ML20214S933

From kanterella
Jump to navigation Jump to search
Unclassified Digest of Classified Rept Entitled, 'Nrc Task Force Rept,770425,' Re Muf for B&W,740401-760818
ML20214S933
Person / Time
Issue date: 04/25/1977
From:
NRC
To:
Shared Package
ML20214S915 List:
References
FOIA-86-568 NUDOCS 8609300141
Download: ML20214S933 (29)


Text

f . . .

zeo >a-,

o . .-

90/e94./ ,

An Unclassified Digest Of A. Classified ~ Report Entitled. 4 NRC /'" "'""\

TASK FORCE (M)

'q'y REPORT .

- \

APRIL 25,1977

' ACCUMULATED MATE,RIAL UNACCOUNTED FOR (MUF)

HIGH ENRICHED URANIUM BABCOCK & WILCOX COMPANY NUCLEAR MATERIALS DIVISION - PENNSYLVANIA FACILITIES APRIL 1,1974 - AUGUST 18,1976 E -- __ _ _-

n .

^*{

TABLE OF CONTENTS 9

- Page Foreword 11 NRC Task Force Conclusions 1 NRC Task Force Background ,

3 Summary of the NRC Task Force Findings 'S Analysis of Past MUF Performance 7 Analysis of Cumulative MJF in Relation to the Cumulative LEMUF 10 -

NRC Task Force Scope of Operations 14 Scope of BW Task Force Operations 21 BW Upgraded Organization for Material Control and Accounting 24 BW Actions Resulting from NRC Task Force Recomendations 25 1

JUN t8 nn

' n .

FOREWORD I

On April 22, 1976, Babcock and Wilcox Company forwarded to NRC the results of the material balance for the highly enriched erations at Babcock and Wilcox Company, Nuclear Materials uranium Division op(B8W, NMD), Materials Plants (Pennsylvania) for the sixty day period ending March 1,1976. This report showed an excessive accounting loss (Material Unaccounted For - MF) for the period. Since April 1,1974, B8W had determined and reported.

  • twelve material balances in accordance with the NRC req'irements u

for increased physical inventory frequency, which became effective in May 1974. For the twenty-three month period ending March 1, i .1976, ten of the twelve material balances resulted in an account-ing loss of highly enriched uranium (HEU).

On May 4 and 7,1976, the Commission was briefed by the Staff.

l These briefings resulted in the approval of an Action Plan and a Task Force approach to investigate the apparent WF's. After a meeting with the staff in Bethesda on May 11, 1976, NRC Region I (Philadelphia) developed the Task Force Plan. This Plan was approved on May 14, 1976. On that same date, the NRC Chairman met with the President of Babcock and Wilcox in order to inform him of the NRC's concern about the MF issue and to apprise him of the Comnjission's intended course of action.

The NRC Task Force has completed its study of the highly enriched uranium cumulative material unaccounted for at the Babcock and Wilcox Pennsylvania facilities (Apollo and Parks Township). -

The Task Force was composed of 25 members,13 from NRC Region I, 8 from NRC Headquarters, 2 from Brookhaven National Laboratory (BNL), I from the National Bureau of Standards, and 1 from the i

Energy Research and Development Administration (ERDA). This staff effort required 963 man-days in developing this report and

, its supporting data. This effort does not include time required

to coordinate all phases of the Task Force on the part of NRC l management.

The NRC Task Force reviews, audits, measurement tests, and quan-titative findings remained independent of the results of parallel work by 88W. The NRC findings in this report are based upon independent measurements and separate audit of accounting records.

11 i

JUN 2 81977

--_-m r,. _-_ . _ , ~-, .-

.u w : ,

In addition to reporting the Task Force findings and conclusions, this report describes the activities perfonned by the NRC Task Force and .by the licensee, The Babcock and Wilcox Company.

Using the Task Force's estimate of WF, the cumulative EF is less than 0.5% of the highly enriched uranium processed through the i, facility.

Applying the maximum probable adjustment, (consideration of all identified error sources connected with the Task Force estimate in a manner.to reduce the MF to the maximum extent) the Task Force ,

findings accounted for the major portion of the cumulative MF.

The reaining MF was less than the uncertainty in the accounting and measurment systems.

There was no evidence of any diversion of highly enriched uranium.

i Recommendations are presented for improved nuclear material control and accounting at the licensee's facilities, based on the Task Force study. t l

i I

9 iii i

JUN 2 8137?

i l

... n . .

(O.H ~

g NRC TASK FORCE CONCLUSIONS The program for safeguarding special nuclear materials at the Pennsylvania facilities of the Babcock and Wilcox Company consists of a physical security system, and a system for material control and accounting. The NRC Task Force has completed a comprehensive

. review of these systems for highly enriched uranium operations during the period April 1,1974-August 18, 1976. The Task Force reached the following conclusions: .

1. Considering the . total quantity of highly enriched uranium processed by the two BW plants during the time interval studied by the Task Force, the amount of material remaining .

unaccounted for is estimated by the Task Force to be less '

than 0.5% of the highly enriched uranium processed through the facility.

i i 2. While the possibility of diversion cannot be precluded, the Task Force has no basis for concluding other than that the l cumulative MUF during the study period can be attributed to l uncertainties in the measurement systems and analytical procedures, and to errors in the accounting system.

