the TMI-1 control room from a human factors standpoint and is implementing adequate improvements.

Recommended Decision, at 1 , infra. We also note that the NRC Staff, based on its own human factors review, concluded that, with the corrections required prior to restart and prior to escalation beyond five percent power, the potential for operator error leading to serious consequences as a result of human factors considerations in the conttol room is suffi-ciently low to permit restart and full power operation of

-139-

TMI-1. Staff Ex. 15, at 12. See also Tr. 21,431-33 (Bax*9r).

Finally, we find Licensee is not ignoring the environmental items specifically called out by Ms. Aamodt (See Lic. Ex. 23, at Section III), but rather has considered them and has plans to implement them, some of them prior to restart. See Lic. Ex.

I 33, at 3. Therefore, the Board finds no merit in the Aamodt position that Licensee is negating the upgraded retraining program by failing to adequately deal wi.a control room human factors design.

192. The second prong of the Aamodt contention is that Licensee's operator training is insufficient because Licensee does not have an exact replica simulator and trains for insufficient hours on B&W's simulator in Lynchburg.

Aamodt, ff. Tr. 12,931, at 4-5. Ms. Aamodt maintains an exact replica simulator is so important that even though the Licensee 1

has committed to an exac*. replica simulator and the lead-time is from three to four years for delive:y and installation alone (Tr. 12,145 (Long)), she would not have TMI-l reopen until the simulator is available. Tr. 13,121 (Aamodt).

193. Licensee's licensed operators are sent to B&W's Nuclear Training Center in Lynchburg, Virginia for simulator training. Long, et al., ff. Tr. 12,140, at 29. The Center includes a control room simulator for a plant which is similar to, but not a replica of, the TMI-l plant or control room.

Lic. Ex. 27, at 104. Dr. Christensen pointed out that he had observed the use of the B&W simulator by TMI-l operators and

-140-

o /

determined that at the functional level, the similarities between the B&W simulator and the TMI-l control room were very _

great and the training was quite valuable. Tr. 12,475-76 (Christensen). The limitations, he pointed out, were with the transfer of reflexive responses, since the B&W simulator uses some devices which are different in the TMI-1 ccntrol room, l

I such as push buttons rather than J-handles or pistol grip

, handles. Id. Licensee oper tors at TMI-l receive extensive on-the-job training in the actual TMI-1 control room to

, familiarize themselves with the actual equipment which they will operate when licensed. Tr. 12,228-30 (Newton, Ross).

194. Licensee's operators receive training on the B&W simulator during initial licensed operator candidate training, during annual requalification training, and for l

specific training periods such as following the TMI-2 accident.

Initial licensed operator candidates in 1980 participated at l

B&W's simulator in eight weeks of training. Long, et al., ff.

Tr. 13,242, at 30. Annually, Licensee's operators receive one i

week of simulator training during which they train as a shift.

Tr. 12,126 (Long). Following the TMI-2 accident, all of the licensed TMI-l operators attended two different training l

sessions at the B&W training center, focusing on loss-of-feed incidents in the first, and integrating multi-failure scenarios in the 32-hour second visit. Long, et al., ff. Tr. 13,242, at 29-30.

195. Ms. Aamodt's view of the necessity of an exact

. replica simulator for TMI-l is founded on: (1) a report

-141-

(Licensee Ex. 27) done by a group which conducted an independent review of Licensee's OARP and which recommended, i

inter alia, that Licensee consider a simulator for TMI-1; and (2) on the fact that other units have their own simulators (Aamodt, ff. Tr. 12,931, Attachment 1). Aamodt, ff. Tr.

12,931, at 4-5. The reviewers of Licensee's OARP, upon which Ms. Aamodt relies, did recommend a replica simulator, but "for the long-term" and not as a condition of restart. Lic. Ex. 27, at 144. Moreover, Licensee is following the recommendation and expects to have a simulator by 1985. Tr. 12,145 (Lcng). In the meantime, Licensee has increased its attention to B&W's simulator program by assisting the B&W staff in relating the simulator performance to actual TMI-l plant experience; Licensee also has underway part-task simulator programs which will be used pending availability of the replica simulator.

Long, et al., ff. Tr. 13,242, at 30; Tr. 12,258-63 (Long);

Christensen, ff. Tr. 12,409, at 7. Although Ms. Aamodt was informed of these interim measures, she nevertheless maintains that a replica simulator be available at TMI-l prior to startup because other plant: have them. Aamodt, ff. Tr. 12,931, at 4-5; Tr. 13,121 (Aamodt).

196. The Aamodt quarrel with the adequacy of the number of hours Licensee's operators spend at the simulator also is founded on a misunderstanding of the report of the independent review of Licensee's OARP. Lic. Ex. 27, at 99. As reported in the OARP Report, Dr. Johnson of TVA reported that

-142-

- . _ _ . _ ~ _ _ . _

l each TVA operator spends two weeks on their simulator each year. Id. He estimates a need for 1000-1500 hours of simu- _

lator time annually for every six nuclear units, and other experts, the Report further states, feel that a two-unit plant needs 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br /> of simulator time per year. Id. Licensee's t witness, Dr. Christensen, who authored this portion of the report relied on by Ms. Aamodt, explained that Dr. Johnson's recommendation was for about a week's simulator training per year. Tr. 12,742 (Christensen). Dr. Long, Licensee's head of training, stated that Licensee's simulator training is compara-

ble to the TVA training recommendation. Tr. 12,154-56 (Long).

Moreover, Mr. Kelly, Licensee's witness who has been involved i

in licensed operator training for some twenty years (Kelly, ff.

Tr. 12,409, Appendix A) testified that one week of annual j simulator training is the present industry-wide standard and 1

l that the standard is adequate. Tr. 13,743 (Kelly).

197. The Board finds that Licensee's present simulator training is adequate.

198. The third aspect of Aamodt Contention 2 is that Licensee's training and testing methods, including the reviews of Licensee's program by indeper. dent consultants, is inade-quate. See, ?.g., Aamodt, ff. Tr. 12,931, at 1-2. The Aamodt view is that expressly set out in the Contention itself, i.e.,

that Licensee's personnel performance should be certified by an i

independent engineering firm, including 1000 test performance with provision for retraining and retest and discharge of poor

-143-

performers. Id. at 1. This concern is directed at licensed operators and embraces, in Ms. Aamodt's opinion, Licensee's

~

training curriculum, instructors and physical facilities, as well as the adequacy of testing methods and the sufficiency of reviews of Licensee's training efforts by independent reviewers. Id., at 1-2, 5-6, 9.

199. Licensee's overall' training program was described in considerable detail in the preceding section on Training. A brief synopsis, however, is necessary here to place in perspective the Aamodt concerns. Following the accident at TMI-2 and the focus on TMI operators, Licensee committed to have all its TMI-l operators undergo intensive retraining and sit for NRC-administered operator exams (a unique situation since ordinarily operators receive on any one facility only one NRC-administered exam in a lifetime, thereafter being subject to NRC-audited, but utility-administered, requalification examinations). Kelly, ff. Tr.

12,409, at 3-4. The Licensee developed and administered to its entire complement of TMI-l licensed operators during the period August, 1979 to March, 1980, an intensive training curriculum, embodying previously taught subjects as well as TMI lessons learned subjects. Long, et al., ff. Tr. 12,140, at 38-39.

This intensive, one-time program, the OARP , was revieweo by independent groups including t.e OARP Review Committee, of which Dr. Eric Gardner and Dr. Julien Christensen were members.

Id. at 41-42. See generally Lic. Ex. 27. Following the OARP

-144-

1 l

l and in anticipation of NRC license exams then scheduled for April 1, 1980, Licensee required OARP participants to sit for mock-NRC exams administered by an independent consulting firm,

-f PQS. Long, et al., ff.-Tr. 12,140, at 40. See generally Kelly, ff. Tr. 12,409, at 6-7. At the same time, PQS adminis-tered special category T exams on lessons learned from th TMI accident, required by NRC to be passed by all licensed opera-tors (not only at TMI, but at all B&W plants). Kelly, ff. Tr.

12,409, at 9; Tr. 20,705 (Crocker). When the schedule for restart of TMI-1 slipped, however, NRC deferred examinations for the operators and the PQS exams were viewed as the annual requalification examinations; under the then-existing NRC grading criteria, no one failed the exam. Tr. 20,610-11 (Newton). All operators and trainees thereaf ter participated in Licensee requalification training programs, including annual simulator traiaing and another round of mock-NRC exams, this time by Associated Technical Training Services, preparatory to sitting for the NRC written examinatior , given finally in April, 1981. Tr. 20,581, 20,584-85 (Newton). NRC, as is their customary practice, does not anticipate giving the oral portion of the examination to the TMI-l operators until about 30 days prior to restart of the unit. Boger (Aamodt Contention #2),

ff. Tr. 12,770, at 2; Tr. 12,785-86 (Boger). i 200. With this brief recap, we now turn to the Aamodt concerns. The Aamodts, as evidenced by Ms. Aamodt's testimony, generally fault the training program at TMI on two diverse, if

-145-f

- ---,--,-wv,-<,mm,w--.--a--n---,,..w,,n ,,m-- ,.-,,m,,w,,-- , ,,-,,,-,w-,-w,,y,-ew----m-n-,-w--v-.-

not contradictory, grounds--first, for not measuring up to college level standards in faculty, facilities or curriculum and second, for not adequately preparing trainees for the actual conditions they will face as operators in the control room. As to the first, Ms. Aamodt testified based on input from her husband (who is an engineer) that an adequate thermo-dynamics course for the TMI-l operators could only be taught by a Ph.D. (Tr. 12,988-99, 13,000 (Aamodt)), and applying her own experience in education at the college level (one year as a graduate assistant), she inferred from the OARP Review Committee's Report that the facilities, faculty and curriculum utilized during the OARP were major faults. (Aamodt, ff. Tr.

12,931, at 5). As to her second reason for objecting, Ms.

Aamodt concentrated, for example, en the lack of phones ringing during the training courses or the lack of communications demands as indicative of training's failure to portray actual conditions with which operators would have to cope during an emergency. See, e.g., 12,509-10; 12,522; 12,544-48 l (Christensen, Kelly).

201. The Board is under no allusions that Licensee (3 l training program is a college curriculum. It is a commercial training program geared to equip its participants with knowl-edge sufficient for them to understand TMI-l and safely operate that plant. See Long, et al., ff. Tr. 12,140 at 3; Tr.

12,225-26 (Long). The management of the training department I

has been injected with experienced educational professionals.

-146-

I l

l See, e.g., Long, et al., ff. Tr. 12,140, at 7-9, 13-15. The faculty has been upgraded both in quantity (from 7 to 45) and in quality, including the requirement that each instructor attend a one-week instructor development program. Id. at 3, 10-11, 54, 57; Tr. 12,213 (Newton). The facilities which were criticized by the OARP Review Committee and detracted by Ms.

Aamodt have been vastly improved since the OARP was conducted and the facilities reviewed in 1979. See, e.g., Tr. 12,321-22 (Long). Moreover, there is under construction adjacent to TMI a million-dollar building which will serve as a training facility. Long, et al., ff. Tr. 12,140, at 12.

202. Similarly, we are under no allusions as to the propriety of the Aamodt concerns. Ms. Aamodt never visited the training facilities; she never met, nor interviewed, any of the instructors; and she has never viewed or otherwise studied the course curricula or lesson plans. Tr. 12,975-76 (Aamodt). Her reliance for faulting these elements rests almost totally on the OARP Review Committee Report. Tr. 13,001-04 (Aamodt).

Under these circumstances, the Board must rely on the views of those on whom Ms. Aamodt herself relies, namely the OARP Review Committee members, two of whom appeared and testified that the training program was adequate. Tr. 12,744 (Gardner and l Christensen); Lic. Ex. 27, at 141. We also take note of the additional training which Licensee's operators have received since that OARP review was conducted in 1979. See generally Long, et al., ff. Tr. 12,140; Tr. 20,584-85 (Newton). Finally,

-147-

)

I with respect to the adequacy of Licensee's training to prepare operators for their actual surroundings, the Board notet the significant ingredient of "OJT" which trainees receive, the fact that many of TMI-l operators are veterans of actual plant operations, and that the simulator training by Licensee's operators has included raal stimuli, including telephone communication demands being placed on the operators. See, e.g., Tr. 12,226 (Newton); 12,741-42 (Christensen); Tr.

12,250-55 (Long, Ross).

203. One element of this aspect of Aamodt Contention 2 remains, namely, the 100% testing requirement by an indepen-dent engineering firm.29 Aamodt, ff. Tr. 12,931, at 2.

Licensee's training program involves frequent quizzes and examinations covering the material. Long, et al., ff. Tr.

12c140, at 22-23, 27-28. NRC licensed operator exams (as well as mock-NRC exams administered by Licensee or its consultants) now are graded more stringently than prior to the TMI-2 accident. Whereas prior to the accident an operator was required to pass by achieving an average of 70% over all of the subject areas, now a 70% grade is required in each of the subject areas and ar. 80% average over all the subjects is i

29 Even though Ms. Aamodt does not agree that NRC's reviewing l of Licensee's training programs coupled with NRC's examination i of licensed operators meets her criterion, it is in the end success or failure on the NRC-administered exams which will be determinative of Licensee's operator preparedness. Additionally, the Staff in its testimony recounted, in addition to the testing, i the broad spectrum of its requirements which led it to oppose

, the Aamodt view. See Crocker and Allenspach, ff. Tr. 12,653.

i

-148-i

o required to pass. See Boger ( Aamodt Contention #2), ff. Tr.

12,770, at 3; Kelly, ff. Tr. 12,409, at 7. Moreover, while the exam draws from each of the subject areas which an operator needs to know and seeks from the examinee a demonstration of his knowledge in each area, it is not fathomable to construct a test to determine all of an operator's knowledge in every area.

See Tr. 12,854-56 (Boger). No witness other than Ms. Aamodt could subscribe to the 100% testing requirement, viewed either as a grading requirement (i.e., 100% required vs. 70/80%) or as a realistic criterion for foemulating the exam (totally comprehensive vs. audit). See, e.g., Crocker and A11enspach, ff. Tr. 12,653, at 15; Boger (Aamodt Contention #2), ff. Tr.

12,770, at 4-5; Gardner, ff. Tr. 12,409, at 8-9. The Board finds that Licensee's training and testing techniques are adequate to prepare licensed operators and give reasonable assurance the operators will perform adequately under both normal and upset plant conditions.

204. The fourth subset of the Aamodt Contention is a concern whether the licensed operators will be prepared to i

t withstand the stresses associated with "a really tough critical j situation." Aamodt, ff. Tr. 12,931, at 6. Aspects of operator stress as perceived by Ms. Aamodt stem from human factors considerations (see 11 190 to 191, supra), simulator training inadequacies (see 11 192 to 147, supra), and general training curriculum inadequacies (see 11 198 to 203, supra). Licensee l

and the Staff presented several witnesses who specifically

-149-

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

i addressed the potential for stress in licensed operators under accident conditions. Licensee's witness, Dr. Gardner, observed that although no psychological test exists nor is there any psychologist who can certify that any specific individual will behave in a stable fashion in all stressful conditions, initial screening of applicants, simulator training, providing checks and back-up elements among the personnel in the control room, and access for control room personnel to outside knowledgeable personnel are all steps that minimize that potential. Gardner, ff. Tr. 12,409, at 7-8. Dr. Christensen reviewed the factors which minimize or alleviate stress operators would experience during emergencies. Christensen, ff. Tr. 12,409, at 9-11.

First, he observed that adequate training plays a role, and in his view, Licensee's program provided the necessary training.

