ML19248C963

From kanterella
Jump to navigation Jump to search
Responds to 790419 Request to Evaluate NRC Program to License Power Plant Operators.Provides Examples of Operator Error & Suggestions for Improving Training & Qualifications
ML19248C963
Person / Time
Site: Calvert Cliffs, Rancho Seco, Crane
Issue date: 05/15/1979
From: Staats E
GENERAL ACCOUNTING OFFICE
To: Schweiker R
SENATE
References
REF-GTECI-A-17, REF-GTECI-SY, TASK-TF, TASK-TMR ACRS-CT-373, B-127945, NUDOCS 7907180130
Download: ML19248C963 (16)


Text

,

es

.w::

  • D.... ".,

COMPRCLA.ER GENERAL. CF THE UNITED STATES q _ *M).

~%.

M ASHINGTCN. 3.C.

. dss 2

/.-

,,(% '

',tAY 151978 S s.'IfAs=

1 The Honorable Richard S. Schweiker United States Senate

Dear Senator Schwei<er:

This report responds to your April 19, 1979, request for a prompt analysis of the Nuclear Regulatory Commission's program for licensing nuclear powerplant opera:ces.

As you indicated, information from the Three Mile Island nuclear accident and from other nuclear plants has raised questions concerning the Commission's program.

In responding to your request, we have answered the specific questions that you had asked.

Moreover, our review has resulted in questions beyond those you asked which we believe must be answered to assure tha: this important ele-ment of nuclear powerplant operation is being carried out safely.

These questions are discussed in the text o f this report although we were unable to answer them because of the short time f rame for carrying out your request.

But we believe the questions should be raised in a public forum now so :nat other parties--the President's recently appointed COnmission investigating the Three Mile : Eland accident, the Nuclear Regulatory Commission, and congressional committees--

can consider them in carrying out their studies o f the Three Mile Island accident and related matters.

HUMAN / OPERATOR ERRCR AT THREE MILE ISLAND AND CTHER MUCLEAR PCWERPLANTS The specific events that caused the nuclear accident at the Three Mile Island powerplant are being carefully evaluated by the Commission, i.c particular the Office o f

nspection and Inforcement.

However, the Chairman o f the Ccmmission and the Director of the Office of Nuclear Reac-

or Regula:10n have pu=licly stated ::a: human error v.c a ma:ce contrihu :: to :he acciden: a: Three Mile :slanc.

Because of ficials of the Commission's Of fice of :nspection and Inforcement are currently performing their o f ficial inves:igation, :ney refused to ma<e availacle :: us any 3:ecific information ceing generated.

n their Opinion,
Orir investigation would e impeded.f specifi:

IMc-79 4 7907180/3@

V e'

i

~

1 JcJ

' a 'T

3-12.'945 facts were released prematurely.

Commission officials did offer to brief us continuously on the status of their inquiry.

While the principal causes of the Three Mile Island nuclear accident are tentative, documentation shows that human / operator error has occurred at other commercial nu-clear powerplants.

According to the Commission's statistics, human error accounted for 13 percent of all reportable inci-dents in 1973, with specific operator error accounting for one-third of that percentage.

Human error could involve errors that were caused by a nuclear facilities' management, maintenance, and other technical personnel who are not re-quired to be licensed by the Commission.

Operator error relates only to those personnel who are licensed to operate a nuclear reactor.

Although we cannot provide specific information on the human /operrtor errors at tne Three Mile Island powerplant, we have described two other examples of human / operator errors at powerplants to illustrate the nature of such errors.

Rancho Seco Nuclear Plant, California, Maren 1978 On March 20, 1973, Rancho Seco experienced a severe cool-down caused by the loss of electrical power to a suo-stantial portion o f the nonnuclear instrumentatien.

T*e loss of electrical power was caused when a control room operator began replacing a burned-out light bulb on one of the control consoles.

To change the light bulb, the light as.sembly was pulled out from the panel and flipped down, exposing the bulbs.

During the anange, a bulb was dropped into che open light assembly cavity, creating an electrical short.

Leter investigations by the utility showed that approx-imately two-thirds of the nonnuclear instrus.entation signals were affected by the power loss.