3. BW has upgraded their physical security sytem and has taken actions' toward substantive program improvements in material control and ~ accounting. These efforts, respectively, have improved BW's capability to detect any covert attempt to divert highly enriched uranium and will provide for in-

! creased capabilities for the control of high enriched uranium. .

4. Uncertainties in the material control and accounting system, including unidentified biases, resulting in Material Unaccounted l For (MUF), may continue to exist because of the nature and i complexity of the processes and certain equipment limitations.

BW has committed to capital improvements in processing i

equipment, accounting and measurement systems, and security

. systems. These should result, when fully implemented, in a safeguards system that is accurate and sensitive to the diversion of material.

l l

l 1 JUN 2 81977 I_ _ _ _ _ _ _ _ _ _ _ _ __ _._____ _. _ .-

1 _

In the opinion of the Task Force, the BW Safeguards Systen, 5.

with improvements already initiated is adequate to allow continued operations of BW's highly enriched uranium facilities.

e e

e

.1 i

i 2 l

l JUN 28 N77 l

~. . - _ . . - - .

, . , , NRC TASK FORCE BACKGROUND

. . Between April 1,1974 and March 1,1976. NRC Safeguards inspecto' rs identified twenty-four items of noncompliance during twenty-two j inspections of. the material control and accounting system. In addition to written Notices of Violation during this period, the Inspection and Enforcement program also included the imposition .

of a civil penalty, an Immediate Action Letter concerning scrap management, and several Enforcement Meetings with Babcock and  ;

Wilcox Corporate Officers. Although incremental improvements in the accounting system were achieved through correction of noncom-

- pliance itens, the licensee's performance in reducing MIF was not appreciably improved.

As a result of the continuing problems on April 23,1976, an Action Plan to deal with the B&W MUF issue was developed through 4 -

meetings in Bethesda on May 3 and 4,1976 betweeen the.0ffice of Inspection and Enforcement and the Office of Nuclear Materials Safety and Safeguards.

On May'4 and 7, 1976, the Commission was briefed by the Staff.

These briefings resulted in the approval of the Action Plan and its Task Force Approach. After a meeting with the staff in Bethesda on May 11, 1976 Region I developed the details of the Task Force Plan. This Plan was approved on May 14, 1976.: On  !

that same date, the NRC Chairman met with the President of Babcock and Wilcox in order to inform him of the NRC's concern i with the MUF issue and to apprise him of the Commission's course of action.

On May 19,1976, members of the NRC Task Force met in Phila-

! delphia for briefings by the Region I Safeguards staff. On May

, 20, 1976, the B&W, NMD, Materials Plant Manager and his staff l

visited Region I to present their plans for a parallel effort in i re-examining the accounting system for highly enriched uranium.

' l The NRC Task Force Plan involved two major programs: an in-depth I re-examination of highly enriched uranium operations and the

! material control and accounting system, and an analysis of possible '

! diversion paths from the highly enriched uranium facilities.

I 3

JUN 2 8 B77 4

l

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

In addition to reporting its findings, the Task Force was charged with providing:

An estimate of confidence that the findings do or do not provide an explanation for the cumulative MF;

- Reconnendations concerning NRC follow-on actions.

in order to make a quantitative estimate of the impact of any identified accounting discrepancies upon the declared WF, the Task Force initially chose to examine available accounting and production data for the period April 1,1974 - April 28,1976.

April 1974 was chosen as the starting date because it marked the time at which 88W began conducting physical inventories each sixty i days in accordance with new NRC requirements made effective on

May 7,1974. As described in this summary, the period of study 1

was extended to include the period from April 1,1974 through August 18, 1976, a period of approximately twenty-nine months including fifteen sixty-day material balance periods.

i i

}

1

.i

) 4 i

JUN 2 81977 i

SUMMARY

OF THE NRC TASK FORCE FINDINGS Specific Findings ~ -

The NRC Task Force detemined that Material Unaccounted for (MUF) for the highly enriched uranium operations at the Babcock and

Wilcox Nuclear Materials Division, Materials Plants, has been consistently overstated in the Safeguards accounting system. The  ;

Task Force identified the following significant~ areas in which the accounting system failed to completely reflect plant operations:

Losses of highly cariched uranium through comuningling with low enriched uranium in the incinerator and drain su.ap were not included in the accounting records system; I Liquid and gaseous effluents discharged from the incinerator scrubber system were not correctly documented in the account- ,

ing records system;

- Losses of highly enriched uranium via the transfer of contam-i inated clothing to the separately licensed 88W laundry facility j were not included in the accounting records system; Shipments of highly enriched uranium wastes for burial were understated in the accounting records system; Highly enriched uranium present in the ventilation system and process equipment was not included in the accounting records

. system; Liquid effluents to the holding ponds were not included in the accounting records system; I

- Equipment failures, component leaks, and miscellaneous spills i resulting in highly enriched uranium being embedded in the process area floors were not included in the accounting )

records system; I

5 JUN 2 81977 4

,,,,----..---._-,..-_~-,-n-. - - _ _ , _ _ _ , _ , , _ _ , __ , , . . _._.,__,.m.,

_ __ -- -n.g,-.----,.v-,, , , . _ , - .

.j .

t

- Some MUF entered in the accounting records during the period April 1,1974 - August 18,1976, (the Task Force study period) was attributable to operations prior to April 1,1974;

- Liquid effluents were incorrectly documented for highly enriched and low enriched uranium plant operations in the accounting records system;

- Receipts of highly enrfched uranium were under' stated in the accounting records system;

- Residues from the recovery of BW scrap at'a scrap recovery

contractor were understated in the accounting records system.

