ML20136C501
Text
,
~~
DISTRIBUTION
^
i
?
L. Barry RM R. Scrohns, RM E. Triner, RM/B R. Shumway, RM/B N0y 2 31992 J. Clark, RM/BMA S. Feld, RM/BMA RM/B R/F (6)
RM/BMA R/F (2)
MEMORANDUM FOR: Thomas E. Murley, Director i
Regional Operations and Generic Requirements Staff Office of the Executive Director for Operations FROM:
Edwin G. Triner, Director Division of Budget and Analysis -
Office of Resource Management
SUBJECT:
COST ANALYSIS GROUP'S (CAG) COMMENTS ON REGULATORY IMPACT ANALYSIS (RIA) FOR USI A-43, CONTAINMENT EMERGENCY SUMP PERFORMANCE i
Per your request, attached are the Cost Analysis Group's (CAG) comments on the Regulatory Impact Analysis (RIA) for USI A-43, Containment Emergency Sump Performance. As suggested by Walt Schwink, our major focus of attention has been upon the cost of radiological exposure resulting from replacing insulation. Additionally, we have, as a matter of lower priority, given consideration to the length of shutdown time required for the reinsulation.
I believe this early direct involvement with a specific CRGR action will prove to be helpful in enabling us to focus more clearly on the types of issues and analyses we will be dealing with in the future.
I am confident that with this experience already under our belt, the (CAG) planning and development effo*ts currently underway will be enhanced.
I I
This review has helped to identify a number of procedural matters which, if adopted, would improve CAG's oversight role in these types of reviews. As a general rule, it would be helpful if more background material on the proposed action were provided to the CAG. Also, the identification of the principal author of the cost analysis would be valuable and an allowance for a greater turn-around time is essential.
Finally, the CAG is in the process of developing certain scoping worksheets and checklists which, once finalized, should be completed by the RIA author and submitted with CRGR's i
j request for our review.
l Original Signed by L Edwla G.Triner Edwin G. Triner, Director Division of Budget and Analysis Office of Resource Management
Attachment:
x As stated N
/
5~"'
cc w/att:
V. Stello, Jr.
/'
5,i.. RM/ B..M..A...
i
- w. WhKnk' M....f.'}.d j,j,s,,, g,,,,,,,,,,,,,,,,,y.. L...RM/..B MA. g,,i..
i
. M/
RM i.
R
......L...Ba rr
,,, j,,,,,Cl,a,1,,,,,,,,, [, S, h,umway,,l,,Triner,..Scroggins/Barry
,,, M,,/82 l
{
ll/li/82 11/ /82 l
11/, /82 11/,,,/82
,,, hd 1
l
,,I-----
8511210102 851022 PDR FOIA DELL 85-361 PDR
QQC}
COST ANAL.YSIS GROUP'S (CAG)
COM4ENTS ON REGut.ATORY IMPACT ANALYSIS (RIA) FOR US t
CONTAINMENT EMERGGCY SUMP PERFORMANCE i
The following review addresses the treatman't of occupational radiological exposure cost and the length of time required for reinsulation in the USI A-43 Containment Emergency Sump Performance. Regulatory Impact Analysis (RIA), and i
makes specific connents regarding other elements of the RIA.
In general, the Cost i
Analysis Group's (CAG's) concerns are viewed as minor in nature and it is our judgment that this RIA represents a good overali effort.
Cost of Radioloofcal Exposure i
,i For those plants requiring insulation replacement, the RIA estimates an a occupational radiological exposure of 50 man-ree per plant. ~ This estimate is based on discussions with Surry site staff who estimated the exposure cost of insulation replacement at Surry at 50 man-rem or 7 x 10-3 2
man-rem /ft.
Applying the square foot exposure cost to the specific nuclear v11ts under review, the j
RIA estimates exposures at individual units ranging from 8 to 99 man-rem.
