NUREG-1065, Rev. 2, 'Acceptable Standard Format and ... · Rev. 2 Acceptable Standard Format and Content for the Fundamental Nuclear Material Control (FNMC) Plan Required for Low-Enriched

NUREG-1065Rev. 2

Xcceptable Standard Format and,Content for the, FundamentalNuclear Material Control(FNMC) -Plan Required forLow-Enriched Uranium Facilities

J.S . Nuclear Regulatory Commission

Office of Nuclear Material Safety and Safeguards

~. R_ Joy

0 ~ ~ ~ ~ ~ ~ .S ucerReuaor omiso

0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

C;-~ ~ ~ ~~~~~~. .

0fc fNcerMtra aeyadSfgad

AVAILABILITY NOTICE

Availability of Reference Materials Cited in NRC Publications

Most documents cited in NRC publications will be available from one of the following sources:

1. The NRC Public Document Room, 2120 L Street, NW., Lower Level, Washington, DC20555-0001

2. The Superintendent of Documents, U.S. Government Printing Office. P. 0. Box 37082,Washington, DC 20402-9328

3. The National Technical Information Service, Springfield, VA 22161-0002

Although the listing that follows represents the majority of documents cited in NRC publica-tions, it is not intended to be exhaustive.

Referenced documents available for inspection and copying for a fee from the NRC PublicDocument Room include NRC correspondence and internal NRC memoranda; NRC bulletins,circulars, information notices, inspection and investigation notices; licensee event reports;vendor reports and correspondence; Commission papers: and applicant and licensee docu-ments and correspondence.

The following documents in the NUREG series are available for purchase from the GovernmentPrinting Office: formal NRC staff and contractor reports, NRC-sponsored conference pro-ceedings, international agreement reports, grantee reports, and NRC booklets and bro-chures. Also available are regulatory guides, NRC regulations in the Code of Federal Regula-tions, and Nuclear Regulatory Commission Issuances.

Documents available from the National Technical Information Service include NUREG-seriesreports and technical reports prepared by other Federal agencies and reports prepared by theAtomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.

Documents available from public and special technical libraries include all open literatureitems, such as books, journal articles, and transactions. Federal Register notices, Federaland State legislation, and congressional reports can usually be obtained from these libraries.

Documents such as theses, dissertations, foreign reports and translations, and non-NRC con-ference proceedings are available for purchase from the organization sponsoring the publica-tion cited.

Single copies of NRC draft reports are available free, to the extent of supply, upon writtenrequest to the Office of Administration, Distribution and Mail Services Section, U.S. NuclearRegulatory Commission. Washington DC 20555-0001.

Copies of industry codes and standards used in a substantive manner in the NRC regulatoryprocess are maintained at the NRC Library, Two White Flint North,1 1545 Rockville Pike, Rock-ville, MD 20852-2738, for use by the public. Codes and standards are usually copyrightedand may be purchased from the originating organization or, if they are American NationalStandards, from the American National Standards Institute, 1430 Broadway, New York, NY10018-3308.

NUREG-1065Rev. 2

Acceptable Standard Format andContent for the FundamentalNuclear Material Control(FNMC) Plan Required forLow-Enriched Uranium Facilities

Manuscript Completed: November 1995Date Published: December 1995

D. R. Joy

Division of Fuel Cycle Safety and SafeguardsOffice of Nuclear Material Safety and SafeguardsU.S. Nuclear Regulatory CommissionWashington, DC 20555-0001

Abstract

This document provides a revised structure forNUREG-1065, Revision 1, dated 1985, whichcontains information that the licensee or applicantshould provide in its fundamental nuclear materialcontrol (FNMC) plan. The revised structure hasbeen patterned after the acceptable standard formatand content for the FNMC plan required for lowenriched uranium enrichment facilities,NUREG/CR-5734, KIITP-415, dated 1991. Thatdocument was structured in a manner to serve as adirect outline for licensees preparation of theirFNMC plan while NUREG-1065 was structured tomore closely follow and elaborate on the contentsof 10 CFR 74.31.

This revised structure for NUREG-1065 continuesto implement the requirements of 10 CFR 74.31.It applies to NRC licensees (other than produc-

tion or utilization facilities licensed pursuant to10 CFR Part 50 or Part 70 and waste disposalfacilities) that are authorized to possess and usemore than one effective kilogram of unencapsulatedSNM of low strategic significance. Unlike thestructure of the enrichment guide, however, thisrevision for NUREG-1065 includes a complete setof affirmations required from licensees. These arethe same affirmations that appeared in Revision 1 ofNUREG-1065 but in some instances the languagehas been modified to clarify intent.

All other modifications involve format and editorialchanges designed to provide clarifications andfacilitate preparation or revision of the requiredFNMC plan.

NUREG-1065 Rev. 2..

ContentsPare

Abstract ......... iii

Acronyms .ix

Introduction ........................................................ -.. I

Performance Objectives, System Features and Capabilities, and Affirmations . 3

Performance Objectives . 3Affirmations Pertaining to Performance Objectives . 5Affirmations Pertaining to System Capabilities . 5

1 Organization . 11

1.0 Regulatory Intent ...... ........... ......... 111.1 Corporate Organization ............ 1112 Plant or Site Organization ........... 111.3 MC&A Organization ............................... 11

.1.3.1 Responsibilities and Authority .......................... 111.32 MC&A Procedures .......................... 12

1.4 Training and Qualification Requirements - ........................ 121.5 MC&A System Description .............. ....... . ...... 121.6 Acceptance Criteria ..... ......................... 131.7 Affirmations ............ ................. 14

2 Measurements .15

2.0 Regulatory Intent ...... ............................ 152.1 Measurement Points ............... 1522 Measurement Systems ....................... ...... 15

22.1 Bulk Measurement Systems ............................ 15222 Analytical Measurement Systems . ........................... 1622.3 NDA Measurement Systems -... 1622.4 Other Measurement Systems .... 16

2.3 Measurement Uncertainties .......... .................................... 162.4 Measurement Procedures ............................... 1.....6 .............. 162.5 Acceptance Criteria ................... 172.6 Affirmations .... ........... .... S 17

3 Measurement Control Program .19

3.0 Regulatory Intent .. 193.1 Organization and Management .. 19

3.1.1 Functional Relationships .193.12 Procedures .193.1.3 Contractor Program Audits and Reviews .19

32 Calibrations ................... . 203.3 Control Standard Program .................. 213.4 Replicate Program ................... 223.5 Control Limits .. ............................. 22

3.5.1 Measurement Control Data Analysis. ............................... 233.52 Response Actions ............................... 23

3.6 Acceptance Criteria ............. 233.7 Affirmations ............. 25

v NUREG-1065 Rev. 2

4 Statistics ......................................... .27

4.1 Determination of Measurement Uncertainties .................................... 274.2 Determination of SEID . .......................................... 2743 Bias Corrections .......................................... 28

S Physical Inventories .......................................... 29

5.0 Regulatory Intent ................................5.1 General Description ..............................52 Organization, Procedures, and Schedules ..............5.3 Typical Inventory Composition ......................5.4 Description of Typical Item Strata ...................5.5 Conducting Physical Inventories .....................5.6 Inventory Difference Limits and Response Actions .......5.7 Acceptance Criteria ..............................5.8 Affirmations ...................................

292929293030323233

6 Item Control .......... 35

6.0 Regulatory Intent ...................................6.1 Organization ......................................62 General Description .................................6.3 Item Identity Controls ...............................6.4 Storage Controls ...................................6.5 Item Monitoring Methodology and Procedures .............6.6 Investigation and Resolution of Item Discrepancies ..........6.7 Acceptance Criteria .................................6.8 Affirmations ......................................

353535353636363637

7 Shipper-Receiver Comparisons ....................... 39

7.0 Regulatory Intent .................................7.1 Receiving Procedures ..............................72 Determination of Receiver's Values ...................7.3 Evaluation of Shipper-Receiver Differences ..............7.4 Resolution of Significant Shipper-Receiver Differences .....7.5 Acceptance Criteria ...............................7.6 Affirmations ....................................

8 Assessment and Review of the Material Control and Accounting Program .....................

39393939404040

41

8.0 Regulatory Intent ......................................................... 418.1 General Description ....................................................... 4182 Report of Findings and Recommendations ..................................... 428.3 Management Review and Response to Report Findings and Recommendations .... ....... 428.4 Acceptance Criteria ....................................................... 428.5 Affirmations ........................................................... 43

9 Resolving Indications of Missing Uranium .45

9.0 Regulatory Intent ................... .............................. 459.1 Methods and Procedures for Identifying Indicators .............. ................. 459.2 System and Procedures for Investigating and Resolving Indicators ..... ............... 459.3 Response Actions for Unresolved Indicators ........... ......................... 459.4 Documentation Requirements .............................................. 469.5 Acceptance Criteria . ................................................ 469.6 Affirmations . ................................................... 46

NUREG-1065 Rev. 2

........................

........................

........................

........................

........................

........................

........................

........................

........................

..............

..............

..............

..............

..............

..............

..............

..............

..............

.......................

.......................

.......................

.......................

.......................

.......................

.......................

vi

10 Informational Aid for Assisting in the Investigation and Recovery of Missing Uranium ..... . . . . . 47

10.0 Regulatory Intent ............. ... .. ... .. ... .. .. ... .. .. ... .. .. ... . 4710.1 Information Aid ............. .. .. .. .. ... .. .. .. .. ... .. .. .. .. .. .. . 4710.2 Acceptance Criteria .............................................. 4810.3 Affirmations .................................................. 48

11 Recordkeeping ...................................................... . 49

11.0 Regulatory Intent ............................................... . 4911.1 Description of Records ............................................ . 4911.2 Program and Controls for Ensuring an Accurate and Reliable Record System .5.0... . . . . . . 011.3 Acceptance Criteria ............. .. .. .. ... .. .. .. .. ... .. .. .. .. .. ... . 5011.4 Affirnations .................................................. . 51

Glossary ............................................................. . 53

NUREG-1065 Rev. 2vii

Acronyms

Code of Federal RegulationsDetection quantityDetection thresholdFormula kilogramFundamental nuclear naterial controlHigh-enriched uraniumInternational Atomic Energy AgencyItem control areaInventory differenceLow-enriched uraniumMaterial balance areaMaterial control and accountingNondestructive assayNuclear Materials Management andSafeguards System

NRCSEID

SNMSRDUU-233U-234U-235U-238U0 2

U30aUF6wt

U. S. Nuclear Regulatory CommissionStandard error of the inventorydifferenceSpecial nuclear materialShipper-receiver differenceUraniumUranium-233Uranium-234Uranium-235Uranium-238Uranium dioxideUrano-uranic oxideUranium hexafluorideWeight percent

NUREG-1065 Rev. 2

CFRDQDTFKGFNMCHEUIAEAICAIDLEUMBAMC&ANDANMMSS

ix

Introduction

An applicant's fundamental nuclear material control(FNMC) plan must demonstrate how the basiccapabilities specified in 10 CFR 74.31(c) areachieved and maintained and how such capabilitiesare used to achieve the performance objectiveslisted in 10 CFR 74.31(a). After accepting anFNMC plan and imposing it as a condition oflicense, the U.S. Nuclear Regulatory Commission(NRC) will judge the adequacy of a licensee'smaterials control and accountability (MC&A)performance by inspecting for compliance vithcommitments and practices described in the plan.

Because 0 CFR 74.31 is, for the most part, aperformance-oriented regulation, the emphasis is ondefining objectives rather than the means forachieving them. Thus, applicants and licenseeshave many alternatives with regard to how theirMC&A system and program are designed,managed, and operated. Hence, this documentcannot begin to cover all possible methodologiesthat a licensee might use to achieve the desiredobjectives. Instead, this document providesexamples of acceptable MC&A approachescommonly or typically used by low enricheduranium fuel fabricators. This document isintended for use by applicants, licensees, and NRCsafeguards licensing reviewers. For the most part,recommended criteria are not to be regarded asrigid, fixed standards. That is, a lowereffectiveness of one feature relative to a particularaspect can be tolerated if there is a compensatingfeature, or combination of features, that provides anoverall effective safeguards system. Although notlikely, a licensee could fail to meet most of thecriteria recommended in this document and stillachieve the regulatory objectives. In the finalanalysis, an NRC reviewer must make a judgmentas to whether the applicant or licensee can achievewith high probability, without going beyond itsFNMC plan commitments, the objectives stated in10 CFR 74.31(a). The recommendations providedherein pertain to both applicant submitted FNMCplans and any revisions made to existing approvedplans.

Pages 3 through 10 of this introduction sectiondescribe the basis of the three general performance

objectives of 10 CFR 74.31 and the MC&A systemfeatures and capabilities needed to meet theobjectives. The chapters that follow incorporateand expand on (1) the general performanceobjectives, and (2) the system features andcapabilities. The first eight chapters are arrangedin a sequence corresponding to the systemcapabilities required by 10 CFR 74.31(c)(1) through(8), of which (c)(5), and hence chapter 5, dealswith the first performance objective of confirmingthe presence of SNM. Chapters 9 and 10 addressthe second and third performance objectives ---"resolving indications of rissing uranium' and aidin the investigation and recovery of nissingmaterial". Chapter 11 deals with the recordkeeping requirements of 10 CFR 74.31(d).Together these 11 chapters provide applicants orlicensees an outline for an FNMC plan.

The body of an FNMC plan, corresponding toChapters 1 through 11, shall be a condition oflicense, and compliance with the FNMC plancommitments, the required affirmations, and relatedprocedures will be inspectable. Explanations anddiscussions appearing in the body of the plan shouldbe detailed and precise, as opposed to general andvague, so that all parties concerned (i.e., NRClicensing reviewers, NRC inspectors, and licenseepersonnel responsible for plan execution) have aclear and nonconflicting understanding of the what,how, and when aspects of each plan commitment.

The annex (or appendix) of an FNMC plan shouldprovide supplementary and general informationabout the facility and the MC&A system e.g.,copies of blank record forms, site map, processdiagrams, an example standard error of theinventory difference (SEID) calculation, etc.]. Theannex will not be incorporated as a condition oflicense and will not be the basis for inspection.Thus, descriptions presented by the applicant orlicensee to satisfy regulatory intent must be in theplan itself, rather than the annex, and must provideadequate detail so as not to be largely dependent onexamples or supplementary information in theannex for proper understanding.

NUREG-1065 Rev. 21

Performance Objectives, System Featuresand Capabilities, and Affirmations

Performance Objectives

The general performance objectives and the systemcapabilities to be addressed by the FNMC plan areset forth in 10 CFR 74.31. The basis on whicheach of the three general performance objectivesand eight system capabilities were formed isdescribed in the following paragraphs of this sectiontogether with identification of the related requiredlicensee affirmations.

General Performance Objectives

1. Confirm the presence of special nuclearmaterial

The purpose of this objective is to verify thepresence of or to detect the occurrence ofany significant loss or theft of special nuclearmaterial (SNM). To naintain currentinformation of the SNM in their possession,licensees should have in place a program thatprovides timely, accurate, reliableinformation about the quantity and location ofmaterials in their possession. Accurateinformation means that item quantities forboth the element uranium and the isotope U-235 are based on measured values or onreliable factors. Reliable information meansthat the quantity of material in an item andthe location of all items is known (with thepossible exception of items that have beencreated, transferred, or consumed within thepast 8 hours) and that the locationdesignations are specific enough to providefor the retrieval of the items in a promptmanner. Reliable informtion also-meansthat the quantities and locations of all classesof material and'items listed in the'accountingrecords are correct and verifiable (with thepossible exception of items that have beencreated, transferred, or consumed within thepast 8 hours).

The licensee or applicant should accuratelyaccount for all SNM that is received andshipped by maintaining reliable records basedon accurate measurements. When a shipmentis received, the licensee should beginmonitoring movement and location of the

material within the facility using item controlprocedures (1) to monitor the location andintegrity of items until they are introduced tothe process and (2) to ensure all SNMquantities of record associated with receipts,shipments, discards, and ending inventoryare based on measurements.Recommendations for meeting the

- performance objectives for the item controlprogram are provided in Chapter 6 of thisdocument, and for measurements andmeasurement control programs are providedin Chapters 2 and 3, respectively.Monitoring the material in process mayinvolve the use of process or material controldata. A detailed and accurate recordkeepingsystem for the generated data that providesknowledge of the material's location on atimely basis should be maintained to supportthis function.

The licensee should conduct total plantphysical inventories at an average interval ofno more than 12 months, with no singleinterval (between any two consecutivephysical inventories) being greater than 13calendar months. Each physical inventorymust be conducted in a manner that providesat least a 90 percent power of detecting anyactual loss or theft of a detection quantity(DQ) that may have occurred since the lastyearly inventory. A DQ is a site-specificquantity of U-235, the magnitude of which isdiscussed in Chapter 4 of this document.

The icensee should verify the presence of allSNM currently possessed by the facility, asstated in its accounting records. Thisverification is normally accomplished by ashutdown and cleanout of processingequipment, measurement of cleanoutmaterials and measurement of any materialsnot previously measured in their existingform, visual verification (on a 100 percentbasis) of the presence of all possessed SNMitems (by means of unique item identities),and confirning the SNM quantitiesassociated with unencapsulated and unsealeditems on ending inventory. However, adynamic (i.e., nonshutdown) inventory ofsome or all processing equipment may beutilized if the measurement uncertaintyassociated with the total material balance (for

NUREG-1065 Rev. 23

Objectives and Capabilities

the inventory period) is within the 0.25percent of active inventory constraintspecified in 10 CFR 74.31(c)(4).Recommendations pertaining to physicalinventories are detailed in Chapter 5 of thisdocument. In summary, a total plantphysical inventory involves:

(1) verifying the presence, on a 100percent basis, of all uniquely identifiedSNM items listed in the accountingrecords

(2) measuring (by direct measurement or,if direct measurement is not feasible,by indirect measurement) all bulkSNM quantities on hand (i.e., allSNM not in item form)

(3) measuring any items not previouslymeasured

(4) verifying the identity and integrity ofall encapsulated items and itemsaffixed with tamper-indicating seals

(5) measuring a SNM related parameterfor a sample of randomly selectedunencapsulated and unsealed items,based on a statistical sampling plan, toverify the previously measuredquantities of SNM contained in suchitems

The physical inventory program should bemanaged and maintained independent of theproduction or operations organization butshould not be excluded from using processmonitoring and production control data.

2. Resolve Indications of Missing Uraniun

The licensee or applicant should have aformalized program to resolve any indicationthat SNM is missing. Resolution of suchindicators means that the licensee has made adetermination that a theft or loss of SNM hasnot occurred. Only indications that suggest apossible loss of items or material from itemsequal to or greater than 500 grans of U-235need be investigated; a possible loss of adetection quantity detected'during a materialbalance closure needs to be investigated andresolved in accordance with 74.31(c)(5).

Multiple loss indications from items and/orprocess equipment within a material balanceperiod must be resolved even though the lossfrom each event is less than 500 grams ofU-235. This resolution need not occur untilthe total potential loss equals or exceeds 500grams of U-235.

Resolution of an indicator depends on thetype of indicator. Anomalies at plants canindicate a number of scenarios from simpletheft to complex diversions. The resolutionprocess in some cases should begin with athorough review of the MC&A records tolocate blatant errors. These errors mightinclude omissions of entire items, incorrectentries to computer programs or records,transcription errors, incorrect estimates of theamount of holdup in equipment, orcalculational errors. A detailed examinationof the MC&A records for each material typeshould identify gross errors. The next stagein the resolution process would be to isolatethe process or storage area that appears to becausing the anomaly. Once this isaccomplished, all of the information thatcontributed to the SNM quantities for thatlocation should be verified. If resolution stillis not accomplished, the licensee shouldremeasure and sample material in the processor storage areas to verify quantities. If theinvestigation of an indicator results in adetermination that an actual loss or theft hasoccurred, the loss or theft must be reportedto NRC pursuant to 10 CFR 74.11.

