United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-452/R-95-006
October 1995
Air
& EPA ESCALATION INDEXES FOR
AIR POLLUTION CONTROL COSTS
-------
Acknowledgments
It has become customary for every author to state, solemnly:
"This report/book/article, etc., never could have been written
IA without the help of the following individuals..." Most of the
PJ time, we'd hope, these acknowledgments are sincere. But in other
r*. cases, they are motivated more by politeness or political
C~> expediency than by an honest need to thank those who should be
;,p. thanked.
fr
^ Not in this case, though. Without the kind assistance of
' those named below, this report certainly could not have been
( written. What's more, I doubt if any worthwhile report could
\^ have been composed, for the information these individuals gave
formed the very foundation of this independent study. Without
these data, the study would have been nothing more than a
theoretical treatise, aptly entitled "How to Develop Air
Pollution Control Cost Indexes If You Can Get Enough Data".
Fortunately, I was able to get enough data, and then some. And I
will always be grateful to the good people who took time from
their busy schedules to give me what I requested—and sorely
needed. Their names and affiliations are:
i Mike Beltran (Beltran Associates)
•k Tom Betsock (Bureau of Labor Statistics)
i- Mark Brinkley (Bureau of Labor Statistics)
ir Darrell Bump (ABB Air Preheater)
ir Tom Butzbach (Graham Manufacturing)
•^ Robert Chironna (Croll Reynolds)
•^ Nicholas Confuorto (Research-Cottrell/REECo)
i Roger Haley (Paul Mueller Company)
^ Jorgen Hedenhag (Air Pol, Inc.)
* Craig Howell (Bureau of Labor Statistics)
•fc Gary Pashaian (Monroe Environmental)
* Theo Powell (NAO, Inc.)
^ Ellen Rafferty (Chemical Engineering magazine)
i( Scott Sager (Bureau of Labor Statistics)
^ Paul Sengupta (VARA International)
•fc Scot Smith (John Zinc Company)
•fr Richard Vanhoy (M & W Industries)
^ Richard Waldrop (Edwards Engineering)
I am also grateful to those who took the time to peer-review
the previous (October 1994) version of this report and to offer
their valuable comments and suggests. They are:
* John Bachmann (OD/OAQPS/EPA)
•fc Tom Betsock (Bureau of Labor Statistics)
* Frank Bunyard (ISEG/OAQPS/EPA)
U S Environmental Protection Agency
Region 5, Library (PL-12J)
77°\Vest Jackson Boulevard, 12th Floor
Chicago, IL 60604-3590
-------
11
* Norm Kaplan (AEERL/ORD/EPA)
•fr Scott Sager (Bureau of Labor Statistics)
•k Richard Waldrop (Edwards Engineering)
* Tom Walton (ISEG/OAQPS/EPA)
•fr Michael Wax (Institute of Clean Air Companies)
* Jim Weigold (OD/OAQPS/EPA)
Lastly, I would be remiss if I didn't express my heartfelt
thanks to the B.J. Steigerwald Opportunity for Independent Study
evaluation panel and their support staff. In alphabetical order,
they are: John Bachmann, Ron Campbell, David Kee, leva Spons,
Martha Strobel, Jim Weigold, and Jeanette Wiltse. Without the
opportunity they gave me, this project never would have been more
than a pipe dream.
-------
Ill
Table of Contents
Section
I. Introduction
II. Equipment Cost Indexes
III. Total Annual Cost Indexes
IV. Total Annual Cost Spreadsheet Programs
V. Distributing the Indexes
VI. Future Work
Appendix A: Control Device Component Price Data
References
Page
3
13
33
47
56
59
62
70
-------
I. Introduction
This report documents the work accomplished during the
second Bernard J. Steigerwald Opportunity for Independent Study,
which occurred from April to October 1994." This opportunity was
used to develop a group of quarterly indexes for adjusting
("escalating") air pollution control equipment costs and total
annual costs from one period to another. In all, 16 indexes were
developed, one equipment cost index (ECI) and one total annual
cost index (TACI) for each of eight control device categories.