Based on this sample, a 50 man-rem average value was adopted. (See page 19 of the j
RIA.)
i j
In this limited timeframe, the CAG cannot validate the reasonableness of th l
initial 50 man-rem estimate provided by Surry.
However, the steps taken -- (1)
{
1 seeking an estimate from industry, (2) converting to a cost per square foot an costing out a number of jobs, and (3) deriving a 50 man-rem average based on that sample - all appear to represent a meaningful approach.
i ihere are, however, a couple of concerns that can be raised in regard to the i
i
- reatment of the cost of radiological exposure on the overall value-impact i
- ssessment.
It should be stressed, however, that given the relative' magnitude
) if this impact, it is highly unlikely that these issues would appreciably alter he RIA's bottom-line conclusions.
g DRAn 1.
Occupational radiological exposure is only estimated for the task of I-replacing insulation.
Another proposed task -- the placement of interception screens and other shielding structures within the plant --
would appear to result in additional occupational radiological exposure, yet, no estimate of this impact is provided.
2.
In the value-impact (V-I) ratios that are developed in the RIA (see pages 23 and 25), there is no explicit inclusion of the occupational radiological exposure cost.
This is dismissed based on the reasoning that the averted cost of 23 man-rem on site essentially offsets the 50 man-rem occupational dose due to the averaging methods used and associated uncertainties.
Clearly, this is not an indefensible position.
However, in CAG's judgment, the occupational exposure is probably the most precise man-rem estimate available to the staff and the credibility of the V-I ratio would be enhanced with its explicit inclusion.
CAG reconnends that.the 1500 man-rem / plant value (benefit) be characterized as representing the sum of the 1487 man-rem / plant of public dose averted and the 23 man-rem / plant of on-site dose averted, with the total of 1510 rounded to 1500 due to associated uncertainties and averaging methods employed.
The 50 man-rem occupational dose can then be deducted from the 1
1500 man-rem value to produce the net benefit of 1450 averted man-rem per i
plant.
Thus, for example, the V-I ratio appearing on the bottom of page 23 would take the following form:
y,g, (1500 man-rem / plant)(5 plants) - (50 man-rem / plant)(5 plants), 2266 man-rem
$(.7+2.5)M
$ million Also, it might be more meaningful to convert the V-I expression to
$/ man-rem instead of expressing the resrlt as man-rem /$.
Such a conversion would make the result directly comparable to the safety goal of
$1000/ man-rem, and would show, for example,.that in this particular instance, a man-rem can be averted for about $440 as ccmpared to a target of $1000,
.,.s_..,
4 h
Following are some additional comments concerning other elements of the RIA.
1.
On the bottom of page 21, the estimated present worth of plant cost due to
[
a core-melt accident is placed at $1.65 billion.
The $1.65 billion 1
estimate has been identified as the dollar outlay to physically replace a
}
nuclear unit that has been destroyed by a core-melt accident.
CAG
[
questions the inclusion of this infonnation on several grounds:
l 4
.j a)
The entire RIA addresses the value of the proposed action in tenns of f
averted man-ress.
Suddenly this new value is introduced with no i
l explanation as to why or even what it is supposed to represent.
If a l
dollar est,imate is deemed appropriate, the averted man-rems can be translated to dollars based on a $1000-per-man-rem conversion factor.
b)
If the author feels 1.t is necessary to identify additional costs due to a core-melt accident, there is no basis for stopping with the cost i
to replace the unit.
For example, why not include the expected
{
off-site dollar cor6 sequences of the accident, i.e., property losses,
{
l health and life effects, economic disruption, interdiction costs, etc.
1' 4
j c)
TechnicalJy, the cost of physically replacing a damaged nuclear unit may not necessarily be a cost of an accident.
The loss of the nuclear unit itself 1,s of no consequence as it is a sunk investment.