Erroneous signals prov:.ded faulty information to both the control room and the inte -

grated (co=puteri:ed) control system.

The integrated con-trol system cut off all main feedwater flow in response to faulty signals.

The cutoff in feedwater flow caused the reactor cooling system pressure to increase and the reactor to trip or snut down.

n the period following the reactor shutdown, the oper-ators still were hampered by the lack of instrumentation and cy equipment responding to inaccurate signals.

These false signals had several effects.

: was difficul: for the oper-ators to ascertain vnich indica ans were valid, given :ne W

m-

/

t j

L,#

w

3-127945 changing plane conditions and the wide variety of possible errors that were introduced.

Cnly a select few parameters were known to be valid readings, and the operators had to control the plant cased on that information.

The second effect was that false signals were fed into the integrated control system, so equipment was operated without regard to actual conditions.

Power was finally restored to the nonnuclear instrumen-tation 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and 10 minutes after the reactor trip, which permitted proper operator response to plant conditions.

By this time, the reactor cooling system had dropped to around 285 degrees Fahrenheit.

This meant that the reactor had cooled approximately 300 degrees Fahrenheit per hour, which was well in excess of tne technical specifications.

!=me-diate action was taken to return the temperature to the permissible heat range.

Following this accident, engineers from 3abcock and Wilcox--designer s o f the plant--investigated the matter and recommended that a closer icok be made of operator training as it relates to loss of nonnuclear instrumentation power.

Subsequently, the Commission agreed on March 24 that the plant could restart power for co=mercial oceration if the Baccock and Wilcox recommendations were followed.

Calvert Cliffs Unit 1, Lusev, Marvlanc. Cecemcer 1975 A plant operator discovered that a manually operated water supply valve to two auxiliary feedwater pumps was shut; thus, both pumps did not have a supply of water.

Se-cause this was a manually operated valve, there were no con-trol canel indications that the valve had been closed.

At least one auxiliary feedwater pump must operate to remove heat from the reactor if the main feedwater system should fail.

The utility analyted this " occurrence" and concluded that a plant operator erred about 2 weeks ear ier when valve positions were chang?d.

The utility also concluded that had operation of the au.siliary feedwater pumps been required, "it is highly probable

  • that the operator would have noticed the lack of water supply to the pumps prior to any serious damage Occurring.

u 2.

n--,J S e F _ a... C n.u..,.

..e v 3, e-

.m CPIRATCR LICINSING PRCGRA21 Personnel with various levels of qual fications form the cegani:ation that operates a commercial tuclear power-plant.

The operating atility casists of onsite perronnel

!I vUv

3-127945 s

concerned with the day-to-day operation of the pinnt, maintenance, and certain related technical servicas.

The levels of the operating organa:ation normally in :1ude man-agers, supervisors, professional-technical staff, control room operators, technicians, and repairmen.

Howe'. e r, the Commission licenses only those persons who physical'y manip-ulate reactor controls (operators) and those persons who direct the activities of licensed operators ( senior operators).

The Atomic Energy Act of 1954 established the require-ment that individuals who ocerate nuclea: reactors not owned by the Government must be l'icensed.

Section 107 of that act required that the Atomic Energy Cc= mission prescribe uniform conditions for licensing individuals as operators, determine the qualification of such individuals, and issue licenses ;o such individuals in such form as the Commission may prescribe.

The implementing regulations that werc developec re-quired that the cont:cis of a nu: lear powerplant would be manipulated only by individuals who were licensed under title 10 of the Code of Federal Regulations, Part 55, rela -

ing to operators' licenses.

This regulation estaclished the procedures and criteria for issuance of licenses to operators.

Part 55 was amended in 1963 to require certain individuals to hold senior operator licenses and was amended again in 1973 to require operators and senior operators to participate in requalification programs as a condition for license renewal without reexamination.

The Energy Reorga-ni:ation Act of 1974, in amending the Atomic Energy Act, retained the requirements for licensed operators and seniot operators and authori:ed the Nuclear Regulatory Commission to issue operator and senior operator licenses.

ELIGI3:LITY RECU!REMENTS FCR CPERATORS The Commission has no minimum eligibility requirements for either type of operator.