These indexes—collectively known as the Vatavuk Air Pollution
Control Cost Indexes (VAPCCI)—can be used to escalate costs from
the initial ("base") period (first quarter 1994) forward to any
quarter. So far, final indexes have been calculated for the
second, third, and fourth quarters of 1994; preliminary indexes,
for first quarter 1995. Indexes will be computed for subsequent
quarters as soon as the required input data become available.
These indexes are unique, as no others have been developed
for such a wide array of control devices. The only indicators
similar to the VAPCCI are two equipment Producer Price Indexes
(PPI) compiled by the U.S. Department of Labor's Bureau of Labor
Statistics. The latter will be described in this report, as
well. But, first things first...
Why Do We Need the VAPCCI?
For that matter, we could ask, "Why do we need the Consumer
Price Index, or the Producer Price Index, or any other cost/price
index?" Cost/price indexes are needed for two reasons: (1) to
record changes in costs or prices over time and (2) to escalate
costs or prices from one date to another. The Consumer Price
Indexes are good examples. Compiled monthly by the Bureau of
Labor Statistics (BLS) since 1913, they have been used for
decades throughout the public and private sectors to adjust
prices and (with another BLS index, the Employment Cost Index)
wages and salaries. Consequently, they are oft-quoted in the
media and frequently cited in the literature.
Nonetheless, not even such indexes as the CPI perfectly
reflect the marketplace. At best, they provide a cloudy mirror.
The CPI are comprised of certain "mixes" of goods and services
that may or may not correspond to the mix(es) for which prices
are needed. In other words, there is no substitute for current
price information obtained from suppliers of those goods and
a A "final" report was completed in October 1994, but not
peer-reviewed. This report incorporates peer review comments and
index values updated through first quarter 1995.
-------
services.
For instance, if we wanted to determine the price increase
in widgets from 1947 to 1994, the most accurate way to do so
would be for us to survey all the widget manufacturers and
analyze the data they provided to come up with an average price
increase. But suppose we didn't have the time or the resources
to do that? We could get the data we need from the Producer
Price Index for widgets. Alas, there is 220 such PPI. Instead,
we could look up the Producer Price Index for frabbits, which (as
everyone knows) is the category of devices that resembles widgets
most closely. Now, the PPI for frobbits wouldn't give us the
exact price history for widgets over this 47-year period, but it
would provide a close approximation. What's more, it wouldn't
take nearly as long to use this PPI as it would to survey all (or
even a representative sample) of the dozens of widget
manufacturers presently in business.
In this respect, air pollution control devices are no
different from widgets. Ideally, we wouldn't need an index to
track the change in air pollution control costs or to escalate
them. We just could contact air pollution control equipment
vendors for price information. Wouldn't that suffice?
Unfortunately, it isn't easy to obtain price data from vendors.
Vendors are usually busy preparing quotations for potential
(read: paying) customers and have little time or interest to work
up price quotes for those who would use them for (in their view)
"academic" purposes. Even a vendor who would agree to provide
quotes would almost always require a written request and, once
he/she receives it, often take weeks to respond.b How much
easier would it be to estimate the current cost via one of the
VAPCCI, using this simple equation:
Costnew = Costold (VAPCCIjinew/VAPCCIj>old) (1)
where: the subscript "j" denotes the VAPCCI for control
device "j"
To illustrate, suppose that the 1988 equipment cost for a
fabric filter (baghouse) were $100,000 and that we wanted to
estimate how much this baghouse would cost in 1994. Further,
suppose that the 1988 (old) and 1994 (new) values of the VAPCCI
were 100 and 130, respectively. Then, the 1994 estimated
equipment cost would be:
Costnew = $100,000 (130/100) = $130,000.
b Or months, as we discovered when awaiting vendor replies
for this project.
-------
In other words, according to the VAPCCI for fabric filters,
the cost of this baghouse would increase by an estimated 30%
during this six-year period." Clearly, this VAPCCI would provide
a very quick and easy method for updating costs. Using this
index would save analysts time and effort, especially if they
needed to escalate costs over several different time periods.
