ML20205J197

Matl Control & Accounting Insp Rept 70-1100/85-08 on 851209-13.Violation Noted:Failure to Maintain tamper-safe Records,Per Fundamental Nuclear Matl Control Plan

ML20205J197
Person / Time
Site:	07001100
Issue date:	01/21/1986
From:	Della Ratta A, Huff D, Keimig R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20205J166	List: ML20205J162 ML20205J179 ML20205J197
References
70-1100-85-08, 70-1100-85-8, NUDOCS 8601300119
Download: ML20205J197 (4)

Text

r. - . .,

tA i

U.S. NUCLEAR REGULATORY COMMISSION REGION I' Report.No. 70-1100/85-08 Docket No. 70-1100 '

License No. SNM-1067 Safeguards Group: III' .

l s Licensee: Combustion En'gineering, Inc.

P .O. 4 Box 500 Windsor, Conne'eticut 06095 l

l Facility Name: Combustion Engineering - Nuclear Fuel Manufacturing Inspection At: . Windsor, Connecticut Inspection Conducted: December 9-13, 1985 Date of Last Material Control and Accounting Inspection: November 4-8, 1985 l Type of Inspection: Announced, Material Control and Accounting Inspectors: NdI/&

A. Della Ratta, Safeguards Auditor f ,h -f(

date b2[ . /-J'-4 Op. Huff,Kteri fs$htrol and Accounting date Specialist, Saf uVds and Materials Programs Branch, E l Approved by: .

f;:=E/ /-J * - 8 l R. . Keimig Chief afeguards Section date l uclear Materials afety and ,

l Safeguards Branch Inspection Summary: Inspection on December 9-13, 1985 (Report Number 70-1T00/85-08)

Areas Inspected: Nuclear. material control and accounting, including:

facility: operations and internal controls; measurement control and statistics;

' and, physical inventory. The inspection involved 58 hours6.712963e-4 days <br />0.0161 hours <br />9.589947e-5 weeks <br />2.2069e-5 months <br /> on-site and 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> off-site by two NRC inspectors.

Results: One violation was identified: failure to maintain the' tamper-safe

! records as described in the licensee's NRC approved Fundamental Nuclear Material Control Plan (Paragraph 3). j h

8601300119 86012T

{DR ADOCK 07001100 PDR-

=

DETAILS'

1. Key Persons Contacted

• 0. Parks, Manager, Nuclear Materials and Security -

• R. Sheeran, Manager, Nuclear Lt. censing, Safety, Security and Accountability G. Kersteen, Manager of Production H. Cohen, Supervisor, Nuclear Material Accountability J. Vollaro, Health Physics Supervisor D. Masse, Nuclear Accountacility Specialist

• Present at exit interview

-The inspectors also interviewed other licensee employees involved with nuclear material controls and production.

2. 30703 - Exit Interview The inspectors met with the licensee representatives indicated in paragraph 1 at the conclusion of the inspection on December 13, 1985, and summarized the scope and findings of the inspection.

At no time during this inspection was written material provided to the licensee by the inspectors.

3. 85205 - Facility _ Operations and Internal Controls The inspectors audited the licensee's system of tamper-safing containers.

The seal logs (tamper-safing records) were reviewed on December 11 1985 to determine if the logs coatained information regarding the seal number, item being sealed, date and time of application or disposal of the seal, and the signature of the individual responsible for applying and/or re-moving the seal, as required by paragraph 7.4.7 of the licensee's.NRC-approved Fundamental Nuclear Material Control Plan-(FNMCP).. The inspec-tors noted that, between the period of November 11, 1985 and December 11, 1985, the licensee did not record in the seal logs, the time of applica-tion of the seal and the signature of the individual responsible for applying the seal. The licens'ee had recently revised the procedure con-cerning-tamper-safing and apparently overlooked these requirements of the, FNMCP. Failure to record the required information was identified as a violation of the FNMCP and NRC requirements (70-1100/85-08-01). j

4. 85209 - Measurement Control and Statistics As a follow-up to some inspection findings at the 002 feed. plant for this facilit) (CE-Hematite), the following areas involving measurement control dnd statistics were inspected: Control chart limits, response to exceeded i

e .