4 Ten of the above eleven findings represent areas in which the material control and accounting program systematically failed to l reflect true plant operations over an extended period of time.

! , The last finding (concerning the measurement of residues) is not-i attributable to the BW accounting system, but is due to measure-

I ment uncertainties of a scrap recovery contractor.

1 e

6

\

l l

I f

1 6

l i

j JUN 2 81977 I

i

--v--,r,,,m,--, ,--e_ , , - - m- ,e ec,

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

i, ANALYSIS OF PAST MUF PERFORMANCE In order to gain insight into BW performance as reflected by the MUF, an analysis of MUF, as reported by BW vs the NRC estimate of MUF over the twenty-nine month period was performed. This time span includes fifteen' material balance periods of approx-imately sixty days each.

Licensee action following the declaration of a MUF can be sun-marized as follows:

- If the reported value. (either positive or negative) of MUF exceeds its associated statistical limit of error (LEMUF), and 300 grams of U-235 contained in HEU, a report of investigation and corrective action must be submitted to the NRC; If the reported value (either positive or negative) of MUF is greater than 0.75%, but less than 1.0% of throughput, an imediate' plant re-inventory must be conducted;

- If the reported value (either positive or negative) of i MUF is greater than 1.0% of throughput, the plant must i te shut down and a re-inventory conducted.

Under the accounting rules in effect during this period, the HEU operation was divided into three plants for accounting purposes.

It was required that a MF for each plant be detemined for each i

two-month period. NRC criteria for allowing continued operations i were based on the relationship of MUF to throughput for each of s, the three plants. Each plant throughput, like WF, was determined (

) separately. For example, the Conversion and Finishing Plant -

! throughput was based on the amount of material converted. When this material was transferred ~ to the fabrication facility for further processing, it' counted as an additional (separate)

throughput for the fabrication plant. The plant W F's, under the then effective three plant concept, when compared to the individual

, plant throughputs, generally met the NRC criteria for continued l operations without requiring a re-inventory or plant shutdown.

However, the average MUF loss per inventory period was 1.09% of

the average period throughput in tems of'the present single l plant system. For the twenty-nine month period under study, the l NRC estimated MUF represents an accounting loss of less than 0.5%

i of feed receipts. The consolidation of the three plants into one, for accounting purposes, imposed more stringent MUF controls i

upon the licensee.

7 l JUN 2 81977 l I

The cumulative MUF should normally consist of individual period MUF's which indicate periodic accounting gains as well as account-ing losses. In a well controlled plant, there should be no long-term build-up in the cumulative WF. For BW, however, eleven of the fifteen material balances completed during the period under study indicated an accounting loss and the buildup of a cumulative MUF.

Effective August 18, 1976, all HEU operations were re-designated as one plant for accounting purposes eliminat.ng multi-plant throughput in the detemination of the regulatory LEMUF limit.

Subsequent MUF performance has been based upon its relationship to this more restrictive throughput determination which is based l

upon the larger of feed receipts or product shipments during the

!,, material balance period.

I Eight-months of operations between August 18, 1976 and April 22, l . 1977, have resulted in additional WF. In terms of the total i eight-month throughput, this MUF represents 0.3% of material .  ;

i I processed. j i

l Results of inventory balances subsequent to the task force period indicate that some measurement bias or unidentified loss streams may still exist in the processing of HEU. However, given the complex nature of the process and the fact that the current scrap ,

recovery capacity cannot keep pace with the wide variety of scrap materials generated as the result of low product yields, there is

) evidence that the existing accounting and measurement system may not be able to achieve the desired sensitivity of proposed require-

~

ments.

In recognition of the preceding, the Office of Nuclear Materials Safety and Safeguards (NMSS), by letter dated March 2,1977, i requested that BW submit the fo,llowing to the NRC:

l

- A written plan to upgrade the material control and accounting system, with particular emphasis on assuring prompt recovery of ' scrap and measures to avoid the commingling of low enriched uranium (LEU) with HEU.

8 l

' JUN 2 81977

,.,-..-=.--3 - . , . _ - - - - ~ - - e , ~ .-, - ,,. _..,_ _ ,. ~ , _,_ ~ , , , - . , . ,

    • , = . .

i .

t I l

- An analysis of' changes in operations which would be necessary in order to establish a material control and i

accounting system capable of detecting diversion of a significant quantity of HEU by a single- employee during a period of one year.

On March 31, 1977, BW responded to the first portion of the request with a written plan to' upgrade the material control and accounting system. In addition to planned improvements in the  ;

accounting records system.. internal controls, and measurements, the plan included cosuitments for capital expenditures to upgrade ,

plant systems and equipment as follows:

- Improve scrap recovery capacity;

- Install separate highly enriched uranium scrap incinera-tion equipment;

- Purchase new equipment for the measurement of burial f shipments;

! - Modify and improve the liquid and gaseous effluent piping.

I The scope of the BW plan was approved by NMSS on April 8,1977.

The details of this plan are to be included in the BW Fundamental Nuclear Material Control Plan which is to be submitted by June 8,

1977.

t l

The BW reply to the second portion'of the letSS request is to be .

) submitted in June,1977.