Of course, if we did use one of the VAPCCI to escalate the
costs, we would have to make sure that we were using the index
correctly. That is, if an index were designed for escalating
baghouse equipment costs, we certainly shouldn't use it to update
costs of electricity, labor, materials, or other items that are
needed to keep a baghouse system in operation. By the same
token, if a VAPCCI were designed for escalating equipment costs
of, say, a wet scrubber, it shouldn't be used to update
incinerator, adsorber, or other equipment costs.
Why Can't We Use Other Equipment Cost Indexes?
Before developing the VAPCCI, this author used certain
published price indexes to escalate control equipment costs.
These have been the Chemical Engineering Plant Index (CEPI),
computed by and published in Chemical Engineering magazine, and
the Marshall and Swift Equipment Cost Index (M&SI), also
published in Chemical Engineering. While better than most,
overall, these indexes have not been adequate air pollution
control cost indicators.b
Developed in 1963, the CEPI has been used mainly to escalate
chemical process plant construction costs. It encompasses such
process equipment as heat exchangers, pipes and fittings, pumps,
and compressors.1 While some of these items are used in control
devices, most are not. Moreover, the CEPI does not cover such
control device components as fabric filter bags and electrostatic
precipitator collector plates.
Updated quarterly (like the VAPCCI), the M&SI compiles cost
data on an industry category basis. The 47 industries covered
include "electric power," "mining and milling," "refrigerating,"
and "process industries". The last encompasses such categories
as cement, chemicals, clay, and rubber products. A separate
• The operative word here is "estimated," as indexes are
intended for predicting current costs. They are not meant to be
substitutes for them.
b However, the CEPI can be used to escalate the costs of
flue gas desulfurization (FGD) systems, Claus sulfur recovery
plants, sulfuric acid plants, and other stand-alone chemical
processes that are often used to control air pollution.
-------
index is developed for each of these industries. Unfortunately,
each index reflects the specific mix of equipment for that
particular industry, as well as costs for other commodities, such
as labor.2 Undoubtedly, some of this equipment is control
equipment. However, the contribution of control equipment to the
total equipment cost in, say, the portland cement industry, would
be relatively small. Changes in total cement industry equipment
costs would overshadow any changes in the control costs incurred
there. Thus, the M&SI is too broad-based to use in accurately
tracking or escalating air pollution control costs.
Newer—and more applicable—indicators are the two Producer
Price Indexes mentioned above, namely those for "fabric filters"
and "mechanical collectors". Before discussing them, we should
present some background on the PPI, as several of them are key
inputs to the VAPCCI.
The Producer Price Indexes at a Glance
Calculated and published monthly by the Bureau of Labor
Statistics, the PPI measure average changes in sales prices
received by domestic commodity producers in all "stages of
processing". The BLS presently tracks prices for about 3,200
commodities and processes 80,000 quotations each month. There
are three primary systems of indexes within the PPI program: (1)
stage-of-processing indexes; (2) indexes for the net output of
industries and their products; and (3) commodity indexes.
Stage-of-processing indexes track price changes for products
organized by "class of buyer" and "degree of fabrication". They
are grouped into three major categories: "finished goods"
(commodities that will not undergo further processing);
"intermediate materials, supplies, and components" (commodities
that have been processed but require further processing); and
"crude materials for further processing" (products entering the
market for the first time that have not been manufactured or
fabricated and that are not sold directly to consumers). Each of
these categories encompasses several subcategories, covering such
products as foodstuffs, fuels, consumer goods, and capital
equipment. Also, each stage-of-processing index has a reference
base of 1982 = 100.