Instead, the C:mmission, as part of a regulatory guide to the nuclear industry, endorses a standard established bv. the American Nuclear Society _1/

pertaining to selection and ::aining of nuclear powerplant personnel.

The standard is intended as a guide and does no preclude anyone f:0m not conforming to it.

The

_1/ A nono. r :fi t c.rofessional society interested in fur:nerinc.

nuclear standards throughou: the world.

4

,' '. j.

s

3-127943 followt.q recommendations are made in the standard for individuals requiring licenses.

Cperator:

a.

Eich-school graduate or equivalent.

b.

Two years of powerplant experience or its equivalent, provided that a minimum of 1 year is at a nuclear powerplant.

Senic operator:

a.

High-school graduate or equivalent.

b.

Four years of responsible powerplant experience, of which a minimum of 1 year must be nuclear powerplant experience.

A maximum of 2 years of the remaining 3 years of powerplant experience can be fulfilled by academic or related technical training on a one-for-one basis.

In addition to recommendations concerning education and experience, the standard says minimum health requirements shall be established for operating personnel.

The Commis-sion requires each applicant for an operator or a senior operator license to complete a certificate of medical his-tory.

Besides asking the applicant to answer for himself sucn questions on the certification as "Have you ever seri-ously considered committing suicide?", the applicant's e.yes, ears, hear:

blecd pressure, and pulse must be examined by a physician.

GAO cuestions 1.

The Commission has no minimum eligibility require-ments but endorses recommendations made by the American Nuclear Society.

Should the Commission have minimum eligibility requirements?

If so, what should those requirements be?

2.

Is a person with a high schoo'. education suited to operate the controls of a nuclear powerplant?

Should that person be better educated?

3.

The term

  • equivalent
  • high school education is s.ot defined.

Should it have a specific meaning?

4.

"Powerplant experience

  • can pertain to that experi-ence acquired during any stage of a powerplant's s

W

3-127945 life including the design and the construction.

Should "powerplant experience" be more specifically defined?

5.

The term " responsible powerplant experience," when referring to a senior operator, is not defined.

Should it have a specific meaning?

6.

Should medical examinations for nuclear pcwerplant operators be more stringent?

Should psychological pr0 files be developed for these operators, analy:-

ing thei: response capacilities in stress situations?

TRAINING PRCGRAM FOR CPERATORS A ::aining program, together with a training scLedule prior to reactor startup, is developed by the utility and sucmitted to the Commission for a paper review and approval.

Usually the training program for app;icants with no previous nuclear experience starts 2 years bef ore fuel loading and covers a period of 14 to 17 months.

Applican s who have previous nuclear experience are phased in at the proper imes in accors snce with their experience.

Typ ically, tuere are four phases to the training program.

In phase I, a basic course which normally lasts 12 weeks is usually presented to the applicants by a local university.

The course includes approximately 10 weeks of basic study, which includes nuclear physics, health physics, chemistry, and plant technology.

The study program is fol-lowed by 2 weeks of practical operational training on a :.u-clear training or research reactor, where the applicants participate in various experiments and manipulate the con-

ols during 10 reactor startups.

In phase II, tha applicants receive coth observation and simulator training.

Cbservation ::aining involves ob-serving the day-to-day operation of a nuclear powerplant.

This training is conducted by the plant operating organi-

stion.

During the observation training, the applicant observes various operations, surveillance testing, and the practical aspects of the radiation protection program.

The

2ining period varies from 1 to 3 months.

In conjunc fon with plan: Observation, the applicant receives ::aining on a powerplant simulator.

Simulator training varies from 2 to 3 months and is usually ;;ven oy the manufacturer of :he reactor or by the utili:y,

.f i: has ::s own simulator.

While a: the 5

' (\\ [k i

  • )

VOV

'U/

3-127945 simulator, the applicant observe: and participates in the various phases of powerplant operations (such as reactor startups and power-level changes) and learns to use normal procedures and, to a lesser extent, other procedures to cope with abnormal and 2mergency conditions.

In phase III, training consists of a lecture series, given by the plant operating organi:ation, to familiari:e the applicant with the design features of his plant.

This phase normally takes 6 weeks.