Id., at 9. Second, he observed that exposure to a wide variety of possible emergency situations engenders confidence and in i

his view the simulator training was designed to accomplish this. Id., at 9-10. Third, Dr. Christensen noted that to respond adequately under stressful conditions, it is important that opert ars have available to them readouts that allow them to track automatic system actions and that the TMI-1 control room has such information displays. Id., at 10. Fourth, he noted that stress can be evidenced in the members of a team who are unaccustomed to working together and viewed favorably Licensee's effort to maintain the same operators as members of each shift team. Id., at 10-11. Fifth, Dr. Christensen noted

[

, -150-l

-,..-...,,,-.,,,_,--..,...-,.v.,,..m,,. ,,.r...,,,,,,_,-y-,,,

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

5 that environmental stressors, such as noise and lighting, could play a role, but further noted that Licensee, working with him and with other consultants, is addressing environmental factors. Id., at 11. Finally, Dr. Christensen observed that Licensee's inclusion of a decision analysis course which provides the operators with a model that can be followed to l deal especially with stress-inducing events is an important element to avoiding stress. Id., at 11. NRC's training witness, Mr. Boger, echoed these observations, noting that operator candidates cope with stress which must be overcome to successfully complete NRC examinations and that previous operating experience, simulator exercises, and training on anticipated critical situations all help overcome the stress which may be associated with abnormal situations. Boger (Aamodt Contention 2), f f . Tr . 12,7'/0, at 6.

205. The Board takes as a given the Aamodt belief that operators at TMI-1 (as at any plant) will experience some stress when faced with emergencies. Moreover, we agree with 1

l Dr. Gardner that prior certification that an individual will behave stably under virtually any stressful situation is not possible. Stress, however, and its potential impact on operators has not been ignored by this Licensee, but rather has been consciously factored inte its program for preparation of operators. We regard as sufficient the measures Licensee is taking to alleviate or minimize the potential for stress in operators under critical situations.

-151-

1 h, ) . - .* ,

o a 206. The fifth element in Aamodt Contention 2 is i operator attitudes. This element, as reflected in Ms. Aamodt's testimony (Aamodt ff. Tr. 12,931, at 8-9), was more in the f

nature of a recommendation that Licenace should maintain an awareness of operator attitude, foster morale and ensure an

, appreciati.:n of the significance of the actions of the opera-i tors, than a specific complaint. This does not to the Board appear to be a problem. The testimony is uncontroverted that Licensee persottr.al enjoy enthusisatic, confider.t attitudes and have appreciation for the importance of their functions. Dr.

Christensen testified that there appears to be a general attitude of enthusiasm and dedication to the task at hand that is infectious, and, if maintained, should do much to assure that TMI-l will be operated in a safe, efficient manner.

Christensen, ff. Tr. 12,409, at 12-13. Mr. Kelly of PQC cited his experience with the operators' reaction to Licensee's decision to require all operators to retake NRC examinations; l he termed as remarkable their positive response which he attributed to both the operators' commitment to Licensee and i

their pride and confidence in their capabilities. Kelly, ff.

Tr. 12,409, at 4. He further observed what he characterized as sincere and enthusiastic participation in training by the operators. Id., at 10. Mr. Wegner, whose consulting firm BETA 1

conducted over more than a year an independent review and assessment of the management capability and technical resources of GPU as related to the TMI-l restart, which included l

( -152-

i l

j l

interviews of over 150 individuals, spoke of a sense of pride and accomplishment in TMI-1 personnel and knowledgeable, interested and actively involved managment personnel. Wegner, ff. Tr. 13,284, at 1, 6, 18 and 27. While BETA did not address in its review the actual state of qualification of licensed l

operators (Id. , at 12), Mr. Wegner's views on employee atti-tudes are instructive. Finally, Dr. Long, who heads GPU's training department, testified that responsible attitude training is a part of General Employee Training which all individuals working at TMI receive; that it has been the focus of management directives which are discussed by supervisors with individual teams; that it is discussed with employees in employee meetings; that responsible attitude, as well as a concern and a knowledge of the responsibilities for not only their own radiation protection, but as well the protection of fellow workers and the general public, is part of the training and briefings by supervisors as part of the Licensee's inte-grated effort of presenting to its employees their responsi-bilities to the company, to each other and to the public; that it is not possible at this point for anybody associated with TMI to not understand the consequences and potential consequen-ces of a serious accident; that it is a fundamental element of Licensee's training program that its personnel understand the

! possible consequences of a nuclear accident both in terms of l risk to Licensee and risk to the public; and that operators i

specifically are trained thoroughly in the details of plant l

-153-

operation and the consequences of an accident. Tr. 12,303-04, 12,308-12 (Long).

207. The final element in the Aamodt contention is a concern with the adequacy of NRC's licensed operator tests.

NRC's principal witness on licensed operator testing was Mr.

Boger, who described NRC's testing techniques. See Boger (Aamodt Contention 2), ff. Tr. 12,770. He explained that NRC license exams since the TMI-2 accident have been revised in accordance with the NRC Action Plan, NUREG-0660, Item I. A.2.

Id., at 3. The scope of the exams has been increased to include a new category dealing with thermodynamics, heat transfer, and fluid flow. Id. Additionally, time limits now exist in the exam, oral examinations for SRO's have ceen instituted and the grading criteria have been increased. Id.

NRC does not believe it is necessary for each individual to be l examined on 100% of the job description or receive a 100% score l

on the NRC exam for the following reasons:

The new criteria provide reasonable assurance that an applicant who can achieve an exam score of at least 80% overall with no category less than 70% and who can success-fully pass the operating te.St can perform licensed duties safely and competently.

l l

The licensed operator is not alone at the facility. We require two licensed i operators and two licensed senior operators l on each shift.30 Hence, plant safety is not l

l 30 NRC's initial requirement for two SRO's and two RO's was modified, following a reexamination of the requirement, to require one SRO and two RO's, as are presently rec; ired at other operating plants. Staff Ex. 14, at 22-23.

-154-

I l

dependent solely on the knowledge and understanding of one individual.

The operating test varies from one individual to another so that we can assess the overall effectiveness and scope of the training program. While ecsch operating test must cover a minimum number of plant systems, operating procedures and transients, the specific topics will not be the same for each applicant. If the exam results indicate that there is a subject or system which has not i

been adequately covered in the training program, we will inform facility management to provide additional training in the weak area.

Id., at 5 (footnote added). The NRC position is that their examinations are structured around those items that NRC considers important to reactor safety as enumerated in Sections l 55.20 through 55.23 of 10 CFR Part 55, Operators' Licenses, and l

l are designed to give reasonable assurance that the individuals can safely and competently operate the facility. Id., at 2.

I 208. We reject the Aamodt attack on the adequacy of NRC's license examinations. In so doing, we take into account not only Mr. Boger's testimony but as well the opinions of those of the calibre of Dr. Eric Gardner (who reviewed with favor Mr. Kelly's examinations which were fashioned after NRC examinations), who unanimously agreed that successful com-pletion of such exams coupled with training sufficient to allow success on those exams was indicative of a capable licensed operator. See Lic. Ex. 27; Gardner, ff. Tr. 12,409, at 14; Christensen, ff. Tr. 12,409, at 11-13; Kelly, ff. Tr. 12,409, at 10.

y -155-e 1

1

- :.._ . ? , .-. .- - . .

V. VIEWS OF NRC INSPECTORS l

209. CLI-80-5, Issue (3), states:

What are the views of the NRC inspectors regarding the quality of the management of TMI Unit 1 and the corporate management, staffing, organization and resources of Metropolitan Edison.

210. This question was answered primarily by the Office of Inspection and Enforcement through the introduction

into evidence of NUREG-0680, Supp. 1, the management safety evaluation report (SER), which contains a section devoted to the inspections conducted by IE since the accident at TMI-2.

In addition, the Staff presented three witnesses on this l

l subject who have played major roles in IE's inspections at TMI since the TMI-2 accident: Mr. Donald R. Haverkamp, Senior Resident Inspector, TMI-1; Mr. Richard R. Keimig, Chief, Reactor Projects Section No. 1, Reactor Operations and Nuclear Support Branch, Region I of IE; and, Mr. Norman C. Moseley, Director of the Division of Program Development and Appraisal, i IE.

211. IE has been expending more than double the amount of inspection time at TMI-l that IE is spending at other plants, with a senior resident inspector and two resident inspectors assigned to the unit, plus a health physics inspector, a supervisor, and regional-based inspectors visiting l the site approximately every week. Tr. 12,115 (Keimig). This degree of surveillance was established by iE during its

-156-

investigation into operational, radiological and emergency i

response actions taken by Licensee during the TMI-2 accident l and in recognition of the major changes being taken or consid- -

ered to support the restart of TMI-1, the recovery and cleanup of TMI-2, and the operation of Oyster Creek. This surveillance specifically addressed Licensee's implementation of their revised organization and management systems. Staff Ex. 4, at i

9 14.

212. In addition to the regular inspections conduc-ted at TMI-l unrelated to the Commission's August 9,1979 Order items, IE has conducted numerous special inspections over the more than two years that have transpired since the TMI-2 accident. In summarizing its ef forts in these areas in the -

management SER, IE divided its activities into three cate-gories: (1) investigations into the TMI-2 accident; (2) three extensive special inspections conducted by the Performance and Appraisal Branch of IE to appraise and evaluate the status and adequacy of Licensee's !.mplementation of certain management control systems and programs. These appraisals were conducted in the areas of overall management capability, health physics, and a "near term operating license" (NTOL) review; and (3) other inspections of relevance to TMI-l's restart. Staff Ex.

4, at 15 and Appendices A-C; Tr. 12,067-83 (Haverkamp). The l 31 IE also provided in the management SER various informa-l tional uascriptions of its inspection programs. Staff Ex 4, at 18, and Appendix D.

-157-

findings and conclusions reached in the IE section of the management SER were introduced into evidence by Mr. Haverkamp, TMI-l senior resh ant inspector; however, each of the TMI-l inspectors and L am leaders, where ' appropriate, supervisors of those individuals up through and including the branch chiefs of the various responsible branches in the region, the Deputy Director and the Director of Region I concurred in the IE input to the management SER. Tr. 12,085 (Haverkamp).

i 213. IE did not reach a final conclusion in the management SER because of the procedural and equipment changes

! required for TMI-l restart which were incomplete at the time the SER was published in November,1980. However, IE did i i

conclude that in general, the improvements described by NRR in the management SER concerning Licensee's management capability had been confirmed during the conduct of the various IE l

investigations, appraisals, and inspections. In addition, l after reviewing Licensee's noncompliances related to the TMI-2 accident and evaluating Licensee's commitments for corrective action, IE concluded that each item had been or would be corrected; in fact, these corrective actions had been closed out by the time Mr. Keimig testified. Staff Ex. 4, at 18; Tr.

11,982 (Keimig).

214. While the Board initially was concerned about the open items identified by IE in the management SER, its concern was mollified by IE's recognition of its responsibility to bring to the Board's attention any unsatisfactory responses 1

-158- ,

l

by Licensee to IE, and has been removed by the absence of any such notifications to us by IE. Tr. 12,077-81 (Haverkamp); Tr.

12,092 (Keimig). To the contrary, in its March, 1981 supple-ment to the management SER, IE concludes that the corrective measures proposed by Licensee, when fully implemented, are ,

sufficient to resolve t.1e management concerns identified during past IE inspections. Staff Ex. 13, at 7. Similarly, both the senior resident inspector, Mr. Haverkamp, and his regional supervisor, Mr. Keimig, testified in March,1981 that they were fully satisfied with Licensee's management capability to operate TMI-1. Tr. 12,028 (Haverkamp); Keimig, ff. Tr. 11,946, at 15-16; Tr. 11,956-57, 11,998 (Keimig).

215. Finslly, the Board received evidence from Mr.

Norman C. Moseley, Director of the Division of Program Development and Appraisal of IE, on the subject of information flow during the accident at TMI-2. Our interest in th'.s subject was limited to the implications which could b<. drawn, positive or negative, regarding Licensee's credibility, reliability and general sense of commitment to its license obligations. Mr. Moseley currently is responsible for the development of the inspection program for all types of licen-sees regulated by NRC, and for the execution of the licensee performance appraisal inspection program. Professional Qualifications of Norman C. Moseley, ff. Tr. 13,024. However, Mr. Moseley ecently completed his responsibilities as the team leader in IE's 1980 investigation of information flow during i

-159-

o the TMI-2 accident. This investigation, conducted in response to a request from NRC Chairman Ahearne, followed the initial IE investigation of the accident in 1979, as well as a subsequent inquiry devoted to the subject of information flow. Both of these earlier reports concluded that there was no direct evidence suggesting intentional withholding of information by Licensee's management. Staff Ex. 5, at 1. Mr. Moseley's team similarly concluded that while the flow of information was inadequate on March 28, 1979, information was not intentionally withheld by Licensee from the NRC or the Commonwealth of Pennsylvania.32 Staff Ex. 5, at 10-11; Tr. 13,031-33 (Moseley). The Board is satisfied that IE has conducted an exhaustive examination of this subject.33 32 The manner in which Licensee has responded to these infor-l mation flow problems is discussed in our findings on emergency l planning. See Tr. 13,776-79 (Giangi) .

33 The Board is aware of the investigations by bodies other than IE which are devoted to or include a discussion of the subject of information flow during the TMI-2 accident and which concur with IE's findings on this subject. Included in that list are Three Mile Island, a Report to the Commissioners and to the Public, at 159-60; Vol. II at 894-913 (1980) (Rogovin Report); Report of the President's Commission on the Accident at Three Mile Island at 19 (1979) (Kemeny Commission Report); " Nuclear Accident and Recov-ery at Three Mile Island," Report Prepared by the Subcommittee on Nuclear Regulation for the Committee on Environment and Public Works, U.S. Senate, at 13 (July 2, 1980) (Hart Committee Report);

Memorandum forwarding Report to Chairman Ahearne from Mitchell Rogovin et al. Regarding Questions Submitted by Congressman Udall (March 4, 1980) (Supplemental Rogovin Report). Tr. 13,070-71 (Blake). There has also been a report by the majority staff of the House Committee on Interior and Insular Affairs (Udall Committee),

entitled, " Reporting of Information Concerning the Accident at Three Mile Island," (January 29, 1981) which disagrees with IE's conclusion regarding the intentional withholding of information.

Tr. 13,052-54 (Moseley). To the best of Mr. Moseley's knowledge, (continued next page)

-160-l . . - .

s 216. In conclusion, without exception, the views of the NRC inspectors regarding the quality of the management of TMI Unit 1, the corporate management, the staffing, the organization and the resources of Licensee are favorable. This opinion was shared by IE's supervisory staff.

VI. HEALTH PHYSICS 217. CLI-80-5', Issue (4), asks:

Whether the Unit 1 Health Physics program is appropriately organized and i

staffed with qualified individuals to ensure the safe operation of the facility.

4 218. In the area of health physics, witnesses were presented by the Licensee and by the NRC Staff. No intervenor witnesses appeared and intervenors conducted no cross-examination. The Board, however, on its own, conducted extensive examination of both Licensee and Staff witnesses.

219. Licensee's witnesses consisted of a panel of five individuals from Licensee's organization, an independent consultant, and an individual General Employee Training (GET) instructor. The panel was comprised of Richard Heward, William Potts, Ronald Knief, Jesse Brasher and Richard Dubiel. See Heward, et al., ff. Tr. 16,292. Mr. Heward is the Vice-President of GPU Nuclear Corporation for Radiological and (continued) no information was relied upon in the Udall Committee majority staff report whic'a was not included in the data base of other investigations and reports, including those conducted by IE.

Tr. 13,043-44, 13,065 (Moseley).

-161-

, ,-e --v w .< u-,-.w. . - - , - ,w--,-,-,w e - %, ,----r,-.--,re--,mw ww.-..---,-------,w-*.-Sm,ww.-,,-.wr,.-w,-..----.-,.ww,-+e-.mww%. -

v.ww w- , - , - + - -

o Environmental Controls with 14 years of experience with GPU and 10 years with NEw York Shipbuilding, involving both radiolog-ical control and projects management. Id. (Attached quali-fications of Richard Heward, Jr.) He holds a B.S. degree from Swarthmore College, is a registered professional engineer and has been certified both at the Oak Ridge School of Reactor Technology and in the Reactor Safety Course, V.K.A.E. A.

l Harwell, England. Id.; see also Arnold, ff. Tr. 11,434, at 14.