The second system of indexes includes the PPI for the net
output of industries and their products. These industry price
indexes are grouped according to the Standard Industrial
Classification (SIC) and the Census product code extension of the
SIC. They are also compatible with other economic time series
organized by SIC codes, such as data on employment, wages, and
productivity. Each index is classified according to a four-digit
"industry code," followed by a two- to five-digit code number
denoting the specific product class, product, and (in some cases)
subproduct. Consider this example:
-------
Industry: "Industrial and commercial fans and blowers and
purification equipment" (3564)
Product class: "Dust collection and other air purification
equipment for industrial gas cleaning systems"
(3564-6)
Product: "Particulate emission collectors" (3564-651)
Subproduct: "Fabric filters" (3564-65113)
Thirdly, the PPI commodity indexes structure organizes
products by similarity of end-use or material composition. These
eight-digit indexes track commodity prices irrespective of the
industries in which they are produced. Some of the commodity
indexes correspond to the industry price indexes, the differences
being in their reference bases and index levels. While most of
the commodity indexes have a base of 1982 = 100, most of the
industry indexes have different bases, each corresponding to the
year and month that an index was introduced. (For instance, the
reference date of the "Fabric filters" subproduct index is Jane
1989.) As a result, the absolute values of the indexes are
different. Nevertheless, the relative, month-to-month changes in
the respective commodity and industry price indexes are
identical.
All three classes of indexes are published in the monthly
ELS periodical Producer Price Indexes. For each index, the
following information is listed: index name, code number, base
(month-and-year), index values, and percent changes. The index
values are provided for the current period, preceding period, and
for the period four months previous. The percent changes are
given to the current period both from the previous period and
from the period twelve months before, as shown in Figure 1. Note
that the percent changes are listed as "unadjusted". This means
that the index values have not been "seasonally adjusted" to
account for price movements resulting from normal weather
patterns, regular production and marketing cycles, model change-
overs, seasonal discounts, and holidays. However, to cost
analysts, marketing specialists, purchasing agents, and similar
professionals, the unadjusted index data are of primary interest,
for these data are generally cited in escalating longterm
contracts, such as purchasing agreements. For this reason, the
VAPCCI have been based on the unadjusted PPI.
To supplement information in their Producer Price Indexes
periodical, the BLS will, upon request, distribute computer
printouts of the various PPI. (See Figure 2.) If the requester
so desires, the printout can list the PPI from the initial
reporting period to the latest month for which the PPI are
available. In the case of the "Metals and metal products"
commodity price indexes, the initial period was January 1926, so
-------
Figure 1. Sample Producer Price Index Listings in Producer Prices
Table 5. Producer price Indexes for the net output of selected Industries and their products—Continued
Industry •** product"
Speed changers, industrial high-spew! drive*, and gears— Continued
Other secondary products I.................... .................. »».»..•.«.•.* «...
PowOf transmission eQuipmenl. n-W.c, ........................... — .
Electric industrial furnaces, ovens, and kilns, excluding induction .„
Electric furnaces . ™-.™.,..«.«-™__. ..,,-„,,.,..,......- ......,..,.,,.,............,.,,,...
Metal processing and heat treating (such as annealing, hardening.
Electric industrial ovens and kins, including infrared
Fuel-fired industrial furnaces, ovens, and kilns .™™»...... „ ..
Fuel 'fired1 ovens and kilns •< •• • •••
Electrical heat equip, lor industrial use. n.e.c. (exc. soldenng
irons) and parts and attach ™-
industrial electric heating units and devices, except heating units
for electric furnaces ™. ... «-..«™™. «
All other Industrial heating units and devices, ind. immersion
Parts and attach, for ind. turn, and ovens, including electnc
heating units ..
Plain bearings and bushings, except automotive and aircraft
Plain bearings and bushings, unmounted, machined, excluding carbon
and graphite ................. . „.._,„,.._.„„„..„.,.... -
Power transmission equipment, except speed changers, dnves, and
Fricuon type - ............
Ftextote court tnos
1-inch nominal bore and over, gear type
1 -inch, nominal bore and over, other than gear type
t_ess than mnch nominal bore ..........—......«..«...«...-....«...............
Chains for sprocket dnves
Sprockets
Pulleys ...
Pulleys
Single drive
Other power transmission equipment, except aircraft, automobile,
Bad joints, drive and flexible shafts, and drive snarl parts
Mechanical valve operators and transmissions, exd. manne
transmissions
All other mechanical power transmission equip., except aircraft.