What one j
loses is the energy that could have been generated by that nuclear I
t
)
unit over its remaining life and the potentiial savings in generation 3
costs associated with that output, i.e., the difference in the nuclear j
unit's generating costs and the variable costs of the source of j
replacement energy.
It is only when a new nuclear unit can be shown j
j to be the preferred source of replacement energy that a case can be -
s made for including the capital cost of the new unit as a part of the i
cost of an accident.
And even under such circumstances, about 10 i
{
years of replacement energy from other sources will still be required j
because of the long lead time required to bring a new nuclear unit i
on-line.
l 1 -
Qp"}
~...
2.'
There should.b'e some recognition that the benefits (averted man-rems) are 1
distributed over a 23-year period whereas the costs are incurred in the present.
It is recognized that there is some sensitivity to the use of discount rates in these analyses, but it would be helpful to explicitly i
acknowledge that a zero real discount rate has been assumed.
1 3.
CAG believes it might be helpful if the value impact analysis were to be i
prepared as a " stand alone" document.
This could be achieved by including a brief introductory section which identifies the problem, major 4
assumptions used in the analysis, and the recomended solution.
4.
The V-I ratios on the bottom of page 23 and top of page 25, i.e.,
page 23:
i Y-I = (1500 man-rem /per plant)(5 plants), 2344 man-rem
$(.7+2.5)M
$ million l
page 25:
)
or (1500)(5)/(.7 +.4) = 6818 man-rem 5 million 4
are potentially confusing to the reader because of the differing notation used in each equation (second equation adopts a much more abbreviateu specification), and because the manner'in which they are presented suggests that a mathematica.1 transformation has occurred in going from the first to the second equation.
In actuality, the equations adopt different scenarios
),
on the cost (first equation assumes refrisulation is required on five units, whereas the second equation assumes that the less costly fixes (screening) can resolve the problem).
This latter point is never acknowledged in the RIA nor is it ever explicitly stated that the alternative fixes being 1
considered are either/or options.
i
. =
.- \\
nG UiL.g C 1 a
1 5.
The CAG recogni::es that the term " plant" is frequently used as a proxy for the more precise tems -- generating unit and generating station.
The difference between a unit and a station is signifidant and as the i
definitions below suggest, it is more comon to view a plant as a synonym for a generating station, Throughou't this RIA, impacts are expressed on a per plant basis where the true intent is to provide' per unit measures.
The use of the, at best, ambiguous term " plant" adds unnecessary confusion and can cause some readers to underestimate the calculated impacts by a factor of two or three.
1 l
1.ength of Time Needed to Reinsulate A.
NRR has surveyed 18 operating reactors and identified 4 as potentially requiring insulation replacement.
They are:
Salem Unit 1, Maine Yank'ee, Ginna, and Millstone Unit 2.
Based on this ratio, NRR estimates there may be a total of. 6 to 10 reactors requiring this correction.
i B.
The length of time required to reinsulata can vary greatly depending not only on the quantity of insulation involved, but also the perceived' urgency.
of the repair.
That is, if the licensee is not allowed to schedule the repair to coincide with a planned outage,.but must incur a costly and othemise unnecessary shutdown to perform the work, he will initiate st:eps j
to reduce the period of perfomance--for example, very careful planning to insure a smooth and timely repair, and the use of more workers and j
i additional shifts and weekends to complete the work.
In the analysis that "From Edison Electric Institute's " Glossary of Electric Utility Tems," EEI Publication No. 61-31, September,1961, pp. 28, 3.9.
Generating Station (Generating Plant) - A station at which are located prime movers, electric generators, and auxiliary equipment for converting mechanical, chemical and/or nuclear energy into electric energy.
Generating Unit - An electrical generator together with its prime mover.
4
r follows, it is assumed that these steps would be adopted to reduce shutdown costs ranging anywhere frem about $150,000 to about $1,000,000 per day for replacement energy.
C.
The steam generator replacement at Surry serves as the basis for the man-hour requirement for reinsulation work.