Mr. Potts is the Manager Radiological Controls of TMI-1. He j has been involved in TMI-1 activities since 1972, and has a broad nuclear power plant background including technical support, licensing, quality control and health physics.

Heward, et al., ff. Tr. 16,292. (Attached qualifications of l William E. Potts.) He holds a B.S.E.E. degree from Penn State, has attended various educational and training courses on health physics, quality assurance and management skills and has held an NRC SRO license. Id. Dr. Knief heads up training at TMI-l and haJ extensive nuclear education background including university faculty positions, training experience and publica-tions in the nuclear field. See Long , et al . , ff. Tr . 12,140, at 13-15. Mr. Brasher is the Director of Radiological Controls i

at TMI-2. Howard, et al., ff. Tr. 16,292 (Attached quali-fications of J. W. Brasher). He has some 20 years of experi-ence in health physics, having been Director, Radiation Control, of Litton Industries' Ingalls Shipbuilding Division and a supervisor in the health physics section of E. I. Dupont I

1 1

l

-162-

_ ~ _ .. _ , _ ._.._ .-.-_-..__ _ _ __ .-.__. __ _ .. _ _ _ _ . _ - _ .._ _ _ _ .___ _ ._

+ . e , . .. .

o I de Nemours at the Savannah River Plant. Mr. Brasher graduated from Millsaps College with a degree in chemistry, and has been certified by the American Board of Health Physics since 1964.

Id. Mr. Dubiel heads up the TMI-l Radiological Engineering group in Mr. Potts' organization. He is a graduate of Fairfield University with a B.S. in physics and a graduate of Georgia Institute of Technology with an M.S. in Nuclear Energy; he has about 10 years of experience in radiation protection, eight of those at TMI. Broughton, et al. (Instrument Ranges (In Plant)), ff. Tr. 7509, attached qualifications of Richard W. Dubiel. The other Licensee witness who appeared at the Board's request was one of Licensee's individual instructors, Ms. Sheila McAlister. See Tr. 16,392-404 (McAlister). Ms.

McAlister, who teaches GET courses, holds a B.S. degree from St. Andrews Presbyterian College, has one year of secondary teaching experience and worked as an HP technician at Surry Nuclear Power Plant and at TMI for two years before joining GPU's training staff. Tr. 16,392-93 (McAlister).

220. Licensee's consultant witness was Mr. Murray Miles of Basic Energy Technology Associates, Inc. (BETA). S(:

generally, Wegner, ff. Tr. 13,284; Tr. 13,293-99, 16,378-91 (Miles). Mr. Miles has more than 25 years of health physics involvement. Wegner, ff. Tr. 13,284, Attachment 1, at 1-2. He is a graduate of Cornell University with a degree in Engineer-ing Physics. Id. From 1955 through 1979, he was a member of DOE's Naval Reactors program, the last tnirteen years as

-163-

- . _ .. . _ . - - . . . _ , - - _ _ - - . - . - - - - . - _ . - _ _ . . . - - . - - - , - - - _ . . .._.2

! Associate Director for Nuclear Technology. Id. In this pos.ttion, Mr. Miles was responsible to Admiral Rickover for establishing and monitoring radiological control procedures.

He developed the procedures and methods now followed in the naval program to control radioactive discharge and radiation exposure. Id.; Tr. 13,293 (Miles). His reports and methods have become world-wide standards. Wegner, ff. Tr. 13,284, Attachment 1, at 2. He has personally conducted more than 200 inspections of radiological controls at various facilities.

Tr. 13,293 (Miles). Since October of 1979, he has been actively involved in evaluating GPU's radiological controls i

organization and procedures, including in January, 1981,

! completing a detailed assessment of TMI-1. Wegner, ff. Tr.

13,284, at 26-28.

l 221. The Staff's witness on health physics was Mr.

Donald Neely, an inspection specialist in the Performance Appraisal Section of NRC IE's Division of Program Development and Appraisal. Neely, ff. Tr. 16,450, at 1. See generally Tr.

16,448-65 and 20,663-86 (Neely). Mr. Neely has been involved in health physics for about 20 years, having been employed by i GE in providing radiation protection services, ty Chem-Nuclear Systems as a supervisor and instructor of health physics technicians, and since joining NRC as an inspector of i

NRC-licensee health physics programs. Neely, ff. Tr. 16,450, (Attached Professional Qualifications).

222. The objective of the TMI Radiological Program is to control radiation exposure, to avoid accidental radiation E -164-I

-w--v,- ,, , = -,,+,,,,,-<--,-w.-- c --rw ,-mr,. .-.-------,-------,--.----e--we - - , - -,--.w.w >-e-r,-=-,--,-w.-,--.-,.---,,w---.-,v--e-cwwe.---ww,-,-,-,,-rw,---,,,--

l . .*

c /

l exposures, to maintain exposures within the regulatory requirements, and to keep exposures to workers and to the general population as low as is reasonably achievable. Heward, j et al., ff. Tr. 16,292, at 4. To meet these objectives, Licensee's Radiological Controls department has been reor-ganized, staffing at TMI-1 has been increased from approxi-mately 9 to 79, and the TMI-l Radiation Protection Plan and implementing procedures have been rewritten. Id; Wegner, ff.

Tr. 13,284, at 19-20, 23, 26. TMI-l's radiological control organization, headed by the Manager, Mr. Potts, reports not through the operational chain but rather to a Vice-President (Mr. Heward) of Radiological and Environmental Controls. See Arnold, ff. Tr. 11,434, at 13-14. This organization allows independence, but close coordination with plant management is maintained. Heward, et al., ff. Tr. 16,292, at 4-5; Wegner, ff. Tr. 13,284, at 20-22.

223. The TMI-l Radiological Controls Depart ;nt, under its Manager and a Deputy, has three groups: a Radiological Engineering group, a Radiological Technicians group, and an Administrative group. Heward, et al., ff. Tr.

16,292, at 5; Attachment 1 (organization chart). The Radiological Engineering group under Mr. Dubiel, an experienced health physics supervisor, has about six engineers, in compari-son to the single engineer formerly available to TMI-1. Id.;

Wegner, ff. Tr. 13,284, at 22-23. The radiological engineers are responsible for anticipating and solving technical radio-logical controls problems; planning and development of the

-165-

Radiological Controls Program; providing technical support to the TMI-1 organization; and assessing the performance of the radiological controls organization to ensure continuous improvement in the Program. Id. The Radiological Technicians group consists of a Manager, six foremen, and 30 technicians, an increase by a factor of about four over just the past year.

Heward, et al., ff. Tr. 16,292, at 5; Wegner, ff. Tr. 13,284, at 23-24. These technicians provide around-the-clock coverage of radiological work. The Administrative group performs most administrative and clerical work, maximizing the time available for the engineers and field group to perform their functions.

Heward, et al., ff. Tr. 16,292, at 5-6. In Mr. Miles' opinion, to which the Board accords considerable weight in view of his experience and the standards to which he has been accustomed, the organization is larger with more talent than at most other commercial licensed nuclear power plants and enjoys a very heavy management commitment to particularly high standards of radiological controls. Tr. 16,379 (Miles). See also Wegner, ff. Tr. 13,284, at 19, 27-28.

224. The operational readiness of the TMI-1 Radiological Controls Department has been demonstrated.

Heward, et al., ff. Tr. 16,292, at 10-11. The most challenging periods for radiological control are during shutdown when most radioactive work is performed. Wegner, ff. Tr. 13,284, at 26.

i In 1980, over 13,000 Radiation Work Permits were issued at i

TMI-1; total manrem exposure was 201, significantly below the

-166-

projected total of 320. A number of complex jobs were safely and successfully performed on radioactive systems in relatively high radiation areas. Heward, et al., ff. Tr. 16,292, at 10-11; Wegner, ff. Tr. 13,284, at 23, 26. The highest individual occupational exposure for 1980 was 1.005 rem, which is less than 10% of the NRC allowable exposure; no individual working in TMI-1 received more than 1% of the limits estab-lished for internally deposited radioactivity. Heward, et al.,

ff.' Tr. 16,292, at 10-11. Licensee has also instituted successfully a major effort to reduce radwaste volumes at TMI-l and to clean up contaminated areas of the plant. Heward, et al., ff. Tr. 16,292, at 11; Wegner, ff. Tr. 13,284, at 25-26; Tr. 16,388-91 (Miles).

225. In the summer of 1980, NRC conducted a special evaluation of Licensee's health physics program at TMI-1.

Neely, ff..Tr. 16,450, at 3. NRC reviewed the program in extreme detail, utilizing six persons who spent about six weeks essentially full-time, plus several others who were utilized part-time. Tr. 13,294 (Miles). NRC identified a number of problem areas which were documented in a report to the Licensee. Staff Ex. 4, Appendix B. The Licensee's reaction to l NRC's identification of weaknesses was, according to Mr. Miles, to respond actively and interestedly (Tr. 13,294); Mr. Neely, too, characterized the Licensee as cooperative. Tr. 20,679 (Neely). Initially, the Staff was prepared to find Licensee's program adequate based on Licensee's commitments to correct

-167- i E _ _ _ _ - -

l l

l l

NRC's identified weaknesses and the Staff's commitment to the Board to review implementation of the corrections prior to restart; however, the Board felt it needed more assurances.

With the Board's encouragement, the Staff conducted a follow-up inspection in April,1981, of the deficiencies previously

. .,ntified to verify Licensee's implementation .of corrective actions. See Neely (Memorandum), ff. Tr. 20,572. Even though most of Licensee's corrective actions were not scheduled for completion until late May or August of 1981 (Neely, ff. Tr.

16,450, Attachment G; Heward, et al., ff. Tr. 16,292, Attachment 2), the results of this NRC inspection were impor-tant to the Board's ultimate confidence in Licensee's

Radiological Controls Program. See Neely (Memorandum), ff. Tr.

20,572. The Staff concentrated on areas such as organizational relationships and staffing, bioassay program, exposure control, personnel dosimetry, and radwaste management at TMI-l which the Board had identified as particularly important; at the same time, they examined Licensee's corrective actions with respect to items of noncompliance previously identified by the Staff in its inspection conducted immediately following the accident at TMI-2 and reported in NUREG-0600 in 1979. Id,. The results of the Staff's April, 1981, inspection were provided to the Board by Mr. Neely, whom the Board questioned on virtually every i

item. Tr. 20,663-86 (Neely). Out of the thirteen open items related to organization, the Staff closed out seven in April.

They did not have time to complete inspection on four of the

-168-

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

thirteen items, and only two remained for further discussion with Licensee. Id., Attachment A, at 1-2; Tr. 20,663-71 (Neely). Of the eleven weaknesses related to exposure control earlier identified by the Staff, nine were completely resolve'd, one awaited solely confirmatory measurement by the Staff, leaving only one which required further discussion with Licensee. Id., Attachment A, at 2-3; Tr. 20,671-76 (Neely).

In the area of radioactive waste management, all items previously identified as open were found to have been cor-rected. Id., Attachment A, at 3; Tr. 20,676-77 (Neely). Mr.

Neely did not foresee problems closing out the few remaining open items; in fact, he anticipated that the organizational items would all be closed out in about a week. Tr. 20,671, 20,674, 20,678-79 (Neely). As for the NUREG-0600 items from the 1979 inspection, every item examined by the Staff was closed out, and Licensee reported it was ready to be inspected on all other items. Id., Attachment B.

226. In summary, according to Mr. Miles, Licensee has no weaknesses that need to be improved as a prerequisite to restart of TMI-1, although improvement toward higher standards of excellence in radiological controls is a never-ending challenge. Tr. 16,379 (Miles); Wegner, ff. Tr. 13,284, at 27-28. Mr. Miles' opinion reflects years of experience in this field and a year of constant involvement with TMI-l radiolog-ical control, capped by a detailed assessment in January, 1981, with which NRC Staff is satisfied. Tr. 13,293 (Miles); Wegner,

-169-

ff. Tr. 13,284, at 21, 26-27; Tr. 20,668 (Neely). Recognizing ,

that the Staff was not normally given to use of the term,

" excellent," Mr. Neely characterized Licensee's program at TMI-l as better than just adequate. Tr. 16,464 (Neely).

227. The Board finds on the basis of this record that the TMI-l health physics program is appropriately organized and staffed with qualified individuals to ensure safe operation of the facility.

VII. FINANCIAL / TECHNICAL INTERFACE 228. CLI-80-5, Issue (6) asks:

whether the relationship between j Metropolitan Edison's corporate finance and technical departments is such as to prevent financial considerations from having an improper impact upon technical decisions.

229. The Staff's findings on this issue are brief.

In essence, the Staff identified the fact that technical decisions related to TMI-1, as well as GPU's other nuclear facilities, are made through the operational chain of the GPU Nuclear Corporation (or Group), headed by Mr. Arnold.

Financial matters, on the other hand, are under the purview of a completely separate chain of command. If any monetary constraints are required to be placed on the nuclear-related budget, they are only introduced into the budget development process at the highest management level. The Staff was unaware of any instances during the period since the TMI-2 accident where financial considerations have had an improper impact on

-170-i

, , - - . - - - . . . , , .-..,,.__.-..-,,,,_%,.--.--_,y. , . . . ,,__-,_,,-,......._,-,.y. --, -w . . . - - , , - . , , , , , , , . _ , - - - . . - , - . . , - . . , -

-...._.--,-.---w._

technical decisions regarding TMI-l startup. Furthermore, it was the Staff's view that there is no indication of undue influence of financial considerations on TMI operations before -

the TMI'-2 accident, as revealed by any of the various post-accident investigations. See also, Dieckamp, ff. Tr.

13,437, at 3. The Staff therefore concluded that for TMI-1, the relationship between the corporate finance and the tech-

nical departments is such that financial considerations should I

not have an improper influence on technical decisions. NRC i

Staff Ex. 4, at 26-27.

230. The process that the Staff very generally described and approved in its management SER was explained in detail by Mr. Herman Dieckamp, Chairman of GPU Nuclear Corporation, and President of GPU and the GPU Service Corporation. Dieckamp, ff. Tr. 13,437, at 1. In addition, Mr.

Dieckamp reviewed GPU's perspective and policies on the relationship between corporate finances and operation of GPU's nuclear facilities.

231. Safe operation of nuclear facilities is very demanding of resources. Experience continues to show expanding requirements which result in increased manning levels, larger operations and maintenance budgets, and additional capital for plant modifications. The rapid escalation in the price of oil has resulted in a large differential in cost between nuclear fuel and replacement power. In sum, the overall economic incentives are to provide the resources necessary to ensure

-171-

operability of the nuclear facilities. In Mr. Dieckamp's view, since operability and safety are directly linked through reliability and through regulation, the economic incentives are _

supportive of safety, not necessarily competitive. Dieckamp, ff. Tr. 13,437, at 1-2.

232. In any event, even if the financial incentive were absent, GPU would subscribe to the view that safety takes precedence over economics by virtue of the overriding require-ment to protect the health and safety of the workers and the

public, to satisfy regulatory requirements, and to constrain operation within the limits of the nuclear plants' technical specifications. Mr. Dieckamp firmly states that it is the obligation of management to reinforce the preeminence of safety, not only through the allocation of resources, but also by its practices and policies. Dieckamp, ff. Tr. 13,437, at 2.

233. While Mr. Dieckamp believes, and we concur, that there is no hard and fast dollar figure, manning level, or organizational framework which will ensure that financial considerations do not inappropriately interfere in technical, health and safety matters, a review of GPU's resource appli-cation to its nuclear plants in relationship to industry practice is relevant in ;eaching a judgment on the issue.