Industry
code
3S67
3568
Product
code
3566-SM
3566-M
3566-Z89
3566-S
3566-SSS
3568-S
3567-P
3S67-1
3567-11
3567-118
3567-19
3567-2
3567-29
3567-5
3567-55
3567-559
3567-59
3567-SM
3S67-S
3567-SSS
3568-P
3568-1
3568-11 5
3568-3
3S68-3A
3566-311
3568-319
3568-3B
3568-321
3568-322
3568-324
3568-3C
3566-332
3566-335
3568-30
3568-3E
3568-351
3568-3F
3568-361
3568-3G
3568-393
3568-394
3566-399
3568-SM
3568-M
3568-Z89
3568-S
Index
07/84
07/84
12/83
12/83
12/83
06/81
06/81
06/81
12/86
12/86
12/86
06/81
12/86
06/81
06/81
06/81
06/81
06/81
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
12/84
06/89
06/89
06/89
06/B9
02/85
02/85
12/84
Index
July
1093*
138.4
125.0
138.1
149.7
128.0
146.1
143.0
146.5
116.0
116.5
122.1
137.4
O
150.4
169.1
137.0
166.1
167.6
128.3
126.9
116.5
114.7
129.1
133.9
127.8
142.0
1229
93.2
141.9
130.7
120.8
114.8
125.9
122.8
119.4
119.4
137.0
136.8
113.7
110.1
1207
1107
1301
1246
135.8
Oct.
1893'
138.4
12S.O
143.9
161.7
128.6
147.4
144.3
149.4
119.0
122.3
122.4
138.2
O
151.3
169.1
139.7
166.6
168.5
128.8
127.3
117.8
115.9
129.4
133.9
127.8
142.0
124.6
93.7
144.8
130.7
120.9
115.0
125.9
124.5
118.9
118.9
137.0
136.8
113.7
112.0
118.8
1107
130.1
124.6
136.8
Nov.
19931
138.4
125.0
143.5
160.8
128.6
147.5
144.4
149.4
119.0
122.3
122.3
(°)
O
151.5
169.1
140.4
167.5
168.5
128.7
127.3
117.0
115.0
129.6
133.9
127.8
142.0
124.0
937
1436
130.7
120.9
115.0
125.9
O
118.9
118.9
137.0
136.8
114.2
112.0
121.0
1107
1301
ft
136.8
Unadjusted
percent change
10 Nov. 1993 from —
Nov. 1992
4.7
ft
12.1
23.6
2.0
1.9
1.9
2.8
3.5
5.1
.8
fl
O
2.2
1.7
3.2
2.5
2.1
2.3
2.4
13
1.7
2.6
1.3
.7
2.3
1.2
-4.9
4.3
2.0
2.5
4.0
1.0
O
6.4
6.4
42
5.5
2.9
24
.7
1 8
15
O
2.0
Oct 1993
0.0
0
-J
-.6
0
.1
.1
0
0
0
-.1
ft
ft
.1
0
A
.4
0
-.1
0
-.7
-A
2
0
0
0
-£
0
-A
0
0
0
0
O
0
0
0
0
.4
0
19
0
0
P)
0
-------
Figure 2. Sample Producer Price Index Printout
UNITED STATES DEPARTMENT OF LABOR
BUREAU OF LABOR STATISTICS
IABSTAT SERIES REPORT
SURVEY PC
SERIES PCU3443I111
BEGIN DATE (0/03
TUT — JUIDATfrTTEIRUARY]
(9 176.4 178.2
90 1(2. ( 1(3.1
91 190.5 1(9.2
92 190.2 190.4
93 192.0 192.3
(3) (3)
94 192.2 192 2
SURVEY PC
SERIES PCU3443I115
IEOIN DATE SO/03
Bar*
END
HARCHI
177.6
1(3.4
1(9.4
191 .6
192.3
(3)
192.7
Fin
END
tub* h«at *xchang*rs (003
DATE 94/03
ATRTTT HAYr JUNE 1
176.5 177.2 179.7
1(3. ( 1(4.6 1(4.5
190.1 1(9.0 1(9.5
191.6 191.6 191.6
192.7 192.7 192.7
tub* h*at *xchang*r» (003
DATE 94/0]
NOT SEASONALLY
DATE OF LAST UPDATE
1(0.1 1(0.2 1808 1(0. ( 1(1.4 1(2.2
1(4.5 1(4.7 1(5 ( 185.8 185 7 187.5
1(9.5 1(9.5 1(9.5 t((.9 188 9 188.9
192.0 192.0 191 9 191 9 192.0 191.9
192.7 192.4 192.4 192.2 192.2 192.2
HOT SEASONALLY
DATE OF LAST UPDATE
ADJUSTED
04/08/94
ANNUAL
179.2
184 7
1(9.4
191 .5
(3)
192.4
ADJUSTED
04/08/94
VRI JANUARY 1 FEBRUARY!