3. Aid in the Investigation and Recovery ofMissing Material

If the NRC and/or other government agenciesdeem it necessary to conduct an investigationrelating to actual (or highly suspected) eventspertaining to missing material, the licensee isto provide, without specifically being asked,any information deemed relevant to therecovery of material involved in a loss, theft,or diversion . The burden shall be on thelicensee to provide (without being asked) allinformation that it recognizes as beingrelevant, as opposed to providing onlyinformation that the investigators request.Additional information and recommendationspertaining to providing information to aid in

NUREG-1065 Rev. 2 4


investigations are provided in Chapter 10 ofthis document.

The licensee is required to make affirmationswith respect to these three performanceobjectives and the eight system capabilities, andthese must be stated (without any modification)in the FNMC plan. The affirmations are listedbelow:

Affirmations Pertaining to PerformanceObjectives

* An MC&A system will be maintained that iscapable of confirming, at least annually, thepresence of all SNM expected to be present(at a given time) based on the ac'countingledgers, with the possible exception of wastematerials transferred to (via DOE/NRC Form741 transactions) and stored in holdingaccounts.

* An expeditious investigation will be promptlyinitiated for all indications of significantlosses (500 grams of U-235) of SNM andeach alegation or indication of SNM theft.

* Information will be provided to appropriateFederal Authorities to aid in their investiga-tion of indications of missing naterial and inthe recovery of SNM in the event of a loss,theft, or unauthorized diversion.

* A cause or probable cause that is based onobjective evidence will be assigned to eachindication of possible loss that isinvestigated.

* The results of all investigations of allegedthefts, and any indications of a significantloss (500 grams U-235) which remain'unresolved after 30 calendar days shall bereported to the'appropriate NRC MC&Alicensing authority.

Affirmations Pertaining to System-Capabilities

To meet the three general objectives, the MC&Asystem depends on features and capabilities asrequired by 10 CFR 74.31(c)(1) through (c)(8) and74.31(d). These system capabilities and their'related affirnation statements are identified below,and are discussed in Chapters I through 8 and 11 ofthis document.

(1) Establish, document and naintain amanagement structure which assures clearoverall responsibility for material control andaccounting functions, independence ofMC&A management from production

-responsibilities, separation of keyresponsibilities and adequate review and useof critical material control and accountingprocedures.

In its FNMC plan,-the licensee must makethe following affirmations with respect tomanagernent structure:

* Responsibility for the overall MC&A systemmanagement is assigned to a position that isseparate from production responsibilities orany other responsibilities that may give riseto conflict of interest.

* The responsibility for each MC&A functionis assigned to a specific position in theorganization, and the organization isstructured in a way that the key functions areseparated or overcheck one another. Theposition descriptions are available in writingto the personnel affected.

* The facility organization and the MC&Apolicies and procedures are documented.

* All critical MC&A procedures, and anyrevisions thereto, are reviewed and approvedprior to their implementation.

* Management policies are established,documented and maintained to ensure that allcritical MC&A procedures are adhered toincluding measurement procedures used foraccountability purposes.

(2) Establish and maintain a measurement systemwhich ensures all quantities in the materialaccounting records are based on measuredvalues.

NUREG-1065 Rev. 25


The licensee must make the followingaffirmations with respect to the.accountability measurement system:

* A program of measurement procedures andmethods is maintained for all SNM receipts,removals and inventory items, and allquantities of SNM in the material accountingrecords are based on measured values.

* Measurement systems that are the keycontributors to the total measurementstandard error will be identified. The listwill be reviewed annually and updated asnecessary. These are considered as keymeasurement systems and their standarddeviations are monitored and controlled bythe measurement control program.

(3) Follow a measurement control programwhich assures that measurement bias isestimated and significant biases areeliminated from inventory difference valuesof record.

The licensee must make the followingaffirmations with respect to measurementquality:

* A measurement control program is followedby which all measurement biases associatedwith key measurement systems, are estimatedand any significant biases are eliminatedfrom inventory difference (ID) values andshipper/receiver differences.

* The calculation of the measurementcontributions to SEID are traceable to theappropriate measurement error data and tothe calibration standards used.

* The total measurement uncertainty iscontrolled so that twice its standard error foreach material balance period will be less thanthe greater of 9,000 grams of U-235 or 0.25percent of the active U-235 inventory for theinventory period. (NOTE: "Twice thestandard error associated with totalmeasurement uncertainty" can also bedenoted as LEID", i.e., "the limit of errorof the inventory difference").

* The measurement systems have adequatecalibration frequencies, sufficient control ofbiases, and sufficiently small standarddeviations to achieve the requirements of74.31(c)(4). A measurement controlprogram is used by both in-house and by any

utilized contractor to assure that the qualityof the measurements is maintained on a levelconsistent with the regulatory requirements.

(4) In each inventory period, control totalmaterial control and accounting measurementuncertainty so that twice its standard erroris less than the greater of 9,000 grams ofU-235 or 0.25% of the U-235 active inven-tory, and assure that any measurementperformed under contract are controlled sothat the licensee can satisfy the requirements.

(5) Unless otherwise-required to satisfy 10 CFRPart 75, perform physical inventories at leastevery 12 months (with no more than 13calendar months between any two consec-utive physical inventories); and, within 60days after the start of the inventory, reconcilcand adjust the book inventory to the resultsof the physical inventory, and resolve, orreport an inability to resolve, any inventorydifference which is rejected by a statisticaltest which has a 90% power of detecting adiscrepancy of a quantity of U-235established by NRC on a site specific basi..

The licensee must make the followingaffirmations with respect to physicalinventories:

* An MC&A system will be naintained that ,capable of confirning, at least annually, thepresence of all SNM expected to be present(at a given time) based on the accountingledgers, with the possible exception of watematerials transferred to (via DOE/NRCForm 741 transactions) and stored in holdingaccounts.

* Unless otherwise required by FacilityAttachments that satisfy 10 CFR Part 75,physical inventories will, on the average, beperformed at least every 12 months (with nomore than a 13 calendar month intervalbetween consecutive physical inventory dates)and will be used as the basis for reconcilingand adjusting the book inventory which isdone within 60 days after the start of eachphysical inventory.

* For each physical inventory, inventoryprocedures are clearly written and arereviewed and approved by the individual

NUREG-1065 Rev. 2 6


responsible for the conduct of the physicalinventory.

* The individual responsible for the conduct ofthe physical inventory is either free frompotential conflicts of interest or isoverchecked sufficiently to preventcompromising the validity of the physicalinventory.

* Each physical inventory listing will includeall SNM possessed (on the inventory date),with the possible exception of waste materialsassigned to holding accounts (in accordancewith DOE/NRC Form-741 instructions) andresidual holdup that can be regarded as zeroby definition, and that all such listed SNMquantities shall be based on measurements(for SNM quantity).

* Within 60 days after the start of eachphysical inventory, the inventory differencewill be determined. Any inventorydifference, which is rejected by a statisticaltest that has a 90% power of detecting adiscrepancy of a quantity of U-235established by the NRC on a site-specificbasis, will be reported to the appropriateMC&A licensing and MC&A inspectionorganizational units at NRC Headquarters.

* Discrepancies in the identity, quantity orlocation of items, objects or containers ofSNM that are detected during a physicalinventory will be corrected.

* Inventory difference values will be correctedfor (1) accounting adjustments resulting fromprior period activity, and (2) significantbiases that have not been'previously takeninto account. (NOTE: See the definition of.significant bias' in Chapter 4 of this Plan).

* Adjustments made to reconcile the bookinventory to the physical inventory are inaccordance with standard accountingpractices and are traceable and auditable inthe MC&A records.'

* Whenever a finalized U-235 ID (afterapplying any appropriate bias corrections andprior period adjustments) is greater than the

U-235 detection threshold (DT) and is notresolved within the 60-day reconciliationperiod, all SNM processing will be haltedunless otherwise authorized by the NRC.[NOTE: This applies to both positive andnegative ID values.]

* The results of all physical inventories and ofinvestigations and resolution actions follow-ing any excessive U-235 ID are recorded andauditable. An excessive ID is one (regard-less of the algebraic sign) that exceeds both3.00 times SEID and 9,000 grams U-235.

(6) Maintain current knowledge of items whenthe sum of the time of existence of an item,the time to make a record of the item, andthe time necessary to locate the item exceeds14 days. Store and handle, or subsequentlvmeasure, items in a manner so thatunauthorized removals of substantialquantities of material from items will bedetected. Exempted are items individuallycontaining less than 500 grams of U-235 upto a total of 50 kilograms of U-235, solution.with a concentration of less than 5 grams ofU-235 per liter, and items of waste destinedfor burial or incineration.

The licensee must make the followingafirmations with respect to maintainingcurrent knowledge of items and detectingunauthorized removals:

* For items not exempted from item controlprogram coverage, a record system ismaintained to provide a knowledge of thecurrent status of such items for which thesum of the elapsed time from initialgeneration of the item plus the time requiredto locate the item, including any timerequired to complete or update item records,would exceed 14 calendar days. For itemssubject to this commitment, the item controland records system provides the capability topromptly, locate and confirm the existence of

y . i_t..em.any specific item or group of items upondemand. The item record system is securedin such a manner that the record of an item'sexistence cannot be destroyed or falsified bya single'individual without a very highprobability of detection.

NUREG-1065 Rev. 27


* Each item (unless it is an exempted item) isstored and handled in a manner that enablesdetection of or provides protection againstunauthorized or unrecorded removals ofSNM; otherwise knowledge of the SNMcontent is assured by seals or maintaining theitem as a sealed source (i.e., as encapsulatedmaterial).

* All incidents involving missing orcompromised items or falsified item recordsare investigated. (A compromised item isone for which there is evidence of tamperingor which is found outside its assignedcontrolled access area.)

* The contents of a compromised item or anunsealed, unencapsulated item located after ithas been missing will be redetermined bymeasurements (i.e., by NDA or by weighing,sampling and analysis).

(7) Resolve on a shipment basis and, whenrequired to satisfy Part 75 of this chapter,also on a batch basis, shipper/receiverdifferences that exceed both twice thecombined measurement standard error forthat shipment (and/or batch) and 500 gramsof U-235.

The licensee must make the followingaffirmations with respect to resolvingshipper/receiver differences:

* Each shipment received is inspected for lossor damage to the container or seals todetermine if SNM could have been removed.If the integrity of the shipping container isquestionable, the presence of all items thatwere packaged in the container will beverified.

* Except for those materials specificallyidentified as exempted in Chapter 7 of thisFNMC plan, measurements of the quantity ofSNM received in each shipment areperformed and the shipper/receiver differenceis tested for statistical significance.Occurrences of significant shipper/receiverdifferences in excess of 500 grams U-235and missing items are reported to the shipperpromptly.

* For SNM received, shipper/receiverdifferences that are statistically significantand also greater than 500 grams U-235, on atotal shipment basis (and also on a batchbasis when subject to 10 CFR Part 75), aredetected within 30 days of receipt except forthose materials specifically identified inChapter 7 of this FNMC plan as beingexempted from shipper/receiver evaluations.

* Measurement results for shipments andreceipts are corrected for biases that aresignificant at the 0.05 level (i.e., for any biasthat exceeds two times the standard errorassociated with a mean), and which impactindividual items by more than their roundingerror in terms of U-235 and/or uraniumcontent.

* A significant shipper/receiver difference,e.g., one that exceeds both twice thecombined measurement standard deviationfor that shipment (and also any batch whensubject to Part 75) and 500 grams of U-235,is promptly investigated and resolved on ashipment basis and also on a batch basiswhen required to satisfy 10 CFR 75.4(d).

* Significant shipper/receiver differences, asdefined above, are reported to theappropriate NRC safeguards licensingorganizational unit by telephone or facsimilewithin 5 calendar days of determining suchsignificant difference.

(8) Independently assess the effectiveness of thematerial control and accounting system atleast every 24 months, and documentmanagement's action on prior assessmentrecommendations.

The licensee must make the followingaffirnations with respect to independentassessment and evaluation of the MC&Asystem effectiveness:

* The capabilities and performance of theMC&A system will be reviewed and itseffectiveness will be independently assessedat least every 24 months. That is, thenominal elapsed time from the completion ofone review/assessment to the completion ofthe next will not exceed 24 calendar months.

NUREG-1065 Rev. 2 8


* Reviews and assessments are performedeither by qualified individuals from outsideor qualified individuals from inside thefacility organization whose work assignmentsand positions within the organization will notinpair their ability to nake objectivejudgements of the MC&A system capabilitiesand performance.

* Each overall review and assessment will beconducted and completed in a time frame thatis short with respect to the time for changesto have occurred in the MC&A system andwill include any such changes made duringthe time the review/assessment is beingconducted.

* The completion date for anyreview/assessment is defined as the datewhen the team submits its final written report(of findings and recommendations) to plantmanagement. The start date is the first dayin which one or more team members actuallyinspect records and/or interview MC&Apersonnel, and such start date will bedocumented.

* The assessment team leader will have noresponsibility for managing or performingany of the MC&A functions.

* The results of the assessment andrecommendations for corrective action, ifany, will be documented and reported to theplant manager and other managers affectedby the assessment. Management will reviewthe assessment report and take the necessaryactions to correct MC&A systemdeficiencies. Such corrective actions (if any)that pertain to daily or weekly activities willbe initiated within 40 calendar days followingthe submittal of the review/assessment finalreport.

* Management's response to recommendationsfrom the review and assessment, includingany corrective actions ordered bymanagement and the expected time frame forcompleting such actions, will be documentedwithin 30 days following the submittal of theteam's report.

The MC&A recordkeeping requirements of 10 CFR74.31 (d) are as follows:

(1) Each licensee shall establish records that willdemonstrate that the system capabilitiesrequirements have been met and maintainthese records for at least three years, unless alonger retention time is required by Part 75of this chapter.

(2) Records which must be rnaintained pursuantto 10 CFR Part 74.may be the original or areproduced copy or a microform, if suchreproduced copy or rnicroform is dulyauthenticated by authorized personnel and themicroform is capable of producing a clearand legible copy after storage for the periodspecified by Commission regulations. Therecord also may be stored in electronic mediawith the capability for producing legible,accurate, and complete records during therequired retention period.' Records such asletters, drawings,'and specifications mustinclude all pertinent information such asstamps, initials, and signatures.

(3) The licensee shall 'naintain adequatesafeguards against tampering with and loss ofrecords.

The licerLsee must make the followingaffirrmations with respect to recordkeeping:

* A record retention system is maintained forthose records necessary to show that theMC&A system requirements of 10 CFR74.3 1(c) have been met. Such records areretained for at least thre'e years. The recordsreferred to in 10 CFR 75.22 and 75.23 andgenerated during any period that the facilityis under International Ato'rnic Energy Agency(IAEA) safeguards will be retained for atleast five years. Records of the followingwill be maintained current and will beretained for at least three years:

- Management structure, MC&A jobdescriptions, and MC&A policies andprocedures;

- Accounting source data records(accounting source data normally

NUREG-1065 Rev. 29


consist of shipping and receivingforms, physical inventory forms, andthe forms used for initially recordingmeasurement and measurement controldata. After an item is destroyed, theitem location record needs to beretained for an additional 14 days butthen may be destroyed);

Records of shipmeints and receipts andinvestigations of significantshipper/receiver differences plus theinformation used to resolve them;

Measurement data for receipts,shipments, discards, and inventory;

Calibration of measurement systems,measurement control data, biasestimates, and the statistical analysesof the measurement control data;

* Sufficient protection and redundancy of therecord system is provided so that an act ofrecord alteration or destruction will noteliminate the capability to provide a completeand correct set of SNM control andaccounting information that could be used toconfirm the presence of SNM, resolveindications of missing material, or aid in theinvestigation and recovery of missingmaterial.

* Ready traceability will be provided for allSNM transactions from source data to finalaccounting records.

The following chapters of this document incorporateand expand on the performance objectives and onthe system features and capabilities of 10 CFR74.31. The chapters are arranged in a format andsequence to provide applicants and licensees anoutline for the FNMC plan.

Data used to demonstrate that themeasurement system performanceachieves the standard deviation limitsrequired by 10 CFR 74.31(c)(4);

Physical inventory listings andinventory work sheets;

Calculations of detection thresholds forexcessive IDs of a safeguardssignificance (i.e., ID alarm point fordetecting a loss equal to or greaterthan DQ);

Calculations of the standard error ofthe estimated ID and information usedto reconcile an excessive ID (i.e., anID that is > 3.00 times SEID);

Reports of investigations andresolution of indications of loss ofSNM; and

The results of independent assessmentsand management action taken tocorrect any deficiencies identified.

NUREG-1065 Rev. 2 10

1 Organization

1.0 Regulatory Intent

The intent of 10 CFR 74.31(c)(1) is to requirelicensees to implement a management structure thatpermits effective functioning of the MC&A system,and assures that the MC&A program performancewill not be adversely affected by the plantmanagement structure. Documentation, review andapproval of the procedures, and the assignment ofthe key functions to specific positions eliminatesambiguities about what is to be done by whom.The management structure is meant to separate keyMC&A functions from each other in order toprovide overchecks that increase MC&A systemreliability and counter defeat of the system throughdeceit and falsification. It is also meant to freeMC&A management from conflicts of interest withother major functions such as production.

1.1 Corporate Organization

The corporate structure'should be described, and allcorporate organization positions that have responsi-bilities related to MC&A at the licensee's siteshould be identified. A description of thecorporate-level functions, responsibilities, andauthorities for MC&A program oversight andassessments should be provided. At least onecorporate official should have responsibilitiespertaining to the control and accounting of all SNMpossessed by the licensee.

1.2 Plant or Site Organization

A description of the site's management structureemphasizing MC&A should be provided. The sitemanagement structure should be described to theextent that it can be clearly shown that the MC&Aorganization is independent of potentiallyconflicting responsibilities. This description shouldalso indicate how responsibilities are assigned forthe following functions:

(1) overall MC&A program or system

(2) SNM custodianship

(3) receiving and shipping of SNM

(4) analytical laboratories

(6) sampling operations

(7) measurement control program

(8) physical inventories

(9) on-site SNM handling operations

A brief description should be provided for eachsite-level position, outside of the MC&Aorganization,'that has responsibilities relating toMC&A activities (e.g., sampling, nassmeasurements, analytical measurements, andmeasurement control). For each position, thefunctions, responsibilities, and authorities should beclearly described.

1.3 MC&A Organization

An organizational chart and position-by-positiondescription of the entire MC&A organization shouldbe provided. An individual should be designated asthe overall manager of the MC&A program and theFNMC plan must demonstrate the assurance ofindependence of action and objectivity of decisionfor the MC&A manager. Two options for meetingthe organizational independence are (1) reportdirectly to the plant or site manager or (2) report toan individual who reports directly to the plant orsite manager and who has no productionresponsibilities.