Based on Federal Energy Regulatory Commission reports, industry manpower surveys sponsored by the Edison Electric Institute, and informal communications with other utility companies, and recognizing the uncertainties to which this data is subject, 172-e r y--- ,-? ,e v---e-e-my gyg-,----ar-%--o,,,--- w,,,.-y w--y----, w w.,w-%weww---ew w-er-w--m y -r w p - n- --ye -em-e-emmwe--, wee-~~- -p w w ., w w ge w e - ere - w--

0 f Mr. Dieckamp testified that (1) the manpower levels at the GPU  !

nuclear plants are among the highest in the industry; (2) GPU spends more for operations and maintenance (O&M) at its nuclear plants than the industry average; and (3) GPU follows a practice of devoting a larger share of its budget to in-house manpower than the industry in general, based on its belief that l

l there is benefit from the familiarity, continuity of experi-l ence, dedication, and loyalty that derive from a strong in-house staff. Id., at 3-6.

234. Using another economic indicator as a basis for evaluating GPU management's attitude toward the allocation of resources to its nuclear plants, GPU has more than doubled its expenditures (O&M and capital improvements) at Oyster Creek and TMI-l in 1980 (and budgeted for 1981) over the expenditures averaged over the 1976 through 1979 period, despite the fact that GPU's financial health is clearly poor, with no access to external sources of capital. GPU Nuclear Corporation's manpower levels have also sharply increased. Id., at 6-8, Table 5, and Figure 5.

235. While these statistics of themselves do not prove funding adequacy, they do provide evidence that GPU's management, even in times of financial stress, has recognized the unique demands of its nuclear cbligations, and has shifted available resources to meet those obligations. Moreover, it is clear that GPU's financial commitment to its nuclear plants is high relative to the industry norm.

1

-173-

/ ,

y -

236. In addition to looking at the resource amount GPU allocates to its nuclear plants, it is important to consider the manner in which GPU allocates the budget for its nuclear activities. This process begins at the plant sites, within the various technical organizations, which are assigned the responsibility of identifying the work effort and the l physical modifications necessa;y for safe operation of the facility, e.g., TMI-1. The resulting task list is the basis for manpower requirements, materials or contractor support to operations and estimates for the cost of plant modifications.

These composite requirements are reviewed by the management of GPU Nuclear Corporation (or Group) to ensure completeness, priority, and planning adequacy. With the resulting definition of required manpower, materials, and plant modifications, GPU Nuclear seeks authorization from the plant owners for the financial resources necessary to support safe operation. This latter process formerly took place within each board of directors of the plants' owning companies af ter review by the operating utility and GPU corporate management; however, as part of the extensive organizational changes associated with the formation of the GPU Nuclear Group, the budget approval process has changed. Today, the Nuclear Management Oversight Committee, which will beccme the GPU Nuclear Board of Directors upon formal regulatory approval of GPU Nuclear, approves GPU Nuclear's budget. ?he individuals who make up this oversight committee collectively have more than 125 years of experience

-174-l

-ww----- w-g, - -, * - -w-,- w-M-'e- -w --m-- - - ' - + y :-w-s-w----w--- * - = - --ev-w+-y- p ,gw- pw- w- ew*--- w--rww e---w-g y -g .in---=<-^-T~wT wT-s~w t---'

l I

i with nuclear technology, and have senior management responsibility for either the overall GPU System and its operating utility companies, or of the GPU Nuclear Corporation.34 In the course of reviewing the operating performance of the GPU nuclear activities, the oversight committee visits the plants, and comes into contact with a r'nge of personnel such that operating concerns, including the impact of any budgetary constrainte, can be brought directly to its attention. In this manner, the oversight committee can effectively observe the needs of the nuclear plants and all supporting activities, and can directly coordinate those needs with the overall financial constraints cf the company.

Dieckamp, ff. Tr. 13,437, at 9-11; Tr. 11,462-76 (Arnold).

237. The Board has reviewed the dollars and manpower GPU is now dedicating to its nuclear activities and finds that this resource amount and allocation represents a strong commitment by GPU to operate and maintain its r.uclear plants in a manner which will protect the public health and safety. The Board finds no organizational link which will cause financial 34 The members of the Nuclear Management oversight Committee are: W. G. Kuhns, Chairman of GPU, GPU Service Corporation, Jersey Central Power & Light Corporation (JCPL), Met Ed Company and Pennsylvania Electric Company (Penelec); R. C.

l Arnold, President of GPU Nuclear Corporation; P. R. Clark, Executive Vice President of GPU Nuclear Corporation; Dr.

S. Bartnoff, President of JCPL; W. A. Verrochi, President of Penelec; B. H. Cherry, Vice President of Planning, GPU Service Corporation, and H. Dieckamp, President of GPU and GPU Service Corporation, Acting President of Met Ed, and Chairman of GPU Nuclear Corporation. Dieckamp, ff. Tr. 13,437, at 10.

i

-175-

/ , e * . .* .

o I considerations to have an improper impact upon technical decisions. To the contrary, the evidence suggests that the review process which Licensee has adopted for obtaining corporate approval of its nuclear budget can lead to increases in that budget at the corporate level despite the severe economic constraints under which GPU is functioning. Tr.

I 11,588-91 (Arnold).

VIII. SAFETY REVIEWS AND OPERATIONAL ADVICE 238. CLI-80-5, Issue (7) asks:

Whether Metropolitan Edison has made adequate provision for groups of qualified individuals to provide safety review of and operational advice regarding Unit 1.

239. The Staff's management SER includes a descrip-tion of Licensee's safety review process, and the method by which operational advice is provided to TMI-l Operations personnel. Staff Ex. 4, at 19-21, as modified by Staff Ex. 13, at 13. Licensee's description of its safety review program was provided by Mr. Philip R. Clark, Executive Vice President of GPU Nuclear Corporation. Clark, ff. Tr. 11,759. In addition, the Board was able to discuss this issue with management personnel in key positions of responsibility with respect to Licensee's safety review program, including Mr. R. C. Arnold, Mr. P. R. Clark, Mr. J. Thorpe, Mr. J. Herbein, and Mr. M. A.

Nelson. Tr. 11,531-36, 11,539-63, 11,584-86 (Arnold);

11,758-852 (Clark, Thorpe, Herbein, Nelson). In general, the

-176-

l l

Staff is satisfied, as is the Board, with the organizational method used by Licensee to carry out the safety reviews which are mandated by Commission regulations, and which are necessary to properly operate and maintain TMI-1. Staff Ex. 13, at 5-6, 8; Staff Ex. 14, Table B-1, at 8. A summary of Licensee's safety review program follows.

240. GPU Nuclear Corporation has instituted major organizational and staffing changes in order to provide additional safety review and operational advice regarding TMI-1. These changes have been developed by senior technical management with many years of experience in nuclear activities within GPU and in a variety of other organizations including the Navy nuclear power program, NSSS suppliers and architect engineer firms. The development of Licensee's revised safety review program included consideration of the TMI-2 accident experience, the many investigations of that accident, developing NRC requirements and nuclear industry standards and practices which specify safety review and operational advice criteria.35 Clark, ff. Tr. 11,759, at 1-2.

35 Included in this list of proposed or recently finalized requirements and guidelines are NUREG-0731, " Draft Guidelines for Utility Management Structure and Technical Resources,"

(September 1980), at 15-16, see insert ff. Tr. 11,820; Task I.B.l.2 of NUREG-0694, "TMI-Related Requirements for New Operating Licenses," (June 1980) and NUREG-0737, " Clarification ,

of TMI Action Plan Requirements," (November 1980), see Tr. 11,840 '

(Dornsife, Clark) and Staff Ex. 4, at 19; Second Proposed Revision 3 to Regulatory Guide 1.33, " Quality Assurance Program Requirements (Operation)," (November 1980), see Staff Ex. 4, ,

at 19; and, Draft 5 of ANS 3.2, " Administrative Controls and Quality Assurance for the Operational Phase of Nuclear Power Plants," (August 1980), see Staff Ex. 4, at 19.

l

-177-

241. GPU Nuclear Corporation's safety review and operational advice programs are designed to assure that activities are performed in accordance with company policies and applicable laws, standards, policies, rules, regulations, licenses, and technical requirements; that proposed plant, test, and procedural modifications receive independent review; i

that events, including those that require prompt reporting to the NRC, are investigated and corrected in a manner which r . duces the probability of recurrence of such events; and that trends which may not be apparent on a day-to-day basis or by consideration of individual items are detected and appropriate action taken. Id., at 2.

242. The safety review and operational advice i

programs are structured so as to assess not only individual items but also whether the work of the functional organizations is being done properly and effectively from a safety standpoint, and to identify any needed improvement in how that work is being done. Safety reviews and operational advice are performed in addition to Licensee's substantially expanded Quality Assurance program, which includes all aspects of nuclear activities which are important to safety and which audits all such activities, including the safety review and operational advice programs themselves. See 11 24 to 32, supra. Clark, ff. Tr. 11,759, at 3-4. It is Licensee's intent to avoid certain potential pitfalls associated with extensive safety review and operational programs. These pitf alls include

-178-

relying on "the other guy" to perform a safety review, and therefore not performing the job properly in the first place, or ignoring principles of safety in performing the job; and, diverting personnel with significant line functions from their primary responsibilities to reviewing large volumes of proce-dural and other changes. Tr. 11,766, 11,842-43 (Clark).

243. Licensee's safety review and operational advice programs have been so completely modified since the TMI-2 accident that an element by alement comparison with earlier programs is not practical. For example, Licensee's new safety r.eview program involves the elimination of a previous on-site safety committee, called the Plant Operating Review Committee l (PORC), with the functions of that committee being assumed to some degree by a new onsite group, but primarily assumed by j various functional groups within GPU Nuclear. Tr. 11,760-62 (Clark); Clark, ff. Tr. 11,759, at 9. We will therefore focus our attention on Licensee's new programs.

i 244. Safety Review Program. In its safety review program, Licensee seeks to emphasize its philosophy, with which we concur, that the first line of reliance on safety is having whoever is doing the job feel responsible for safety and l accountable for it, and then to provide a complete check on the job. Thus, at TMI-1, the initial safety evaluation is prepared by the individual assigned the task. The reviewer then checks the safety evaluation report or the determination that a safety evaluation is unnecessary. Tr. 11,767-68 (Clark); Tr.

-179-

11,556-67 (Arnold). The first element of the safety review program, then, is a one hundred percent review by someone other than the individual doing the work. This review is performed prior to implementation of the change on activities important to safety, including design work or changes, plant operating, emergency and alarm procedures, radiological control procedures and plant maintenance procedures. Responsibility for assuring adequate review of particular activities is assigned to the appropriate GPU organization, such as Technical Functions or TMI-l Plant Engineering. Id. at 4; Tr. 11,767, 11,781, 11,824-26 (Clark); Tr. 11,532-33 (Arnold).

245. The second element of the safety review program is the Independent on-site Safety Review Group (IOSRG) . There is a separate IOSRG for TMI-l responsible solely for that unit.

Tr. 11,772 (Clark). The TMI-l IOSRG has no line responsi-bilities or functions and is devoted solely to safety matters.

It is independent of the plant staff and reports off-site to the Manager of the Nuclear Safety Assessment Department (NSAD),

which is part of Licensee's Nuclear Assurance Division; however, IOSRG advises the Vice President of TMI-l and the TMI-l Operations & Maintenance Director, as appropriate. NRC Staff Ex. 4, at 20, as amended by NRC Staff Ex. 13, at 13; Clark, ff. Tr. 11,759, at 4. IOSRG provides before the fact review of such items as proposed Technical Specification changes, unreviewed safety questions, and other items it chooses or which are referred to it by GPU Nuclear's divisions,

-180-

such as Technical Functions or the TMI-l staff. It also provides an af ter the fact overview of activities important to safety to assess both the adequacy of individual matters and the effectiveness of the preparation and review by the func-tional groups. This responsibility includes. evaluating, on an ongoing basis, the. technical adequacy and clarity of procedures l important to safe operation of the plant. Id. at 5; Staff Ex.

4, at 50, as amended by Staff Ex. 13, at 13.

246. The IOSRG consists of a Safety Review Manager, Mr. Max Nelson, who has a B.S. in physics and thirteen years of professional experience in nuclear power. Clark, ff. Tr.

11,759, at 5. The staffing plan is for at least three techni-cally qualified individuals.

247. The third element of the safety review program i

! is the Nuclear Safety Assessment Department (NSAD) which, under the direction of its Manager, reports directly to the Vice l

President of Nuclear Assurance. Clark, ff. Tr. 11,759, at 5.

NSAD is a new concept employed by GPU to ctrengthen the safety of its nuclear activities. In addition to the safety review groups customarily employed by licensees and required by NRC to independently conduct reviews of specified plant activities, this Department has incorporated within it the resources and the freedom to conduct on their own initiative assessments of the safety implications of any and all facets of plant design and operation, to consider their potential for compromising nuclear safety and to provide management with recommendations

-181-

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

for improvements. Arnold, ff. Tr. 11,434, at 20-21; Tr. 11,532 (Arnold). NSAD has no specific tasks which it is required to do. Tr. 11,775 (Clark). Moreover, its activities are not required to fulfill any regulatory requirements. Tr. 11,584-86 (Arnold). However, in addition to overseeing the IOSRGs for l TMI-1, TMI-2 and Oyster Creek, the NSAD, through its headquarters staff, is charged with the following general responsibilities: (1) to maintain an overview of activities affecting or potentially affecting safety; this broadly defined i

responsibility makes it possible for NSAD to assess the adequacy of Licensee's entire safety review program and identify areas for improvement; (2) to serve as a corporate ombudsman accessible on a confidential basis to anyone in the company having a safety concern he or she considers is not being adequately addressed; the ombudsman is empowered to investigate the mateer, identify any needed action, seek resolution of the matter, and reply to the individual who raised the concern; (3) to evaluate on an ongoing basis the I

technical adequacy and clarity of all procedures important to safe operations of TMI-1; and, (4) to provide staff support to the fourth element of the safety review program, the General Office Review Board (GORB). Clark, ff. Tr. 11,759, at 5-6; Tr.

11,809 (Clark). The Manager of NSAD is Mr. Robert Whitesell, who has a B.S. degree in electrical engineering, an M.S. in engineering science, and a Ph.D. in mec!!anical engineering.

Mr. Whitesell's extensive experience in the nuclear power field

-182-

incIttdes three years at General Electric's Knolls Atomic Pcwer Laboratory working in the areas of reactor physics and advanced submarine power plant design; Manager of Nuclear Safety and Analysis for the Electric Power Research Institute (EPRI); and, Washington, D.C. representative for the Nuclear Power Division of EPRI. Mr. Whitesell has also worked on industry-wide

! standard setting committees, and published extensively in his field. Tr. 11,795-96 (Herbein). As the Manager of NSAD, Mr.

Whitesell will be assisted by approximately five technical staff during 1981. Tr. 11,781 (Clark). In order to facilitate NSAD's function of providing staff suppo'rt to the GORB, the Manager of NSAD will be its permanent Vice Chairman. Staff Ex.

4, at 20-21, as modified by Staff Ex. 13, at 13; Tr. 11,554-55 (Arnold).

248. GORB is the final element of GPU Nuclear's i safety review program. In fact there are three GORBs, corre-I sponding to the three GPU nuclear plants, with a common, full-time Chairman and a number of common members on the three boards. Clark, ff. Tr. 11,759, at 7. The TMI-l GORB is made up of eleven senior level individuals with diverse technical backgrounds. Lic. Ex. 26(a-k); Tr. 11,787-89, 11,791 (Thorpe).

