ML20003E297
| ML20003E297 | |
| Person / Time | |
|---|---|
| Issue date: | 03/24/1981 |
| From: | Lancaster L NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Abramson L, Elbassioni A, Vesely W NRC OFFICE OF MANAGEMENT AND PROGRAM ANALYSIS (MPA), Office of Nuclear Reactor Regulation, NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| References | |
| NUDOCS 8104020816 | |
| Download: ML20003E297 (5) | |
Text
5.
, * ^ :
e fa asov UNITED STATES
,,(
..g NUCLEAR REGULATORY COMMISSION y
,e g
W ASHINGTON, D. C. 20555 3
y or t
49 VAR 2 41081
'f/
e ell [ll*
g MAR 2 61937 g bu MEMORANDUM FOR:
W. E. Vesely, RES L. R. Abramson, MPA q E4 d
i-A. El-Bassioni, NRR D. Rubinstein, MPA J. W. Johnson, RES q
/
F. F. Goldberg, RES FROM:
Leslie E. Lancaster Methodology and Data Branch Division of Systems and Reliability Research, RES
SUBJECT:
MINUTES OF THE PROBABILITY #iD STATISTICS RESEARCH REVIEW GROUP (PSRRG)
This meeting was held on the 18th of March 1981 in the Willste Building with the following members in attendance:
W. E. Vesely L. E. Lancaster J. W. Johnson L. R. Abramson D. Rubinstein A. El-Bassioni This discussion was 1.
The general charter of the PSRRG was discussed.
a continuation from the 6th of February 1981 meeting. The discussed version of the Proposed General Charter is contained in the agenda announcement for the meeting (attached).
The members agreed that the tasks as outlined in this Proposed General Charter will be too time consuming to accomplish. Therefore, the members agreed to spend more time on the wording. For example, the wording should imply that certain tasks can be cut off from further discussion at the discretion of the members, but guided by Research.
Hence, each member will send a handwritten version to the undersigned two weeks before the next meeting. The undersigned will combine f
the different versions, again as a Proposed General Charter, and the discussion will continue at the next meeting.
l Priorities on action items and group accomplishments were discussed.
l 2.
On the average, The priorities should be set by Research as they arise.
we will have a meeting every six weeks. Before discussing any topic, When necessary at least one member should be familiar with the topic.
we should invite the project director to discuss the report prior to the discussion.
1Rb s104020
,s
- y.i ;.. :.
'I 3.
The ASA Ad Hoc Committee on Nuclear Regulatory Research was discussed.
We set up a meeting with the chaiman, Dr. Ccri Bennett, 0.' the 30th of March 1981, to discuss the June ASA meeting. At the 30 March meeting with Dr. Bennett, we would like to discuss, forWe example, the details of having ASA review the Zion report.
would have to narrow the scope and have them look at the methodology in general and specific applications of the Bayesian techniques.
4.
Appendix A to the NPRDS report by Idaho was discussed. The project director, Bob Dennig, sat in on the discussion. The discussion centered in cn how to handle the case with unequal exposure times.
The Kansas State Report's on Bayesian techniques were discussed.
5.
f Dr. Dallas Johnson and Dr. George Milliken from KSU will visit l
on the 20th of April 1981 for an all day question and answer workshop. It was decided to have them presant an overview on their on-going projects for this fiscal year.
l On the 24th of March,1981 Dr. Milton Sobel will discuss a proposal 6.
to the PSRRG on an application of some of his work to systems reliability. It was decided to leave the discussion with Dr. Sobel On a request by one of the members for some open and informal.
t of his work, Dr. Ram Uppuluri recommended that we look at their
{
1979 book, " Selected Tables of IMS," Volume 4.
This book contains i
83 pages of introduction and examples of tables.
t 7.
Atwood's common cause reports were discussed.
C. Atwood has applied i
his computer package to pumps and to diesel data.
It was commented that a paper by Flemming would be a helpful reference.
The next meeting was tentatively set for either the 20th of April 8.
1981, after the KS'J visit or the 22nd of April 1981 from 9:00 until 11:30 AM. For the next meeting, the tentative agenda is:
i o
Charter o
KSU o
In Plant Data o
ASA The Zion Report l
o Easterling's draft on the Estimation of Accident Sequence o
'I Probabilities I
i
-42.> r. M Leslie E. Lancaster Methodology and Data Branch Division of Systems and Reliability Research I
Enclosure:
As Stated E
cc:
R. Minogue, RES
[
R. Bernero, Res Public Document Room
.p nnea,
u o
UNITED STATES 53 j
NUCLEAR REGULATORY COMMISSION t'
3 c,;.