1.3.1 Responsibilities and Authority

A description that clearly indicates theresponsibilities and authority of each supervisor andmanager should be provided for the variousfunctions within the MC&A organization. Thedescription should indicate how the activities of onefunctional unit or individual serve as a control over,or checks on, the activities of other units orindividuals. The FNMC plan should explain howcoordination is achieved and maintained betweenthe MC&A organization and other plant organizational groups that perform MC&A-relatedactivities' A'definitive statement should be madespecifying how the MC&A manager assuresappropriate review and approval for all written pro-cedures pertaining to MC&A-related activities, andto any future revisions thereto, that are issued bothwithin and outside of the MC&A organization. In

(5) bulk and NDA measurementsI1I NUREG-1065 Rev. 2

Organization

addition to the MC&A manager function, thefunctions to be addressed should include, as aminimum, the following:

(1) nuclear material accounting


(3) item control program

(4) statistical applications

Whenever more than one key MC&A function isassigned to the same person, the FNMC planshould clearly describe the checks and balances thatpreclude the following:

(1) performance of accounting or record controlfunctions by individuals who also generatesource data

(2) assignment of sole authority to any individualto overcheck, evaluate, or audit informationfor which he or she is responsible

1.3.2 MC&A Procedures

Critical MC&A procedures to be described arethose written procedures which, if not performedcorrectly, could result in a failure to achieve one ormore of the performance objectives of 10 CFR74.31(a) and the system capabilities of 10 CFR74.31(c). All critical MC&A procedures should beidentified in the body of the FNMC plan. TheFNMC plan also should contain a definitivestatement that the procedures will be followed.This set of critical MC&A procedures should, as aminimum, adequately address the following topics,regardless of which facility organizational group isresponsible for the particular topic:

(1) accountability record system

(2) sampling and measurements


(4) item control program


(6) investigation and resolution of loss indicators

(7) determination of SEID, active inventory, andinventory difference

(8) providing information to aid in investigations

(9) MC&A recordkeeping system

(10) independent assessment of the effectivenessof the MC&A program

1.4 Training and QualificationRequirements

This section of the FNMC plan should describe thetraining programs to be established and maintainedto provide qualified personnel and to provide forthe continuing level of qualification with respect topersonnel assigned to MC&A responsibilities.Training procedures and qualification criteriashould be discussed in definitive statements.Minimum qualification requirements should bestated for each key MC&A position.

1.5 MC&A System Description

The length of this section and its level of detail willbe dependent on the information provided in theprevious sections of this chapter. The overallMC&A organization should be described in amanner that explains how the general performanceobjectives of 10 CFR 74.31(a) and the capabilitiesof 10 CFR 74.31(c) will be effectively achieved.

The individual who has responsibility for each ofthe following MC&A-related functions should bespecified by title:

(1) overall MC&A program management (Note:This individual should have no major nonMC&A-related responsibilities.)

(2) measurements [Note: Responsibility may bedivided on the basis of type of measurements(e.g., analytical laboratory measurements,NDA measurements, bulk measurements, andsampling).]

(3) measurement control and statistics

(4) accountability records


Organization

(5) item control


(7) custodial responsibilities (e.g., SNM storageand movement controls)

(8) investigation and resolution of indicators thatsuggest possible loss of SNM

(9) receiving and shipping of SNM

(10) analytical laboratories

(11) MC&A recordkeeping system and controls

The information in this chapter should include adescription of the policies, instructions, procedures,duties, responsibilities, and delegation of authorityin sufficient detail to demonstrate the separation ofduties or overchecks built into the MC&A system.

This section should also include a discussionrelating to critical MC&A procedures including asummary of the review, approval, andimplementation control and requirements. CriticalMC&A procedures and all changes to them shouldundergo technical review by cognizant members ofthe staff and be approved by line managementdirectly affected and by a level of managementabove the level responsible for executing theprocedures (but not beyond on site plantmanagement).

1.6 Acceptance Criteria

A judgment that the applicant's or licensee'sapproach for establishing and documenting aneffective MC&A program that will be maintainedby an acceptable organizational (management)structure, as described in the FNMC plan, will bebased on (but not limited to) the following criteria:

* The authorship, approval authorizations, andeffective dates of MC&A policies andprocedures will be documented, and willinvolve appropriate management andtechnical staff.

* The responsibilities and authorities for eachposition assigned a function having asignificant impact on SNM control andaccounting (including all positions authorizedto control SNM movement, generate sourcedata, define and/or implement measurement

control requirements, and conduct dataanalysis) are clearly defined in a writtenposition description that defines theresponsibilities for that position.

* The qualifications and experience requiredfor each position assigned an SNM controland accounting function will be sufficient to

permit adequate performance of the dutiesrequired of that position.

* The descriptions (in the FNMC plan) of themanagement structure and assignment ofduties and authorities show that thoseresponsible for each MC&A function willhave sufficient authority to perform thefunction in the intended manner.

* The MC&A organization is separate from theproduction organization and is also separatefrom organizations that generate source data,if practical; otherwise, independence of thefunctions is attained by suitable controls andoverchecks.

* The responsibility for MC&A programmanagement is designated to an individual atan organizational level sufficient to assureindependence of action and objectiveness ofdecisions.

* No two key MC&A functions are assigned tothe same person unless adequate checks andbalances are provided. As a consequence ofthis criterion:

- Individuals who generate source data,such as performing measurements, orperform shipping and receivingactivities, do not perform anyaccounting or record control functionsunless suitable overchecks areprovided to prevent falsification ofboth source data and accountingrecords, and

- No individual has the sole authority toovercheck, evaluate performance, oraudit information for which he/she isresponsible.

* Critical MC&A procedures and all changesto them which directly affect the licensee'sability to detect the loss of SNM or to

NUREG-1065 Rev. 213

Organization

resolve indications of missing SNM, as per10 CFR 74.31(a), undergo technical reviewby cognizant members of the staff and willbe approved by line management directlyaffected and by a level of management abovethe le-vel responsible for executing the criticalprocedures (but not beyond on-sitemanagement).

All current critical MC&A procedures aremade easily accessible to all affectedindividuals, and are maintained to show foreach procedure (1) the revision number, (2)date issued, (3) who prepared the procedure,and (4) who approved the procedure (asindicated by signature and date signed).

1.7 AffirmationsThe five affirmation statements related to MC&Amanagement structure, given on page 5 undersection (1), must be stated (without modifcation) inthe FNMC plan.


2 Measurements


The intent of 10 CFR 74.31(c)(2) is that licenseesmust maintain a system of measurements to assurethat all quantities of SNM (both element and fissileisotope) in the accounting records are based onreliable measurements. The measurementuncertainty associated with the values entered intothe accounting records must be sufficiently small soas to assure that the limit specified in 10 CFR74.31(c)(4), for the total MC&A measurementuncertainty (associated with a physical inventorymaterial balance), is not exceeded. Except forsealed sources, samples, and other certainmaterials, all SNM receipts'are to be measured forthe purpose of performing shipper-receiverevaluations. In the absence of any significantshipper-receiver difference, a licensee may bookeither its measured values or the shipper's measuredvalues. When recording shipper's values (for SNMreceipts), the measurement uncertainty associatedwith the shipper's values needs to be known andused in the determination of SEID. It is alsointended that a licensee's measurement programprovide bias estimates to be used in correctinginventory difference values and shipper-receiverdifferences for significant measurement biases. Theestimation of measurement bias is discussed inChapter 3 (Measurement Control Program).

2.1 Measurement Points

The FNMC plan should identify and describe eachmeasurement that is used for accounting purposes.Measurements (1) establish the quantities in eachcustodial area, material balance area (MBA), oritem control area (ICA) and in the facility as awhole, and (2) contribute to the desired capabilityto localize losses and to generate and to assessalarms. Measurement points and/or sampling -stations should be selected to provide quantitativeinformation about material flows and inventoriesthat will permit detection and localization of anyloss or diversion, or to confirm that no diversionhas occurred. Typically, three functional types ofMBAs and ICAs are present: (1) processing,--(2)storage, and (3) receiving and shipping. Theidentification and definition of measurement pointsfor processing MBAs are necessary because of thephysical or chemical changes of the nuclearmaterials that occur in these MBAs. The storageand the'receiving and shipping areas are typicallyICAs. Typical processing MBAs include (1)processing areas, (2) decontamnination and recoveryareas, (3) laboratory areas, and (4) feed andproduct sampling and transfer areas.

2.2 Measurement Systems

The FNMC plan should describe in detail eachmeasurement system used for nuclear materialaccounting purposes. The principal elements andoperations involved in the measurement systems forMC&A encompass mass (or weight) or volumedetermination; sampling; chemical analyses forelement and isotope; and nondestructive assay(NDA). Each measurement system should also bedefined or identified by its unique set of thefollowing parameters: (1) measurement device orequipment used, (2) standards used for calibration,and (3) standards used for control. Additionally,for analytical laboratory measurements, thefollowing'also should be identified: '(1) samplingtechnique and equipment used, (2) samplealiquoting tech-nique, and (3) sample pretreatmentmethodology. Elements of the measurement controlprogram (e.g., standards traceable to a nationalsystem) used for validating and determining controllimits, precision, and accuracy levels for eachmeasure-ment system used for accountability aredescribed in Chapter 3.

The FNMC plan should provide descriptions foreach measurement system associated with bulk,analytical, and NDA measurements, and shouldidentify, where applicable, any other measurementsystems used for accounting urposes that do notfall within these-categories. These descriptionsshould provide sufficient information to demonstratehow the systems are implemented to ensure thecapability to meet the precision and accuracy limits.The following sections provide examples of thetypes of information necessary for selectedmeasurement systems.

2.2.1 Bulk Measurement Systems

For each weighing system, the applicant or licenseeshould specify the type of weighing device, the typeof container(s) weighed, material within thecontainers being weighed, capacity of the weighingdevice, range to be used, sensitivity of the device,and the calibration frequency. The descriptionshould include the capacity and the sensitivity of thescale (e.g., capacity not to exceed X kilograms, andsensitivity to be at least as good as Y grams).

For each volume measurement system, the FNMCplan should identify the vessel (e.g., tank, column,etc.), capacity of the vessel to which themeasurement applies, the material being measured,the volume measuring device and instrumentation,the sensitivity of each device and system, the range

NUREG-1065 Rev. 215

Measurements

of operation and/or calibration, and the calibrationfrequency.

2.2.2 Analytical Measurement Systems

For each analytical measurement system, theFNMC plan should specify the following:

(1) type of material or chemical compound (e.g.,UF6, uranium alloy, UO2, uranyl nitratesolution) being sampled and measured

(2) sampling technique

(3) sample handling (i.e., pre-analysis samplestorage and treatment)

(4) analytical method used

(5) characteristics measured (e.g., grams ofuranium per gram sample, U-235 isotopicconcentration)

(6) measurement interferences

(7) expected measurement uncertainty

(8) types of calibration standard(s) andcalibration frequency

2.2.3 NDA Measurement Systems

For each NDA measurement system, the FNMCplan should identify the following:

(1) the NDA equipment package (i.e., type andsize of detector, and type of associatedelectronics and computer interface, asappropriate)

(2) the type of container measured

(3) SNM material type within container

(4) attribute measured

(5) measurement configuration (including sourceto detector distance)

(6) calculational method

(7) expected measurement uncertainties

2.2.4 Other Measurement Systems

If applicable, the FNMC plan also should identifyany other measurement systems used for accountingpurposes that do not fall within the three categoriescovered by subsections 2.2.1, 2.2.2, and 2.2.3.

2.3 Measurement Uncertainties

The expected measurement uncertainties of thedescribed measurement systems should be provided.Variance components for calibration, sampling,random, and systematic error for each measurementsystem should be stated. The units in which theerrors are expressed should be clearly identified.

2.4 Measurement Procedures

The licensee or applicant should define howassurance is demonstrated that measurementprocedures (i.e., methods) are established,approved, and maintained. This can beaccomplished by (1) making a definitive statementthat an approved measurement procedures (i.e.,methods) manual, or set of approved manuals, areestablished and maintained, (2) stating whichorganizational units are responsible for thepreparation, revision, and approval of measurementprocedures, and (3) defining the requirements forperiodic review of the procedures.

A clear statement should be made defining how thefacility assures that a measurement procedurecannot be used for accountability purposes withoutdocumented approval. Each procedure should beapproved by the overall MC&A manager and bythe manager of the organizational unit responsiblefor performing the measurement. Measurementprocedures also should be approved by themeasurement control program manager.

The FNMC plan should provide a definitivestatement that all SNM quantities in the materialaccounting records are based on measured valuesand that measurement systems are maintained forthe measurement of SNM associated with thefollowing:

(1) additions to inventory (e.g., receipts)

(2) removals from inventory (e.g., shipments andmeasured discards)

(3) material on ending inventory


Measurements

For receipt of material, the licensee nay useshipper's measured values rather than its ownmeasurements, provided that (1) a shipper-receivercomparison, based on attributes or confirmatorymeasurements, shows no significant shipper-receiver difference (SRD) [as defined by 10 CFR74.31(c)(7)]; (2) in the case of a significantdifference between shipper and receiver, nosignificant difference exists between shipper's valueand the umpire value used to resolve the difference;or (3) the material in question is exempted fromshipper-receiver comparison requirements (e.g.,sealed sources and samples). However, whenbooking shipper's values, the shipper'smeasurement uncertainty should be used whendetermining SEID.


A judgment as to the acceptability of the applicant'sor licensee's system for accountability measure-ments, as described in the FNMC plan, will bebased on (but not limited to) the following criteria:

* A basic description or summary of each keymeasurement system which is utilized togenerate SNM quantities for accountabilitypurposes is provided. A measurementsystem being defined as any instrument ordevice, or combination of devices, used toderive (1) an element concentration, (2) anisotope quantity, (3) a U-235 enrichment orisotopic distribution, (4) a bulk material mass(weight), or (5) a bulk material volume, andwhich can be characterized by its randomand systematic error components.

* The set of key measurement systems, basedon recent (or anticipated) measurementcontrol data and modes of processoperations, is expected to account for at least90 percent of the total measurementuncertainty contribution to the standard errorof inventory difference.

* The recalibration frequency for eachmeasurement system is compatible with itsexpected stability. Recalibrations for allmeasurement systems should be performed atfrequencies compatible with widelyestablished, or licensee demonstrated,stability for each particular system.

* All calibrations are made with the use ofprimary standards or primary referencematerials (certified and issued by NIST or

NBL, or equivalent organization) or withreference standards traceable to primarystandards. The standards used forcalibrations need not be representative of theunknowns to be measured by the systemunless it is to be regarded as a bias-freesystem that is calibrated during each time ofuse, in which case the calibrations standardsmust be representative.

When determining an SNM quantity by weighing,sampling, and analyses, the net weight of materialin each item within an uniform material batch (orlot), such as blended U02 powder or sintered U02pellets, must be determined by direct massmeasurement. However, the element and/or isotopeconcentrations for the batch need not be determinedfor each container, but instead may be derived bysampling procedures, including:

- Analysis of composite samples ormeasurements of representative items,objects or samples selected bystatistical sampling; or

- Use of concentration and/orenrichment factors determined fromhistorical averages, controlled inputspecifications values, or empiricalrelationships where such values orrelationships are periodically tested,their uncertainties or bounds have beendetermined to be within 2.00 percentof the factor value, and wherediversions with material substitutionare improbable. However,heterogeneous materials, such asammonium diuranate, may not beassigned common factors unless thequantities are small, such as less than500 grams of contained U-235 (permaterial type, per inventory period).Justification for any materials assignedcommon factors withoutbatch-by-batch verification analysesmust be presented in the FNMC plan.

2.6 Affirmations

The two affirmation statements, given on page 6under section (2), relating to the measurementsystem capability required by 10 CFR 74.31(c)(2)must be stated (without modification) in the FNMCplan.

NUREG-1065 Rev. 217

3 Measurement Control Program


The intent of 10 CFR 74.31(c)(4) is thatmeasurement systems used to establish SNMaccountability quantities be controlled by a formalmeasurement control program that results in a U-235 total measurement standard error that is within4500 grams U-235 or 0.125 percent of the U-235active inventory quantity (whichever is greater).That is, in terms of a 95 percent confidence level,twice the standard error associated with a materialbalance total MC&A measurement uncertainty (forU-235) must be less than the greater of 9,000grams U-235 or 0.25 percent of the U-235 activeinventory. It is also intended that the programprovide bias estimates to be used for adjustinginventory difference results and correcting shipper-receiver measurements for significant measurementbiases.

3.1 Organization and Management

The organization and management of themeasurement control program should be describedin sufficient detail to show how the measurementquality assurance function is assigned and howindependence from the analytical laboratory andother units performing either sample taking ormeasurements is maintained. The measurementcontrol program manager should be at amanagement level that is sufficiently high to ensureobjectivity and independence of action. Thus, themeasurement control program manager could eitherreport directly to the overall MC&A manager or, ifin a different organizational unit, be on the samelevel as the MC&A manager.

The licensee's measurement control program shouldbe properly managed to ensure adequate calibrationfrequencies, sufficient control of biases, andsufficient measurement precision to achieve thecapabilities required by 10 CFR 74.31(c)(3) and(4).

3.1.1 Functional Relationships

The relationship and coordination between themeasurement control program manager, theanalytical laboratory, and other measurementperforming groups needs to be clearly defined.Adequate assurance should be provided that themeasurement control program manager has theauthority to enforce all applicable measurementcontrol requirements.

3.1.2 Procedures

The measurement control program proceduresshould be established and maintained in a manualthat is kept current and readily available. Thismanual should contain all the currently applicablewritten procedures pertaining to measurementcontrol and measurement quality assurance.Responsibility for preparation, revision, andapproval of manual procedures should be specified.Individual measurement control procedures shouldhave documented approval by the measurementcontrol program manager. The procedures shouldaddress the following:

(1) calibration frequencies and methods

(2) standards used for calibration (i.e.,description and storage controls)

. (3)

(4)

(5)

standards used for control (i.e., method ofobtaining or preparation, and traceability)

control standard measurements

replicate sampling and replicatemeasurements

(6) verification of process controlinstrumentation through comparison withother process instruments

-(7)

! (8)

control limits and control responses

generation and collection of control data

(9) recordkeeping controls and requirements

3.1.3 Contractor Program Audits andReviews

If measurement services are provided by an outsidecontractor or off-site laboratory, the reviewprogram used to monitor the off-site measurementsshould be described. Such reviews are to ensurethat the contractor or off-site laboratory has'anacceptable measurement control program to theextent that use of the contractor's measurementswill not compromise the licensee's ability to meetany measurement or measurement control requirement contained in either 10 CFR 74.31(c) orin its FNMC plan. An initial review of thecontractor's measurement control program shouldbe conducted before licensee use of measurementsperformed by the contractor or off-site laboratory.

NUREG-1065 Rev. 2

: I

Pikent th2 ue nf the rnntrarknr'.q mf--q.,,urL-,mt-nt-q

19

Measurement Control

All contractor or off-site laboratory assessmentfindings and recommendations should bedocumented and submitted to both the measurementcontrol program manager and the overall MC&Amanager within 30 days of completion of thereview. The two managers should arrive at anagreement on corrective actions that should betaken, based on their evaluation of the report, andshould transmit these findings to the contractor oroff-site laboratory in writing. The licensee shouldnot use measurements performed by suchcontractors or off-site laboratories until they haveverified that the corrective actions have beeninstituted.

The persons who conduct a contractor review neednot be employed by the licensee; but they shouldnot be employed by, or in any way associated with,the contractor or off-site laboratory so that theindependence of the conclusions may bemaintained.

3.2 Calibrations

The FNMC plan should sumnarize the licensee'scalibration program and confirm that the licenseehas-written procedures covering the followingtopics:

(I) calibration frequency for each measurementdevice or system

(2) identification of the standards used forcalibration of each measurement device orsystem

(3) protection and control of standards used tocalibrate measurement systems to maintainthe validity of their certified or assignedvalues

(4) the range of calibration for eachmeasurement device or system and theminimum number of calibration runs(observations) needed to establish acalibration

Unlike control standards, standards used forcalibrating measurement systems need not berepresentative of the process material or items to bemeasured by the calibrated device or system. Ifpractical, the standard used during the calibrationprocess should be subjected to all the steps involvedin the measurement process that the processunknowns are subjected to (e.g., samplepretreatment), but this need not always be the case.