During phase IV, the appli-cant must successfully complete a Commission approved on-site training program that covers information on the plant for which he seeks a license.

In addition to classroom training, the applicant wi_1 engage in the day-to-day activ-ities, such as procedure writing, construction check-out, and pre-operational testing, for a period of approximatelv 1 year prior to fuel loading.

The time spent in this phase varies according to the experience of the applicant; how-ever, the minimum required time is 6 months.

After the plant has become operational, it may be necessary for the utility to train replacement applicants because of turnover in personnel.

These applicants would go through the same four-phase training program already men-tiened, except they do not necessarily go through the obser-vation and simulator program.

The Commission believes that the replacement applicants have probably been at the plant for a couple of years and participated in all the pre-critical check-out of the equipment and startup testing of the plant.

Thus all their P:aining would be received at the plant site.

Normally cne training program for replace-ment applicants covers 4 period of 6 to 3 monens.

GAC cuestions 1.

The Commission basically performs a paper-review'of a utility's training program.

Should the Commis-sien establish its own minimum training require-ments?

Should the Commission have its staff per-sonally inspect the training program?

2.

The plant operating organi:ation is very much in-volved in training operator applicants.

Should the Commission review and approve the individuals who give this training?

3.

Cnly about six otilities have their own simulators.

Should the Commission require that each utility

5. ave a simulator onsite so that ocera:ces can con-tinually improve their skills?

~

4 I

u c.;

3-127945 4.

It appears that there are no specific criteria on what the simulator training should include.

Should such criteria exist and what should the criteria be?

5.

Abnormal conditions are given lesser attention in the simulator training program.

Should this be expanded?

6.

Replacement applicants do not necessarily participate in the simulator training program.

Should at be a requirement that they do so?

CPERATOR EXAMINATICNS Title 10 of the Code of Federal Regulations, Part 55, stsces that an application for an operator's license will be approved if the Commission finds that among other things, the applicant has passed a written examination and operating test as may be prescribed by the Commission.

These examini-nations will determine whether the applicant has learned to operate, and in the case of a senior ocerator, to operate and direct the licensed activities of licensed opera:ces in a competent and safe manner.

The written examination for reactor operators consists of seven ategories 1/ and generally requires 6 to 3 hours3.472222e-5 days <br />8.333333e-4 hours <br />4.960317e-6 weeks <br />1.1415e-6 months <br /> to complete.

Written and oral examinations are revised each time they are given at a particular nuclear facility.

How-ever, the questions are always selected from the same stand-ard body o f questions.

Most o f the questions require essay type answers.

The written examination for senior operators consists of the same seven categories as for the reactor operator plus an additional five.

Approximately 4 to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> are required to complete the five senior categories, which include reactor theory and specific operatlag charac-teristics.

The principal difference between the reactor operator and senior operator examinations is that the five senior categories are more difficult and more indepth about powerplant operation.

The typical operating test takes from 4 to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and proceeds as follows.

First, the examiner ex lores the a:-

plicant's '<nowledge of reactivity effects, t.heory of cpe'a-r tion, and radiation protection practices and procedures.

1/ Examples nclude principles of reactor operation and features of fac 11:7 design.

a n'.

('; A ) \\g,r 7

m

3-127945 The major portion of the o; -cating test, however, is conducted in the control rcom.

At a minimum the examiner will have the applicant talk through the startup, indicating controls and instrumentation used in taking the reactor to criticality.

The examiner also determines the applicant's knowledge of how to operate the facility under emergency conditions.

This is accomplished by postulating symptoms of an incident to the applicant.

From the symptoms the appli-cant must determine the type of incident that has occurred and indicate the immediate actions required by procedure.

The final phase of the operating test is touring the plant with the applicant.

During the plant tour the appli-cant must review local procedures and demonstrate his knowl-edge and understanding of local plant operations.

Typical systems explored include electrical control centers, diesel generators, engineered safety features, plant instrument air systems, and selected operating equipment.

GAO cuestions 1.

The difference between the examination given to the senior operator and the operator appears 111-defined.

Questions asked of in applicant for a senior operator's license supposedly are more difficult and more indepth.

Should specific cri-teria be developed addressing the difference in degree of the difficulty and complexity?