Five of these individuals are not employed by GPU and thus provide independent input and insight. GORB reports to and takes general direction from the Office of the President of GPU Nuclear Corporation; however, GORB members have direct access to the Chief Executive Officer (Mr. Dieckamp), and the Board of i

! -183-

o Directors. The GORB charter is broadly defined to encompass all matters potentially affecting safety so as to foresee potentially significant nuclear and radiation problems. This includes the functioning of the safety review process and the 4

adequacy of the Quality Assurance program. Like NSAD, GORB has no assigned tasks; it is therefore able to devote attention to identifying previously unrecognized safety questions or underlying issues. Meetings of the TMI-1 GORB are held about once every three months. Clark, ff. Tr. 11,759, at 6-7.

249. While Licensee's allocation of its resources to various safety review functions is not identical to NRC's recommended framework, e.g., with respect to the functions of the so-called Independent Safety Engineering Group (ISEG) identified in NUREG-0731, see insert following Tr. 11,820, the differences improve rather than detract from Licensee's safety review program. Tr. 11,544-45 (Arnold); Tr. 11,770-73, 11,819-825 (Clark). In general, Licensee's safety review l

program appears to be a carefully thought out system designed by GPU to effectively utilize its unique organizational i

i framework to obtain the appropriate deiree and number of reviews, i.e., both the on-site TMI-l s taff, and the on and off-site technical support staff of GPU Nuclear Corporation.

I See, e.g., Tr. 11,550-53, 11,5;5-62 (Arnold). Licensee has i

avoided burdening its line functional employees with extensive review responsibilities. Tr. 11,761-65, if,770, 11,805, 11,825 (Clark); Tr. 11,830-31 (Nelson). Licensee has selected and 1

-184-

l 1

requires that any new personnel wit.S safety review responsibilities be highly qualified to perform these reviews, _

thereby assuring that the recommendations of its safety review grotss are based on sound technical judgment. Clark, ff. Tr.

11,759, at 9-10; Tr. 11,791 (Clark). In addition, the respon-sibilities and organizational status of t':e various entities which comprise Licensee's safety review program have been defined so as to assure that the safety review groups can operate independently, and with the requisite authority to function effectively. Id. at 10.

250. Operational Advice. In addition to the safety l review functions described above, Licensee is taking specific

[

measures to assure the proper collection, evaluation, and dissemination of plant operational experiences throughout the corporate structure and at TMI-1. NRC Staff Ex. 4, at 21, as modified by NRC Staff Ex. 13, at 13. Like the safety review program, the operational advice program supplements the responsibility of the line function organizations. These line function responsibilities are assigned to plant staff, such as Operations engineers and Plant Engineering, as discussed in 11 48, 58-59, supra; shif t technical advisors, as discussed in 1 43, supra, and 1 286, infra; and, the Plant Safety Analysis section of Technical Functions, as discussed in 1 286, infra.

It is the Plant Safety Analysis group which reviews all licensee event reports from not only GPU plants but other nuclear plants as well, and assigns to responsible line

-185-

l c

organizations the specific review and assessment of corrective action for all operating information considered to have ,

applicability to TMI-1.36 Clark, ff. Tr. 11,759, at 7-8; Wilson , ff . Tr . 11,722, at ' 7-8; Tr. 11,738-39, 11,746-52 (Wilson); Tr. 11,778 (Clark); Tr. 11,536 (Arnold).

251. In addition to the line functional organiza-tions with responsibility for providing operational advice to TMI-1, Licensee's operational advice program includes review by the TMI-l IOSRG of Technical Specification violations, report-l able events and other operating experience from within and l

outside of GPU Nuclear, with emphasis on the adequacy of corrective actions needed and taken or planned for TMI-1.

Clark, ff. Tr. 11,759, at 8. It also includes the overview responsibility of the GORB to assess the effectiveness of Licensee's utilization of operating experience from within and outside GPU Nuclear to improve TMI-l operations and activities.

l$+

252. In conclusion, based on the record presented by l Licensee and the Staff on Licensee's safety review and opera-tional advice p;ograms, we find that Licensee has made adequate provision for groups of qualified individuals to provide safety.

review of and operational advice regarding TMI-1.

36 This function is one of the review functions which NUREG-0731 designates as the responsibility of the Independent Safety Engineering Group (ISEG), but which Licensee has chosen to delegate to one of the sub-groups within its Technical Functions Division. Tr. 11,819 (Clark). See,

  • 249, supra.

-186-

A, e , . .'

j' ' .

i o /

I IX. VALIDITY OF COMPARING TMI INFRACTION, I

LER AND OPERATING EXPERIENCE HISTORY WITH INDUSTRY-WIDE STATISTICS .

253. In Issues (8) and (9) of the Commission's March 6, Order, CLI-80-5, the Board is asked to consider the following issues:

Issue (8):

What, if any, conclusions regarding Metropolitan Edison's ability to operate Unit 1 safely can be drawn from a comparison of l the number and type of past infractions of l NRC regulations attributable to the Three Mile Island Units with industry-wide infrac-tion statistics.

Issue (9):

. What, if any, conclusions regarding i

Metropolitan Edison's ability to operate Unit 1 safely can be drawn from a comparison of the number and type of past Licensee Event Reports (LER) and the licensee's operating experience at the Three Mile Island Units with industry-wide statistics on LERs and operating experience.

254. Infraction Statistics. In the management SER, the Staff provided three tables of comparative infraction and civil penalty statistics which it developed in order to draw conclusions, if possible, about Licensee's ability to operate TMI-l safely. The Staff compared the enforcement history at TMI Units 1 and 2 to selected plants and national average data using enforcement statistics compiled since 1975.37 The number 37 To evaluate the enforcement history for TMI, an attempt was made to select plants for comparison that were licensed in a similar time frame, since safety equipment required and the associated licensing requirements for this equipment are (continued next page)

-187-

of &_ap actions, noncompliances and civil penalties, and the severi y of the noncompliances were compared. In one table, '

the Staff showed the enforcement data of 16 plants and TMI-l by year (1975-1978), the noncompliances by category of severity (severity 1 being the most serious), the total number of noncompliances, and the number of inspections. In addition, the ratio of noncompliances to the number of inspections was included in the table to provide a " normalization" of the data, since IE inspection results have shown through the years that noncompliances are dependent on the inspection manhours applied. NRC Staff Ex. 4, at 27-28, Table III.I.l. The table also provides average statistics, based on all plants licensed to operate during the years shown. The Staff's interpretation of these data led the Staff to conclude that TMI Unit 1 l enforcement statistics prior to the TMI-2 accident were very close to the national average and slightly better than typical l

for the units shown. As an added positive consideration, the Staff noted that there were no severity 1 noncompliances for TMI-1. NRC Staff Ex. 4, at 28, Table III.I.l. In its second comparison, applicable to TMI Unit 2, the Staff used dcta (continued) to some degree dependent on the time the facilities were licensed. The TMI-l comparison sample included PWRs manu-factured by Babcock & Wilcox (4 units), Westinghouse (10 units), and Combustion Engineering (2 units). The TMI-2 comparison used plants licensed in a similar time period and included history for only the first year of operation. This list included two Babcock and Wilcox-designed units and three Westinghouse-designed units. NRC Staff Ex. 4, at 27.

-188-

o /

similar to that used in the first table except that the statistics were limited to the first year of operation of the selected units, and the average data points shown were those for the average of the six units listed, not the national 4

average. NRC Staff Ex. 4, at 28, Table III.I.2. With respect to this second table, the Staff concluded that TMI-2 's noncom-i pliance performance was typical of that for comparable units.

Again, there were no severity 1 noncompliances at TMI-2 prior to the accident. Id. Finally, to complete its comparison of TMI noncompliance statistics to those for other plants, the Staff tabulated the number and dollar amount of civil penalties i

that have been levied against licensees from their inception in 1973 through 1978.38 The Staff found that civil monetary penalties were levied against Met Ed on two occasions at TMI-l for physical security weaknesses; that the imposition of these two penalties spaced two years apart and the cumulative amounts, coupled with the noncompliance history of this license, did not indicate a cause for regulatory concern; and, that no order was issued to TMI for correction of a safety problem. NRC Staff Ex. 4, at 28, Table III.I.4. Overall, the 38 In the past, IE used the civil penalty primari y as a signal to management that repeated or significans ron-compliances have produced a regulatory concern, and that prompt attention must be given to correct the r .oblem. These penalties were imposed to show stronger concer n than the Notice of Violation letter conveyed. An order would be used in place of a civil penalty where the threat to public health and safety was iminent and immediate corrective action was called for. NRC Staff Ex. 4, at 28.

-189-

l Staff concluded that a comparison of Licensee's enforcement history with industry-wide statistics indicated that Licensee _

has been an average performer. NRC Staff Ex. 4, at 28.

l 255. The Board is not convinced that the data analyzed by the Staff represents a meaningful means by which to assess an individual plant's performance, or the management capability of that plant's licensee. This view was shared by Mr. Norman C. Moseley, a senior official in the Office of Inspection and Enforcement, see 1 210, supra, who stated in reference to the Staff's noncompliance comparisons:

In general, I think that the value of this information is certainly not in the statistical treatment, but rather how well a particular licensee learns from the events that occur, and how much in depth they pursue the causes; not just the apparent cause, but M.e real basic root cause, and what correc-cive actions they take in correcting their l

management systems, or their procedures or i whatever might be the basic root cause.

Tr. 13,081 (Moseley). Mr. Moseley went on to explain how IE's new Systematic Assessment of Licensee Performance (SALP) program is a program which has as one of its goals the ability to rate licensee performance on a national basis. While some of th e enforcement actions, as well as licensee event reports (LERs), are part of the information that is considered in doing this evaluation, it is only a small part of the overall in-depth analysis of licensee weaknesses and strengths. Tr.

13,082-85 (Moseley). The Board specifically asked Mr. Moseley whether, for example, with respect to the comparison of

-190-

noncompliances, one could assume that a severity 2 noncompli-ance was two times weightier than a severity 3. Mr. Moseley answered in the negative. Moreover, he noted that the compari-sons provided in these tables were further obscured by the fact that regional IE offices interpret technical specification compliance differently. Tr. 13,088-89 (Moseley). In addition, because of the statutory limitations placed on the imposition of civil penalties, Mr. Moseley considered the information provided by the Staff in its third table to also be of limited value.39 Tr. 13,089 (Moseley).

256. It is Mr. Moseley's view, and we agree, that if one really wanted to make some assessment of the relative enforcement history of licensees, one would have to summarize the types of occurrences or the types of things about which citations had been made, and then look to see whether there l

were repeats in these areas, where the corrected actions had been taken, whether they were helpful in turning around adverse trends, and then compare the plants with respect to their successes (and failures) in these areas. Tr. 13,088-89 (Moseley).

257. Licensee's response to the Board's inquiry with respect to the usefulness of infraction history statistics in 39 Mr. Moseley explained that this limitation could not have been cured by the Staff's providing a dollar figure of what the civil penalty would have been, but for the statutory limit, because the Staff did not in each case determine what the maximum penalty might have been, and so the information was not available. Tr. 13,089-90 (Moseley).

-191-

i l

assessing Licensee's ability to operate Unit 1 safely was similar to the approach taken by the Staff. Licensee's expert witness on this subject was Mr. Rocert H. Koppe, the Manager of Reliability and Safety Projects with the consulting firm S. M.

Stoller Corporation. Mr. Koppe has had substantial nuclear licensing and safety analysis experience, and, in particular, has worked extensively on operating experience analyses for the Nuclear Safety Analysis Center to identify and examine events at all U.S. nuclear plants with potential significant economic j or safety implications. Koppe, ff. Tr. 13,335 (attached witness qualifications). Using a simple count of noncompli-t ances of units in Region I, which were potentially related to plant safety, and discarding strikingly different plants and irrelevant noncompliances, Mr. Koppe developed a table of comparative statistics. He then derived a second table to show the number of noncompliances per 100-inspector-hours for Region I plants. In both cases, the performance of TMI-1 was exactly or almost exactly average. (Mr. Koppe was of the view that because TMI-2 had been in commercial operation for only three months, the limited amount of data available was not useful; consequently, all of his comparative statistics exclude TMI-2.

Koppe, ff. Tr. 13,335, at 1.) Mr. Koppe concluded that:

l [a] t least in theory you could look at all of the non-compliance reports and read in detail what they examined, to what extent they examined it and what they found, and possibly you could draw some meaningful conclusions.

l But there is no way by counting non-com-i pliances or putting them in different bins or by doing statistical analyses of them that you can draw meaningful conclusions.

Koppe, ff. Tr. 13, 335, at 25-27; Tr. 13,366 (Koppe).

-192-

258. In summary, while both the Staff and Licensee compiled statistical information on infraction histories of plants which could reasonably be compared with TMI, both parties essentially derived little meaning from these statis-tical comparisons. To the extent a conclusion might be drawn i

at all, Licensee appeared to be an average performer.

Probably, the more accurate view, however, is that with respect to TMI, there is no statistically meaningful conclusion that can be drawn concerning Licensee's ability to operate TMI-1 from a comparison of the number and type of past infractions of NRC regulations attributable to the Three Mile Island Units with industry-wide infraction statistics. The Board is of the view that this conclusion is not unique to TMI, but is the inevitable outcome of a purely statistical analysis of licen-sees' infraction and general noncompliance records which involve other subjective judgments.

259. LER Statistics. In its evaluation of LER statistics, the Staff considered three measures of Licensee's l relative performance in the industry, based on an examination of industry-wide LER statistics from 1972 to 1980. First, the l Staff considered management's effectiveness in eliminating the recurrence of causally connected events, based on the premise that once an event has occurred, good management will ensure that it is analyzed and corrected so that it will not recur.

In the case of TMI, the Staff found one series of recurring LERs related to maintaining seals on one of two pressure-lock

-193-

l l doors on the TMI-l containment. While the Staff considered this series of LERs to reflect negatively on Licensee's management, in light of the fact that there was only one such series and that in substance, this particular problem was not as serious as that for series identified on other operating plants, the Staff placed little significance on this finding.

Staff Ex. 4, at 33-34.

260. In its second method of analyzing LERs, the i

Staff tabulated the number of LERs for the first year of commercial operation and for the commercial life of the operating Babcock and Wilcox plants. For these plants during these time frames, the Staff also listed the ratio of so-called

{ Human Error Licensee Event Reports (BELERs) to LERs. O This ratio was considered more accurate than comparing numbers of LERs because while the number of LERs varies greatly according to plant type, size, and age, as well as the vintage of i

Technical Specifications and how they are interpreted by the licensee, HELERs have been shown to be stable in quantity over time, and also represent a stable percentage (20%) of the total number of LERs issued per plant. Also, the ratio of HELERs to LERs eliminates variations between plants that are due to Technical Specifications. Id., at 33-34, Table III.I.S. The 40 The Staff did not consider other PWRs (or BWRs, for that matter,) in its data base in an effort to accurately reflect the human error rate, which would be influenced by plant type, size and age. Staff Ex. 4, at 34.

-194-

Staff's statistical analysis indigated that the safety l

performance of TMI was about average. Id.

261. In its third evaluation, the Staff' compared the 1

percentage of HELERs in a series of "What Went Wrong" cate-gories, which classified the nature of the problem identified in the LER, e.g., safety equipment malfunction, monitoring instrumentation malfunction, safety equipment tripped. This analysis included in its data base the average percentage for 31 PWRs. Again, TMI-l's statistics turned out to be very close i to the average; however, TMI-2 showed statistically significant (higher) variations .in the categories of (1) safety equipment being on the wrong setting or surveillance not being performed on schedule, and (2) surveillance or maintenance not performed en schedule. The explanation offered to account for TMI-2's above average number of problems was that TMI-2 operating data spanned less than a full year during the first year of opera-tion, a period of time which usually shows significantly higher ratios of HELERs to LERs than subsequent years of operation.

See Staff Ex. 4, at 33-35, Table III.I.S.