It is the primary measurement device, not necess-arily the entire measurement system, that needs tobe calibrated, especially when the primary measure-ment device is common to two or more measure-ment systems. For example, the Davies & Graytitrimetric method is often used to analyze samplesfor uranium concentration of two or more differentmaterial types (e.g., UF6 , U0 2 , UNH solutions).In this case, more than one measurement system isinvolved because different sampling and samplepretreatment methods and different controlstandards are used. The potassium dichromatetitrant, however, is common to the systems; thus,the titrant is what is calibrated (or standardized)with a primary reference material such as certifiedK2 Cr2O7 , certified U30, or certified uranium metal.

In the case of nonconsumable standards used tocalibrate measurement systems (e.g., weightstandards), the frequency of recertification ofassigned values should be specified. Therecertification frequency should be dependent onhow often the standards are handled, the standard'sstability, and the adequacy of the controls used tomaintain the integrity of the standards. Biennialrecertification of such standards is usuallyacceptable.

The FNMC plan should contain a definitive state-ment that no SNM accountability value is based ona measurement that falls outside the range ofcalibration. The FNMC plan also should identifythose measurement systems that are point-calibrated. A point-calibrated measurement systemis one in which the following are true:

(I) The entire measurement system is calibratedwith a standard or set of standards that isrepresentative of the process unknowns thatare measured by the system. That is, therepresentative calibration standard(s)undergoes all the measurement steps, and inthe same manner, that the unknowns do.

(2) One or more calibration standards areprocessed and measured along with eachunknown or set of unknowns measured.That is, both the standard(s) and unknown(s)are measured during the same general timeinterval, with the same individual measuringboth the standard(s) and unknown(s).

(3) The measurement values assigned to theprocess unknowns are derived from themeasurement response observed for thestandard(s) that was measured along with theunknown(s).


Measurement Control

(4) The measurement response for each unknownshould fall within plus or minus 10 percentof the response for a standard measured atthe same time as the unknown, or as in thecase of a low concentration unknown, thedifference between the unknown's responseand the standard's response should be lessthan four times the standard deviationassociated with the standard's response.

3.3 Control Standard Program

For those measurement systems that are not point-calibrated, a defined program for the periodicmeasurement of control standards should beestablished and followed. Control standardmeasurements serve the dual purpose of (l)monitoring the stability of a previously determinedcalibration factor, and (2) estimating the averagesystem bias over a period of time (e.g., an inven-tory period). The minimum total number of controlstandard measurements during the time period, aswell as the typical frequency, needs to be specifiedfor each measurement system. Generally speaking,for each key measurement system, a minimum oftwo control standard measurements should be madeduring each week that the system is in use. Forthose key systems that are used less than eightweeks during a given material balance period, morethan two control standard measurements per weekof system use may be necessary in order toprovide:

(1) A minimum of 12 control standardmeasurements for those systems used duringthe material balance period to measurematerial totaling less than 100 kilograms ofU-235

(2) A minimum of 24 control standardmeasurements for those systems used tomeasure a total of 100 or more kilograms ofU-235

Key measurement systems for the current inventoryperiod are any set of designated measurementsystems (of the licensee's choosing) which, basedon the most recent previous period, account for atleast 90 percent of the total measurement variancecontribution to SEID. Included within the set ofkey measurement systems, should be any systemutilized to measure an SNM quantity (during an,inventory period) greater than 25 percent of theactive inventory, regardless of its contribution toSEID. The minimum number of control standardmeasurements for situations (I) and (2), above, can

be reduced to 8 and 16, respectively, for non-keymeasurement systems that measure from 1) to 25percent of the active inventory, and the minimumnumber of control standard measurements forsituations (I) and (2) can be further reduced to 4and 8, respectively, for those non-key systems usedto measure less than 10 percent of the active.inventory quantity.

Control standards should be representative of the.process material or items being measured. To berepresentative, the standards need not always beidentical to the process unknowns; but anyconstituent of the process material, or any factorassociated with a process item, that produces a biaseffect on the measurement should be present to thesame degree in the control standards. For scalesused to weigh very large items, such as UF6cylinders, the control standard weights should beartifact standards (e.g., both empty and full UF6cylinders) of certified mass to avoid a bias effectcaused by buoyancy or point loading.

For each measurement system that is not point-calibrated, the control standards to be used forcontrol standard measurements should be identifiedand/or described. Along with material compositionand matrix factors, biases also can be induced bychanges in (among other things) temperature,humidity, line voltage, and background radiation.Biases also can be operator or analyst induced.Therefore, the scheduling of control standardmeasurements should be based on the followingconsiderations:

(1) Does the variation between analysts oroperators need to be considered and hencemonitored?

-(2) Can environmental variables contribute tomeasurement bias?

(3) Is bias likely to vary with respect to the timeof day?

(4) Is a particular bias likely to be long term,short term, or cyclic in nature?

(5) Is bias a function of the process measurementvalues over the range of calibration (i.e., isthe relative percent bias nonuniform over therange of calibration)?

(6) What controls or procedures are needed to-I ensure that sampling or aliquoting of the

control standard is representative of the

NUREG-1065 Rev. 221

Measurement Control

sampling or aliquoting of the processmaterial?

(7) To estimate the bias for each measurementsystem, how much alike - in terms ofchemical composition, uraniumconcentration, density, homogeneity, andimpurity content - should the controlstandards be relative to the processunknowns?

3.4 Replicate Program

For systems involving sampling, duplicate measure-ments performed on single samples and/or measure-ments of replicate samples are necessary to estimatethe combined analytical plus sampling randomerror. For nonsampling measurement systems suchas NDA and weight measurement systems, therandom measurement variance component can bederived either from (1) replicate measurementsperformed on process items, or (2) the data gener-ated from the measurement of control standards.

The licensee should ensure that replicate samplesare independent of one another. The number ofreplicate samples measured for each analyticalmeasurement system, that is designated as a keymeasurement system, during an inventory periodshould be equal to one of the following:

(1) 100 percent of the accountability batchessampled (when less than 15 batches)

(2) the greater of 15 samples or 15 percent ofthe accountability batches sampled

(3) 50 samples (when 15 percent of the batchesis greater than 50)

For non-key analytical measurement systems, theminimum number of replicate samples to bemeasured during an inventory period should beequal to one of the following:

(4) 100 percent of the accountability batchessampled (when less than 8 batches)

(5) the greater of 8 samples or 10 percent of theaccountability batches sampled

(6) 25 samples (when 10 percent of the batchesis greater than 25)

For each measurement system involving samplingand analysis, the FNMC plan should indicate (1)

how many samples are taken and measured for eachaccountability batch measurement, and (2) howmany analyses are performed on each accountabilitysample. If two or more samples are used and oneor more analyses per sample are performed for eachaccountability batch measurement, replicaterequirements are automatically met. If, however,one sample per batch is normally used foraccountability purposes, the replicate programshould include a periodic taking of a second (i.e.,replicate) sample. For NDA and mass (weight)measurement systems, replicate dst. can beobtained either from the repeat measurements onproduction items or by using the data generatedfrom the control standard program. That is, eachconsecutive pair of control standard measurements(for a given NDA or mass system) can be regardedas a replicate pair. The minimum number ofreplicate measurements performed during aninventory period for a given key NDA or masssystem should be as given in items (1), (2), or (3),above, except that the numbers or percentages arein terms of items measured, rather than batchessampled. Likewise, for non-key NDA and massmeasurement systems, the minimum number ofreplicate measurements should be as given in items(4), (5), or (6) above. The scatter in the repeatmeasurements is used to estimate the random errorvariance using a statistical technique known as theone-way analysis of variance. (The statisticalmethods described in Statistical Methods forNuclear Material Management, NUREG/CR-4604,are recommended by NRC for satisfying thestatistical requirements of 10 CFR 74.31; see alsoChapter 4 of this document.) Replication not onlyimproves the precision of results obtained from thestatistical analysis of the measurement data, it alsocan detect gross errors in the data.

3.5 Control Limits

Both warning and out-of-control limits are to beestablished and used for control standard measure-ments for those measurement systems used fornuclear material accountability. Out-of-controllimits are also to be utilized for replicatemeasurements and measurement of replicatesamples. However, warning limits are optional forthe replicate program. For point-calibratedsystems, the assigned value of the standard(s)measured along with the unknown(s) is assumed tobe valid. If the standard's true value could changebecause of factors such as evaporation, moisturepickup, or oxidation, the value of the standardshould be checked periodically. Therefore, controllimits for the verification measurements associated


Measurement Control

with such standards should be established. This isespecially true for those point-calibrated systemsthat use a single standard, or aliquots from a singlestandard, over any extended period of time.

The warning and out-of-control limits are nornnallyset by the licensee based on a tradeoff between (1)the cost of investigating and resolving incidentswhere limits are exceeded and (2) the cost ofaccepting measurements of poor quality. Warninglimits set at the 0.05 level of significance and out-of-control limits set at the 0.001 level ofsignificance are usually sufficient. When a systemgenerates a control measurement that falls beyondan out-of-control limit, the system should not beused for accounting purposes until it has beenbrought back into control (i.e., within the upperand lower warning limit).

Control limits should be recalculated at apredetermined frequency and modified if required.The FNMC plan should clearly explain how controllimits are established and the frequency forredetermining them.

3.5.1 Measurement Control DataAnalysis

Measurement control data such as control standardmeasurement results and the-differences betweenmeasurement values of replicate pairs should beplotted manually on graphs or entered into acomputer data base for generation of control charts.All control charts should be reviewed at least onceevery two weeks uiless a measurement system wasnot used during that period. The review shouldassess the frequency of control data exceedingeither the warning or the out-of-control limits andalso provide an evaluation for, any significanttrends.

3.5.2 Response Actions

Either the analyst or the operator performing acontrol measurement or their supervisor shouldhave the responsibility for promptly reporting anycontrol measurement that exceeds an out-of-controllimit. Such reporting should be made to themeasurement control program manager (or his orher designee), who should have the responsibilityand authority to carry out or-direct the necessaryresponse and corrective actions.

Minimum response and minimum corrective actionrequirements should be clearly defined. Inaddition, the measurement control manager (or his

or her designee) should be responsible for, andhave the authority for, determining and executingadditional response and corrective actions asdeemed appropriate.

The minimum response to a reported incident of acontrol measurement exceeding an out-of-controllimit should consist of.the following:.

(1) verifying that the measurement system inquestion has been taken out of service withrespect to accountability measurements

(2) documenting the occurrence of the event

(3) performing at least two additional controlmeasurements

(4) performing additional control measurements,if results of item 3 do not show the system tobe back in control, using a different controlstandard or different replicate sample (asappropriate) or recalibrating the measurementsystem, or making any necessary system.-repairs

For those measurement systems that make asignificant contribution to the SEID, the response toan out-of-control condition also should include theremeasurement of any samples (or items) that weremeasured before the out-of-control condition butafter the last within-control measurement. Thevalidity of the previous' measurements can beestablished without a complete remeasurement of allthe samples (or items) involved if remeasurementon a last in, first out" basis is used.- That is, thelast sample (or item) measured before theout-of-control measurement, should be the first tobe remeasured, and continuing in reverse orderuntil two consecutive remeasurements are found tobe in agreement with their initial measurement atthe 95 percent confidence level.


A judgment as to whether the applicant's orlicensee's measurement control program, asdescribed and committed to in the FNMC plan, isacceptable will be based on (but not limited to) thefollowing criteria:

The description' of the measurement controlprogram showsJthat the measurement systemsthat are the key contributors to the totalmeasurement standard error will be routinelyand adequately monitored for both bias and

NUREG-1065 Rev. 223-

Measurement Control

random error. The standard deviations ofmeasurement systems are estimated fromreplicate data from measurements made inthe same manner as made routinely ontypical process samples and items. Ifstandard deviations are based on replicatedmeasurements of standards (for NDA ormass measurement systems), data is collectedthat demonstrates that the standard deviationestimates do not differ significantly fromthose based on replicated process itemmeasurements. [NOTE: For analyticalchemistry measurements, the combinedanalytical plus sampling random error mustbe derived from the measurement of replicateprocess samples rather than based onmultiple measurements of a standard.]

All reasonable and probable sources ofmeasurement error, such as the effects ofsampling, instruments, environmental factors,and variability between operator/analysts areincluded in the estimates for standarddeviations, either directly as experimentalvariables in an analysis of variance or bybeing included in the sample of measurementcontrol data from which the standarddeviations are determined. It must be shownthat it is reasonable to expect that themeasurement systems that could significantlyimpact the inventory difference will be incontrol (when used for accounting purposes)so that twice the total measurement standarderror will be within the limits stated in 10CFR 74.31(c)(4) when calculated at the timeof a physical inventory.

* Bias tests are made by measurements ofrepresentative control standards whoseassigned values are traceable to nationalmeasurement systems. As reasonablyachievable, the control standards shouldclosely resemble the unknowns to which themeasurement is applied, and themeasurement procedures and conditions ofmeasurement must closely resemble those oftypical measurements made on processunknowns.

* A record of bias estimates for each keymeasurement system (not defined as a bias-free, point-in-time calibrated system), asderived from control standard measurements,is maintained. The basis for determining thetime frame associated with each significantbias needs to be provided (so that thequantity of measured SNM to which that bias

applies can be determined). Bias corrections(expressed as both grams element and gramsisotope) are derived for each significant bias(based on the quantity of measured SNM towhich the bias applies). The FNMC planmust confirm that each significant bias iseither applied as corrections to items listed inthe accounting records (if the correction foran individual item is greater than therounding error for that item), or included inthe net bias correction to inventory difference(on line 7 of NRC Form 327).

* Schedules and frequencies of replicate andcontrol standard measurements are designedso that the estimates of standard deviationsand measurement biases will be based onmeasurement control data collected under thesame measurement circumstances and overthe same time span corresponding to that ofthe SNM accounting measurements to whichthe standard deviations and bias estimateswill be applied. The standard deviation andbias of each key measurement system shouldbe evaluated periodically. The frequency ofsuch evaluations should typically be at leastevery four months. When determining theaverage bias and standard deviation of aparticular measurement system for an inven-tory period, pooling of data from previousdeterminations may be used only if statisticaltests show that the standard deviations andbiases from prior determinations do not differsignificantly from that of the current period,and further provided the pooled data do notinclude any data enerated more than 24months prior to the current determination ofsuch bias or standard deviation.

* The effort expended by the licensee inmonitoring and controlling the bias andstandard deviations of each measurementsystem is shown to be consistent with itsimpact on inventory difference and the totalmeasurement standard error. The number ofdegrees of freedom for estimating themeasurement standard deviation may begraded according to its contribution to thetotal measurement standard error.

* Waming limits for a change in bias (for thosesystems that are not point-in-time calibrated,bias free) will be set at the 0.05 level ofsignificance (or tighter), unless adequatejustification for less stringent limits isprovided. Warning limits are optional,however, for the monitoring of replicate data


Measurement Control

(for standard deviations). If a control datumexceeds this limit, the individual responsiblefor the measurement control program will benotified (this normally should occur within72 hours) and a data review will be initiatedto find the cause, and corrective action willbe taken when appropriate. Such reviewsand corrective actions are completed anddocumented within two weeks.

Unless adequate justification for less stringentlimits is provided, out-of-control limits forboth standard deviation and bias are set at the0.001 level of significance for all keymeasurement systems, except that no biascontrol limits are needed for bias-free,point-in-time calibrated systems. If a controldatum exceeds this limit, the system inquestion shall not be used for MC&Apurposes until corrective action andresolution is completed, and the system isback in control within the upper and lowerboundaries of the warning limit. Likewise,any measurement values generated betweenthe last within control datum and theout-of-control datum shall not be used forMC&A purposes until their validity has beenconfirmed. Such confirmation can beaccomplished by remeasurement of theinvolved items or samples on a last in, firstout" basis until two consecutiveremeasurements are found to be in agreementwith their initial measurement at the 95percent confidence level. It should be notedthat other criteria for initiating correctiveaction relative to potentially out-of-controlmeasurement systems may be accepted whereit can be demonstrated that the licensee'scapability to meet the 0.25 percent of activeinventory limit for total MC&A measurementuncertainty (as specified in 10 CFR74.3 1(c)(4)) will not be jeopardized.

published measurement performance insimilar applications.

- The calculation of the total measurementstandard error is performed in accordancewith a recognized error propagation method.Such methods have been published by Jaech(1973), Lumb and Tingey (1981), and theIAEA (1977).

* The licensee will confirm that theaccountability measurements provided by acontractor are controlled by a measurementcontrol program, and that the licensee willconfirm that the contractor's measurementcontrol program is adequate by conductingaudit and assessment reviews of thecontractor's program at intervals not toexceed 24 months.

3.7 Affirmations

The four affirmation statements, given on page 6under section (3), relating to the measurementcontrol system capabilities required by 10 CFR74.31(c)(3) and (4), must be stated (withoutmodification) in the FNMC plan.

The approach used for bounding the totalmeasurement standard error for a typicalmaterial balance period meets the followingcriteria:

- All reasonable and probable sources ofmeasurement error affecting inventorydifference are included.

- Any assumed measurement standarddeviations are shown to be reasonable.They may be shown to be reasonableby comparison to either records of thelicensee's past performance data or to

NUREG-1065 Rev. 225

4 Statistics

4.1 Determination ofMeasurement Uncertainties

To achieve the objectives and capabilities of 10CFR 74.31, each licensee or applicant shouldinstitute a statistical program that evaluates theMC&A data to ensure that () the measurementdata are analyzed in a rigorous manner and (2)hypotheses concerning the status of the nuclearmaterial possessed are appropriately tested. NRCsponsored the development of a comprehensivereference that specifically addresses the statisticaltreatment of measurement control and accounting -

data. The statistical methods described in thisreference, Statistical Methods for Nuclear MaterialManagement," NUREG/CR-4604, are recom-mended by NRC for satisfying the statisticalrequirements of 10 CFR 74.31.

The FNMC plan should

(1) contain a detailed discussion of theprocedures and methodologies for estimatingmeasurement variance components

(2) discuss how biases are determined and howbias corrections are applied, including:

(a) how often biases are estimated-(b) how the bias' effect on the measured

quantity of material in an item isdetermined

(c) when and how bias corrections toitems are made

(d) how their effect on inventorydifference is determined

(e) when and how bias corrections areapplied to the ID

(3) describe the procedure and means fordetermining active inventory

(4) provide all relevant information regarding thedetermination of the SEID

(5) specify the nominal (typical) DQ, and themethod for determining the actual DQ for agiven material balance period. Also confirmthat DQ will not exceed 1.30 percent of thelarger of annual additions to or removalsfrom process, except when 1.30 percent ofadditions to or removals from process is lessthat 25 kilograms of U-235, in which casethe DQ need not be less than 25 kilograms ofU-235 [NOTE: For those facilities that donot utilize any chemical processing, DQ

should be limited to 0.90 percent of annualadditions to or removals from process, butneed not be less than 25 kilograms of U-235]

(6) specify the methodology for determining IDthreshold values to be used to provide a 90percent power of detecting a loss of a DQ, asrequired by 10 CFR 74.31 (c)(5). (SeeChapter 5 for additional information oninventory difference limits and responseactions.)