2.

The examiners who prepare, give, and evaluate the examinations are not all Commission employees--they are often part-time consultants who work full-time for the national laboratories.

Often these part-time examiners themselves have not taken commercial powerplant licensing examinations, and do not hold licenses.

Many have not had experience in commer-cial nuclear powerplants.

Many have not been through simulator training for nuclear powerplants.

Is this appropriate?

Can this lead to examination problams?

3.

An average score of 70 percent overall is passing on the written examination.

However, a person could fail one or more categories and still pass overall.

s this apptopriate?

4.

A person wne fails one or more parts of the written examination out passes overall does not have to re-ceive addi:icnal training on those parts that s. e failed.

s this appropriate?

9

(> O U

3-127945 5.

Approximately 90 percent of those persons who take the written examination pass on the first try.

Is the examination too easy and should it be revised?

CPERATOR RETRAINING AND RELICENSING If an individual is denied his senior operator applica-tion, an evaluation is made to determine if he should be li-censed as an operator.

If he passed the operator's written examination and demonstrated sufficient knowledge and under-standing at an operator level, he is issued an operator's license.

If an individual f ails to pass his written exam-ination, the operator test, or both, he may file a new ap-Flication to be retested 2 months after the date he was

" enie d a license.

An applicant may file a third applica-ticr 6 months af ter the date of denial of his second appli-canon, and may file further successive applications 2 years af ts.

the date of denial of each prior application.

At license renewal time (a period of 2 years), the licensee suomits an application supported by his utility management and the Commission issues a renewal license pro-vided there is evidence in the application that the person has (1) actively and extensively engaged as an operator or as a senior operator under his existing license, (2) has discharged his responsibilities competently and safely, and (3) is capable of continuing to do so.

The phrase

  • actively engaged,' in this context, is interpreted by the Commission as having reported to the plant on a daily basis.

Also, there must be evidence that the licensee has successfully participated in a requalification program.

This program is administered by the utility and audited by the Commission for its quality.

It must include an oral and written examination of each licensee.

On the written examine". ion the Commission requires that anyoody who gets less than 30 percent in a given category should go to a pre-planned lecture on that subject.

The triteria for additional training in the program is that if an individual gets less than 70 percen t in the annual written exam or has an unsat-isfactory performance on the oral examination, he must go into accelerated training.

If lacking in both areas, he is prohioited from performing licensed duties.

Cther parts of the requalification program require tha t each operator manipulate the controls at least 10 times in a 2-year period and participate in walk-through-type d r ill s, including emergency drills.

Manipulating tne con-trols means startups, orderly snutdowns, and power changes.

10

/On

'77 isOO L1 J

3-117945 Lastly, the Commission requires documentation that each licensee review procedure changes, license requirements, and design changee.

GAO cuestions 1.

The Commission requires that a nuclear powerplan:

operator undergo examination once a year.

Is 1 year or a much shorter period appropriate?

For example, the Federal Aviation Administration re-quires that airline pilots be reexamined e;ery 6 months.

2.

Nuc_aar powerplant management, maintenance, and other technical personnel are not required to be licensed.

Only the operators are required to hold licenses to manipulate the con trols of a powerplant.

Since virtually many, if not all, of the unlicensed personnel may critically affec: plant operation, should other plant personnel also be required to hold licenses?

3.

Licensee event reports identify errors or other proclems that develop in reactor operations.

When an operator makes an error, it has to be reportet to the Commission through a. licensee event report.

However, the reports do not provide the names of operators who commit the errors.

Therefore, it appears that the Commission cannot maintain opera-tional error records based on what specific oper-ators committed the errors.

How effective is this?

How can the Commission effectively monitor operator errors?

4.

In completing licensee event reports, the ctilities have considerable discretion in how they classify eacn event.

Should the C0mmission require more specific details so that it can clearly distin-guish human / operator error from a technical design proclem?

5.

To a large extent, the Commission relies on utility management to certify that an operator should have his license renewed.

Should the Commission inde-pendently check this certification?

  • f 4

/i I O

s

3-12~945 ENFORCEMENT CF TSE OPERA"0R LICENSING PRCGRAM After a control room operator has been licensed, he can be removed from that position if it was found that he committed a deliberate and willful act in viciation of the Commission's regulations.