(9
90
91
92
93
94
142.4
149.0
157.0
161 .3
161 .0
(3)
161 .6
141 .0
148.1
156.9
161 .3
161 .3
(3)
162.0
HARtHI
146.0
152.9
157.0
161.0
161.1
(3)
160.5
APRIL!
148.8
153.0
156.3
161.7
161 .1
HAYl
145.7
153.4
156.6
160. (
160.7
JUNE!
145.7
156.0
156.5
160.9
161. (
JULY!
145.7
156.2
156.5
161 .6
161.7
AUGUST! SEPT EMBER!
145.8
156.3
156.5
161 .(
160.3
146.1
156.3
158.8
161 (
160.4
OCTblERl NbVEhBERl DECEMBER!
147.3
156 9
158.8
161 .0
160.0
148.5
156 9
159.0
161 0
160.2
148. (
157.2
160.7
161 .0
(3)
161 .3
ANNUAL
146 8
154.3
157.6
161 .3
160.8
3) PREllNINARY DATA
-------
10
that nearly 70 years of data are available. As Figure 2
indicates, a PPI printout lists both monthly and annual indexes.
Each annual index is simply the arithmetic average of the 12
monthly indexes for the year in question. Also, for the most
recent (usually three) months, the index values are listed as
"preliminary". These index values are subject to revision by BLS
one time, four months after being published as "preliminary," to
reflect the availability of late reports and corrections by
respondents. Finally, the letters "NA" mean that no index was
reported for the month indicated.
The foregoing discussion on the PPI was taken from a BLS
technical note, "Brief Explanation of Producer Price Indexes".3
If the reader would like more background on the PPI, he/she
should consult this reference or "Producer Prices, " chapter 16 in
the BLS Handbook of Methods.4
The PPI and Control Equipment Cost Escalation
Of the hundreds of PPI, only two have potential use for
escalating air pollution control equipment costs. These are the
"Fabric filters" and "Mechanical collectors" subproduct indexes,
discussed above. Like the other PPI, each value reflects the
average of data from at least two of three respondents in the BLS
survey pool.5 These respondents are control equipment
manufacturers. Because these indexes were first compiled in June
1989, they are newer than most of the PPI. Figure 3 displays
these PPI from their inception through July 1994.
Although it would seem that both PPI are ideally suited to
adjusting control equipment costs, they have two shortcomings.
First, neither index specifies such important facts as the sizes
and designs of the control equipment to which the index applies.
As we'll see in Section II of this report, these specifications
are important to the formulation of an escalation index. Second,
the "Mechanical collectors" index is of limited value, simply
because mechanical collectors (cyclones) are rarely used as
primary particulate control devices. Rather, mechanical
collectors mainly are used upstream of fabric filters,
electrostatic precipitators, or wet scrubbers to remove larger
particles from the waste gas. Thus, they are auxiliary
equipment, like screw conveyors, spray chambers, fans, ductwork,
and the like.
Despite these weaknesses, these PPI—especially the "Fabric
filters" index—are certainly more relevant to escalating control
costs than are the Chemical Engineering Plant and Marshall &
Swift Indexes. At the very least, they can be compared to the
VAPCCI and, in some cases, augment them.