,4 ; E WASHINGTON. D. C. 20$55
- f4EfoN
+....
MAR 2 31981 ME!!ORAf!DUit FOR:
W. E. Vesely, RES L. R. Abramson, !!PA A. El-Bassioni, flRR D. Rubinstein, ftPA J. W. Johnson, RES F. F. Goldberg, FIS FRO:1:
Leslie E. Lancaster
!!ethodology and Data Branch Division of Systems and Reliability Research, RES
SUBJECT:
AGENDA FOR THE PROSA3ILITY Afl0 STATISTICS RESEARCH REVIEW GROUP l-tEETIllG 1.
The next meeting will be held on Wednesday, the 18th of March 1981, in the Willste Building, room 130 from 9:00 AM until 11:30 AM.
2.
The agenda or discussion items for this meeting, time permitting, will be as follows:
a.
Ch arter_
We shall continue the discussion from the last meeting, and, as agreed, a first cut at the Proposed General Charter is as follows:
Prcoosed General Charter
'l. To review reports en selected methodologies, approaches, techniques, and formulation of data bases and models used for probabilistic, statistical, and risk analyses written in-house or by contractors for the fluclear Regulatory Research Office and available for use to the overall flRC agency.
2.
To check the'vilidity of"thelalues of assessed statistical parameters assigned in prcbabilistic, statistical, and risk analysis as recommended by the fluclear Regulatory Research Office.
3.
To review contractor proposals utilizing probabilistic, statistical, and risk techniques, formulations, and approaches to be carried out by the Nuclear Regulatory Research Office and to comment on a selective basis.
w
=
. 4.
When appropriate, to publish documents, including minority opinions, authored as a group or write-ups on revieu group discussions for the Nuclear Regulatory Rasearch Office and the public document room.
b.
Priorities We need to discuss the pricrities of the action items given in the general charter. What procedures should be set up so that the review group can act in consonance?
c.
ASA Ad Hec Committee on Nuclear Regulatory Research The cc=mittee meets with NRC cn 11-12 June 1931 for the third meeting. 'ie need to discuss a proposed agenda for this meeting.
The discussicn should be aimed at an agenda for focusing the committee efforts tcwards solving some of our problems.
Any ideas en a mechanism or procedure for a good working interface between the committee and the NRC Research Program? Carl Sennett has mentioned sever &l times that this committee is a new experience for ASA.
d.
Accendix A to the NPROS Recort by Idaho Read pages 44-48. Of concern is the case with unequal exposure times.
Are other techniques available? Can related work be found in the literature? So far Idaho has not tested the' canned packages on any risk data. Therefore, does anyone know of any problems that -
could arise or are there any obvious mistakes in the report that could cause problems?
e.
The YJU 3ayesian Project We hope to get both Dallas Johnson and George Milliken from KSU to
~
give us a question.and answer oorkshop on the project in mid-April.
What questions need to be anyt". J at this workshop? Of interest is the choice of conditional priors, simulation studies, and various
- diagnosties.
It may' be helpful to KSU to give them a list of questions beforehand..
In the draft NUREG/CR-KSU-2075c on page 54 f f you work thru the mathematics for the case when Tj = T for all j, then you will find that the estimates are constrained by 51.> E'.
However,.since F.
has a Poisson distribution, S'p = T.. NR6 is working on a modified version using.the harmonic mean a..d it worked well in a current simulation-study. Does anyone know of other techniques or other problem areas
-in the KSU report?
e e, b omm e e rb 9
1,
, f.
Dr. Milton Sobel Dr. Sobel will give a 15 T.inute presentation followed by a questien and answer session on Tuesday the 24th of March 1981 from 1:00 to 3:30 Pit. He will disevis his proposal on an application of his work on hypergeometric analogues of in-complete Dirichlet integrals to a reliability analysis of series-parallel systems. The meeting will oe held in the Willste Building, rocm 110.
Any ideas on how to focus the discussion or shall we leave it wide open?
g.
Common Cause The ASA Ad Hoc committee has some favorable comments on Atwood's fWREG. However, they also mentioned that in applications it was always considered after the fact and that we need some practice to see if it is a reascnable model. Any ideas ?
Also, of some concern would be to _ have a discussion on how Atwood could write an executive. summary on his report (EGG-EA-5289) so a lay-person could easily read it or use the computer package.
3.
Some items for future meetings a.
The Zion report b.
Easterling's draft on the Estimation of Accident Sequence Probabilities.
c.
In-Plant report i
d.
Prescribed Procedures.
g g pmW-Leslie E. Lancaster Methodology and Data Branch Division of-Systems and Reliability Research, RES s emaa
%AM g