262. Overall, despite the general disclaimer that the methods adopted by the Staff in these analyses to evaluate the quality of Licensee's management have not been proven accurate since the premises and assumptions upon which they are based have not been validated, the Staff concluded that the LER datti do not show any statistically significant or substantial anomalies for the management of TMI-l compared to that for i

other plants. Id., at 34'-35.

-195-i i.__ _. _.__,_-

l l

1 263. Licensee's LER analyst, Mr. Koppe, did not consider a purely numerical analysis of LERs to be of any value, including statistical efforts to increase the reliability of these numbers, e.g. , by ratios. Koppe, ff. Tr.

13,335, at 23; Tr. 13,354-56, 13,377-78, 13,393-395 (Koppe).

Rather, in light of Mr. Koppe's view that the level of

! reliability one can expect from comparing LERs depends directly on one's ability to isolate individual factors and their influence on performance, Mr. Koppe read every TMI-l LER, summarized it for himself, classified it according to system and problem area, relisted them all, and then reviewed his notes for patterns, dividing the LERs topically into (1) personnel errors; and (2) loss of key safety system functions, defined as systems the failure of which can result in signifi-cant core damage if the system is not restored fairly rapidly af ter the initiation of a transient.41 Koppe, ff. Tr. 13,335, i

l at 27-44; Tr. 13,383 (Koppe). In his personnel error analysis, Mr. Koppe took the entire five-year history of TMI-1 and compared it with the experience of PWRs4 for the first eight 41 The exact amount of time available depends upon the tran-sient and the system involved. Koppe, ff. Tr. 13,335, at 37-42.

42 For example, since thousands of failures of safety-related components are reported every year, it would have been impossible for Mr. Koppe to perform a careful analysis and interpretation of all failures. Therefore, he focused on the relatively few occurrences of complete failures of safety systems which '

have been experienced. Mr. Koppe gave four reasons why 'ocusing on these system failures is particularly useful: (1) it is the reliability of safety systems, rather than the reliability of individual components, which determines the (continued next page)

-196-

.e, --w,,. ,-ny-.-...,-..,,,m._,,v,,- ,,Jw---,- - , - ,,,,,,,-,,-v,-~,,_-,,,,_,,,,,,,,_,w,,,-,,e,,,,,m,--,_w,,_,_,_,,m,,-,,-,---,,s,,,,-.

o months of 1980, thereby manageably reviewing a sample of LERs in detail. Koppe, ff. Tr. 13,335, at 29-30. Mr. Koppe's safety system analysis focused upon complete failures of safety systems, and varying time periods and different groupings of units for the system failures he analyzed, e.g. , PWRs and BWRs in the case of diesel generator system failures, but only PWRs in the case of high pressure injection system failures, in order to improve the reliability and meaningfulness of the results of his comparisons. Koppe, ff. Tr. 13,335, at 34-42; Tr. 13,375-77, 13,385-88 (Koppe).

264. On the basis of his very detailed LER study, Mr. Koppe concluded that the pattern of personnel errors at (continued) safety of the plant; (2) many system failures result from causes other than random component failures; therefore, the failure rate for systems is generally higher than would be t

calculated using the failure rates for individual components; (3) system failure rates depend not only on the rate of failure of individual components but also 'on the frequency with which these components are removed from service for preventive maintenance. Most units do not report preventive maintenance on safety components unless failures of redundant components result in a system failure; and (4) a number of system failures result from personnel errors. While most plant personnel errors have a minor effect on plant safety, those which cause system failures represent a substantial contribution to the total " unsafety" of the plant. The effect of plant management and personnel on the rate of failure of systems is perhaps the most meaningful measure of their effect on plant safety. Moreover, while the complete failure of a given system at plant "x" is an unlikely event, the reliability of such events for purposes of comparison is based on the fact that it represents a level of deficiency which is sufficiently great that one can feel assured that it will be reported in an LER; and, while it is unlikely that a given event in one particular system will occur in any one plant, it is likely that one such event will occur in one of the systems analyzed in a given plant. Koppe, ff.

Tr. 13,335, at 34; Tr. 13,379-80 (Koppe).

-197-

l l

l TMI-l was typical of industry-experience, i.e., a decreasing rate as the unit matured. The average rate of reported errors at TMI was slightly above average; in his view, this was almost certainly due to a greater willingness at TMI to blame person-nel error for component malfunctions. Tr. 13,361, 13,378 (Koppe). However, the most serious type of personnel error, involving tagouts and lineups, occurred less frequently at TMI than at the average PWR. Koppe, ff. Tr. 13,335, at 33. With respect to the failure rate of safety-related systems, Mr.

Koppe's comparison of the failure rates of four key systems 43 to the failure rates / system unavailabilities used in the WASH 1400 evaluations, as well as with failure rates on an industry-wide basis, led him to find that in all cases, actual experience was better (by a very large factor) than WASH 1400 probabilities, even if one ignored the lower risk of system failure which would result if one accounted for the ability of operators to restore systems to service. Mr. Koppe considered the fact that industry experience was significantly better than

( WASH 1400 levels to be an Andication of the more than adequate level of safety which is being achieved today. Since TMI-l was average or somewhat-above-average when compared against the industry statistics, he concluded that on the basis of his LER 43 Mr. Koppe analyzed system failures of the diesel generator system, the high pressure injection system, the auxiliary or emergency feedwater system, and the low pressure injection system. Koppe, ff. Tr. 13,335, at 38-41 3,

-198-

~, .m,,,....____ , . , _ , ._v.,.., ... -,m

l i

analysis, the operations of 'TMI-l prior to March,1979 were such that they provided an adequate level of safety. Koppe, ff. Tr. 13,335, at 42-44, 265. The Board is convinced that a purely numerical analysis of LER data will not result in a meaningful basis for l comparing industry-wide statistics with those of an individual plant, such as TMI-1. As in the case of noncompliance history, this view was shared by Mr. Moseley, from IE, whose personal view was that purely statistical analyses of LER data are incapable of providing a basis for reaching a conclusion with respect to Licensee's management. Tr. 13,095-96 (Moseley). In his view, while statistics should be looked at to assess whether there is an extreme problem or apparent problem, they cannot be relied upon absent introducing judgment calls, and j other frequently subjective factors. Id.; Tr. 13,099-100 (Moseley). While the Board cannot independently verify the

! conclusions reached by Mr. Koppe, who introduced into the 1

analysis some of the factors to which Mr. Moseley alluded, we are satisfied that Mr. Koppe's analysis provided us with no basis to suspect that there were any serious shortcomings in TMI-l LER history which would cause us concern about Licensee's present management capability. We therefore find that conclu-sions drawn from a purely statistical comparison of Licensee's LER history, including the number and type of past LERs, with industry-wide LER statistics, are of limited value. We conclude that on the basis of Mr. Koppe's very detailed

-199-

. - e' ,

o s quantitative and qualitative analysis of this data, there is no reason to question Licensee's ability to operate Unit 1 safely.

266. Operating History. The Staff relied upon its LER analysis in responding to this portion of CLI-80-5, Issue

9. Staff Ex. 4, at 33-36. Licensee's analysis of TMI-l's operating history compared the capacity factor of TMI-l with three other groupings of nuclear units: Babcock & Wilcox (B&W) units other than TMI-1, PWR units other than TMI-1, and all nuclear units other than TMI-1.44 Koppe, ff. Tr. 13,335, at 5-8, Table 1. In order to account in his analysis for the varying years of nuclear experience gained at each plant, Mr.

Koppe first weighted and averaged the lifetime capacity factors of each of the units in the group, using data from the first full month of commercial operation for each unit through the end of March, 1979. Id., at 7. Since TMI-l and the B&W units have nuclear system designs that are generally very similar, Mr. Koppe noted that the strongest similarity of performance should be evident in these two statistics. (Generic problems that affect the nuclear systems of B&W units will probably also affect TMI-1. Conversely, problems affecting the nuclear systems of units designed by a different vendor, but not affecting the B&W units, will probably also not influence the 44 In this comparison, Mr. Koppe included performance statistics as of March 31, 1979 for all large modern nuclear units (all units rated 400 MW or larger), which consisted of seven B&W units, 37 PWR units, and 56 nuclear units in total, excluding the TMI units. Koppe, ff. Tr. 13,335, at 7-8, Table 1.

-200-

i

, performance of TMI-1.) Because all PWR units have many design similarities inherent in their reactor systems, one would expect that performance of PWR units would also be similar to I

that of B&W units and TMI-1. Id., at 8. In fact, the lifetime capacity factor of TMI-l is considerably higher than the average lifetime capacity factors for other B&W units, for other PWR units, and for all other nuclear units both as a weighted average over the plants' lifetimes and, in Mr. Koppe's second analysis, when comparing TMI-l's first five years of commercial operation with the corresponding averages of other units. This is the case even though the experience for other PWR units and other nuclear units in general includes pro-portionally more experience during the later, more mature years of operation than does TMI-l when nuclear units chstacteris-tically operate with a higher average capacity factor than young plants (i.e., plants with up to four years of commercial operation). Moreover, the largest differential (72.3% v.

58.7%) in lifetime capacity factors is apparent between the performance of TMI-l and other B&W units, which Mr. Koppe attributes to TMI-l's ability to avoid many of the problems that caused outages at these other units. Koppe, ff. Tr.

13,338, at 8-10, Table 1, Table 2.

267. Licensee also compared TMI-l's operating performance with other units on a more detailed level by reviewing the systems or components which have caused plant outages or load reductions in the industry, and evaluating

-201-

l l

l whether these problems affected TMI-1. Koppe, ff. Tr. 13,325, at 10-23, Tables 3 and 4. Mr. Koppe concluded that all the j systems at TMI-1 have performed as well or better than corre-i sponding systems at similar units. Id., at 23.

268. The Board delved into the question of whether a utility could increase or maintain a high capacity factor intentionally, by disregarding necessary work at the plant, such as preventive maintenance. Mr. Koppe was of the view that while a plant's management could achieve this result over the short term, it sould not be possible over an extended period of time because of technical specification limits. Tr. 13,350, 13,358 (Koppe).45 Secondly, since refueling outages are responsible for more than one-third of all plant outage time, in Mr. Koppe's view, good refueling outages are prooably the way in which good management is most effective in improving capacity factor. Mr. Koppe viewed Licensee's success in this regard as a positive indicator of management competence. Tr.

13,353 (Koppe); Koppe, ff. Tr. 13,335, at 23. In addition, Mr.

Koppe concluded that TMI-l's ability to minimize the impact of hardware problems, not so much in any one instance but over a whole range of incidents to which many plants were subject, by consistently responding effectively to these individual l

45 At TMI-1, it is clear that Licensee's operations and main-tenance expenditures have been consistently above the industry norm, see 1 233, supra, whic. suggests a steady attention given to maintenance activities.

-202-

problems, created a cumulative impression of good management.

Tr. 13,356 (Koppe); Koppe, ff. Tr.13,335, at 22-23.

269. The record is clear cut with respect to the high capacity factor of TMI-l over its operating lifetime. The Board is satisfied that this achievement was not the result of a disregard by Licensee of safety in its efforts to resolve equipment problems and minimize the lengths of outages. While a good operating history does not in itself prove that a licensee is able to operate a plant safely, the Board is convinced that in the case of TMI-1, a review of that plant's operating history indicates that a good capacity factor and proper regard to safety were two sides of the same coin, and that both results required good management practices. See Tr.

13,351-52 (Smith, Koppe).

X. LICENSEE'S MANAGEMENT RESPONSE TO THE TMI-2 ACCIDENT 270. In Issue (10) of CLI-80-5, the Commissioners l asked the Board to consider the following question:

1 Whether the actions of Metropolitan l Edison's corporate or plant management (or l any part or individual member thereof) in

! connection with the accident at Unit 2 reveal deficiencies in the corporate or plant management that must be corrected before Unit 1 can be operated safely.

1 271. In view of the breadth of Issue (10), it is not surprising that the Board received evidence on various sub-issues in response to this question. In its management SER and in supplement two to NUREG-0680, NRC Staf f Ex. 13, as well as

-203-

i l

in discussions with the Board, the Staff Locused upon two issues related to the TMI-2 accident: the flow of information during the accident, and an on-going Department of Justice i

investigation into past procedure adherence practices at TMI.

Staff Ex. 4, at 36-37; Staff Ex. 13, at 9-10; Tr. 13,025-72 (Moseley). Licensee provided a chronological description of the actions taken by GPU management in the days and weeks following initiation of the TMI-2 accident. Keaten and Long, ff. Tr. 13,242. Mr. William S. Lee, ? resident and Chief Operating Officer of Duke Power Company, also testified with regard to his perspective on what Licensee's response to the accident revealed about Licensee's management capability. Mr.

Lee's views were based upon his work at TMI af ter the accident when he was called by Mr. Dieckamp to assist in overseeing on-site activities, as well as his familiarity with Licensee's top management and organizational structure. Lee, ff. Tr.

13,251, at 1-3; Tr. 13,254, 13,273-75 (Lee). Licensee's I

testimony by Mr. William Wegner provided additional information on this subject, based on Mr. Wegner's deep invcivement in TMI activities since the TMI-2 accident. Wegner, ff. Tr. 13,284, at 1-8, 33-35; see n.5, supra.

272. In its consideration of Issue (10), the Board was anxious to avoid a minute-by-minute recap of the TMI-2 accident scenario or other topics which were reviewed in extraordinary aetail in tne numerous post-accident investiga-tions, or to review the multitude of Licensee responses to the

-204-

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

accident, which essentially consist of averything that has happened at TMI since March 28, 1979. Uowever, we wern interested in determining whether any particular actions on the part of Licensee reflected positively or negatively on Licensee's management competence, and whether any of the l witnesses considered there to be more subtle factors, such as management attitude, which were deficient in Licensee's management during the events following the accident. Included in this latter inquiry was our interest in Licensee's responses to external stimuli, such as the Staff'a inspections, as well as Licensee's internal reactions to the TMI-2 accident, such as whether Licensee evaluated and, if so, in wh'at manner, the performance of individuals in the company during the accident.

l See Tr. 11,597-608 (Arnold).

t 273. The NRC has conducted numerous special inves-tigations into the TMI-2 accident, from various vantage points.

Included in this list is IE's extensive investigation of the accident, NUREG-0600, in which the Staff provides an account of the actions and management decisions undertaken by those members of Licensee's management who were called to the site to provide emergency direction to cope with the operational aspects of the accident. Staff Ex. 4, at 36, Appendix A.

NUREG-0600 also addresses the additional support that was provided through Licensee's organization and by other parties to support the onsite operational activities. Id. In his cover letter forwarding the escalated enforcement action which

-205-

resulted from the IE investigation findings, the Director of IE stated that "the NRC Staff will consider the effectiveness of actions Taken in response to this correspondence in developing its position on readiness for re' start before the Atomic Safety i

and Licensing Board constituted to consider the restart of Unit 1." Staff Ex. 4, Appendix A, at 1. As a result of the IE investigation findings and subsequent enforcement action, Licensee responded in a number of ways including questioning some of the conclusions reached by IE, paying a fine, reaching an agreement with the Staff with respect to acceptable correc-tive actions, and transmitting additional responses to document corrective actions taken or to be taken for each item of r

noncompliance identified in the original Notice of Violation.

Ij., at 1-7. As a result of Licensee's responsiveness, there 4 are no remaining items raised by IE's investigation of the accident where the Office of Inspection and Enforcement considers Licensee's response to be inadequate, or where there remain differences between the positions of IE and Licensee.

Tr. 11,982 (Keimig). The Board considers this fact to be a positive factor in our assessment of Licensee's response to the TMI-2 accident, in particular, and in general, an indication of a responsible management attitude by Licensee towards its nuclear-related activities.