The programmatic and capital improvements described above are now in the process of implementation by BW. However, some capital l

improvements will require approximately eighteen months for comple-I tion. In the interim, the Task Force considers the current l' safeguards system of physical security and material control and accounting to be adequate to allow continued highly enriched uranium operations under existing requirements.

I i

i a

i 9

JUN 2 81977 l

L.- -. - - - _ - _ . _ -. _--

1 ANALYSIS OF CtMULATIVE MUF IN RELATION TO THE CtMULATIVE LEMUF

Background

Individual measurements of material- (each consisting of weight or volume, uranium concentration and isotopic enrichment deterinina-tions) represent the source inputs into the accounting records which are reconciled upon completion of each physical inventory.

Material Unaccounted For (MUF) represents that amount by which

, the accounting records fail to balance following a physical inventory. The MUF reported for a particular period is an esti-

, mate of the true MUF. ,

Each individual measurement has an associated " uncertainty" because of the variety of both random and systematic errors i associated with the weighing and chemical measurement of the ,

material. These uncertainties are estimated from measurements of I process material and standards. The proper statistical combina-tion of these uncertainties provides an estimate of the measure-ment uncertainty associated with the estimated PflF. This is called LEMUF. Mathematically, the LEMUF is twice the measurement i standard deviation of the MUF. If the measurement systems are the only sources of this uncertainty in the MUF, then the esti-mated MUF is considered to come from a distribution with a mean '

, of zero and an estimated uncertainty, LEMUF. )

l Upon completion of a material balance, MUF and LEMUF are compared. l

! If the est'inated MUF exceeds the LEMUF, than it is unlikely that

the true MUF is equal to zero. It is an indication that factors other than measurment uncertainty should be suspected as contrib-utors to the estimated MUF. These factors could be human error. l systen failures, unrecognized measurmant biases or diversion. If '

the estimated MUF does not exceed the i.EMUF, it can not be inter-

preted as an indication that the true M'lF is equal to zero.

! Rather, it implies that there is no evioence which indicates that l the true MUF is different from zero.

If the true MUF were equal to zero, the probability that the estimated MUF would exceed the LEMUF is approximately 5 percent.

If the true MUF were not equal to zero, there would be an unde-termined, yet greater than 5 percer.t probability that the esti-mated MUF would exceed the LEMUF.

l l

l l

l 10 JUN 2 81977

If the true MUF were equal .to zero, then the probability that the -

estimated MUF will lie within the LEMUF is approximately 95 l percent. It does not imply that there is a 95 percent robability that the true MUF is equal to zero when the estimated MJF lies within the LEMUF. If the true MF were not equal to zero, than there would be an undetermined but less than 95 percent probability that the estimated MUF would lie within LEMUF. The measure of that probability is dependent upon the difference (unknown) of the true MUF from zero. That probability increases as the differ-

ence decreases. As a result, instances can occur where factors other than measurement uncertainty affect MF resulting in a true MUF different from zero, but the probability is high that the estimated MUF will lie within LEMUF implying that there is no evidence to indicate that the true MF is different from zero.

1 Implied in the analysis of WF in relation to LEMUF is the assump-

  • tion that the LEMUF accurately represents the plant activity.

Instances of unrecognized measurement bias, nonrepresentative I

sampling, errors in measurement documentation, and other discrep-

! ancies will result in a calculated LEMUF which is not representa- i j

tive of the true measurement uncertainty in the plant operation. "

) In this case, the relationship of EF to LEMUF will be unknown.

1 Detemination of Cumulative LEMUF

- In order to analyze the B&W residual cumulative MUF, the Task l Force undertook the project of constructing a cumulative LEMUF for the period April 1,1974 - August 18, 1976.

1 The construction of a meaningful long tem LEMUF proved to be a difficult task. It required the detemination of uncertainties associated with each of the following elements:

-- The uncertainty in the beginning inventory on April 1,1974;

-- The uncertainty associated with all feed receipts, product i' shipments, scrap shipments, shipments of waste to burial, and measured discards (liquid and gaseous effluents) during the twenty-nine month period; i -- The uncertainty in the ending inventory on August 18, 1976.

a i

! 11 1

JUN 2 81977

'_ _ __ _ ~ _ _ _ __. _ . ,_ _

The following further complicated the detemination of a meaning-

- ful LEMF. A licensee's detemination of LEMF for material

, balances prior to April 1,1974 generally suffered in quality because of the lack of comprehensive measurement control programs (the source data input to LEMF). Specific requirements for actions necessary to ensure quality in measurement controls were instituted in May,1974. The development of a meaningful measure-ment control program over the past three years has been a continuous, evolutionary process. The fact that the NRC identified noncom-p11ance with measurement and measurement control requirements i during inspections in February,1977, is evidence of the fact i that the Task Force study period of April 1,1974 - August 18, 1976, did not have an adequate program of measurement controls.

The licensee's new comprehensive formal Measurements Control Plan ,

is currently being reviewed by the NRC.

f

.. In deterinining a cumulative LEMF based upon a single 29-month period, the Task Force concluded that, in general, the BW repre-sentation of measurement uncertainties was understated, particularly  :

during the early portion of the Task Force study period. For  ;

some measurement types, the records incorrectly indicated that l accuracy was better in April,1974, than in August,1976. Since comprehensive back-up measurement control data were not available, I a reconstruction of error estimates could not be accomplished to make a quantitative estimate of the factor by whig.h LEEF data was understated.