4.2 Determination of SEID

As defined in 10 CFR 74.4, the standard error ofinventory difference' means the standard deviationof an inventory difference that takes into account allmeasurement error contributions to the'componentsof the ID. For strategic SNM facilities,nonmeasurement contributors to the ID are not tobe included in the SEID calculation. However, forfacilities possessing only SNM of low strategicsigrificance, the NRC allows licensees to takelimited credit for nonmeasurement contributors. Itis not really possible to quantify thenonmeasurement contribution to' the uncertaintyassociated with a given ID, but 10 CFR 74.31licensees are permitted to assume that the totalnonmeasurement'contribution to SEID equals thetotal measurement error contribution. When.including only measurement uncertainty, SEID (for-either U or U-235, as applicable) can be expressedas follows:

kSEID = I E (Gj)2 {(a;),2 + (a;)r2 /n}I"2

i=lwhere

k = number of measurement systemsGi = total grams uranium (or U-235)

;measured during inventory-periodby measurement system i

(u1) = systematic-error standard deviationfor measurement system i

(a.), = random error'standard deviation formeasurement system i

n = number of batches (items) measured 'by measurement system i

When taking the maximum allowable credit fornonmeasurement contribution (which assumes a 1:1ratio of measurement to nonmeasurementcontributions),

kSEID = 2E (G1)2 ((or,). +(u),21n}r'2

i=l

NUREG-1065 Rev. 227.

Statistics

In theory, SEID provides the uncertainty, at the 67percent confidence level, of the ID.

The FNMC plan must provide all relevantinfornation regarding the determination of SEID.There should also be a comrnitment that at least twoindividuals independently verify the correctness ofthe SEID calculation for each total plant materialbalance. If the SEID value is calculated by acomputer, the verification by two or more personsinvolves a checking for correctness of the input dataused by the computer to calculate SEID.

4.3 Bias Corrections

From a statistical perspective, biases that are notstatistically significant (at the 95 percent confidencelevel) should never be applied as adjustments(corrections) to the accounting records. If onewishes to obtain the best estimate of the trueinventory difference value, it can be argued thatsuch insignificant biases should be applied as anon-accounting adjustments to the initiallycalculated inventory difference (as obtained fromthe ID equation: ID = BI + A - R - El). Suchpractice is not deemed necessary, however, formaterial balances pertaining to SNM of lowstrategic significance, and thus is purely optional.

For biases that are statistically significant (at the 95percent confidence level), it is common practice toadjust the accounting values for individual items ifthe bias effect (as grams element and gramsisotope) on the item is more than the rounding errorfor that item, and if less than the rounding error, toapply the bias as a non-accounting adjustment to theID. Under a well designed and well managedmeasurement control program, bias corrections tothe accounting records should seldom, if ever, benecessary under the above mentioned approach.Although the effect on an individual item from astatistically significant bias should be negligible, theeffect of that bias across hundreds or thousands ofitems (whose SNM values were derived from thebiased measurement system) could have a verysignificant impact on the ID value.

Nevertheless, in view of the very large quantity ofSNM (of low strategic significance) that is of asafeguards significance, NRC acceptance criteriadoes not normally call for applying bias correctionsto either the accounting records or as an adjustmentto ID unless the effect of a single significant bias orthe net sum of all significant biases is unusuallylarge.

As a minimum, to meet NRC acceptance criteria, abias correction for a single key measurementsystem should be considered 'significant," and thusapplied either as corrections to the accountingrecords or as an adjustment to the inventorydifference, if (1) such bias is statistically significantat the 95 percent confidence level, and (2) either orboth of the following are also true:

1. Applying the correction would cause theID to exceed its DT value, or

II. The bias is greater than 0.0100 percentrelative and also impacts the ID value bymore than 1,000 grams U-235.

Additionally, the net algebraic sum (expressed asgrams U-235) of all statistically significant (95percent confidence level) biases, from keymeasurement systems not defined as bias-free, thathave not been applied as a correction/adjustmentunder condition I and/or II, above, is considered tobe significant' and is to be applied as a netadjustment to the ID if either or both of thefollowing are true:

III. Applying such correction would cause theID to exceed its DT value, or

IV. The net correction impacts the ID value bymore than 5.00 percent of the licensee'sDQ, or 10,000 grams U-235, whichever islarger.


5 Physical Inventories


The principal method of confirming the presence ofSNM is to perform a physical inventory andcompare it to the book (record) inventory. If allSNM is included, the expected difference betweenthe book inventory and the physical inventory iszero plus or minus the measurement uncertaintyassociated with both the physical and bookinventories. In any actual case, the size of theestimated ID depends on measurement errors, aswell as various nonmeasurement contributors suchas recording errors, unmeasured losses, unmeasuredholdup, etc. The intent of 10 CFR 74.3 1(c)(5) is torequire licensees to perform annual physicalinventories so to confirm that a loss or diversion ofa safeguards significant quantity (i.e., a DQ) of lowstrategic SNM has not occurred.

5.1 General Description

The applicant or licensee should provide a generaldescription of how physical inventories of the plantwill be planned, conducted, assessed, and reported.

5.2 Organization, Procedures,and Schedules

The FNMC plan should explain the makeup andduties of the typical physical inventoryorganization. The individual having responsibilityfor the coordination of the physical inventory effortshould be identified by position title. The FNMCplan also should indicate how the preparation andmodification of inventory procedures are confrolled.

The FNMC plan should contain a definitivestatement that specific inventory instructions areprepared and issued for each physical inventory.

5.3 Typical Inventory Composition

The typical expected in-process inventory within theequipment for both uranium and U-235 at the timeof the physical inventory should be specified. Atypical composition of SNM as stored items at thetime of a physical inventory 'also should bepresented. Plants may be (but are not required tobe) divided into a number of MBAs and ICAs toreflect the functional activities as follows:

The FNMC plan should contain a definitive - - - Processing - An MBA in which occurs (1) routinestatement that physical inventory functions and transfers of nuclear material from one container toresponsibilities are reviewed comprehensively with - another, (2) changes in chemical assay, or (3)the involved individuals before the start of each changes in chemical or physical form. Variousphysical inventory, - measurements are required to define materials flows

- through the process and to perform physicalA book inventoiy listing, derived from the MC&A inventories so that periodic material balances can berecord system, should be generated just before the completed for the MBA. Because theseactual start of each physical inventory; such listing measurements have associated uncerainties, ashall include all SNM that the'records indicate' processing MBA will normally have a nonzero IDshould be possessed by the licensee at the inventory for each inventory. Of the total plant MBAs andcutoff time. - -ICAs, a relatively small number might be

The inventory differenc ad rprocessing MBAs. Examples are the'Me inventory difference and related inforniation decontamination and recovery operations; analyticalassociated with each physical inventory of low- laboratory; and material rebatching, blending, andenriched uranium (LEU) (i.e.,'-DOE/NRC material ' - sampling operations. Physical inventories for thecode 20-El and 20-E2) are to be reported to the decontamination and recovery operations are theNRC, pursuant to 74.17(a), on an NRC Form 327 'most complex and involve the most coordinationas the results of a physical inventory. In addition, and careful timing.it any matenal associated witn matenal coaes 2u-E.3and 20-E4 (i.e., high-enriched uranium), 70 - ' Storae'-ICAs in which all materials are within(U-233), 50 (Plutonium), or, 83 (Plutonium-238) is ' containers with measured values and are beingpossessed by the licensee; such material must also 2 stored for future processing or shipment. Somebe subject to the physical inventory,' and each ' . rminor sampling of containers can occur in a storagematerial code inventory difference and associated ' * ICA. Because nuclear materials in a storage ICAinformation must be reported on separate 'NRC 327 are primarily accounted for on an item basis, a trueForms.

NUREG-1065 Rev. 229

Physical Inventories

storage ICA typically will have a zero ID for eachinventory period when all items are accounted forand their integrity and previously documentedmeasured values are confirmed.

Receiving and shinpin - An ICA from whichmaterials are shipped or into which materials arereceived from off-site. This ICA type willnormally serve as an interim storage area and willsee more activity (i.e., changes in currentinventory) than the typical storage ICA. At somefacilities, sampling and rebatching of items mayoccur in this type of ICA.

5.4 Description of Typical ItemStrata

The FNMC plan should describe the expected itempopulation in terms of the following:

(1) type of item (i.e., stratum)

(2) expected range of the number of itemswithin each stratum

(3) the average uranium and U-235 contentof the items within each stratum

(4) the expected rate of item generation andconsumption for each stratum

5.5 Conducting PhysicalInventories

A description of the methodology, including cutoffand inventory mirimization procedures, should bepresented; and all measurements (includingsampling) should be identified. The FNMC planshould contain sufficient information to show howthe total in-process inventory for both uranium andU-235 is obtained.

The means for measuring or estimating residualprocess material (i.e., holdup) should be addressedin detail. The change or variation in such depositedholdup from one physical inventory to the nextshould also be discussed.

A detailed inventory notice should be prepared foreach physical inventory. The notice should beissued to all involved parties and should containinstructions that define the timing and performanceof various inventory steps and conditions underwhich the inventory is to be taken. Specificsampling points throughout the process and

instructions on data submission to the accountabilityorganization should be identified. The instructionsshould highlight any required deviation from normalinventory procedures contained in the plant'soperating procedures.

A description of the procedures and methodologiesassociated with performing physical inventoriesshould be provided in sufficient detail todemonstrate that valid physical inventories areconducted. Such description should include ageneral outline of the following:

(1) organization and separation of functions

(2) assignment of inventory teams and theirtraining in the use of uniform practices

(3) obtaining, verifying, and recording ofsource data

(4) control of inventory forms

(5) assurance that item counts verify thepresence of each item while preventing anyitem from being counted more than once

(6) implementation of cutoff and materialhandling procedures

Decontamination and recovery is a complexoperation involving the disassembly anddecontamination of failed pieces of processequipment, and recovery of uranium from varioustypes of scrap materials. The basic inventoryprocedure should involve establishing a cutoff ofmovement of materials into the area and processingall materials to a measurable form, such ascontainers of solution or oxide. Except for thedecontamination enclosure in which in-processsolutions are mixed, sampled, and measuredvolumetrically, the inventory process should involveemptying and flushing of process systems andpiping, which then could be measured using NDAtechniques to establish levels of residual holdup, ifsuch holdup is significant.

Special item storage and handling or tamper-indicating methods, which are used to ensure thatthe previously measured and recorded SNMcontent values can be used for inventory purposeswithout remeasurements, should also be described.In addition, the FNMC plan should provide adescription of how item identities are verified andhow tampering with the contents of items will bedetected or prevented.



Items that are not encapsulated, affixed withtamper-indicating seals, or otherwise protected toensure the validity of prior measurements needspecial attention. The basis for determining whichitems are to be measured at physical inventory timeand the justification of anyIproposed altematives tomeasurement of any SNM included in the inventoryshould be presented. If statistical sampling isproposed as an alternative method to 100 percentverification, the FNMC plan should describe thesampling plan. Such description should include:

(I) the method of segregating the types ofitems to be smpled (i.e., selected forremeasurement)

(2) the procedure for calculating the samplesize (i.e., the number of items) for eachstratum

(3) the parameter to be measured (e.g.,gross weight or total U-235 content)

(4) the quality of the measurement methodsused to verify original measurementvalues (for the parameter beingmeasured)

(5) the procedure for reconcilingdiscrepancies between original andremeasurement values, and forscheduling additional tests andremeasurements

(6) the basis for discarding an original 'SNMvalue and replacing it with aremeasurement value

One acceptable means for establishing the numberof items (to be randomly selected forremeasurement) from a giver stratum to give therequired 90 percent power of detecting a loss of adetection quantity is given by the followingequation:

n = N [ - (0.10)"!

where

n = number of itens to be remeasuredN total number of itens in the stratumx = maximum U-235 content per item

(kilograms)g = DQ = detection quantity (kilograms

U-235)

When using such a statistical sampling plan toconfirm the validity of prior measurements, theremeasurement value obtained for each item (amongthe n items remeasured) must be compared to itsoriginal value. If the difference for a given itemexceeds some pre-determined limit (usually threetimes the standard deviation of the measurement, or ,3 sigma), that item is designated as a defect". Toachieve the 90 percent power of detection capabilityfor detecting a loss of a DQ, there must be at leasta 90 percent probability that one or more defects'will be encountered among the items remeasuredacross all involved stata if an actual loss of a DQhas occurred. If,,across all strata, one or moredefects are encountered, a second set of n randomlyselected items (or all remaining items if n 2 0.5 N)from each stratum should be remeasured. If one ormore defects are encountered (across all item stata)while performing any second round of remeasure-ments, all unsealed and unencapsulated items notyet remeasured are to be remeasured. Any item,regardless of whether there are any defects, whoseremeasured value differs from its original measure-ment by more than two sigma (2a) should have itsaccounting value revised to reflect its remeasuredquantity.

The FNMC plan also should contain a definitivestatement that all items on ending inventory thathave not been previously measured are measuredfor inventory purposes.

The decision rationale for determining when theelement and isotope factors for items, objects, orcontainers are measured directly for inventory andwhen they may be based on other measurementsshould be presented in the FNMC plan. Forexample, if the U-235 contained in liquid wastebatches is derived by applying an averageenrichment factor to the measured uranium elementcontent, the rationale for such practice (as opposedto measuring each batch for both uranium andU-235 content) should be discussed, and the methodfor establishing the average enrichment factorshould be described.

If the content of items is established through priormeasurements and those items are sealed withtamper-indicating devices or access to them iscontrolled, the SNM quantity in those items may';be based on those measured values. Otherwise,verification of SNM content can be achieved byreweighing either (I) all items within a givenstratum or (2) randomly selected items from thestratum based on a statistical sampling plan. Astatistical sampling plan will not be acceptable ifthere is any likelihood of any significant change in

NUREG-1065 Rev. 231


the uranium concentration (or weight fraction) or inthe uranium isotopic distribution because of suchfactors as oxidation, change in moisture content,commingling with materials of differentenrichments, or different compositions.

5.6 Inventory Difference Limitsand Response Actions

Each licensee should have a well-defined system forevaluating total plant IDs and taking actions whenIDs exceed certain predetermined thresholds. As aminimum, there should be two response levels, andthree levels are recommended, for excessive IDs.The following would be an acceptable approach forthree increasing levels of response actions withrespect to physical inventories:

Warning-level IDU-235 ID 2 1.7(SEID) + 500 grams, andU ID 2 1.7 (SEID) + 10 kg U

Significant ID problemU or U-235 ID 2 3 (SEID)

Major ID problemU-235 ID 2 DQ - 1.3 (SEID)

All of the above limits are expressed in terms ofabsolute values of ID without regard for algebraicsign. The minimum response for a warning-levelID should be a documented licensee investigationconducted by the MC&A organization. Such aninvestigation should provide a conclusion for theprobable cause of the excessive ID and giverecommendations for avoiding recurrences. Whena warning-level ID is positive, it should be regardedas being equivalent to an indicator of a possible lossthat requires investigation and resolution.

For a significant ID problem, an extensiveinvestigation by the licensee should be conducted.If a significant ID problem cannot be satisfactorilyexplained, a reinventory may be needed.

For any unresolved ID determination that remains amajor ID problem (without regard to algebraicsign), the licensee should conduct a plant-widereinventory and investigation. The NRC considersa positive ID large enough to be a major IDproblem as a very serious condition.

The FNMC plan should fully describe in definitivestatements the minimum response actions for eachID action level.


A judgment that the applicant's or licensee's planfor physical inventories is acceptable will be basedon (but not linited to) the following criteria:

* The FNMC plan contains adequatecomnitments to assure that each physicalinventory will be organized and coordinatedso that all involved persons are instructed inthe use of uniform procedures of checkingSNM quantity and recording observations.The means for conducting the inventory mustinsure that no SNM currently on hand (withthe possible exception of holding accountwaste materials and residual holdup followingcleanout that can he regarded as zero bydefinition) will be omitted and no quantitywill be counted more than once.

* The inventory procedures provide forconfirming the presence of all items by directobservation and the presence of all quantitiesof SNM, that are neither encapsulated nortamper-safed, by direct measurement or anacceptable alternative. The proposedalternatives to remeasurement must satisfyone of the following criteria:

- The SNM content is verified bystatistical sampling and measurementof representative items, objects orsamples of the material. The totaloverall sampling plan shall support thecapability for detecting any loss inexcess of the current DQ with 90percent (or better) probability.

- The previous measurement results areaccepted because the items are storedin a controlled access enclosure thatprovides protection equivalent totamper-safing.

- Residual holdup (that is too large to beregarded as zero by definition)remaining after cleanout or draindownmay be estimated if the estimate isbased on previously measured values,and it is periodically verified orvalidated.

- For material whose SNM content hasbeen previously measured, and there isno likelihood of any significant change



in the uranium concentration (orweight fraction) or in the uraniumenrichment due to such factors asoxidation, change in moisture content,commingling with materials ofdifferent enrichment or differentcomposition, etc., the previouslydetermined uranium and U-235 contentmay be accepted without verificationof SNM content provided the grossweight and/or net weight of all itemswithin the population is confirmed by(1) a 100 percent reweighing of allsuch items, or (2) reweighing anadequate number of randomly selecteditems (based on a statistical samplingplan) to provide a 90 percent (orbetter) probability of detecting a lossequal to or greater than the currentDQ.

* As an additional altemative to remeasurement(of unsealed SNM) at physical inventorytime, a program of routine processmonitoring will be acceptable when thecombination of the process monitoringprogram and the inventory procedures willachieve the same level of loss detectioncapability as that provided by a physicalinventory in which all unencapsulated itemsare either tamper-safed or remeasured.

* Any previously measured, but unsealed (orunencapsulated) SNM that is on hand at thetime of the physical inventory, and which isto be introduced into subsequent processingsteps prior to inventory reconciliation shouldbe remeasured or have its prior measurementvalue confirmed (by an acceptablealternative) before the subsequent processingis initiated.

* The DQ (in kilograms U-235) for any giveninventory period will be no greater than 1.30percent of that period's throughput forfacilities involved in chemical processing(such as UF6 conversion, scrap recovery,oxidation or reduction processes, etc.), andno greater than 0.90 percent of throughputfor facilities where material only undergoesphysical changes (such as pressing UO2powder into pellets and/or loading pelletsinto fuel rods), unless the 0.90 or 1.30percent of throughput (as appropriate) is lessthan 25 kilograms U-235, in which case theDQ need not be less than 25 kilograms. TheU-235 throughput is defined as the greater of

.additional to process" or -removals fromprocess" during a 12-month inventory period.

* The information in the FNMC plan showsthat the DT for an excessive ID will result ina 90 percent (or better) probability ofdetecting a discrepancy (i.e., an apparentgain or loss) equal to or larger than theU-235 DQ for the inventory period inquestion. In general, a licensee may assumethe ID distribution approximates a normaldistribution, and therefore:

DT = DQ - 1.30(SEID)

Acceptable methodology for calculating themeasurement error contribution to the SEIDby error propagation is found in Jaech(1973), IAEA (1977), and Lumb and Tingey(1981). Special attention is given toinclusion of all applicable and measurablesources of error to avoid underestimating theSEID.

* In addition to the DT ID alarm limit[required by 10 CFR 74.31(c)(5)], there is atleast one excessive ID waming level limitthat when exceeded will require aninvestigative response action. The resourcesand level of effort to be committed to theinvestigation of an excessive ID will beproportional to the magnitude of the ID, butwill be sufficient to reassess the results of thephysical inventory, the accounting records,and the measurement control program data;to confirm the relevant calculations and dataanalysis; and when necessary, to carry outsearches for unmeasured inventory such asholdup and discards. Investigations are to becompleted within 60 days after initiating theinventory (except when additional time isgranted by the NRC for extenuatingcircumstances).

