ML20059E168
| ML20059E168 | |
| Person / Time | |
|---|---|
| Issue date: | 08/31/1990 |
| From: | Taylor J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | |
| References | |
| TASK-PII, TASK-SE SECY-90-310, NUDOCS 9009100063 | |
| Download: ML20059E168 (7) | |
Text
,
o
+f#g u.g%
l RELEASED TO THE PDP, l
r I
l Tl19l90
- g...../
l 6 ate '
indts
........................l POUCVT880E August 31, 1990 (Information)
SEcy.go_310 f.g:
The Commissioners f.r.QW :
James M. Taylor Executive Director for Operations Subiect:
PERFORMANCE INDICATOR PROGRAM THE STATUS OF CAUSE CODES COMPARISONS Puroose:
io inform the Commission of the status of the Performance Indicator (PI) Program Cause Code Comparison Development, Backaround:
In the Secretary's memorandum of August 10, 1989, on SECY-89 211, Performance Indicator Program - Cause Codes, the Commission approved the inclusion of cause code trends in the NRC Quarterly Performance Indicator Report. However, prior to approving the display of cause code information in a manner that compared plants, the Commission requested further analysis regarding the comparability of the licensee event report (LER) data among licensees.
In SECY 89-280, Performance Indicator Program - Cause Codes, dated September 5, 1989, the staff discussed developmental activities which address the issue of comparability of LER data among licensees, and committed to provide the results of those activities and to mal;e recommendations for plant-to-industry comparison methods to the Commission by August 1990.
This document describes the status of, and future plans for, the development of plant-to-industry comparison methods.
Recommendations for changes to the performance indicator program are pending completion of a test program.
Discussion:
Develoomental Effort in the spring of 1989, the staff initiated a study at the Oak Ridge National Laboratory (ORNL) to determine
Contact:
NOTE:
TO BE MADE PUBLICLY Donald Hickman, AE00 AVAILABLE IN 10 49-24361 WORKING DAYS FRO.M THE DATE OF THIS PAPER h O lf0 00 43
]ff Y ik Qq
.r i
l l
The Commissioners 2
i l
l l
appropriate " peer groups" for comparing reported event data i
among licensees. During this effort, it was verified that cause.codedatawe[ecyclicwithaperiodapproximatingthe refueling interval.
l With this information, the staff initiated a study at the Idaho National Engineering Laboratory (INEL) to evaluate the effects of a plant's operating cycle on reported event data.
l These programs are described in the Enclosure while the.
staff's current findings and planned activities are described below.
I Findinas i
l Operatina Cvele Study i
l A major finding of this developmental effort was that~ the operating mode of the plant can have as great an effect on I
the frequency of events.as design considerations.
The staff examined the operating cycle to determine those phases of operation in which event reporting varies significantly.
(
The following five phases were proposed for further evaluation: (1) startup'from refueling, (2) power operation, (3) non refueling outages, (4) the period preceding the refueling outage, and (5) the refueling outage.
Peer Grouo Study In the Peer Group Study the staff examined factors affecting event re)orting excluding the operating cycle effects. Of these ot1er factors, design characteristics ~were. determined I
to be the most important,'although the age of the plant (operating license issuance before or after the Three Mile Island [TMI) accident) and type of technical specifications (standard or custom) are also significant.
Peer groups were established based primarily upon Nuclear. Steam Supply System l
(NSSS) vendor and model, or product line, with allowances made for operating licensing date, the type of technical i
specifications, and the unique features of several~ plants.
This resulted in nine proposed peer groups, generally grouped as follows: (1) all Babcock and Wilcox (B&W) plants;.
(2) Combustion Enq > ering (CE) plants with reactor protectiun systems nich do not include core protection calculators (CPCs); (3) CE plants with CPCs- (4) General Electric-(GE) Boiling Water Reactor (BWR) BWR/2, BWR/3, and older (pre TMI) BWR/4 plants; (5) newer BWR/4.and all BWR/5 I
ORNL/NDAC 254. " Development of an Analytical Tool for Investigating Plant Peer Groups " oak Ridge National Laboratory, Nuclear Operations Analysis Center. November 3,1989.
4 I
_._,m., _ _. _,. _.
~.,.. _ -
y_
The Commissioners 3
l and BWR/6 plants; (6) Westinghouse'(W) two loop plants;' (7)
. older W three loop plants; (8)-older W four loop plants; and (9) newer W three loop and four loop plants.
[pture Activities The methodology for calculating the performance indicators-on an operating phase basis is currently under development.
Using.at least two cycles of operating history, the.
performance indicators will be calculated using the proposed cycle phases and_the candidate set of peer groups. These-l results will be evaluated against the current met'ods to determine if they provide improved capability to identify significant plant trends. The staff expects to complete this program by June 1991-including recommendations for changes to the performance indicator program..
/'.