The job was characterized by NRR as very orderly and well planned.
The effort involved the replacement of about 7200 square feet of insulation and required 3400 man hours to remove the old insulation and 2600 man hours to install the new insulation, for a total of 6000 man hours.
This results in about 0.83 man hours per square foot of insulation.
D.
The following table presents the quantity of replacement insulation required and as.sociated man hours for the four units that have bein identified as likely requiring this fix.
Insulation to be Man hours per Total man hours Unit 2
Reolaced (ft )
Souare feet Reouired Salem Unit 1 13,200 0.83 11,000 Maine Yankee 6,700 0.83 6,000 Ginna 1,000 0.83 1,000 Millstone Unit 2 1,300 0.83 1,000 E.
To estimate the actual length of time associated with this work effort, the staff must make certain assumptiens concerning the work schedule and the number of workers involved in the effort.
The staff assumes that if additiorjal shutdown time were contemplated, the licensee would perfonn the work around the clock (3 shifts), including weekends.
Further, given the dimensions of the task, it is assumed that on average 18 men could be I
working on the insulation effort.. This assumes six workers on each steam generator with an average of three steam generators per unit.
Although additional workers might well be cost justified, their presence on common scaffolding and in relatively tight quarters would probably prove very
.1.
j Unn?I inefficient.
These assumptions produce the following lengths of work effort for the four reference units.
Man Hour Man Hours Available Leng'th of UM Requirement Per Day (24 hr x 18 men)
Effort (Days)
U Salem Unit 1
,11,000 432 25 Maine Yankee 6,000 432 Ginna 14 1,000 432 2
Millstone Unit 2
' 1,000 432 2
4 F.
It is NRR's intent that required'reinsulation work can be folded into planned shutdowns and even forced outages that might occur for other reasons.
Given the non-emergency nature of this work (as currently per-ceived) and the historical outage rates for nuclear reactors (in 1980 the average reactor experienced outages during 128 days of which about 95 1
were scheduled and 33 days were unplanned (forced)), it appears reasona to assume that reinsulation work, if required, can be accommodated with l
imposing an incremental shutdown.
G.
In the event specific shutdown were required to perform the insulation task, the CAG estimates the following replacement energy costs.
These cost estimates are based on the Surry man-hour requirements, a 24-hour, 7-days-a-week work: schedule, and a continuing 18-man work force.
Daily replacement energy costs of $150,000 to $1,000,000 are assumed.
These results suggest that costs associated with a shutdown have the potentia significantly overshadow the other cost elements already o
Daily replacement energy costs are derived from estimates developed by U.S. Department of Energy--Attachment on Cost of Licensing Delays--subm i
part of NRC Monthly Report to Congress (Bevill Committee Report).
.... =..
r:..,_
a D RM...,.
i
~
included in the RIA.
These estimates assume there will be no adverse reliability impacts associated with the shutdowns.
Work Effort Replacement Energy Total Replacement Unit (Days)
Cost Per Day ($)
Energy Cost ($)
Salem Unit 1 25 150,000 to 1,000,000 3,750,00dto25,000,000 Maine Yankee 14 150,000 to 1,000,000 2,100,000 to 14,000,000 Ginna 2
150,000 to 1,000,000 300,000 to 2,000,000 Millstone Unit 2 2
150,000 to 1,000,000 300,000 to 2,000,000 H.
The foregoing discussion hinges very strongly on the data base provided by the Surry steam generator replacement.
That experience resulted in a man-hour requirement of 0.83 man hours per square foot of insulation.
A similar repair at the Turkey Point plant raises the possibility that man-hour requirements can be five times greater.
If these data were relied upon in the previous analysis, the range of replacement energy costs would be five times greater than what appears above.
t i
o I
Turkey Point data are being questioned by NRR as being in. error.
At this point in time, CAG has not been able to confirm their accuracy.
_ _ _ _. _ _.