5.8 Affirmations

The 11 affirmation statements, given on pages 6and 7 under section (5), and the first performanceobjective affirmation statement listed on page 5,related to physical inventories must be stated(without any modification) in the FNMC plan.

NUREG-1065 Rev. 233

6 Item Control


The intent of 10 CFR 74.31(c)(6) is to requirelicensees to implement item control procedures thatprotect against unauthorized and unrecordedremoval of items, or of material from items, andthat enable timely location of items. Items areknown quantities of SNM in well-defined anduniquely identified containment such as cans,drums, and canisters, or fixed units such as fuelassemblies. Un-containerized solid SNM, such asuranium metal ingots or buttons, are also items ifthey are uniquely identified. In order. to promptlylocate a given item, sufficient current informationmust be recorded. These item control requirementsapply only to items for which the total length oftime that the item exists plus the length of time ittakes to locate it (after,having been recorded)exceeds 14 days. For example, if an item has beenin existence for 10 days, the licensee must becapable of making a record of the item and locatingthe item within 4 days in order for the item to beexempt from the current knowledge requirement.Conversely, if the time to make a record of an itemplus the time it takes to locate the item equals 4days, then if the item exists for less than 10 days itneed not be formally entered into the currentknowledge system. Also in order to eliminate theneed to keep records for insignificant quantities ofSNM, waste containers and most solutions areexempt. Licensees are given some flexibility incontrolling items by permitting an additional -

exemption of up to a total of 50 kilograms U-235 initems each containing less than 500 grams ofU-235.

i 1 flwnon

(1) items whose time of existence plus the timeto record the item (since its generation) plusthe time to locate the item is less than 14calendar days

(2) any licensee-identified items listed bymaterial type containing less than 500 gramsof U-235 each but not to exceed a plant totalof 50 kilograms of U-235

(3) containers of waste (designated for burial orincineration), and solution containers inwhich the U-235 concentration is less than5.00 grams per liter.

Each item that is not exempt from the item controlprogram should be stored and handled in a mannerthat enables detection of, and provides protectionagainst, unauthorized or unrecorded removals ofSNM. All items, whether or not they are subject toitem control program coverage, should have aunique identity. For items subject to the itemcontrol program, the following are acceptablemeans for providing unique identity:

(1) a unique alpha-numeric identification on atamper-indicating seal applied to a containerof SNM

(2) a unique alpha-numeric identificationpermanently inscribed, embossed, or stampedon the container or item itself

(3) a uniquely prenumbered (or bar-coded) labelapplied to each item having good adhesive.qualities such that its removal from an item

-. would preclude its reuse

Location designations shown by the MC&A recordsThe FNMC plan sbould identify the individual need not be unique, but location designations shouldresponsible for overseeing the item control program be specific enough so that any item may be locatedby position title. Positions of those individuals who within I hour. Longer times may be acceptable buthave significant item control program should be further justified in the FNMC plan. Theresponsibilities' should also be identified.-'' ' MC&A record system should be controlled in such

a manner that the record of an item's existence

6.2 General Description - - . cannot be destroyed or falsified without a highprobability of detection., Each nonexempt item -

- should be stored and handled in a manner that.The applicant or licensee should state that the enables detection of, and provides protectionoverall MC&A system maintains a record of all against, unauthorized or unrecorded removals ofSNM items, regardless of quantity or duration of SNM.existence. In addition, the item control programshould provide current knowledge of the location, - 6.3 Item Identity Controlsidentity, and quantity of all SNM contained in allitems hat are' not exempt from item control. terns ;-'-; : itesthat ae otexemp t fromim iotmronrol.ram - Descriptions should be provided of the item recordsthtcanerae exempt:from item control p showing how items are identified for each material

type and each type of container. If the unique

NUREG-1065 Rev. 235

Item Control

number on a tamper-indicating seal is the basis forproviding unique item identity, the FNMC planshould:

(1) describe the type of seal used

(2) describe how the seals are obtained and whatmeasures are implemented to ensure thatduplicate (counterfeit) seals are notmanufactured

(3) describe how the seals are stored, controlled,issued, and accounted for

(4) describe how seal usage and disposal recordsare maintained and controlled

Similar information should be provided for othermethods of unique item identity (e.g., labels).

6.4 Storage Controls

Item storage areas and controls should be fullydescribed in the FNMC plan. In particular,controls that are used as the basis for ensuring thevalues of prior measurements, as opposed toremeasuring the item at inventory time, should bediscussed in detail and the rationale for acceptingprior measurements explained. Any controls usedto ensure the validity of prior measurements shouldbe equivalent to the protection provided bytamper-indicating seals.

Both administrative controls (e.g., custodianassignments and limiting authorized access tostorage areas) and physical controls (e.g., lockedand/or alarmed doors) should be identified.

6.5 Item Monitoring Methodologyand Procedures

As part of the item control program, a licenseeshould maintain a system of item monitoring that:

(1) verifies that items shown in the MC&Arecords are actually stored and identified inthe manner indicated in the records

(2) verifies that generated items and changes initem locations are properly recorded in theMC&A record system in a timely manner

(3) can detect, with high probability, any realloss of items, or uranium from itemsamounting to 500 grams or more of U-235

The item monitoring system should conduct thefollowing activities at least on a monthly basis:

(1) for each item inventory stratum, compare theactual storage status to the recorded status ofa sufficient sample of randomly selecteditems from the item control program records

(2) for each item inventory stratum, check theaccuracy of the MC&A records for asufficient sample of randomly selected itemsfrom each storage area

(3) check the accuracy of a sufficient sample ofrandomly selected production records ofcreated and consumed items

The actual frequency of the above activities, andthe size of the random sample, should be a functionof the expected discrepancy rate based on priorobservations. The FNMC plan should specify (1)minimum monitoring frequencies associated witheach storage area, (2) discrepancy rates that triggermore frequent monitoring frequencies, and (3)comnitments for resolving discrepancies.

6.6 Investigation and Resolutionof Item Discrepancies

The applicant or licensee should provide definitivestatements of the procedures and controls that willensure that all incidents involving missing orcompromised items or falsified item records will beinvestigated. A compromised item is () onedisplaying evidence of tampering or (2) anunencapsulated and unsealed item assigned to acontrolled, limited-access storage area that is foundelsewhere.

If any unencapsulated and unsealed item is locatedafter having been determined to be missing, or if anitem is found to be compromised, its contentsshould be reestablished by measurement (e.g., byNDA or by weighing, sampling, and analysis).Recommendations on resolution of indicators(Chapter 9) of this document should be used toresolve item discrepancies.


A judgment that the licensee's FNMC plan for thecontrol of items subject to the requirements of10 CFR 74.31(c)(6) is acceptable will be based on(but not linited to) the following criteria:


Item Control

* The licensee's item record system uniquelyidentifies items. The records includeinformation on the chemical form, quantityof material (element and isotope), physicaldescription, identification label or number,and location. The system providesreasonable assurance of detecting falsificationor destruction of records of an item'sexistence. Groups of itens that areproduced, stored, processed, or otherwisehandled together as a unit, such as a batch orsublot of material, may be exempt from therequirements for identification of each item ifthe group is uniquely identified and stored asa separate group under conditions such thatgroup identity, composition and quantity willbe maintained constant.

* The record of the status of an item can becompleted or updated in sufficient time toallow the licensee to meet the requirementsfor promptly locating an item.

* For items that will not be remeasured atinventory time, the item control proceduresprovide reasonable assurance that the SNMcontents stated in the records are valid andthat unauthorized removal of SNM from theitem has not occurred. Remeasurement isnot necessary if the SNM content of the itemwas measured previously and reasonableassurance is provided that the SNM contenthas not subsequently changed.

- Solutions having less than a 5.00grams U-235 per liter concentration;or

- Items individually containing less than500 grams U-235 each, not to exceeda total of 50 kilograms of U-235.

• The FNMC plan makes commitments fordetermining the total quantity of U-235 initems exempted from control on the basis thatthey contain less than 500 grams U-235 each,so as to assure that the total 50 kilogramU-235 limit is not exceeded.

* A current accounting is maintained of thetotal quantity of SNM contained in items thatare exempted from item control. Theaccounts identify the quantities by materialtype category for both controlled andexempted items.

6.8 Affirmations

The four affirmation statements, given on pages 7and 8 under section (6), relating to the currentknowledge requirements of 10 CFR 74.31(c)(6)must be stated (without modification) in the FNMCplan.

* Items that are exempted from item controlprocedures must fall into at least one of thefollowing categories:

- Items that are in existence for only arelatively short period of time, asdefined in 10 CFR 74.31(c)(6);

- Waste destined for burial orincineration;

NUREG-1065 Rev. 237

7 Shipper-Receiver Comparisons


The intent of 10 CFR 74.31 (c)(7) is to require thematerial control and accounting system to promptly'detect and resolve all significant shipper-receiver(SIR) differences. SIR comparisons are importantfor confirming that either shippers' and receivers'values are acceptable for establishing the bookaccounting quantities associated with receivedmaterial, or detecting unacceptable shippers' orreceivers' values.

7.1 Receiving Procedures

Regardless of whether or not formal S/Revaluations are to be performed, the first action tobe taken on receipt of SNM should be theverification of the number of items, the itemidentities, and the integrity of individual items andof tamper-indicating seals. Except for certainspecified types of material, all SNM shipmentsreceived from an external supplier are to be subjectto SIR comparisons. Such SIR comparisons involvemeasurement'of received material by the receiver,or by the receiver's contractor (who is independentof the shipper), and comparing receiver's totalreceipt measurement for element and isotope to thatof the shipper's.'' The FNMC plan should identifywhat material types, if any, are exempted from S/Revaluations. Receipt materials which may beexempted from SIR comparisons include:

(1) shipments containing less than 500 gramsU-235;

(2) individual items containing less than 25grams U-235 when the total shipment is lessthan 2000 grams U-235;

(3) encapsulated items, whose encapsulationintegrity has not been compromised, andwhich are to be retained by the licensee asencapsulated items;

(4) fuel assemblies and fuel rods previouslyshipped by the licensee that are beingreturned, provided that the originalencapsulation has not been compromised;

(5) UF6 cylinders that are empty except for aheel quantity of UF6 ; and

(6) Heterogeneous scrap that must be subject todissolution before a meaningful accountabilitymeasurement can be obtained, and bothshipper and receiver agree to accept the after

dissolution plus residue" measurements foraccounting purposes.

For any SNM received, the licensee must provideall appropriate information on the DOE/NRC Form741 that accompanies'the shipment. (NOTE: SeeNUREG/BR-0006 for instructions and requirementsfor completing DOE/NRC Form 741.)

7.2 Determination of Receiver'sValues

For UF6 , the licensee may establish receiver'svalues by (1) measuring the U-235 isotopicconcentration by NDA, (2) weighing each cylinderto determiine net weight UF6 (gross weight minuscertified cylinder tare weight), and (3) applying ahistorical established percent uranium factor (that isperiodically confirmed or updated by measurement).Such a practice avoids the need to sample UF6cylinders and analyze the samples (except whenobtaining data to' establish or update the historicalpercent U factor). However, it is more desirableand common practice to derive UF6 receipt valuesbased on the measurement of UF6 samples.Arrangements can usually be made with the UF6supplier to provide samples of the shipped UF6 .The receiving facility should have an agent (who isindependent of the shipper) witness the sampling(that represents the UF6 contained in the cylinder(s)to be shipped), apply a tamper-indicating seal to thesample, and verify the unique identify of the filledcylinder.

SNM 'receipts not in the form of UF6 should bemeasured for total quantity (mass), elementconcentration, and isotope abundance, as opposedto using a historical factor for deriving elementcontent.

7.3 Evaluation of Shipper-Receiver Differences

When shipper's measurement uncertainty (orstandard error) information is available, thefollowing should define the combined measurementstandard error:

combined standard error = (4x.)2 + (R) 21 12

where

as = shipper's measurement standard erroraR = receiver's measurement standard error

NUREG-1065 Rev. 239

Shipper-Receiver Comparisons

If the shipper's measurement uncertainty values arenot available, the receiver can assume that theshipper's measurement uncertainty is equal to (butno greater than) its own uncertainty. In thissituation (i.e., both shipper and receiver have thesame measurement uncertainty), the followingbecomes the combined measurement standard error:

combined standard error = [ 2 (R) 21 11= 1.4 14 aR

The difference between the shipper's value and thereceiver's value (i.e., the SRD), in terms of thetotal shipment, must be regarded as significantwhenever the SRD exceeds both 500 grams ofU-235 and twice the combined standard error. Ifsubject to 10 CFR Part 75 (internationalsafeguards), a SRD in excess of both 500 gramsU-235 and twice the combined standard error withrespect to a single batch within the shipment mustalso be regarded as significant.

7.4 Resolution of SignificantShipper-Receiver Differences

The FNMC plan should describe the steps involvedwith the investigation of a significant SRD anddiscuss how such difference is resolved. Thecriteria for defining a resolved SRD also should bepresented. Generally, resolution of a significantSRD involves a referee (or umpire) measurement ofa retainer sample(s) but not of the material weight.The resolution process should specify whose weightvalue is used in the resolution process if shipper'sand receiver's weights differ by more than one-halfof the total combined standard error.

significant SRDs is acceptable will be based on (butnot limited to) the following criteria:

* Each shipping container is inspected within 3working days after receipt. If the integrity ofa container is questionable, the presence ofall items that were packaged in the shippingcontainer are confirmed within 24 hours ofdiscovering the questionable integrity.

* Confirmatory measurements of scrapshipments are performed by the receiver todetermine the amount of element and isotopewithin a time period consistent with theaccountability needs of the shipper.

* The test for significance of a SRD is basedon hypothesis tests.

* The investigation procedure for significantSRDs is sufficiently comprehensive to ensurethat the difference will be resolved.Comprehensiveness is sufficient if thelicensee shows the capability to verifyrecords, resample, perform remeasurements,establish liaison with the shipper, providesamples to a referee laboratory, and performthe statistical analysis needed to evaluate themeasurements.

7.6 Affirmations

The six affirmation statements, given on page 8under section (7), relating to the resolution ofsignificant shipper-receiver differences, as requiredby 10 CFR 74.31(c)(7), must be stated (withoutmodification) in the FNMC plan.


A judgment that the applicant's licensee's FNMCplan for conducting S/R evaluations and resolving


8 Assessment and Review of the Material Controland Accounting Program


The intent of 10 CFR 74.31(c)(8) is to requiremanagement to periodically review and assess theperformance of the MC&A system arid evaluate itsoverall effectiveness. It is intended that the reviewwill be performed by knowledgeable, technicallycompetent individuals free from conflicts of interestand that the deficiencies will be brought to theattention of plant management so that thedeficiencies will be corrected. It should beemphasized that this review process is'intended tobe much more than a routine audit for compliancewith existing procedures and commitments.Conclusions and recommendations relative tooverall program effectiveness and to the adequacyof the program to meet and satisfy regulatory intentare to be made.

8.1 General Description

The capabilities,-performance, and overalleffectiveness of the licensee's MC&A programshould be independently reviewed and assessed atleast every 24 months. The FNMC plan shoulddescribe the assessment and review program'interms of:

(1) maximum interval between assessments

(2) selection procedures for the assessment team

(3) number of team members to be selected

(4)

(5)

qualification and expertise of team members

independence of individual team membersfrom the MC&A responsibilities andactivities they are reviewing and assessing

(6) maximum elapsed time and minimum actualeffort to be used for completion of theassessment and issuance of a final teamreport

The entire MC&A program generally should bereviewed and evaluated during each assessment.When this occurs, intervals between assessmentscan be as much as 24'calendar months. However,if individual assessments only cover part of theMC&A system, individual subsystems should be

assessed at intervals no greater than 12 calendarmonths. Thus, the type of assessment (partial ortotal) and the maximum interval betweenassessments should be specified. "Interval' meansthe elapsed time between either the start of or-termination of successive assessments.

The responsibility and authority for the assessmentprogram should lie at least one level higher in thelicensee's organizational structure than that of theMC&A manager.. Such responsibility shouldinclude selecting the assessment team leader andinitiating corrective actions. Team members maybe selected from the facility staff or from outside,but an individual member should not participate inthe assessment of the parts of the MC&A systemfor which that person has direct responsibility.Hence, the MC&A manager may not be a teammember. Also, a given individual should not assessthe parts of the system that are the responsibility ofanother team member if the other team member isassessing the given individual's area. The leader of-the assessment team should have no responsibilitiesfor managing any of the MC&A elements beingassessed.

The minimum number of individuals on any givenassessment should be dependent on the knowledgeand expertise of the team relative to MC&Aactivities and their experience in conductingassessments. Personnel assigned to the assessmentteam should have a demonstrated understanding ofthe regulatory objectives and requirements of theMC&A program and should have sufficientknowledge and experience to be able to judge theadequacy of the parts of the system they review.The team should have authority to investigate allaspects of the MC&A system and should be givenaccess to all necessary information.

To provide a meaningful and timely assessment, thereview and evaluation process should not beprotracted. The actual review and investigationactivities should .be completed in 30 calendar days,with an additional 15 calendar days allowed forcompleting and issuing a final team report.

NUREG-1065 Rev. 2

.

41

Assessment

8.2 Report of Findings andRecommendations

The areas to be reviewed should encompass theentire MC&A system, and the level of detail of thereviews should be sufficient to ensure that theassessment team has adequate information to makereasoned judgments of the MC&A systemeffectiveness. The team report, as a minimum,should state findings pertaining to:

(1) organizational effectiveness to manage andexecute MC&A activities

(2) management responsiveness to indications ofpossible losses of uranium

(3) staff training and competency to carry outMC&A functions

(4) reliability and accuracy of accountabilitymeasurements made on SNM

(5) effectiveness of the measurement controlprogram in monitoring measurement systemsand its sufficiency to meet the requirementsfor controlling and estimating both bias andSEID

(6) soundness of the material accounting records

(7) effectiveness of the item control program totrack and to provide current knowledge ofitems

(8) capability to promptly locate items andeffectiveness in doing so

(9) timeliness and effectiveness of SRDevaluations and resolution of excessive SRDs

(10) soundness and effectiveness of the inventory-taking procedures

(11) capability to confirm the presence of SNM

(12) capability to detect and resolve indications ofmissing uranium

On completion of each assessment, the findings andrecommendations for corrective action, if any,should be documented. The written report shouldbe distributed to the plant manager, the MC&Amanager, and other managers affected by theassessment.

8.3 Management Review andResponse to Report Findingsand Recommendations

Management should review the assessment reportand take the necessary actions to correct MC&Asystem deficiencies. The management reviewshould be documented within 30 days following thesubmittal of the assessment team's report, and itshould include a schedule for the correction ofdeficiencies. Corrective actions, if any, that pertainto daily or weekly activities should be initiatedpromptly after the submittal of the final assessmentreport.

The FNMC plan should address resolution andfollow-up actions associated with concernsidentified in the assessment report. The individualsresponsible for resolving identified concerns, andthe timeliness of such resolution, should bespecified.


A judgment that the applicant's or licensee'sprogram for independent assessment of the MC&Aprogram is acceptable will be based on (but notlimited to) the following criteria:

* The periodic assessments will becomprehensive and sufficiently detailed toenable the assessment team to rate theMC&A system effectiveness, capability, andperformance by comparison with theexpected and required performance. Theoverall assessment objectives are todetermine that the MC&A system, asdesigned and implemented, is continuing tomeet the overall safeguards goals, and toidentify weaknesses or deficiencies in thesystem design or performance that may needcorrecting.