!* Another complicating factor in'the determination of a cumulative LEMF concerns the quality of the physical inventory listing on April 1,1974. Again, there was no basis for firmly stating the uncertainty in that inventory. However, there was substantial reason to believe that the uncertainty could have been considerably larger than that reflected in the accounting records. The following l examples are presented to illustrate the quality of the inventory

in early 1974

-- Following an excessive WF in April,1974, NRC Region I issued an Immediate Action Letter to BW requiring that all scrap materials,on hand during the April inventory be recovered promptly. This major recovery program, which took place between April and December,1974, eventually resulted in the j recovery processing of difficult-to-measure material. This j scrap program involved over one-half of all highly enriched i

uranium in the inventory on April 1,1974; 1

l 1

l 12 JUN 2 81577

+

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

s

-- In Jarr.ai/,1974 NRC Region I cited BW for noncompliance when the company failed to enter into the accounting records the receipt of some feed material; .

It should be noted that the discrepancies and noncompliances in the material control and accounting system shown by the above examples led to the imposition of a Civil Penalty of $12,170 in August, 1974.

Using data available at BW, the Task Force calculated the LEMUF for the period April 1,1974 - August 18, 1976,. The Task Force determined that the LEMUF values did not accurately represent the uncertainties in the accounting and measurement systems during the period. Based upon its observation during field work since April,1976, and a review of MRC . inspection reports and other data pertinent to the study period, the Task Force concluded that

- the LEMUF calculated using BW measurement data could have been

. understated by a factor of at least two, due to the high uncer-Je 1, tainty in the beginning inventory.

i In December,1976, the Battelle Pacific Northwest Laboratories,

under NRC contract, conducted a special evaluation of HEU account-ing losses at BE. After completing this study, Battelle also
concluded that LEMUF was understated by BW. ,

Concerning the range of MUF adjustments determined by the Task Force, the range approximates a 3a uncertainty band around the

conservative adjustment. The upper range adjustment leaves a MUF i which has not been explained by the Task Force findings. However, .

! this MUF is less than the 2a and 3a bands which the Task Force

~ considers to be representative of the uncertainty in the account-ing and measurement systems during the study period.

Thus, the Task Force findings, using the maximum probable adjust-ment, could account for the major portion of the cumulative MUF, i leaving an unexplained MUF which is less than the uncertainty in the accounting and measurement systems.

i I

e b I i

13 Jim 281977 i

U____

-l NRC TASK FORCE SCOPE OF OPERATIONS

The Task Force began its field work'on May 24, 1976. In accord-ance with the Task Force Plan, the initial review was organized around nine teams, each assigned to review a separate area of plant operations and the control and accounting system. A Team.

Leader from Region I was assigned to each team. Members from the HQ staff and Brookhaven National Laboratory were assigned to review areas in which their backgrounds provided expertise. This review and re-examination of the highly enriched uranium activities a was basically accomplished during the first three weeks of field work between May 24 and June 11. This work by the teams identified

specific areas requiring more detailed review in the following l weeks. During eight of the ten ensuing weeks. Task Force personnel
were onsite to condxt the independent measurements and audits j necessary to quantify the MUF impact for those areas identified in the initial review.

1 Diversion Path Analysis (DPA) Conducted by Task Force i

On May 25-28 and June 1-4, 1976, a Diversion Path Analysis (DPA) of the Babcock and Wilcox facilities at Apollo and Parks Township, Pennsylvania, was conducted by a Task Group composed of represen-tatives from~ the Nuclear Regulatory Commission, the Energy Research and Development Administration,' and the National Bureau of Standards.

The DPA was conducted to identify vulnerabilities in the licensee's security and internal control system and to determine potential channels through which employees or other insiders could covertly -

! .ramove SNM from the facility. No detailed diversion-path scenarios l were included in the Survey findings. However, it was found that l

vulnerabilities and potential diversion paths existed at the facilities. The opportunity for undetected diversion of SNM was possible although no evidence was found that it had occurred.

l l

l 1

I4 JUN 2 81977

~ '

..h The DPA focused attention on vulnerable areas at the facilities where diversion paths were evident and in need .of corrective action. In all cases, the identified diversion paths were closed.

The DPA Report included recommendations for strengthening the BW security and internal control system to minimize the potential for diversion. Since completion of the survey, NRC and BW have taken appropriate actions to include these recommendations in

- their programs for physical security. t HEU Shutdown Due to MUE On July 29, 1976, as the Task Force was completing its data gathering phase, the results of the highly enriched uranium physical inventory taken on June 28, 1976, were reported to the i

NRC. The results indicated a MUF loss.

. During telephone conversations with the Region I Office on July 29, 1976 BW agreed to shut down all high enriched operations in-

, i order to investigate and reconcile the material balance. In a letter dated July 30, 1976, BW confirmed that all operations were shut down. Scrap recovery operations continued in order to reduce scrap materials to a more measurable form in preparation for a re-inventory of all highly enriched uranium on or about August 18, 1976.

In a letter to BW dated August 13, 1976, the NRC confirmed certain agreements which had been reached during telephone con-versations with BW management following the shutdown of operations.