3 control limits, the bimonthly and semiannual analyses of control data, the method of determining random errors, and the treatment of item covariance in the calculation of LEIO. The inspectors found several items relative to the licensee's response to exceeded control limits and the existence of written measurement control procedures which were questionable; however, the licensee is currently pursuing these items in response to a previous inspection (NRC Inspection Report No. 70-1100/85-07). Other questionable items, concerning the bimonthly and semiannual analyses of control data, were found during this inspection and are discussed below.

Section 4.5.1 of the licensee's FNMC Plan requires the performance of statistical analyses to test for trends, randomness, and normality. A review of the measurement control program records revealed that a compu-ter generated frequency bar chart is being used to fulfill this require-ment. However, upon discussing the use of this chart with the person responsible for performing these analyses, it appeared to the inspector that the interpretation of the frequency bar chart is not clearly under-stood, and that few, if any, meaningful analytical conclusions are being realized from its use.

With respect to the use of frequency histograms, the inspector referred the licensee to " Statistical Methods in Nuclear Material Control", by John Jaech, in which it states that although the frequency distribution and the histogram serve to summarize data as a group, they fall to provide any information about time trends that may appear in the data and that a plot of the data versus each such variable is helpful ... as a prelude in using the data for making inferences. The inspector stated that, while frequency bar charts are a good first step in data analysis, they do not provide a quantitative test for data trends. The inspector stated that such quanuitative tests are explained in Jaech's Chapter 2.0; Section 2.8.3 describes cusum plots that can be used to test for trends; Section 2.9.2 describes two tests used to test for randomness; and, Section 2.9.3 describes ths "W Test" for testing the assumption of normality. The in-spector stated further that in order to achieve a meaningful analysis of control data, the:e tests should be performed, in addition to the fre-quency bar charts already being generated. Furthermore, although the FNMC Plan contains no commitment to analyze the appropriateness of control limits, these limits should also be periodically evaluated by calculating the fraction of control measurements that exceed these limits. This ,

should then be compared with the expected frequency for exceeding them to determine if the limits are inappropriate (i.e., either the operating ,

characteristics of the measurement system have significantly changed or the original control limit calculation is questionable). When such a test confirms the possible existence of inapporpriate control limits, this situation should be investigated and the control limits should be changed accordingly.

4 These findings were discussed with licensee management who agreed to consider upgrading their control data analytical methods. This is an Inspector Followup Item (70-1100/85-08-02).

5. 85211- physical Inventory The licensee is required to take a complete physical inventory of the low enriched uranium at six-month intervals. Beginning September 17, 1985, the licensee took a physical inventory, which was observed by the inspec-tor and documented in NRC Region I Inspection Report No. 70-1100/85-06.

Subsequently, the licensee requested a thirteen-day extension to the thirty-day physical inventory reconciliation period. The extension was granted by the NRC's Office of Nuclear Material Safety and Safeguards, Division of Safeguards by letter of October 29, 1985. The inspector noted that the licensee completed the reconciliation prior to October 30, 1985, as required by terms of the extension.

During this inspection, the inspector audited the supporting data used to establish the total low enriched uranium physical inventory. The support-ing inventory data consisted of handwritten inventory sheets used by the inventory teams, computer printouts containing individual items and in-ventory sheet summaries, a record documenting control of inventory tags and sheets, and specific inventory instructions. Each of the line items, which appeared on the handwritten inventory sheets, was compared to the computer printout and a sample of page totals was verified by the inspec-tor. Also, the total of the four inventory areas on the computer print-outs was checked against the booked physical inventory. No discrepancies were identified.

f l

l 1

)