ML19260C756
| ML19260C756 | |
| Person / Time | |
|---|---|
| Issue date: | 08/14/1979 |
| From: | Dircks W NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| To: | Harold Denton Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML19260C757 | List: |
| References | |
| NUDOCS 8001210083 | |
| Download: ML19260C756 (2) | |
Text
'.$ -
- hh.
M<h G.
Bc a
AUG 141579 A
MEMORANDUM FOR: Harold R. Denton, Director Office of fluclear Reactor Regulation FR0ti:
William J. Dircks, Director Office of !!uclear Material Safety and Safeguards
SUBJECT:
COI91Et4TS Oil PROPOSED AGEliCY-WIDE GUIDELItiES FOR EXPAtiDED USE OF QUAliTITATIVE RISK ASSESSMEtiT TECHilIQUES Ue have reviewed the proposed staff guidelines regarding expanded use of quantitative risk assessment techniques in the regulatory process transmitted by your memorandum of July.30,1979. Our comments are enumerated in the enclosure.
%i1Isimity Ei;hui E. h.n.2:.u e
William J. Dircks, Director Office of fluclear Material Safety and Safeguards
Enclosure:
As stated Distribution:
f4 MSS 79-1104 Central File f4 MSS R/F DAtlussbaumer FCTA R/F 1930~275
'MLAu R/F tiLAu HLowenberg TFCarter P.ECunningham g
WJDircks FC Wil d,d@g LCRouse <
C CMS, 60 y
JBMartin y
V TFC ter BMartin
[ [#
JGDavis Dndussbaumer EWLeins pfq f7o 8/
9 8//O/79 FC0hs
- FCr, ry FC
/ t; MSS I4 MSS FCTA
... _.,w 7.
EVANola HLowenberg_
LC Rd'Cunnir[
JGDavis WJDircks 0/7/79
_/
/79 8/ c) D9.
- 8/jp /79_
8/
/70 8/
/79 8
- .,c w I
- ui s. aov =~ c~r =~riaa ome,i,,,,g 0,Q 1 2 9 0 O g],,,
n n ut us e,. o nor cuo
M NMSS COME'.TS ON PROPOSED AGENCY-WIDE GUIDELINES FOR EXPANDED USE OF QUANTITATIVE RISK ASSESSMENT TECHNIQUES 1.
We concur that use of probabilistic, or quantitative, risk assessments should be extended to include the safety aspects of fuel cycle facilities in the near future.
Risk assessment programs have already been initiated for waste repository systems.
2.
We believe that we could refine the general guid'.nce provided and develop a policy expressing our commitment to expanded us/. of quantitative risk assess-ment techniques by October 1,1979.
However,.he development of implementing procedures will require much more time as well as significant resources in view of the following:
a.
NRR may be in a position to develop quickly its own set of specific implementing procedures as a result of the experience gained through acquisition and use of the Reactor Safety Study.
No comparable studies have been performed nor does an appropriate body. of data exist to describe the risks associated with fuel cycle activities (with the exception of the data being developed as referenced in item 1 above).
b.
The typesof risk assessment discussed in your memorandum are for highly engineered (man-made) systems. The major safety systems for high level waste geological repositories are not highly engineered, thus the applicability of past work on risk assessment to geologic repositories is limited.
Specific implementing procedures, including the applicability of quantita-c.
tive risk assessment techniques to the wide range of fuel cycle activities, both front and back end, are inappropriate at this time and must be evolved through experience to be useful when sufficient infonnation is available.
3.
The cost of developing and implementing quantitative risk assessment techniques is expected to be high and will require program requests for additional funds.
For. example, the cost of a methodology development of fuel reprocessing that would include a fault tree data bank, construction and quantification of logic and event trees, and calculation of risk has been quoted at $1.7 million over a< period of three years.
4.
Discussions of failures due to fires and natural phenomena appearin'g on pages 4, 9 and 12 of the memorandum require clarification.
It is not clear whether the difficulties alluded to arise from inability to quantify the initiative event (e.g., fires, earthquakes, floods) or the inability to determine by fault tree analysis failure of in-plant components induced by the initiating event. We believe that there are sufficient data bases to permit quantification of the probability of occurrence of natural phenomena.
Therefore, we would interpret the statements to refer to complexity of fault trees involving externally initiated, internal facility events.
1930 276