In early August, the Task Force increased its observation of the HEU scrap recovery operations. On August 12, BW reported that the re-inventory of all highly enriched uranium would begin on Monday, August 16 and would progress through most of the follow-  :

ing week. Additional task force members arrived on site on l Sunday, August 15, and subsequently observed and tested all l aspects of the physical' inventory. The plant was ultimately '

~

released for startup on August 21, 1976. A task force member was stationed in the plant during all week-day operations and for i portions of nearly all night and week-end shifts.

l 15 JWI 2 81577

i

, From July 30, 1976, essentially all HEU plant activity was dedi-cated to the preparation of the plants for re-inventory. As a result, processing areas were well cleaned and most inentory itens were packaged, measured, tamper-safe sealed, and stored in secure storage prior to the start of the re-inventory.. The Task Force examination of the inventory took place throughout the

shutdown / preparation period as well as during the physical inventory. This examination included independent NDA measure-ments, taking of samples for laboratory verification, independent container weighing, seal checks, and accompaniment of each BW two-man inventory team. By the evening of August 19, 1976, the Task Force had concluded that a satisfactory physical inventory had been perfomed.

~

Return of HEU Plants to Production Work

)

On Friday, August 20, 1976, BW reported that the results of the i inventory were reconciled to the accounting records, and that a ]

significant MUF gain had been realized. By letter dated August 20, 1976, BW requested permission to startup highly enriched '

uranium operations.

]

On Saturday, August 21, 1976, Task Force members met with the j Material Plants Manager and his staff to review the overall '

results of the re-inventory.- Particular attention was paid to

where and what produced the material gain. The gain was found to ,

be consistent with NRC expectations based upon knowledge of  !

material type, inventory, and processing. l This review determined that the overall results of the re-inventory were satisfactory. After a telephone conference call to Task Force Management in Philadelphia, the Materials Plant Manager was l infomed that the plants were released to resume nomal operations.

Written confirmation of the NRC release was provided in a letter j dated Monday, August 23, 1976.

Process Control Upon Return to Full Operations During a meeting on' August 21, 1976, B W proposed that the return to operations be controlled by operating the process on a batch basis with material balances being drawn around each batch. BW proposed that all inputs and outputs for the process be monitored and measured and that a partial clean-out of loose material in 16 JUN 2 81977

J

equipment and transfer lines be conducted between batches. The input and output data would then be used to determine system holdup. This holdup would be compared to historical processing data which reflects the continuing yield of the process between

. physical inventories. BW proposed that the process be shut down if the apparent holdup exceeded that to be expected from a com-parison with historical processing data.

The Task Force approved the plan because it provided for more timely control of HEU. The Task Force considered this to be especially beneficial. ,

During the August 18 - October 18, 1976 material balance period, r BW completed three batch balances. The estimated process holdup was found to be within the limits predicted by processing records.  !

Task Force Study period Extension l I

Shortly after the return of the HEU plants to full operation on August 21,1976. Task Force management concluded that the scope of the MUF examination should be extended to include BW operations between April 28,1976 (the original. end-date of the study) and August 18, 1976. There were several_ considerations behind this decision.

In management's judgement, the period between April 1974 and August 1976 was a logical period of study because it almost coin-cided with a definite set of NRC rules and regulations for SNM Control and Accounting. On May 7,1974 (during the first BW inventory period starting on April 1,1974), new accounting rules were made effective in the Code of Federal Regulations (CFR). At that time, HEU operations were conducted in accordance with the ,

licensee's approved Material Control and Accounting Plan. On l August 21, 1976, additional control and accounting rules were made effective through implementation of the licensee's approved Fundamental Nuclear Material Control Plan and through the compre-hensive revision and upgrading of license conditions. These substantive new rules were considered to be both necessary and reasonable in assuring the safeguarding'of strategic SNM at this facility.

17 JUN 2 81977 l

~

A second consideration in the decision to extend the study period concernad the original Task Force plan (and BW commitment) to segregate and recover the backlog of scrap existing as of the April 28, 1976 physical inventory date. The plan was to credit any material gain resulting from this scrap recovery toward the cumulative MUF. In order to apply only the proper credit (in the event that gains were actually realized), it was important that all pre-April,1976 scrap be strictly segregated from other scrap being generated and recovered during the May-July,1976 effort.

Task Force audits of the pre-April,1976 scrap recovery effort

, revealed that the necessary segregation was not being maintained

] on the operating floor.

I: The Task Force detemined that extension of the MF study period to the August 18, 1976 physicial inventory date would re-create a period in which all MUF components could be isolated a.nd deter- ,

+

mined and in which consistent regulatory requirenents existed.

! Task Force Meetings with BW to Discuss New Requirements Throughout the week of August 30, 1976, Task Force representatives

, from the Office of Inspection and Enforcement and the Office of Nuclear Materials Safety and Safeguards met with BW for detailed discussions on each of the new requirenients contained in the revised Safeguards Amendment License Conditions. During these discussions, emphasis was placed on the requirements for scrap recovery on a current basis, timely accounting of all discrete SNM and upgraded criteria for accounting for all HEU on a single plant basis. This entails strenghtening internal controls to assure that discrete items ~ in Material Balance Areas (MA's) can l be accounted for on a daily basis and that measurement and documen-tation of material transfers between MA's'are accurate and current and. provide for. localizing errors on MUF by area.

Task Force Actions Since August 1976 September - November,1976 l

During September, October, and November,1976, Region I continued to closely monitor BW operations. Particular attention was paid to the actions implemented by BW to correct the material control

and accounting deficiencies found by the Task Force.