- 2 s
or j
E cuti Director for Operations
Enclosure:
As stated DISTRIBUTION:
GPA LSS L
REGIONAL OFFICES l
EDO ACRS ACNW ASLBP ASLAP SECY d
- y m.,
.: 7
- g e
y=
's s
f 1
(
)
a 1
L'.
2 I
1 4
1 i
1 i,
1 i
i, i
i
\\
1 l
1 J
1 ENCLOSURE l-t,
' l I
b 9
l
(
(
k
- t k
k k
k
- r 3
h
- i at Eft f
e v
.t
(
l-9 e
i
'l-3 t
k e
I.
r i
.suquewp sfx,s-i.upp. owAy
.(
y yu. mpg.~
,pmm,ynppq
.4w, 94_ paw,g-M,,,nupsw<ww,vm 4
q 0
1 Enclosure i
i Proaram Descriotions l
Doeratina Cycle Studv Based on the work that identified the cyclic nature of reportable' events, it was also determined that the frequency of events was strongly influenced by the operating mode (refueling, power operations, mid cycle outage, etc.).
For example, it.is' reasonable-to expect that a plant in a refueling outage tends to have more safety system actuations, safety system failures, and maintenance related events due to increased equipment' maintenance and test activities.
o L
The staff, in conjunction with the Idaho National Engineering Laboratory (INEL), is examining methods to account for the variability in event reporting rates between phases of an operating cycle. The objectives are to identify the relationship between the frequency of events and various operating phases and. to develo) the methodology for calculating performance indicators as a function of tiose phases that have significantly different event rates.
Based upon an analysis of the operating experience of several plants and discussions with industry, the' following five candidate operating cycle phases were identified: (1) startup from refueling, (2) power operation, (3) non-refueling outages, (4) the period preceding the refueling outage, and (5) the refueling outage. A data base for each operating nuclear plant was.then constructed from the historical performance indicator data for each of these five phases, for each performance indicator and cause code.
l Trial programs were conducted inder t.Jently by the staff and INEL to test preliminary calculational methods which correct the performance indicators for a plant's operational phase. - These tests included slant self-trends as well as deviations from d e peer group averages, using tie proposed. peer groups.
In addition, the indicators calculated by the preliminary methods were compared to the current indicators. The trial studies identified new insights that could be gained from this approach.
The methodology for calculating the performance indicators for the operating phases and displaying them on a quarterly basis is still under development.
Sensitivity analyses are being conducted to finalize-the operating cycle phases. This study is expected to be completed by April 1991.
I EGG-EAST-9107, " Operational Cycle Effects on the Performance Indicators," Idaho hattonal Engineering Laboratory, June 1991.
)
~
~. _ _.
f a
<N i
2 l
l i
5 2
Peer Groun'Studv f
I The goal of this effort was to establish suitable peer groups for use in comparing each plant's cause code arofile to its peer group average.: Plant-specific and peer group features w11ch' affect event reporting were identified and evaluated. The applicability of these peer groups to the other performance indicators was also considered.
Of the factors other than operating cycle phase which ' affect event reporting, thc staff, with the support of the Oak Ridge National Laboratory (ORNL),
identified plant design as the single most important. Based on design, eight preliminary peer groups were constructed, based primarily on NSSS vendor and i
l model, or product line.
i The staff also had discussions with representatives of the BWR Owners Group I
and the Westinghouse Owners Group.
Both groups were of the opinion that t
operating licensing date (before or after the Three Mile Island (TMI)d) are i
accident) and the type of technical specifications (custom or standar also important. The staff considered these comments and modified the eight preliminary groups into ten groups.
Because of the small number of plants making up certain groups, the staff.
looked into methods for combining peer groups.
Studies indicated that the i
number of peer groups could be reduced to nine and still maintain the basic character of the peer groups. With this grouping, peer group populations ranged from 7 to 23 plants.
The final peer group selection will be validated in a test program expected to be complete by April 1991.
l i
1 ORNL/NOAC 261 " Development of Consnercial huclear Plant Peer Groups for Presentation of Performance Indicator Data." Dak Ridge National Laboratory. Nuclear Operations Analysis Center May 1990.
k.
k k
k k
k k
k 3
TRANSMITTAL. TO:
Document Control Desk, Pl-24 ADVANCE COPY 10:
Public Document Room
- 1 DATE:
9//9 / 94
/f i
FROM:
SECY, Operations Branch' 4
i Attached are copies of SECY papers and related documents.
They are being forwarded for entry on the Daily Accession List and placement in the Public Document Room.
No other distribution is j
requested or required.
i M
i J/0 ll. _
0 2.
YO- $/3 12.
It
/
4
- I/d M
3.
13.
g 4.
14.
S.
- 15..
t
(
6.
16.
4 7.
17.
8.
18.
M j
9.
19.
a
- 10. _
20.
4 i
- PDR is advanced two copies of each SECY paper and one corf of e ch related document.
0 ?l.n -
OF./;2
-a I
.