Usually, utility management makes the initial decision to remove an operator, followed by a Commission investigation.

At that point, the oper-ator's license way be suspended until such time as the oper-ator has successfully completed a reexamination.

Under certain circumstances, the operator's license may even be permanently revoked depending upon the severity o f the violation.

Of the 2,533 licensed operators, the Commission has suspended 1 operator's license, and required 6 other opera ces to be reexamined.

There are two ways by which the Commission audits the requalification program.

In one, personnel from the Oper-ator Licensing Branch visit the facility once every 2 years and look at samples of the requalification exams.

These exams are checked by evaluating the quality of the questions and rescoring several categories of the exams.

In cases where the Commission has been dissatisfied with the requal-ification exams, Commission reexaminations have been given.

In the second phase, personnel from the Of fice o f Inspection and Enforcement visit each f acility once a year and assure that commitments made in the requalification program are being carried out.

They check tha t everybody participated in the requalification training program, manipulated tae controls the minimum acceptable number of times, and completed the yearly examinations and lecture courses, as required.

GAO cuestions 1.

The Commission has found it necessary to suspend one operator's license and require six other oper-ators to be reexamined.

What criteria has the Ccamission established to determine if enforcement action must be taken against an operator?

2.

n light of the apparently low nuscer of enforce-ment actions and high percentage of operator er-rors, should the Commission's criteria for enforce-ment action be strengthened?

2.

How ef fective are the utilities in self-enforcing cperator violations?

em i

k I

0;v m,

v

3-127945 CCNTROL RCCM OPERAT!CN The typical control rocs at a ccamercial nuclear powerplant may oe a room approximately 25 feet wide by 40 feet long.

Covering the length of the room on both sides are control panels with lights and indicators monitoring every aspect of powerplant operation.

Persennel within the contrcl room continually oversee the control panels, check-ing for normal as well as abnormal conditions.

Most commercial nuclear powerplants operate on a five-or six-shift basis per week.

On eyery shift, minimum re-quirements at a single unit plant are one senior operator, two licensed operators, and two auxiliary nonlicensees.

Regulations require only ona licensed operator in the con-trol room at all times.

A majority of whe time, accceding to a Commission official, there are two licensed operators in tne control room.

According to information obtained frca the Commise. en, a profile of a control room operator would be a persen 34 years old with 7 years of operating experience, and a high school education.

GAO cuestions 1.

Control rocms in nuclear powerplants are not stand-ardized.

There are often considerable variations in the controls of the facilities.

If the con-trols were standardized by the Cc= mission, would this make operations easier for the operators?

Would there be less chance for error?

2.

Commission regulations require that only one li-censed operator be in the control ::cm at all times.

Should the Commission amend its regulations and require that a senior operator plus one or more additional operators be continually present?

3.

Commission regulations do not require that a nu-clear engineer be on duty at a nuclear powerplant at all times.

If this were required, there would be one " key" individual availacle at all times to cope with an emergency situation that may r3 quire nuclear engineering knowledge.

Should this be required?

4.

When a new nuclear powerplant cecomes operational, One Commission has statistics which indicate inat approximately 30 personnel are assigned to operate

..1

(

g

}

/

3-127945 s

the facility.

However, the s:stistics indicate that many times as few as six operatoes have nad actual commercial powerplant operating experience.

Is this a sufficient number?

ACT!CNS TAKIN SY TSE NUCLEAR REGULATORY COMMISSICN CN THE NUCLEAR PCWERPLANT OPERATOR LICENSING PRCGRAM The C tmission recently has acknowledged that its powerplant operator liceasing progran needs considerable improvement.

The Director of the Of fice o f Nuclear Reactor Regulation acknowledged tefore the Three Mile Island acci-dent that there were problems with the program, and actions were being taken to correct the problems.

The Commission previously had contracted to have two independe'at evalu-ations of their operator training and licensing program.

Both studies are still underway.

The same Commission official stated that the accident at Three Mile Island has compounded the existing proclems, and clearly shows that the entire operator licensing program needs reexamin tion.