During this period, it became apparent that BW could not comply with the NRC requirement for current scrap recovery which-was

.i I

18 JWl2.6197f

-i made effective in' August. During nomal production operations, the scrap generated exceeded the capacity of the licensee's recovery process. It also was determined that the excess scrap could not be processed at an off-site scrap recovery contractor within 30 days after each BW physical inventory.

In order to attain current scrap recovery, BW requested that the )

backlog of off-specification material be returned to the Energy Research and Development Administration (ERDA) for recovery on a one-time basis. This request was favorably endorsed by the NRC. ,

l However, ERDA was not able to accept the material at that time.

The backlog of off-specification intemediate product was pro-cessed in March 1977. ,

. Another major effort by the Task Force during this period involved a special review of process control records.

December 1976 - April 1977 In December 1976, with the majority of Task Force functions completed, final work was accomplished within the regular IE program of frequent, unannounced inspections by the NRC Region I Office. Inspections conducted between late December 1976 and early February 1977, identified numerous deficiencies in the measurements' program for HEU. Many of these deficiencies were a direct result of new requirements contained in the Fundamental Nuclear Material Control Plan effective August 21, 1976. In

~a ddition, these inspections confirmed that scrap materials were not being recovered on a current basis as required. Following the physical inventory of December 21, 1976, the NRC audits and independent measurement of samples demonstrated that: the t

inventory listing contained accounting errors; the quality of measurements assigned to some scrap materials was questionable; and inadequate attention had been paid to the reconciliation and investigation of MUF in the various M A's.

Based upon the above findings, on February 28, 1977, NRC's Office of Inspection and Enforcement (OIE) issued an Order requiring that special actions be taken to reconcile the physical inventory which had just been conducted on February 21 , 1977. .The Order, which required actions beyond those normally associated with a physical inventory, provided for the following:

19 l

M 281977 l

. ', i

- All scrap materials be recovered and all off-specif-ication, intermediate product materials be verified through remeasurement or recovery; l - The results of the February inventory (MUF and other supporting data) be determined, reconciled and reported to the NRC;

! - Proup't action be taken to correct the itaas of noncom-pliance identified during Region I inspections;

- Normal production operations following the April,1977

- physical inventory not be resumed until OIE has given j written concurrence.

Between March 14 and April 22, 1977, Region I conducted four in- .

spections in order to monitor the quality and progress of BW  ;

actions taken in accordance with the order. On March 21, 1977,  ;

i B W requested that a hearing be conducted with respect to the i t Order. On March 23, 1977, BW filed a request to delay the initiation of hearing procedures since the company was then engaged in discussions with the NRC staff toward achieving an

. appropriate resolution of the matters covered by the Order. In the ensuing weeks, it became evident that all requirenents contained

^

in the Order would be satisfactorily accomplished by BW. On April 22, 1,977, an Atomic Safety and Licensing Board, appointed

' after BW requested a hearing,. approved a Joint NRC-BW Stipulation For Settlenent which stated: "that the licensee has taken remedial actions to identified items of noncompliance, has responded to the NRC request to upgrade the Material Control and Accounting Systen, is in substantial compliance with applicable NRC require-ments and that the matters covered by the Order are not the subject of dispute," thus the hearing proceedings were ended.

The NRC letter of concurrence to resume normal production opera-

~

tions following the April 1977 inventory was delivered to BW on

. April 22,1977.

I l

20 JflN 2 81g77

e SCOPE OF BW TASK FORCE OPERATIONS On January 15, 1976, BW initiated a ~special effort toward re-ducing MUF for highly enriched uranium operations. The "High Enriched Uranium Material Unaccounted For (HEU-MUF) Project" was established under the management of the Manager of Technical Control for Pennsylvania plants. Projects designed to improve MUF performance included:

- A study of staffing requirements;

- Streamlining of the accounting paper flow;

- A study of the analyticai laboratory work load;

- Increase production efficiency in terms of scrap generation;

- Improvements in non-destructive assay (MDA) measurements; All personnel involved in the project were members of the Pennsyl-vania BW organization. The above responsibilities were mainly assigned to the Nuclear Materials Control Manager, the Manager of High Enriched Uranium Manufacturing, and the Regulatory Projects Manager.

The meetings'in Bethesda in early May 1976 between BW and the NRC added new emphasis to the BW efforts in MUF investigation.

Concurrent with the formation of the NRC Task Force, BW created i . a parallel Task Force under the direction of the BW Lynchburg Research Center. The Task Force included Chemical Engineers and a Radiochemist from the Lynchburg Research Center plus professional employees from the Pennsylvania site, several of w.aich were assigned full time duties. The Task Force effort involved three separate segments, each of which was in accordance with verbal

! commitments (later confinned by NRC letter) made by BW during the early May meetings in Bethesda:

- The establiskent of a system of quarterly audits of the material control and accounting system by qualified personnel from outside the local organization; 21 JUN 2 81977

~

l

- A special effort to recover (i.e., process to a form which can be measured accurately) all highly enriched uranium scrap materials on inventory which we're generated prior to April 28,1976 (the initial and point associated with the NRC Task Force analysis of two-year cumulative MUF);

- A study and re-evaluation of all material balance data and ,

procedures supporting the highly enriched uranium elf pres- '

ently documented for the period April 1,1974 - April 28, 1976. .