* The areas to be reviewed encompass theentire MC&A program and the level of detailof the reviews is sufficient to ensure that theassessment team has adequate information tomake reasoned judgments of the MC&Asystem effectiveness which includes:

- Organizational effectiveness andmanagement responsiveness toindicators of possible SNM losses;


Assessment

- Staff training and competency to carryout MC&A functions;

- Soundness of the material accountingrecords;

- Capability to promptly locate items;

- Timeliness and effectiveness of S/Rdifference evaluations and resolution ofsignificant SRDs;

- Soundness of physical inventoryprocedures and practices;

- Effectiveness of the measurementcontrol program to monitor keymeasurement systems, establish biasestimates and measurementuncertainties, and meet therequirements for controlling the totalMC&A measurement uncertaintyassociated with ID;

- Capability to confirm the presence ofSNM; and

- Capability to resolve indications ofmissing SNM and aid in anygovernment led investigation pertainingto missing SNM, and provideinformation that would aid in therecovery of missing SNM.

* Generally accepted auditing principles areused to check each type of record in which arepresentative sample (of a sufficient number)of randomly selected records is examined.

* Personnel assigned to the assessment teamwill have an adequate understanding of theregulatory objectives and requirements of theMC&A system and will have sufficientknowledge and experience to be able to judgethe adequacy of the parts of the system theyare asked to review. The team will haveauthority to investigate any aspect of theMC&A program and will have access to allrelevant information.

-B" is assessing an area under theresponsibility of 'A".

* The entire MC&A system will be reviewedand evaluated during each single assessment(to be completed within an elapsed time thatis short relative to the time between changesin the MC&A system and is demonstrated tobe able to include any such changes madeduring the review/assessment). Conductingtwo or more assessments during a 24-monthinterval, in which only part of the MC&Asystem is covered in each, is not deemedacceptable. That is, reviewing a singlecomponent of the MC&A program at24-month intervals is not very meaningfulunless a knowledge is obtained of how wellthe other components currently interact.Piecemeal review and evaluation ispermissible, however, if each subsystem iscovered at 12-month (or less) intervals.

* The leader of the assessment team will haveno responsibilities for performing ormanaging the functions being assessed.

* The responsibility and authority for theassessment program and for initiatingcorrective actions will lie: (1) at least onelevel higher in the organization than theMC&A manager, or (2) at a level equal tothat of the on-site plant manager.

8.5 Afirmations

The seven affirmation statements, given on pages 8and 9 under section (8), relating to the independentassessment and evaluation of the MC&A systemeffectiveness requirements of 10 CFR 74.31(c)(8)must be stated (without modification) in the FNMCplan.

* An individual team member will notparticipate in the assessment of any part ofthe MC&A system for which he or she hasdirect responsibility. Also, an individual"A" will not assess any part of the systemthat is the responsibility of person "B" if

NUREG-1065 Rev. 243

9 Resolving Indications of Missing Uranium


The intent of 10 CFR 74.31(a)(2) is for licensees toreact to (i.e., investigate and resolve) anyoff-normal or abnormal conditions or situations thatsuggest a likelihood that 500 grams or more ofU-235 may be missing (whether the cause isaccidental or deliberate).

9.1 Methods and Procedures forIdentifying Indicators

The FNMC plan should discuss the means by whichthe licensee will resolve indicators'of missinguranium involving 500 grams or more of U-235.The licensee's resolution program should addressthe possible indicators of missing uranium. The'FNMC plan should enumerate all 'the potentialindicators that can be postulated and developresolution procedures for each.

The following are examples of possible indicatorsof missing uranium:

(1) lack of agreement between a physicalinventory and its associated book inventoryin which the U-235 ID is positive andexceeds 1.7 times the SEID by more than500 grams

(2) determination through the item controlprogram that one or more items are not intheir designated locations and the actuallocations are not immediately known

(3) discovery that an item's integrity or itstamper-indicating seal was compromised

(4) information from the process control systemindicating potential loss of material from theprocess system.

(5) an allegation of theft or diversion

9.2 System and Procedures forInvestigating and ResolvingIndicators

At least one major MC&A procedure shouldaddress the system and practices for investigatingand resolving loss indicators because this topicpertains to one of the three performance objectivesof 10 CFR 74.31(a). Thus, the licensee shouldhave well-defined procedures for investigating

indicators of possible missing uranium (involving500 grams or more of U-235). Likewise, thereshould be established criteria for defining whatconstitutes resolution of an investigated indicator.

Resolution of an indicator means that the licenseehas made a determination that loss or theft has notoccurred and is not occurring. For each type ofindicator, the licensee should'develop detailedresolution procedures and should describe or outlinethem in the FNMC plan.

Any investigation of an indication of a loss or theftshould provide, whenever possible, () an estimateof the quantity of SNM involved, (2) the materialtype or physical form of the material, (3) the typeof unauthorized activity or event detected, (4) thetime frame within which the loss or activity couldhave occurred, (5) the most probable cause(s), and(6) recommendations for precluding reoccurrence.

For indications that a loss or theft of more than 500grams of U-235 may have occurred, the resolutionprocess should include (1) thoroughly checking theaccountability records and source information, (2)locating the source of the problem, (3) isolating theexact reason for the problem within the area orprocessing unit, (4) determining the amounts ofSNM involved, and (5) making a determinationthat the indication is or is not resolved. Theresolution procedures should be prepared in such amanner that no individual that could have beenresponsible for the potential loss also would beresponsible for its resolution. If an investigation ofan indicator results in a conclusion that theindication is true, such conclusion must be reportedto the NRC within I hour of its determinationpursuant to 10 CFR 74.11. The FNMC planshould show the reasonable time allowed forresolution. In general, a time not exceeding 72hours should be adequate.

9.3 Response Actions forUnresolved Indicators

Response actions to unresolved indicators should beclearly defined and should be on a graded scaleappropriate to the level of potential safeguardssignificance. The responsibility and authority forinitiating and executing response actions also shouldbe defined.

For indicators of missing uranium, the level ofsafeguards concern is related to such factors as:

(1) the potential quantity of U-235 involved

NUREG-1065 Rev. 245

Resolving Indications of Missing Uranium

(2) the material attractiveness of the potentialmissing uranium (in terms of fabricating anuclear explosive device) relative to itsenrichment and composition or form (e.g.,U metal, U308, uranyl nitrate solution, UF6,scrap, or waste)

9.4 Documentation Requirements

The FNMC plan should identify all documentationrequirements associated with the licensee's programfor the reporting, investigation, and resolution ofmissing uranium indicators. Review and approvalrequirements and document custodial responsibilityalso should be defined. As a minimum,documentation of the following should be included:

(1) investigation procedures

(2) resolution procedures

individual responsible for resolving theindicator.

No investigation relative to an indication ofa loss or theft of SNM exceeding the currentDQ shall be declared as completed butunresolved without first conducting ashutdown, cleanout inventory in which allunsealed SNM is remeasured for elementand isotope.

9.6 Affirmations

The three affirmation statements pertaining to the10 CFR 74.31(a)(2) performance objective, listed asbullets two, four, and five in the section at the topof page 5 ("Affirmations Pertaining to PerformanceObe 'ctives"), must be stated (without modification)in the FNMC plan.

(3) reporting of indicator to MC&Amanagement including date and time theindicator was reported, name of individualwho discovered the indicator, anddescription of indication

(4) investigation findings and conclusion,including resolution status, date issued,name and signature of principal investigator,and approval signature of MC&A manager

(5) reports made to NRC for unresolvedindicators and for indicators determined tobe real, including date and time the reportwas made, method of communication, andname of NRC individual contacted


A judgment that the applicant's or the licensee'splan for resolving indications of missing SNM isacceptable will be based on (but not limited to) thefollowing criteria:

Adequate commitments are provided to assure ahigh probability that any indicator of missing SNMthat could involve 500 grams or more of U-235 willbe (1) recognized as an indicator, (2) investigated,and (3) resolved.

* Investigation and resolution procedures willprovide for adequate overchecks to assurethat no individual who could have beenresponsible for a possible loss or theft ofSNM would be the sole or primary


10 Informational Aid for Assisting in the Investigation andRecovery of Missing Uranium


The 10 CFR 74.31(a)(3) performance objectivepertains to investigations, relating to actual (orhighly suspected) events pertaining to missinguranium, conducted by the NRC and/or othergovernment agencies. The intent is for licensees tohave ready for and to provide to the investigatorsany information deemed relevant to the recovery ofmaterial involved in a loss, or theft. The burdenshall be on the licensee to provide (without beingasked to) all information that it recognizes as beingrelevant, as opposed to only providing informationthat the investigators are knowledgeable enough torequest.

10.1 Information Aid

The kinds of information that may aid theinvestigation and recovery effort are:

(1) data or observations that led the licensee todetermine that a loss or theft of uraniummay have occurred

(2) data, observations, and assessmentsassociated with attempts to resolve theindication of missing material

(3) the time period during which the materialmay have left the facility

(4) the path and means by which the materialmay have left the facility

Information indicating that a loss of uranium mayhave occurred can come from process monitoringor production yield data, physical inventoryresults, item monitoring audits, and shipper-receiver comparisons. This information couldinclude:

(1) material accountability data records andreports

(2) inventory records

(3) inventory difference and propagation oferror calculations

(4) inventory reconciliation reports

(5) indications of unrecorded or unauthorizedremovals of SNM from storage or processlocations

(6) reports of apparent destruction orfalsification of records pertaining to SNM

(7) records of broken tamper-indicating devicesor compromised item integrity

(8) indications of unauthorized entry into SNMstorage areas

(9) Reports from monthly item statusinspections

(10) material receipt and log-in records

(11) results from shipper-receiver differenceevaluations

(12) process quality assurance or productioncontrol records

(13) documentation relating to an alleged orconfirmed theft

Information associated with resolving indications ofmissing uranium are provided in Chapter 9. Thisinfornation and inforrnation that may be of aid inthe recovery of missing material would include:

(1) the type of unauthorized activity detected

(2) the interval during which the loss may haveoccurred

(3) the amount of material and form of the material involved in the loss -

(4) results of measures to validate indicators

(5) results of extended measures to resolveindicators

(6) results from special inventories (orreinventories) and tests performed

(7) audit results of the SNM accountabilitysource data

(8) assessments of measurement data andmeasurement controls

(9) results from reviews of the material controland accounting program and status ofcorrective actions

(10) history of indicator investigation andresolution activities

NUREG-1065 Rev. 247

Assisting Investigations

(11) indicator investigation and resolutionprocedures and conclusions

(12) probable cause of the loss

(13) any abnormal events that may havecontributed to or caused the loss

(14) the names of individuals who could havebeen responsible for the loss

Much of the backup information necessary to assistin an investigation would be records maintained inthe facility records system described in Chapter 11.


A judgment that the applicant's or the licensee'splan for informational aid for assisting in theinvestigation and recovery of rmissing uranium isacceptable will be based on (but not limited to) thefollowing criterion:

Procedures are in place for the efficient andtimely gathering of relevant information toprovide government investigators so as toaid them in the investigation and recoveryactivities associated with missing SNM.

10.3 Affirmations

The affirmation statement pertaining to the 10 CFR74.31(a)(3) performance objective listed as the thirdbullet in the Affirmations Pertaining toPerformance Objectives' section on page 5, mustbe stated (without modification) in the FNMC plan.


11 Recordkeeping


The intent of 10 CFR 74.31(d) is to require theestablishment, maintenance, and protection of arecordkeeping system that will demonstrate that thesystem capability requirements of 10 CFR74.31(c)(1) through (8) have been met. Recordsare to be retained for at least three years (or longerif specifically required by regulations external to10 CFR 74.31) thereby providing a means forassessing the performance of the MC&A systemand inspecting for compliance with regulatoryrequirements.

11.1 Description of Records

The FNMC plan should identify all records, forms,reports, and standard operating procedures thatmust be retained for a minimum of 3 years, asrequired by 10 CFR 74.31(d). Such records shouldinclude, but are not limited to, the following:

(1) documents that define changes in the MC&Amanagement structure or changes inresponsibilities relating to MC&A positions;

(2) procedures pertaining to any accountabilityrelated measurement or sampling operation

(3) forms used to record or to reportmeasurement data and measurement results,including source data

(4) forms and notebooks used to recordcalibration data'associated with anyaccountability measurement system

(5) forms and notebooks used to recordquantities, volumes, and Iother dataassociated with the preparation of standards,both calibration and control, used in'connection witb accountability measurementsystems

(6) forms and official memos used to'record orreport measurement control program data,control limit calculations, and out-of-controlinvestigations

(7) forms listing and providing instructionsassociated with physical inventories

(8) forms and formal worksheets used in thecalculation of SEID, ID, and active'inventory values

(9) ledgers, journals, and computer printoutsheets associated with the accountabilitysystem

(10) ledgers, journals, and computer printoutsheets associated with the item controlprogram, including seal usage and attestingto' records

(11) DOE/NRC forms 741 and 742

(12) forms, memos, and reports associated withidentification of, investigation of, andresolution of significant SRDs

(13) loss indication and alleged theft investigationreports

(14) investigation reports pertaining to excessiveIDs

(15) official reports containing the findings andrecommendations of MC&A systemassessments and any letters or memospertaining to response actions to assessmentteam recommendations

(16) forms used for recording data associatedwith the item monitoring program

(17) monitoring program status or summaryreports

(18) records of training sessions including dategiven, topics covered, name of instructoroo.names and signatures of those attending

(19) training, qualification, and requalificationreports and records

Examples of the more important MC&A formsshould be provided in the FNMC plan annex orappendix. The retained records and reports shouldcontain sufficient detail to enable NRC inspectors todetermine that the licensee has implemented thesystem features and capabilities of 10 CFR 74.3 1(c)and has met the general performance'objectives of10 CFR 74.31(a).

NUREG-1065 Rev. 249

Rc.ordkeeping

11.2 Program and Controlsfor Ensuring an Accurateand Reliable RecordSystem

The FNMC plan should describe the controls usedto ensure that records are highly accurate andreliable.

The record system also should provide a capabilityfor easy traceability of all SNM transactions fromsource data to final accounting records.

The following topics should be addressed:

(1) the auditing system or program to verify thecorrectness and completeness of records

(2) the overchecks for preventing or detectingmissing or falsified data and records

(3) the plan for reconstructing lost or destroyedSNM records

(4) the access controls used to ensure that onlyauthorized persons can update and correctrecords

(5) the protection and redundance of the recordsystem such that any act of record alterationor destruction will not eliminate the abilityto provide a complete and correct set ofSNM control and accounting informationneeded to achieve the performanceobjectives of 10 CFR 74.31.


A judgment that the licensee's commitments forrecordkeeping are acceptable will be based on (butnot limited to) the following criteria:

* The FNMC plan shows that key materialaccounting and original source datadocuments and relevant reports anddocuments will be retained for three yearsor as long as needed to show continuingcompliance with 10 CFR Part 74. (Forexample, records of the organizationstructure cannot be destroyed if the currentstructure is still the same.)

* The source data will be retained in itsoriginal form until the physical inventoryand any subsequent ID investigations have

been completed. After this time, anyreadable facsimile is acceptable for theremainder of the required retention period.All other records may be retained as hardcopy, microfiche, permanent computerreadable forms, or other permanentlyreadable forms.

* The retained records and reports containsufficient detail to enable NRC inspectors todetermine that the control and accounting ofthe SNM has complied with therequirements of 10 CFR 74.31(c) and hasmet the general performance objectives of10 CFR 74.31(a).

* The records will be retrievable, sufficientlycomplete and detailed to permit auditing allparts of the MC&A system, and traceableback to original source data.

* The records of the data that are the basis ofthe calculated SEID will perrit traceabilityto the sources of the variances due tocalibrations, bias adjustments and randomeffects in the measurements. These recordsmay be summaries of calibrations, bias tests,and variance monitoring data or controlcharts.

* The record system will have sufficientredundancy to enable reconstruction of lostor missing records so that knowledge of theSNM inventory is always available. Theprimary records, as contrasted withduplicate or backup records, will beprovided security against computer failure,fire or water damage, vandalism, and accessby unauthorized persons.

* All retained MC&A records are to bereadily accessible, in order to meet timerestraints relative to their use. In general,the record retention system is to possess thecapability to retrieve records used formeasurement control or accountabilitywithin 24 hours if the record was generatedwithin the past 12 months, and within 7calendar days if generated more than 12months ago. Physical inventory listings areto be available within 24 hours for the latesttwo physical inventories. Item controlrecords are to be retrievable in time tosatisfy the criteria in Section 6.7.

* Overchecks or other controls includingaccess controls for updating and correcting


Recordkeeping

records are provided so as to prevent ordetect errors in the records that would affectinventory difference and item location.

11.4 Affirmations

The affirmation statements relating to therecordkeeping requirements of 10 CFR 74.31(d) aslisted on pages 9 and 10, must be stated (withoutmodification) in the FNMC plan.

NUREG-1065 Rev. 251

Glossary

Glossary

The following terms are defmed in the context of(1) their usage in this document and/or (2) howthey should be used if contained in the fundamentalnuclear material control (FNMC) plans submittedpursuant to 10 CFR 74.31.

ACCOUNTABILITY - The determination of, andcurrent record maintenance of, special nuclearmaterial (SNM) quantities associated with (1)receipts, (2) shipments, (3) measured discards, (4)transfers into or between material balance areas(MBAs) and/or item control areas (ICAs), and (5)total material on current inventory.

ACCOUNTANCY (ACCOUNTING) - Therecords (e.g., ledgers, journals, source documents,etc.) associated with accountability.

ACIVE INVENTORY - The sum of beginninginventory (BI), additions to inventory (A), removalsfrom inventory (R), and ending inventory (El) afterall common terms have been totally excluded. Acommon term is any SNM material value'(or item)that appears in both BI and EI, or both BI and R,or both A and R, or both A and EI,with bothvalues derived from the'same measurement (andthus, does not contribute to the uncertaintyassociated with the current period inventorydifference). The active inventory is used as'anindicator of processing throughput and/or

BEGINNING INVENTORY '-For each materialtype code, the total itemized quantity of SNMpossessed by a facility at the start of a materialbalance period (i.e., inventory period). -Thebeginning inventory quantity for any given materialbalance period is (by definition) exactly equal to theending inventory quantity for the immediatelyprevious period.

BIAS (MEASUREMENT BIAS) A constant,unidirectional component of error that affects allmembers of a measurement data set. Thus, a biascan be estimated from the deviation of the mean ofseveral measurements of a representative standardfrom the reference value (or assigned value) of suchstandard.

CALIBRATION - The process of determining thenumerical relationship between the observed outputof a measurement system and the actual value ofthe characteristic being measured, as based onprimary or reference standards.

CERTIFIED STANDARD - A standard weight,material, device, or instrument having an assignedvalue that is guaranteed to be within specified limitsby a nationally or internationally recognizedorganization (e.g., bureau, laboratory, etc.) that'issues and/or certifies standards.

processing activity.CHECK STANDARD (BENCH STANDARD,

ADDITIONS TO INVENTORY - Quantities of WORKING STANDARD) - A standard, notSNM, of a given material type code, added to a necessarily traceable to a primary standard, that is.plant' inventory and which, prior to such addition, used routinely (e.g., daily or weekly) to check (orwere not part of the plant's total possessed quantity verify) the reliability of a measurement device,for the material type code in question. - instrument, etc. (including those of accountability

measurement systems). Such standards are not,ARTIFACT STANDARD - A container or item, however, used for the actual calibration or controlof certified mass, having a size, shape; and mass of accountability measurement systems.that is representative of a particular type of process-related item or container (e.g., a UF6 cylinder).' COMBINED STANDARD ERROR - An errorWeighing error caused by buoyancy is eliminated '`- --band derived from the respective standard errorby the use of artifact standards for scale i' values associated with each of two measurementscalibrations. (usually independent of each other) performed on a

given material quantity. For both measurementASSIGNED VALUE - A value for rmass, volume, values (of the pair) to be regarded as being inSNM concentration, SNM quantity, etc., assigned agreement, they must not differ from each other byto a standard weight, standard material, etc.:; used more than the calculated combined standard error,for calibrating and/or controlling a measurement' which is normally calculated by taking the squaredevice or system. An assigned value may not root of the sum of squared individual standardnecessarily be a certified value, but if not, should -errors. That is:be traceable to a certified standard. In any event, itis the best estimate of the standard's true value. combined S.E. = [(,)2 + (or)21 'h


Glossary

CONFIRMATORY MEASUREMENT - Ameasurement that confirms (within measurementuncertainty at the 95 percent confidence level) apreviously established parameter, such as netweight, enrichment, etc., associated with an SNMitem (or SNM quantity), but which does notthoroughly verify the previously established elementand/or isotope quantity assigned to such item.Confirmatory measurements are sometimes used asthe basis for concluding that previous measurementvalues for uranium and U-235 (or element andisotope) quantities are still valid.