The Chairman of the Commission also has recently stated that he had thought cefore the accident that operator training was adequate to prevent such inci-dents, but after the accident stated "that is clearly not the case."

As a :esult of the Three Mile Island accident, the Commission has taken the following immediate action to im-prove the operator program.

The Commission instrue:ed the operators of all light water power reactors to review and understand the apparent operational error that led to the Three Mile Island accident.

Specifically, the Commission has instructed the operators not to:

(1) override the automatic action of their engineered planc safety features unless the continued operation of the safety features will r e sul t in unsafe plant conditions and (2) make operational decisions based solely on a single plant parameter indica-tion where one or more confirmatory indications are available.

The Commission has also initiated a comprehensive eval-uation of :ne overall opera:ce licensing program.

This will ce a r eexamination o f all aspects of the program.

The Com-mission has not estaclished a deadline for completing the evaluation, al: hough it indicated that it would take am leas 2everal months to complete.

14

]

,,on

(

a

3-127945

!N LCCKING AT SUMAN/CPERATOR ERRCRS MORE COMPLEX PROBLEMS SHCULD NOT BE CVERLCCKED Although much attention is now being directed at human /oper ato r error s in nuclear powerplants, it is very important to be aware of the possibility of much more ccm-plex problems arising in nuclear powerplants.

Specifically, the possibility exists that there may be technical design inadequacies.

For example, on January 3, 1979, a Commission inspector wrote a memorandum stating that there appeared to be generic safety problems with Babcock and Wilcox designed nuclear plants.

The regional inspector asked that his momorandum be forwarded ec the Atomic Safety and Licensing Boards 1/ for their consideration during licensing hearings.

In response to this memorandum, the Commission's Division of Reactor Cperations Inspection stated that, based on a preliminary evaluation, the warnings of the regional inspector did not

" appear to be new issues or to put a dif ferent light on the issues and, therefore.

. do not meet the intended crite-ria for Board notification."

In essence, the safety con-cerns of the regional inspector were not considered ec be re.evant and material.

However, because the regional inspec-tot insisted that these safety concerns be presented to the Atomic Safety and Licensing Boards, the Division recom-mended that the Boards be so notified.

Cn March 6, 1979, the commission's Assistant Director for Light Water Reactors also recommended that those Atomic Safety and Licensing Boards with jurisdiction over Babcock and Wilcox designed plants be informed of the regional in-spector's safety concerns.

He specifically recom.nended that the Board for the Three Mile Island powerplant be informed.

We were told by Commission officials that the Three Mile Island Board did not receive the regional inspector's safety concerns until Ma,rch 29, 1979--the day after the accident occurred.

We were unable to determine wha t action, if any, has been taken by the Atomic Safety and Liceasing Board.

Since the accident, the Ccemission and those utill-ties ope:ating Fabcock and Wilcox designed plants agreed to

-1/An inde:endent board wnich has a '<ey role in the Commis-sion's licensing and decistenmaking process.

4 :

  1. 9

() U V

3-127945 close the plants until they could determine the specific causes o f the accident.

CCNCLCSICNS Although the Commission is still analyting the causes and ef fects o f the Three Mile Island nuclear accident, it is clear that human / operator errors have been a problem at other nuclear powerplants.

Based upon cuc limited review of the Commission's operator licensing program, and upon the number of human / operator error-related accidents in the past, we believe that the operator liceasing program should be completely reevaluated.

Commission officials nave agreed that a complete reevaluation of the operator licensing program is needed, and have acknowledged that such an evaluation will be made.

The Nuclear Regulatory Commission and the recently appointed Presidential Commission should give attention to the specific questions that we have raised in this letter.

We would like to point out to these organi:ations, however, that their investigations should take special precautions to assure that the potential f or design and other generic weak-nesses is not eclipsed by the emphasis on human error.

As arranged with your office, unless you publicly announce its contents earlier, we plan no further distribu-tion of this report until 3 days from the date of the re-port.

At that time we will send copies to the President's recently appointed Commission investigating tre Three Mile Island accident, the Nuclear Regulatory Commission tc interested congressional committees, and oth2rs upE,n request.

.ely you

.l' teau

{

s Comptreller General of the United States

)

{nn

o :

,