274. The Staff's initial management SER identified two open 1: ems which the Staff was not able to sign-off on at the time that document was issued in November,1980. Those i

l

-206-

i f items were a then-pending IE investigation into information flow during the TMI-2 accident, and a separate investigative

, effort by the Department of Justice concerning adherence to leak rate testing procedures. Staff Ex. 4, at 36-37. With respect to the first open item, the Staff issued NUREG-0760, entitled, " Investigation into Information Flow During the Accident at Three Mile Island," in January, 1981. This report i

identified a number of deficiencies in Licensee's communication

efforts on March 28, 1979;46 however, IE also concluded that 46 The specific conclusions reached in NUREG-0760 were

(1) There was significant information that did not adequately flow either on the site or to the necessary offsite groups on the day of the accident.

L

(2) On the day of the accident, an effective system t

did not exist to ensure adequate information flow; i.e., to t

provide significant information for dissemination and evaluation within the onsite organization or offsite within the Met Ed or GPU organizations as well as the NRC, Commonwealth of Pennsylvania, and other agencies.

1 (3) Those individuals on site failed to understand I

the extent and significance of the problems confronting them on the day of the accident; this contributed to the inadequate flow of information.

! (4) Met Ed was not fully forthcoming on March 28, 1979 in that they did not appraise the Commonwealth of Pennsylvania of either the uncertainty concerning the adequacy of core cooling or the potential for degradation of plant conditions.

(5) Information was not intentionally withheld from the State on the day of the accident.

l (6) Information was not intentionally withheld l from the NRC on the day of the accident.

(7) The NRC did not have an effective system to ensure that information was properly accumulated, evaluated, j (continued next page) l

-207-

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

pertinent information was not inter.tionally withheld by Licensee. Staff Ex. 5, at 10-11. The problems with communica-tions and information flow identified in NUREG-0760 pertain to the adequacy of Licensee's emergency plan and procedures; there are no management, organization, or staffing issues addressed in NUREG-0760 for which additional Licensee action has been identified by IE. Staff Ex. 13, at 9; Tr. 13,071-72 (Moseley).

275. With respect to the second outstanding item in the management SER related to Issue (10), the Department of Justice investigation, the Staf f explained in its March,1981 supplement to NUREG-0680 that the investigation had not been completed, and that the NRC had no information as to when 'it might be completed. However, on the basis of the information thus far obtained by the Staff there appeared to be no direct connection between the incident under investigation and the Unit 2 accident. Further, although the Department of Justice investigation remained open, no indication of practices at Unit 1 similar to those under investigation at Unit 2 were identified. On the basis of these facts, along with the 1

l Staff's review of the numerous actions by Licensee to reenforce 1

its management policy regarding strict adherence to procedures,4 the Staff concluded that those deficiencies in (continued) and disseminated.

(8) Reporting requirements, both to NRC and to l the State, were not sufficiently specific on March 28, 1979.

1 Staff Ex. 5, at 10-11. I 47 These actions taken by Licensee to emphasize its management policy regarding strict adherence to procedures (continued next page) l l

1

-208- -

Licensee's corporate or plant management which were revealed by Staff investigations of corporate or plant management action in connection with the Unit 2 accident had been corrected or identified for correction prior to restart of Unit 1. Staff Ex. 13, at 9-10.

276. A descriptive account of Licensee's on-site and off-site management response to the first day of the TMI-2 accident, and in the days and weeks immediately thereaf ter was provided by Messrs. Keaten and Long, two members of Licensee's

management who were actively involved in post-accident events.

Keaton & Long, ff. Tr. 13,242, at 1-2. While acknowledging that the accident clearly led to major changes in the organiza-D tion and staffing of Licensee's nuclear-related activities as well as its emergency plan, Messrs. Keaton and Long offered a l positive perspective on the actions taken by Licensee's management in connection with the TMI-2 accident. Actions by members of Licensee's management in response to the accident, in the opinion of those witnesses, do not reveal management deficiencies that must be corrected before Unit 1 can be (continued) include issuance of a memorandum from Licensee's Office of the Chief Operating Executive communicated directly by face-to-face discussion between management and plant personnel; incorporation of the policy into the procedure that provides guidance on the conduct of operations; the establishment of a management policy for disciplinary measures to be taken for lack of adherence to procedures; and, the establishment by Licensee of an operations inspection program to verify procedure adherence. Staff Ex. 13, at 10.

-209-

o <

I operated safely. Keaten & Long, ff. Tr. 13,242, at 27-23.

This conclusion is supported by Mr. Wegner of BETA in his testimony, although his basis for doing so is much broader. In Mr. Wegner's view, the accident at TMI-2 shculd be viewed as a means by w'.tich industry problems which were not the result of

, the errors of a few people or a group of individuals, or unique to TMI-1, were made public. The stage was set for the accident at TMI-2 many years ago, according to Mr. Wegner, based on problems applicable to the entire civilian nuclear power industry. Wegner, ff. Tr. 13,284, at 33-34; Tr. 13,320-22 (Wegner). While not all of these deep-seated problems have i

obvious solutions, nor have they been corrected to the point

! where Mr. Wegner is fully satisfied, in his view, GPU has made more progress than other utility companies in implementing changes as a result of the accident and, more importantly, is not now deficient in its corporate or plant management such that Unit 1 cannot be operated safely. He therefore concludes that there are sufficient management and technical capabilities within GPU to permit restart of TMI-1. Wegner, ff. Tr. 13,251, at 34-35.

277. Mr. William S. Lee, President of Duke Power Company and Chairman of INPO, also testified on the subject of Licensee's management response to the TMI-2 accident, both with respect to the events immediately following the TMI-2 accident and Licensee's actions in connection thereto, as well as with respect to his general views on Licensee's management structure l

-210-1

y .

. . .J ,

G /

and capability. See 11 15, 20, supra. With tospect to the events immediately following the TMI-2 accident, Mr. Lee believes that Licensee's management of the unprecedented technical and organizational tasks with which it was faced, as well as the pressures of the event and the demands of the media, was accomplished with great skill and steadfast purpose under conditions which were difficult and trying. Lee, ff. Tr.

13,251, at 5. In general, Mr. Lee believes that Licensee has 1

demonstrated a sound approach to its nuclear-related activities since it began its involvement in nuclear power with the construction of the Saxton experimental unit, followed by Oyster Creek, TMI-1, and TMI-2. Id[. , at 6-9 ; Tr. 13,276-78 (Lee). Mr.1.ee personally knows and highly recommended to us the top managemant within CPU Nuclear. Lee, ff. Tr. 13,251, at 4, 12; Tr. 13,270, 13,278 (Lee). Also, Mr. Lee echoed Mr.

Wegner's view that while it is easy to apply hindsight and characterize the changes which have taken place at TMI as indicia of deficiencies which existcd in GPU's management,

these changes were precipitated, in fact, by the lessons learned from the accident and do not reflect practices in Licensee's management prior to or during the accident which did not exist industry-wide. Lee, ff. Tr. 13,251, at 9-11; Tr.

13,267-70 (Lee).

278. Finally, the Board considered the evidence it received on Licensee's management attitude in evaluating whether the actions of Met Ed's corporate or plant management

-211-

(or any part or individual member thereof) in coanection with the TMI-2 accident reveal deficiencies in the corporate or plant management that must be corrected before Unit 1 can be operated safely. Recognizing the sensitivity inherent in such an inquiry, we nevertheless sought information from Mr. Arnold, in addition to his general views and the views of other l witnesses concerning Licensee's attitude towards its nuclear l

responsibilities, see 11 21 to 22, aupra, on any in-house evaluations conducted by Licensee of its staff's performance during the accident. Licensee did not take the appcoach of looking at a particular individual and evaluating how that individual responded to the accident as it developed. Tr.

11,597 (Arnold). However, Licensee did conduct extensive internal investigations into the accident of sufficient scope

, such that the performance of the people involned during the accident could be assessed. Id. In deciding how to best utilize people in the GPU Nuclear organization, Mr. Arnold concluded that there were three elements related to the TMI-2 accident which he had to consider:

Those three elements were: how did the individual actually perform relative to his responsibilities at the time of the accident and in the first few days, . . . following the event itself. The second (element] was what did we lear- about the performance of individuals in the time period before the accident as to how the facility was being supervised, administered and managed. And the third element . . . was what were the interests of the external organizations, the surrounding communities in particular, and how did their perception of what happened during the accident ar a 'he need for them to

-212-

c j l

view the organization that operated TMI-1, again with [] credibility (,] effect [] those judgments.

Tr. 11,598 (Arnold). Licensee's evaluation included not only the performance of licensed personnel in the control room at the time of the event, but individuals up through the operating organization supervisory and management chain, on and off-site.

( Tr. 11,598-99 (Arnold). Mr. Arnold of course could not speak l

l with regard to judgments that were made about himself or those senior to him in the organization; however, he was able to represent to us that where appropriate, job reassignments have

been made. Tr. 11,599-601 (Arnold). While Mr. Arnold stated j that no judgments had been made to impress the public or the Board, Licensee considered in its assignment of management positions whether the public's confidence in an individual's ability to perform could influence that individual's ability to i perform all of his assigned responsibilities. Tr. 11,602-04 i

(Arnold). In concluding this discussion, Mr. Arnold assured us of his confidence in the appropriateness of the assignments of people within the GPU Nuclear Corporation. Tr. 11,604 (Arnold).

279. In summary, the Board has received various pieces of information each of which pertain to CLI-80-5, Issue (10). With respect to the Staff's position on Licensee's management response to the TMI-2 accident, we find the Staff's resolution of this issue to be reasonable, in light of its detailed investigations into the accident, which are satis-factorily resolved. We find that Licensee has evaluated very

-213-

carefully and has instituted extensive changes in its organization and the manner in which its nuclear activities are conducted as a result of its review of the TMI-2 accident scenario and the lessons learned by the industry as a whole from the TMI-2 accident. Finally, we are satisfied that i

Licensee has conducted a thorough investigation into the conduct of its staff during the accident, and has instituted 4

changes in its staffing, as necessary, in response to that investigation. We therefore conclude that the actions of Licensee's corporate or plant management (or any part or individual member thereof) in connection with the accident at Unit 2 do not reveal deficiencies in the corporate or plant management that have not yet been corrected, which must be corrected before Unit 1 can be operated safely.

l XI. TECHNICAL CAPABILITY AND RESOURCES l

i 280. In CLI-80-5, Issue (11), The Commission asks:

l Whether Metropolitan Edison possesses sufficient in-house technical capability to ensure the simultaneous safe operation of Unit 1 and clean-up of Unit 2. If Metropolitan Edison possesses insufficient technical resources, the Board should examine arrangements, if any, which Metropolitan Edison has made with its vendor and architect-engineer to supply the necessary technical expertise.

281. The Technical Functions Division of GPU Nuclear Corporation, with broad centralized responsibility for opera-tional support and technical review of GPU Nuclear's plants, is

-214-

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

a significant departure from the organization which existed prior to the TMI-2 accident. Wilson, ff. Tr. 11,722, at 2-3.

Prior to the accident, the technical resources within the GPU system were split principally between Met Ed's Generation Engineering Department and the GT J Service Corporation. The Met Ed group, located in Reading, Pennsylvania with a total of i 40 personnel, provided technical support to TMI operations and engineering staff. Id., at 3, Table 1. Occasional technical assistance was provided by the GPU Service Corporation's technical staff, although the primary function of this group of approximately 108 professionals was to support new plant construction, both nuclear and fossi1. Id., at 3, Table 2; Tr.

11,730 (Wilson). The recently organized Technical Functions l

l Division of GPU Nuclear Corporation has greater technical capability and resources devoted to nuclear projects than did the combined Met Ed Generation Engineering Department and GPU Service Corporation as of March 1979. Working in Technical l Functions are some of those personnel previously with Met Ed Generation Engineering, GPU Service Corporation, and profes-sionals formerly at Jersey Central Power and Light Corporation.

In addition, significant staff augmentation has occurred, and continues through Licensee's aggressive pursuit of additional l technical personnel. Wilson, ff. Tr. 11,722, at 4. As of February 1981, Technical Functions now includes approximately 255 personnel, and is authorized to expand to approximately 400 in 1981, if qualified personnel can be recruited. Tr. 11,732 l

-215-

(Wilson). Two comparisons of Licensee's present technical resources to those existent in March 1979 illustrate the extent of expanded technical capability which Licensee has developed since the TMI-2 accident: (1) the total number of in-house professionals and the total manyears of experience has been increased by at least a factor of two; and (2) the number of in-house professionals and the total manyears of experience dedicated to TMI-l has increased by greater than a f actor of four. Uilson, ff. Tr. 11,722, at 4; Tr. 11,732 (Wilson).

Thus, Licensee has substantively redirected and increased the technical resources available in-house to support its nuclear activities. Tr. 11,730 (Wilson).

282. Currently, approximately 155 people, or 60% of Technical Functions, is dedicated to TMI-1. The remaining 40%

is apportioned between Oyster Creek (35%) and TMI-2, Forked River and Saxton. Id., at 5. This allocation of resources represents only the division of GPU Nuclear's in-house tech-nical capability; that is, if one includes Licensee's con-tractor resources in assessing the allocation of Licensee's technical resources, a similar number of professionals is assigned to TMI-1, TMI-2, and Oyster Creek.40 In total, 48 The use of a larger percentage of in-house resources at TMI-1 than at Oyster Creek is a reflection of the early involvement of the GPU Service Company in TMI-l in comparison to Oyster Creek. If you look at GPU's total available technical resources, both in-house and contractor resources, 38-40% are devoted to TMIsl; 25-28% are devoted to TMI-2; and 32-33%

are allocated to Oyster Creek. Tr. 11,733-35 (Wilson).

-216-

Licensee has approximately 750 to 1,000 professionals working on technical matters at any point in time. Tr. 11,732 ,

(Wilson). While the majority of Technical Functions personnel t

are _ located at GPU's Parsippany offices, a number of staff work at each of the nuclear plants, or rotate between these loca-tions. Wilson, ff. Tr. 11,722, at 3.

283. The Vice President of Technical Functions reports to the Office of the President of GPU Huclear Corporation. 53 is responsible for the quantity and quality of the staffing of the Division; the proper functioning of the interfaces of Technical Functions with the operating plants and other divisions of GPU Nuclear; the effectiveness of the licensing and technical support provided to GPU's nuclear facilities; the use and direction of outside engineering resources; and, ultimately, the technical performance of the GPU Nuclear plants. Wilson, ff. Tr. 11,722, at 5; Staff Ex. 4, at 5. GPU Nuclear's educational and experience requirements for the Vice President of Technical Functions exceed the requirements recommended by the Staff 49 and consist of a 49 The Staff has specified for managers of divisions within a company's nuclear organization, such as the Vice President of Technical Functions (or other GPU Nuclear division Vice Presidents),

a Bachelor's degree in a subject of relevance, and eight years of responsible experience of which three years shall be specifically related to the type of function he will perform.

This criterion was initially set forth in the Staff's " Criteria for Utility Management and Technical Competence," at 25 (July 17, 1980), which subsequently was issued as a draft report for interim use and comment as NUREG-0731, " Guidelines for Utility Management Structure and technical Resources,"

(September 1980), in which the a me recommendation is stated at page 14.

-217-

minimum of 15 years of responsible engineering management experience, the majority of which must be in the nuclear -

industry, and an engineering degree in one of the principal engineering disciplines with advanced degrees or specialized training desirable. Wilson, ff. Tr. 11,722, at 5. The current Vice President of Technical Functions is Mr. Richard Wilson, who possesses a B.S. and an M.S. in mechanical engineering, is a graduate of the Oak Ridge School of Reactor Technology, possesses a professional engineering registration, and has approximately 26 years of experience of which almost all is in the field of nuclear technology. Id., at 5-6, attached witness qualifications.

284. Reporting to Mr. Wilson are the Managers of the six Departments wf. thin Technical Functions, namely, Engineering Projects, Systems Engineering, Engineering and Design, Licens-ing and Regulator'f Affairs, Startup and Test, and Engineering Services, each of whom has at least a bachelor's degree in science or engineering and from 9 to 24 years of nuclear l

experience.50 Staff Ex. 4, at 5; Wilson, ff. Tr. 11,722, at 6-12, Figure 1.