CONTROL STANDARD - A standard that (1) isrepresentative of the process material beingmeasured and (2) is itself measured periodically tomonitor for and to estimate any bias associated withthe measurements of the process material inquestion. A control standard must be traceable to aprimary standard or to a primary referencematerial.

CUSTODIAN - A designated individual who isresponsible for (1) the control and movement of allSNM within a specified control area, and(2) maintaining records relative to all SNM that istransferred into or out of the area and that iscurrently located within the control area. Controlareas are usually designated as MBAs or ICAs.From the standpoint of good safeguards practice, asingle individual should not be a custodian of morethan one control area.

DEPLETED URANIUM - Any uranium-bearingmaterial whose combined U-233 plus U-235isotopic content is less than 0.70 wt % (relative tototal uranium elemental content).

DETECTION QUANTITY (DQ) - A site-specificU-235 quantity of SNM for licensees whoseprocessing activities are limited to SNM of lowstrategic significance. The DQ is normally afunction of annual throughput, but for low-throughput low-enriched uranium (LEU) facilities,the DQ need not be less than 25 kilograms ofU-235. The DQ also can be described as a goalquantity, the loss or theft of which must be detectedwith a 90 percent (or better) probability at the timeof a physical inventory.

DETECTION THRESHOLD (DT) - An inventorydifference (ID) limit that will be exceeded (with90 percent or higher probability) by an ID resultingfrom the taking of a physical inventory wheneverthere has been an actual loss of a detectionquantity. The DT is a function of both the DQ and

the standard error of the inventory difference(SEID), as shown by the following equation:

DT = DQ - 1.3 (SEID)

EFFECTIVE KILOGRAM - An effectivekilogram of SNM means (1) for plutonium andU-233, their weight in kilograms; (2) for uraniumwith an enrichment in the isotope U-235 of1.00 wt % (i.e., 0.01 g U-235/g U) and above, itselement weight in kilograms multiplied by thesquare of its enrichment expressed as a decimalweight fraction; and (3) for uranium with anenrichment in U-235 below 1.00 wt %, but above0.71 wt %, its element weight in kilogramsmultiplied by 0.0001.

ENDING INVENTORY - For each material typecode, the total itemized quantity of SNM possessedby a facility at the end of a material balance period,as determined by a physical inventory. The endinginventory quantity for any given material balanceperiod is (by definition) exactly equal to thebeginning inventory quantity for the next period.

ENRICHED URANIUM - Any uranium-bearingmaterial that does not qualify as natural uraniumand whose combined U-233 plus U-235 isotopiccontent is 0.72 wt % or higher relative to totaluranium elemental content.

FORMULA KILOGRAM (FKG)- The number offormula kilograms is computed by the followingequation:

FKG =[(grams U-235 contained in high-enricheduranium) + 2.5 (grams U-233 +plutonium)]divided by 1000

FORMULA QUANTITY - Strategic SNM in anycombination in a quantity of 5.00 FKGs or more.

HIGH-ENRICHED URANIUM (HEU) - Anyuranium-bearing material whose U-235 isotopiccontent is 20.00 wt or more relative to totaluranium elemental content.

INVENTORY DIFFERENCE (ID) -The

arithmetic difference between a book inventory andthe corresponding physical inventory, calculated bysubtracting ending inventory (El) plus removalsfrom inventory (R) from beginning inventory (BI)plus additions to inventory (A). Mathematically,this can be expressed as:

ID = (BI + A) - (El + R)


Glossary

INVENTORY RECONCILIATION - The adjustment of the book record quantity of bothelement and fissile isotopes to reflect the results ofa physical inventory. In the broad sense, inventoryreconciliation also includes the activities ofcalculating (1) the ID for the material balanceperiod in question, (2) the uncertainty (i.e., SEID)value associated with the ID, (3) the activeinventory for the period, and (4) any biasadjustment and/or prior period adjustmentassociated with the ID value.

ITEM - Any discrete quantity or container ofSNM, not undergoing processing, having a uniqueidentity and also having assigned uranium and U-235 quantities.

ITEM CONTROL AREA (ICA) - An identifiablephysical area for the storage and control of SNMitems. Control of items moving into or out of anICA is by item identity and SNM quantity asdetermined from previous measurement.

ITEM CONTROL PROGRAM -A system thattracks (i.e., records) the creation, identity, location,and disposition of all SNM items of certainpredetermined item categories. In addition, itemcontrol programs usually provide a periodicverification of item existence and location for staticitems.

LOW-ENRICHED URANIUM (LEU) -. Anyuranium-bearing material whose U-235 isotopiccontent is greater than 0.72 wt % but less than20.00 wt % relative to total uranium elementalcontent.

MATERIAL BALANCE- A comparison on ameasured basis of beginning inventory plus,additions to inventory to ending inventory plus re-movals from inventory for a given'control area (orcombination of control areas) over a specifiedperiod of time.

MATERIAL BALANCE AREA (MBA) - Anidentifiable physical area for- the physical andadministrative control of nuclear material such thatthe quantity of nuclear material being moved into orout of the MBA is represented by a measured value(for both element and isotope).:

MATERIAL BALANCE PERIOD - The timespan to which a material balance or physicalinventory pertains.

MATERIAL TYPE CODES -Number codes foridentifying basic material types with respect to SM,

SNM and by-product materials. These codes areused by the Nuclear Materials Management andSafeguards System (NMMSS) for tracking U.S.-owned and U.S.-possessed materials worldwide.For SNM, seven material type codeshave beenassigned as follows:

CODE10205070838889

MATERIAL TYPEDepleted UraniumEnriched Uranium ()PlutoniumUranium-233 (**)Plutonium-238 (***)ThoriumUranium in Cascades

'For DOEINRC Form 742, naterial code 20 has 4subcodes to denote enrichment range: El, E2, E3, and E4. ForNRC Form 327. material code 20 has 2 subcodes: LEU andHEU.

**Uranium naterials should be regarded as materialcode 70 if the U-233 isotopic abundance is greater than (1)10.00 wt % relative to total uranium elemental content or (2)both (a) the U-233 isotopic abundance is greater than the U-235isotopic abundance and (b) the U-233 isotopic abundanceexceeds 5.00 wt % relative to total elemental uranium content;otherwise report as material code 10, 20, or 81. as appropriate.

**Plutonium materials should be regarded as materialcode 83 if the plutonium-238 isotopic abundance is greater than10.00 wt % relative to total plutonium elemental content;otherwise, report as material code 50.

MEASURED DISCARD - A batch or quantity ofwaste, whose SNM content has been determined bymeasurement, that (I) has been shipped to a,disposal site, released to the environment, or storedon site, and (2) has been taken off the accountingledgers as part of the current inventory of possessedSNM.

MEASUREMENT - The process of determining a(1) uranium elemental concentration, (2) specificuranium isotopic content, (3) U-235 enrichment(i.e., isotopic abundance), (4) bulk material mass oritem mass, or (5) bulk material volume.Measurement values are derived through acalibration process that establishes the relationshipbetween instrument (i.e., device) response and theparameter being determined.-

MEASUREMENT CONTROL PROGRAM - Amanaged program for monitoring and controllingboth accuracy and precision of SNM accountabilitymeasurements.

NATURAL URANIUM - Any uranium-bearingmaterial whose uranium isotopic distribution has not

= been altered from its naturally occurring state.Natural uranium is nominally 99.283 wt % U-238,0.711 wt % U-235, and 0.006 wt % U-234.


Glossary

NORMAL URANIUM - Any uranium-bearingmaterial having a uranium isotopic distribution thatcan be characterized as being (1) 0.700 wt % to0.724 wt % in combined U-233 plus U-235 and (2)at least 99.200 wt % in U-238. (NOTE: Allnatural uranium having a U-235 isotopic abundancein the range of 0.700 wt % to 0.724 wt % isnormal uranium, but not all normal uranium isnatural uranium.)

PHYSICAL INVENTORY - A determination byphysical means (visual and measurement) of thequantity of SNM on hand for a given material typecode at a specified point in time. The primarypurpose for a physical inventory is to confirm theabsence of (or to detect) a loss, theft, or diversionof SNM. For the conclusion drawn from a physicalinventory to be meaningful, not only does thephysical presence of all material (or at least the vastmajority of the material) need to be confirmed, butthe quantities of material in the inventory also needto be remeasured, or other assurance provided, toverify that prior measurements are still valid.Physical inventories are to be conducted on a.plant' as well as on a material type code' basis.

PLANT - For SNM control and accountingpurposes, a plant is defined as a set of processes oroperations (on the same site, but not necessarily allin the same building) coordinated into a singlemanufacturing, research and development, ortesting effort. Most licensees have only one plantin this context. Independent on-site manufacturingefforts, such as a scrap recovery operation servingboth on-site and off-site customers, should betreated as separate plants.

POINT-CALIBRATED MEASUREMENTSYSTEM - A measurement system in which themeasurement value assigned to an unknownmeasured by the system is derived from theresponse obtained from the measurement of arepresentative calibration standard(s) that wasmeasured along with (i.e., at the same time as) theunknown. The standard(s) must undergo all themeasurement steps (e.g., aliquoting, samplepretreatment, etc.), and in the same manner, as theunknown. Point-calibrated measurement systemscan be regarded as bias free, provided that adequatecontrols are in place to ensure the validity of thestandard's assigned value.

PRIMARY STANDARD - Any device or materialhaving a characteristic or parameter (such as mass,uranium concentration, uranium isotopic distri-bution, etc.) whose value is certified (within aspecified uncertainty) by a nationally or

intemationally recognized bureau, laboratory, etc.,that issues and/or certifies standards.

PRIOR PERIOD ADJUSTMENT - Anycorrection (i.e., adjustment) to an ID value becauseof a correction applied to a component of beginninginventory after the inventory period started. Suchcorrections may be due to resolution of a shipper-receiver difference (SRD) on material receivedduring a prior inventory period, correction of arecording error, etc. Because these types ofcorrections have nothing to do with current periodlosses or errors, and because the official beginninginventory value is not adjusted, an adjustment to theID value (derived from the ID equation) isnecessary to obtain an ID that reflects only currentperiod activity.

PROCESS MONITORING - A system ofmonitoring production data (e.g., flow rates, yields,densities, etc.) and of production control or qualitycontrol measurements (as opposed to accountabilitymeasurements) that could provide early (i.e.,timely) detection of an anomaly that may indicate asignificant loss or theft of SNM or indicateunauthorized enrichment activities.

RANDOM ERROR - The variation encountered inall measurement work, characterized by the randomoccurrence of both positive and negative deviationsfrom a mean value.

RECEIPT - A quantity of SNM in a shipmentreceived by a facility from another facility.

REFERENCE STANDARD - A material, device,or instrument whose assigned value is traceable to anational standard (i.e., primary standard) or nation-ally accepted measurement system.

REMOVALS FROM INVENTORY - Allmeasured quantities of SNM falling within thecategories of (1) shipments, (2) measured discardsreleased to the environment, (3) measured discardstransported off-site, and (4) measured discardsstored on-site and formally transferred in theaccounting records to a holding account via aDOE/NRC Form 741 transaction.

RESOLUTION OF AN INDICATOR - Adefinitive determination (with auditable evidence)by the licensee that an indicated possible theft orloss of uranium was a false indicator.

SHIPPER-RECEIVER DIFFERENCE (SRD) -

The difference between what a sending facility(i.e., shipper) claims was contained in a shipment


Glossary

(of SNM) and what the receiving facility claimswas received, where both shipper's and receiver'svalues are based on measurement.

SPECIAL NUCLEAR MATERIAL (SNM) - (1)Plutonium, U-233, uranium enriched in U-235, andany other material that the NRC, pursuant to theprovisions of Section 51 of the Atomic Energy Actof 1954 (as amended), determines to be SNM; or(2) any material artificially enriched in any of theforegoing. There are three levels of strategicsignificance applied to SNM, depending on the typeand quantity, defined as follows:

SNM OF HIGH STRATEGIC SIGNIFICANCE- Same as FORMULA QUANTITY (seedefinition).

SNM OF MODERATE STRATEGICSIGNIFICANCE - (1) Less than a formulaquantity, but more than 1,000 grams of U-235contained in HEU, or more than 500 grams ofU-233 or plutonium, or more than a combinedquantity of 1,000 formula grams when formulagrams are computed by the following equation:

formula grams = (grams U-235 in HEU)+ 2 (grams U-233 + grams Pu)

or (2) 10,000 grams or more of U-235 containedin LEU enriched to 10.00 wt % or more, butless than 20.00 wt %, U-235.

SNM OF LOW STRATEGIC SIGNIFICANCE- (1) Less than an amount of SNM of moderatestrategic significance, but more than 15 grams of(a) U-235 contained in HEU, (b) U-233,(c) plutonium, or (d) any combination thereof;(2) less than 10,000 grans but more than 1,000grams of U-235 contained in LEU enriched to10.00 wt % or more (but less than 20.00 wt %)U-235; or (3) 10,000 grams or more of U-235contained in LEU enriched above natural butless than 10.00 wt % U-235.

STANDARD - See definitions for CERTIFIEDSTANDARD, CHECK STANDARD, CONTROLSTANDARD, PRIMARY STANDARD, andREFERENCE STANDARD.

STANDARD DEVIATION - The random error(at the 67 percent confidence level) associated witha single value of a data set, which in tum is also ameasure (or indication) of the precision relating to aset of measurements (or set of data) pertaining tothe same item or sample of material. Standarddeviation is calculated as follows:

| E ( ~- )std. dev. = I ,.-

n-I

]h

where

n = number of measurements performedxi= the value obtained for the i th

measurement for i 1, 2, 3 ....... nx = the average value for all n

measurements

STANDARD ERROR - The random error (at the67 percent confidence level) associated with theaverage, or mean, value of a data set derived fromrepetitive determinations on the same item orsample. Mathematically, standard error is thestandard deviation divided by the square root of thenumber of individual measurements used to derivethe mean value.

STANDARD ERROR OF THE INVENTORYDIFFERENCE (SEID) - For licensees subject to10 CFR 74.31 or 74.33, non-measurementcontributions to the uncertainty of ID can beassumed or estimated, but can not exceed themeasurement contribution. Thus for 74.31 and74.33 licensees, SEID is equal to the square root ofthe sum of both measurement and non-measurementvariances associated with the ID.

STANDARD REFERENCE MATERIAL - Amaterial or substance that qualifies as a primarystandard and whose concentration with respect to anuclide or isotope, a chemical element, or chemicalcompound is certified within a specifieduncertainty.

SYSTEMATIC ERROR - A unidirectional errorthat affects all members of a data set. The terms'bias' and systematic error' are ofteninterchanged. However, any determined bias (i.e.,a bias estimated from control standardmeasurements) has an uncertainty value associatedwith it. Thus, after correcting for any estimatedbias, the uncertainty of that bias can be regarded asa systematic error. If an estimated bias is notapplied as a correction, the combination of the biasplus its uncertainty should be regarded as thesystematic error.

NUREG-1065 Rev. 257

Glossary

TRACEABILITY - The ability to relate individualmeasurement results to national standards (i.e.,primary standards) or nationally acceptedmeasurement systems through an unbroken chain ofcomparisons.

VERIFICATION MEASUREMENT - (I) AnNDA measurement of an item conducted to verifythat a previous NDA measurement value for isotopecontent of that item is still valid. (2) There-weighing and re-sampling of an item, batch, lot,or sublot and performing chemical assays of there-sample for element and isotope concentrations soas to verify a previously measured value forelement and isotope content of the item (batch, lot,or sublot). Verification is achieved if the originaland verification measurement values (for elementand isotope quantities) agree within the range ofmeasurement uncertainty (at the 95 percentconfidence level).


NRC FORM 335 U.S. NUCLEAR REGULATORY CO.MMISSION 1. REPORT NUMBER(2-ag) - ~~~~~~~~~~~~~~~~~~~~~~~~(Assignedl by NRC, Add Vol.,NRCM1102 Supp. * Rev..* and AddenidumT Num-3201, 3202 BIBLIOGRAPHIC DATA SHEET bers, If an

(See Instructlons on the reverse) NUREG-1065Rev. 2

2. TITLE AND SUBTITLE

Acceptable Standard Format and Content for the Fundamental Nuclear Material 3. DATE REPORT PUBUSHED

Control (FNMC) Plan Required for Low-Enriched Uranium Facilities MONTH I YEAR

December 19954. FIN OR GRANT NUMBER

5. AUTHOR(S) 6. TYPE OF REPORT

D. R. Joy Technical7. PERIOD COVERED (Inclusive Dates)

8. PERFORMING ORGANIZATION - NAME AND ADDRESS (f NRC, provide Division. Office or Region, U.S. Nuclear Regulatory Commission, andmailing address; It contractor, provide name and mailing address.)

Division of Fuel Cycle Safety and SafeguardsOffice of Nuclear Material Safety and SafeguardsU.S. Nuclear Regulatory CommissionWashington, DC 20555-0001

9. SPONSORING ORGANIZATION - NAME AND ADDRESS (If NRC. type 'Same as above'; If contractor, provide NRC Division, Office or Region.U.S. Nuclear Regulatory Commission, and mailing address.)

Same as item #8

10. SUPPLEMENTARY NOTES

11. ABSTRACT (200 words or less)

This report documents a standard format suggested by the NRC for use in preparing fundamental nuclear materialcontrol (FNMC) plans as required by the Low Enriched Uranium Reform Amendments (1OCFR 74.31). This reportalso describes the necessary contents of a comprehensive plan and provides example acceptance criteria which areintended to commjnicate acceptable means of achieving tbe performance capabilities of the Reform Amendments.By using the suggested format, the licensee or applicant will minimize administrative problems associated with thesubmittal, review and approval of the FNMC plan. Preparation of the plan in accordance with this format will assistthe NRC in evaluating the plan and in standardizing the review and licensing process. However, conformance withthis guidance is not required by the NRC. A license applicant who employs a format that provides a equal level ofcompleteness and detail may use their own format. This document is also intended for providing guidance to licens-ees when making revisions to their FNMC plan.

12. KEY WORDS/DESCRIPTORS (List words or phrases that will assist researchers In locating the report.) 13. AVAILABILITY STATEMENT

Unlimited

Reform Amendments 14. SECURITY CLASSIFICATION

Low Enriched Uranium (This Page)

LEU Fuel Fabrication UnclassifiedMaterial Control and Accounting (This Report)

Special Nuclear Material UnclassifiedSNM of Low Strategic Significance 15. NUMBER OF PAGES

16. PRiCE

NRC FORM 335 (2-89)

Federal Recycling Program