The first quarterly audit of the material control and accounting system was conducted on May 18-19, 1976. The audit team was drawn -

from B8W personnel involved in safeguards and quality assurance at  :

the Lynchburg, Virginia and Barberton, Ohio facilities. The audit report was reviewed by the Task Force. On May 20. during a visit .

to Region I, the Materials Plant Manager briefed Region I on the l preliminary findings. The team noted deficiencies in the system,  ;

but the evaluation was not of sufficient depth to identify specific  !

practices which impact on MUF performance.

An independent audit of the accounting records by an outside firm was initiated in August 1976 and is still in progress. B&W also intends to unie material control and accounting personnel from the B&W Virginia facilities on future audits. These audits are in addition to the program of continuous audits performed by the local Manager of Technical Control.

The second segment of the B&W Task Force effort involved a special effort to recover scrap and other off-specification materials on hand which were generated during the April 1,197.4 -April 28,1976 period under review by the NRC Task Force.

The B&W plan originally called for completion of the scrap re-covery effort by July 31, 1976. This schedule was later modified (after application to and approval by the NRC) for selected small amounts of scrap material. Completion dates of August 31, 1976, September 30, 1976 November 30, 1976 and April 1,1977 were allowed for finishing specific types of material because of: a breakdown in equipment; the need to coordinate the special recovery effort; and the inability to negotiate a suitable recovery contract on one material type.

22 JUN 2 81gry


a .,,, - - - . -

e. -

, , - - - , . - . - - _ ,,-,pg

, _-- - ,-. --,,,,9 i,ie,. , , . - .p, , , , ,r.-- . - -w,c-wmy --_w---- .=------m.,--+---,----w

.: o,.

The third area of BW action, a study and re-evaluation of the highly enriched uranium ac. counting system, was essentially a parallel . effort to that of the NRC Task Force. BW adopted the

- same April 1,1974 -April 28,1976 time frame for its MUF study.

BW progress in the MUF review was documented through weekly letter reports prepared by the Task Force Leader. On a bi-weekly basis during June and July, these reports were reviewed by the NRC Task Force management in meetings held with BW management.

During these meetings, the progress and future plans of each Task Force was reviewed in order to coordinate the variety of opera-tions and tests involved in the elf review.

e I

l i

t 23 JUN 2 819n

~

~. .

~B W UPGRADED ORGANIZATION FOR MATERIAL CONTROL AND ACCOUNTING BW has taken major steps to upgrade the organization for material -i control and accounting. The need to improve inventory management for both financial and Safeguards reasons was a specific objective of a recent re-organization. The former Controller of the Nuclear Materials Division in Lynchburg, Virginia, has been transferred to Apollo as Manager, Nuclear Material Control and Accountability, a position reporting to the Materials Plant Manager. The SNM Control and Accountability functions have been separated under i two managers. The Manager, Nuclear Material Accountability, is ,

basically responsible for the records associated with the accounting system. The Manager, Nuclear Materials Control is basically ,

responsible for physical inventories, internal control of SNM, and the quality of measurement data as inputs to the accounting-systen. These changes were accomplished in September,1976. .

A third professional position, Perfonnance Analyst, was added to the organization responsible for improving the systen through continuous monitoring of accounting data in order to detect trends and problem areas. The internal audit function, as opposed to performance analysis, is housed in a separate department of the organization. This position was filled in October,1976. The new

! organization,also includes a professional radiochemist to concen-trate on improvements in the non-destructive assay (NDA) measure-ment program. This position was filled in December,1976.

For the near-term, the Task Force considers the addition of Safe-guards dedicated personnel in the HEU processing areas to be the major improvement in the organization. A SNM Control Technician is now continuously located on the processing floor to ensure that Safeguards procedures for sampling, measurenent, transferring, ,

etc., are properly followed. While the addition to managerial strength will undoubtedly improve Safeguards performance in the future, the Task Force believes that conscientious step-by-step execution of the system on the operating floor should improve MUF perfonnance in the near-tenn.

24 JUN 2 81977

,e s;

- ~

~

4. +:?\ ,

B&W ACTIONS RESULTING FROM THE NRC TASK FORCE RECOMMENDATIONS The Task Force review of the BW highly enriched uranium material control and accounting systen resulted in the identification of areas in which improvements in the licensee safeguards program should be made. The reconenendations have been submitted separ-ately by the Task Force to NRC and action to implement the recom-mendations has been initiated by the NRC and B8W. In order to carry out the recommendations the following actions have been taken: ,

B8W is recovering all scrap within a 60 day period including that processed by an outside contractor; t B&W has submitted to the NRC-a written commitment for upgrad-ing of the scrap recovery plant including a schedule of completion dates; i -

B&W has 'nstalled new uranium measuring equipment on a liquid effluent line considered an important loss stream by the Task Force; B&W has increased its Material Control and Accounting Staff by 25 people; I -

The fodner Controller of the B&W Nuclear Materials Division in Lynchburg, Virginia, has been transferred to the Apollo facility as Manager of Nuclear Materials Control and Account-ability; B8W has terminated an operation dealing with the. processing of highly enriched uranium that.formerly was outside a material access area; B&W has instituted a program to measure the uranium contained

, in laundry; B&W has installed a refined measurenent control program to measure more precisely the amounts of uranium contained in material destined for disposal; l -

B&W has taken measures 'and is planning additional steps to eliminate unneeded complexity from its material control l systen.

25 JI!N 2 81977 1

1 1

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