There is also a small administrative staff.

The educational and professional qualifications of the Technical Functions staff includes 178 individuals with B.S.

degrees of which 62 have M.S. degrees and 6 are Ph.D.'s. Id.,

50 As of one February, 1981, one Department 6anager had just left the organization, and Licensee was actively seeking a replacement. Tr. 11,723-24 (Wilson).

-218-

l. -

at Table 3. There are 2,315 accumulated years of engineering experience in Technical Functions, of which 1,294 years have _

been in the field of nuclear technology. Id. There are also 57 individuals with nuclear operations experience. Id. In order to understand the breadth and depth of technical capabil-4 l ity which is now available within the GPU Nuclear organization, a brief review of the Departments within the Division is appropriate.

285. The Engineering Project Management Department provides a central project or task management capability for the Division. This Department coordinates the detailed tasks ,

within the various Technical Functions departments, interfaces j directly with the plants and other supporti .ig Divisions of GPU Nuclear Corporation, directs and coordinates the work of outside technical organizations, and assumes the overall technical schedule and monetary responsibility for plant modifications. Wilson, ff. Tr. 11,722, at 6. Each plant modification project is assigned to a project or responsible engineer within the Department, who has responsibility for technical follow-through until modifications are turned over to the facility's Operations staff. Id. At TMI-1, a senior Engineering Project Management Department staff member is presently assigned on a full-time basis to coordinate Technical Functions personnel at the site. I d,. , at 7.

286. The Systems Engineering Department provides a number of important technical support capabilities for GPU's

-219-1 l -.-- -. _ . . . - . - . . - . . - . - . , , - . . _ - - - - - - - - , - - - - - - - - - - ~ - - - - - - - - - - -- -

l nuclear facilities. Systems Engineering directs the activities of the Shift Technical Advisors (STAS), who work on shift with ,

the unit operating staff. The Department includes a sub-group that operates and programs the plant process computers. Id.;

Tr. 11,738 (Wilson). The Plant Analysis sub-group of Systems Engineering, in addition to directing the activities of the STAS, also has two or three professionals who review industry experience as well as GPU's plant experience as reported by LERs and other kinds of communications, such as the computerized NOTEPAD system which connects Licensee to other B&W owners and provides a means by which important plant behavior information can be transmitted quickly between licensees. Wilson, ff. Tr. 11,722, at 7-8; Tr. 11,739, 11,746-51 (Wilson). Plant Analysis personnel perform a third j important function, namely, analyzing the ongoing performance of the plant by, for example, running heat balances on TMI-l I

and assessing the performance of equipment within the plant.

Tr. 11,739 (Wilson). The Safety and Analysis sub-group within the Systems Engineering Department performs safety and tran-sient analysis work for Licensee. The Safety and Analysis sub-1 group includes approximately ten personnel who utilize computer codes and models of both TMI and the Oyster Creek station to carry out fairly sophisticated transient analyses.51 Tr.

11,739-40 (Wilson).

51 Licensee does not rely upon this group to perform the l analyses required by regulation; instead, Licensee uses outside consultants for that purpose who have codes and topical reports which have been accepted by the NRC. Tr. 11,740-46, (continued next page)

-220-

--.-._y, . - - _ ,w . , _ _ _ _ . ..,_,,m ,_~._---,---__,,,..r -

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

o 287. The Engineering and Design (EED) Department of Technical Functions provides a centralized technical capability for engineering work needed on plant systems and components.

Wilson, ff. Tr. 11,722, at 8-9. This function is similar to that prov *ded by an architect / engineering organization, although on a reduced scale. Id., at 9. EED performs the engineering for plant modifications, interfaces with and reviews engineering performed by outside organizations, prepares drawing and specifications, investigates failures of plant systems and components and provides readily accessible technical capability to support general operations or plant outages. Id.

l 288. The Department of Licensing and Regulatory i

Affairs monitors to ensure continuing compliance with licensing I

and regulatory requirements. It is the principal interface with the NRC. Licensing and Regulatory Affairs also admilis-ters and controls environmental permits and other licensing issues which arise in connection with all Federal, State and local regulatory groups. Id., at 9-10, 289. The Startup and Test Department of Technical l

Functions provides operational experience to support nuclear (continued) l 11,751-52 (Wilson). Licensee does, however, use its own transient l

analyses as a cross-check on other work being done, and may challenge conclusions reached by outside experts on the basis of doubts which Licensee's technical personnel feel have i not been fully resolved. Id.

l i

-221-

a- . g .

engineering activities, and plans, directs and maintains responsibility for startup and testing of all significant plant

modifications, using plant Operations personnel. Jd , at 11.

290. The Engineering Service Department is responsi-ble for the technical document control system for Technical Functions, providing a central master file of plant technical data. It also develops, coordinates, and controls all internal engineering scheduling for other departments and provides engineering cost estimating support. Id., at 11-12.

291. In his testimony, Mr. Wilson explained quite thoroughly the circumstances under which GPU seeks additional technical support from its nuclear steam supply system de-signer, B&W, its architect / engineer, Gilbert Associates, or other technical consulting firms. Wilson, ff. Tr. 11,722, at 12-14. In summary, Licensee uses outside technical consultants if it determines that it needs special expertise which does not exist within GPU, the participation by an outside expert is necessary because the consultant has a unique capability or its views are considered important, when an independent viewpoint is desired, when the problem requiring resolution is generic in nature and Licensee can therefore share the cost with other utilities or the work is otherwise most cost-effective when done by an outside technical organization, and when technical resources are required to augment in-house staff during high demand periods. Id., at 12-14.

292. Licensee has assured us of the availability of resources to both respond to TMI-2 recovery efforts and provide

-222-

_i,e--=,-,ww ee%o m m m. . ------.---m-

technical assistance to TMI-1, as well as Oyster Creek. With respect to the potential for TMI-2 needs (bcth manpower and in other costly resources) to divert Licensee from attention its technical staff ought to be paying to TMI-1, Licensee has taken steps to see that this will not occur. First of all, the TMI units are separated organizationally. Arnold, ff. Tr. 11,434, at 23. Thus, the TMI-l and TMI-2 staffs can carry out their respective operations and recovery efforts independently from each other.52 Id. Secondly, within the Technical Functions Division, which provides engineering support to both units, a dedicated technical group devoted to TMI-2 activities has b'ene set aside within the Engineering Projects Department of Technical Functions, comprised of 30 individuals (7 GPU Nuclear employees and 21 individuals from consultant organizations).

Wilson, ff. Tr. 11,722, at 14. In addition, Licensee has contracted with Bechtel Power Corporation to provide a large full-time technical staff to plan, evaluate, engineer, and otherwise support TMI-2 recovery. Id.; Arnold, ff. Tr.

11,454, at 23-24. In the future, the majority of technical resources required for TMI-2 beyond those provided for in the GPU Nuclear Corporation organization will be obtained from Bechtel, thereby assuring that those GPU Nuclear's resources needed to support TMI-l are always available, and are in no way 52 Licensee has assigned some 245 personnel exclusively to the Vice President of TMI-2. Arnold, ff. Tr. 11,454, at 23.

, -223-

5, j , . .* ,

o /

compromised by activities at TMI-2. Wilson,, ff. Tr. 11,722, at 14-15.

293. After reviewir.3 Licensee's technical capability to simultaneously operate TMI-1 and clean-up TMI-2, the Staff was satisfied with the technical resources available to support TMI-1. NRC Staff Ex. 4, at 5-8, 13-14; Keinig, ff. Tr. 11,946, at 9, 13-16. Because the Staff considered Licensee's in-house

technical resources to be more than adequate without recourse i

to outside resources, it did not consider the outside resources available to Licensee. Tr. 11,963-65 (Crocker, A11enspach).

Furthermore, the Staff concluded that cleanup activities at TMI-2 will not have an adverse impact upon the safety of TMI-1.

Staff Ex. 4, at 38; Tr. 11,981 (Crocker).

294. There is no question but that Licensee has enormously strengthened its technical capabilities since the TMI-2 accident, not only by vastly increasing the number of professionals available within the organization to provide i

support to GPU's nuclear plants, but also by centralizing and therefore concentrating its technical activities in one j division of GPU Nuclear Corporation, by diversifying its technical capabilities, and by becoming more involved in the complex technical analyses which its technical consultants perform, such as transient analyst;s. Based on our review of i

Licensee's technical organization and resources, we find that Licensee possesses sufficient in-house technical capability to j operate TMI-1 safely. This finding includes our consideration

-224-

1 l

of resource demands associated with the clean-up efforts on-going at TMI-2 and Licensee's allocation of resources to _

meet those demands without weakening its technical support of TMI-1. Although we do not consider Licensee's in-house technical resources to be inadquate, we nevertheless have

considered Licensee's arrangements with its vendor and architect-engineer, among others, to supply additional tech-nical expertise as necessary, to support its nuclear ac-tivities. These arrangements further support our finding that technical capability and resources to support TMI-1 are i

sufficient.

I XII. OTHER SPECIFIC ISSUES l

295. In Issue (13) of CLI-80-5, the Commission directed the Board to consider:

! Such other specific issues as the Board deems relevant to the resolution of the issues set forth in this order.

296. In order to facilitate organization of the numerous subject areas which were included in our assessment of Licensee's management and technical capabilities to operate TMI-1 safely, the Board has patterned- the foregoing findings after specific issues (1) through (11) idettified by the Commission in CLI-80-5 (March 6, 1980). We have also taken into consideration in our findings the allegations raised by all of the management-related contentions. Although we have not explicitly identified at every step when Licensee's

-225-

j resolution of an issue also satisfies specific TMI-2 lessons learned requirements, including those set forth in NUREG-0694, "TMI-Related Requirements for New Operating Licenses," the ~

, Staff has done so in its Evaluation Report and subsequent supplements to that document.53 Staff Ex. 4, at 39-45; Staff 53 In NUREG-0731, Supplement 1 (the management SER), the Staff lists twelve items which it considers to fall within the management area which are listed in NUREG-0694, "TMI-Related Requirements for New Opern'ing Licenses." Recognizing that in the Commission's view, Licensee should not be considered to fall within the category of so-called Near Term Operating Licenses, see CLI-80-3 (March 23, 1981), at 7, it is a positive reflection on Licensee's management and technical capability to be able to find that in the following respects, Licensee has met the Staff's

, NUREG-0694 requirements or has made a commitment to meet

! these requirements that is acceptable to the Staff:

(1) having an STA present on al'1 shif ts (I. A.l.1);

(2) minimizing shift supervisors' administrative duties (I.A.1.2);

(3) properly manning the operating shif ts (I.A.1.3);

(4) upgrading its RO and SRO training and qualification program (I.A.2.1);

(5) training and retraining its operator instructors properly, thereby appropriately :dministering its training programs for licensed operators (I.A.2.3);

(6) revising the scope and criteria for its licensing exams in accordance with the Staff's revised criteria (I.A.3.1);

(7) improving its management organization, in accordance with NUREG-0731 (I.B.1.2);

(8) establishing necessary shift relief and turnover procedures (I.C.2);

(9) clearly establish and emphasize the command duties and responsibilities of the shift supervisors (I.C.3);

(10) establish proper limits on control room access (I.C.4);

(11) establish procedures for assuring that operating experiences are fed back to operators and other personnel (I.C.5); and (12) develop and implement a training program to ensure that operating personnel are training (continued next page)

-226-

l; Ex. 13, at 12; see also Staff Ex. 14, at 46-47. We are satisfied with the Staff's conclusion that with respect to the twelve management-related requirements included in NUREG-0694, (which encompass the four management-related items in Table B-1 of NUREG-0578) the Licensee has either met each of the specific requirements, or has made an acceptable commitment to meet that requirement. Staff. Ex. 4, at 39-45; Staff Ex. 13, at 12; Staff Ex. 14, at 46-47; Staff Ex. 1, at C8-50 to C8-56.

297. Without reviewing in detail the specific conclusions reached in each of the sections of these findings, l

we conclude that overall, Licensee's management is sufficiently staffed, has sufficient resources and is appropriately organ-ized to operate TMI-l safely. We are confident that facts revealed by the accident at TMI-? do not present outstanding questions concerning management competence which must be i

resolved before Licensee can be found competent to operate Unit i

i 1 safely. We are also convinced that Licensee is capable of operating Unit 1 safely while simultaneously conducting the l

clean-up operation at Unit 2.

XIII. CONCLUSIONS OF LAW 298. In accordance with Commission Orders CLI-70s8, CLI-80-5, and CLI-Sl-3, a?d based on the evidence of record in (continued) in the use of installed plant systems to control or mitigate an accident in which the core is j severely damaged. (II.B.4).

i

Staff Ex. 4, at 39-45; Staff Ex. 13, at 12; Staff Ex. 14, j at 46-47.

i

-227-t - - - - - -

this proceeding and the foregoing findings of fact related to management competence, the Board concludes:

a. That the "short-term actions" recom-mended by the Director of Nuclear Reactor Regulaf:fon related to management competency are necEssary and sufficient to provide reasonable assurance that TMI-l can be operated without endan-gering the health and safety of the public, and should be required before resumption of operation should be permitted; and
b. That the "long-term items" recommended by the Director of Nuclear Reactor Regulation related to management competence are necessary and sufficient l to provide reasonable assurance that the facility can be operated for the long term without endangering the health and safety of the public, and should be required of Licensee as soon as practicable; 299. The Board further concludes:

l c. That Licensee has augmented the retrain-l ing of all Reactor Operators and Senior l

l Reactor Operators assigned to the

-228-

l l

i control room including training in the areas of natural circulation and small break loss of coolant accidents includ-ing revised procedures and the TMI-2

accident. All operators also have received training at the B&W simulator on the TMI-2 accident and Licensee will conduct a 100 percent reexamination of all operators in these areas. NRC will
administer complete examinations to all f

licensed personnel in accordance with 10 C.F.R. 55.20-23;54

d. That Licensee has demonstrated his managerial capability and resources to operate Unit 1 while maintaining Unit 2 in a safe configuration and carrying out planned decontamination and/or restora-tion activities. In reaching this conclusion, we have addressed the adequacy of groups providing safety review and operational advice, the management and technical capability and 54 Commission order Item 1.e is included in Licensee's proposed findings on management, but will be the subject, as well, of findings in Licensee's Design and Procedures findings to be submitted later in accordaace with the Board's schedule for findings.

-229-i . - . . _ - . - . - _ _ - - - . . - . - . - . - - . - - - - . - - - - - - - -

i training of operations staff, the adequacy of the operational Quality Assurance program and the facility procedures, and the capability of important support organizations such as Health Physics and Plant Maintenance.

We have specifically addressed issues (1) through (11) and (13) af CLI-80-5; and

e. That Licensee complies with the Category A recommendations related to management competence (Ite.is 2.2.1.a., 2.2.1.b.,

2.2.1.c. and 2.2.2.b.) in Table B-1 of NUREG-0578 and has made reasonable progress toward completion of the one Category B recommendation related to l

l t

! -230-t

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

management competence (Item 2.2.1.b.) in Table B-1 of NUREG-0578.55 Respectfully submitted, .

SHAW, PITTMAN, POTTS & TROWBRIDGE MI d'I*<[.

George F. Trowbridge Ernest L. Blake,.Tr.

Deborah B. Bauser Counsel for Licensee 1

1 55 See 1 in our findings on Plant Deaign and Procedures for a discussion of the Staff's currently scheduled implemen-tation dates for Category B recommendations.

-231-l l _ _ _ _ _ _ - . . . . _ , . - . _ - _ _ _ . _ _ _ _ - - - - - . _ - - . - - - - - - _ , - - - - - - - - - ~ - - - - - - - - - - - - - - - - - - -