-------
11
Figure 3. Historical Listings of "Fabric Filters" and
"Mechanical Collectors" Producer Price Indexes'
•OKVST PC
EERIE* rcD3S«4tcsii3
•ECXM tOO* 11/Ot
rilt*z« I10(
DID DATE 14/03
DATE Or LAST UPDATE O4/0*S»4
YRI
10
11
12
13
JANUARY! FEBRUARY 1
102. (
105.4
107.3
1O1.*
103.3
IOC. 3
107.3
1O1.(
MARCH)
103.3
104.3
107.3
105.*
APRII.1
103.3
10<.7
107.3
10«.0
MAY)
103.5
10*. 7
10«. 3
107.3
JOKE)
100. O
103.4
107.5
101.3
107.3
JOLYI
100.2
1O3.4
107.5
101.3
107.2
AUGUST 1 SEPTEMBER 1
101.0
104.0
107.5
101. 1
107.5
101.0
104.0
107.5
10*.*
107.5
OCTOBER) NOVEMBER) DECEMBER! AKMUAX.
101.0
104.0
107.5
10*.*
107.5
102. «
1O4.O
107.1
10*.*
107.*
102. (
105.0
107.1
10*.*
(3)
107.5
MA
103.1
10«.»
10* .a
<3>
107.4
94
109.1
109.1 109.1
109.2
109.1 109.1
110
I.T ADJ01TED
•DUE* PC03S44 MS11S
•XCIM DATE «>/0(
tteehanlcal ooll«ctoz< (>0<
BID DATE »4/03
DATE OF UkCT UPDATE O4/OI/04
10
11
•2
90
102. «
10*. 4
110.0
111. S
102. «
101.0
110.0
111.8
103.0
101.0
110.0
111.8
103.0
110.0
110.0
111 .8
103.0
110.0
110.0
111. a
100. 0
103.0
110.0
110.0
111. a
100.4 100.1
103.0 10C.1
110.0 110.0
110.0 110.0
(1)
111.8
100.1 100.1
10C.1 10C.1
110.0 110.0
110.0 110.0
102. (
1OC.1
110.0
110.0
XCKMBCRI
102. «
1OC.1
110.0
110.0
UMOJQ,
HA
104.2
101.3
110. 0
3) PUUMXMARr DATA
' No values are shown for the "Mechanical Collectors" PPI
for 1993, because no data were reported for that year.
-------
12
The Rest of this Report
The next section of this report describes how the eight
control device equipment cost indexes were developed and compares
them to other published indexes. Also presented are "historical"
index data for seven of these devices, which were obtained from
control equipment vendors.
Section III presents the total annual cost indexes (TACI),
which serve to supplement the corresponding equipment cost
indexes. Also presented in this section is a mini-course in
total annual cost estimating, covering cost terminology, input
variables, and related topics. A detailed illustration of the
TACI is also given.
The total annual cost spreadsheet programs are described in
Section IV. These programs were written to facilitate the
computation of the capital and annual cost "weighting factors"
that were used, in turn, to develop the total annual cost
indexes.
In Section V we talk about how the VAPCCI can best be
distributed to known and potential users, mainly cost analysts
and other technical personnel. These will include not only those
in EPA and other control agencies, but also users in industry,
academia, and other sectors.
Section VI discusses future work needed to update, refine,
and expand these indexes. Estimates of resources (both in-house
and contractual) for accomplishing this work also are given.
Appendix A tabulates the control equipment component price
factors which were used in developing the equipment cost indexes.
This is followed by a list of the references cited in the report.
-------
13
n. Equipment Cost Indexes
The foundation of the VAPCCI are the control device
equipment cost indexes (ECI). Each of the ECI allows the user to
adjust equipment costs forward from the first quarter 1994 to any
future quarter. Moreover, the ECI are key inputs to the total
annual cost indexes (TACI), which will be discussed in Section
III.
As stated in Section I, ECI have been developed for eight
categories of control devices. These are as follows:
/ Carbon adsorbers (fixed-bed regenerable)
/ Catalytic incinerators (fixed-bed)
/ Flares (elevated)
/ Gas absorbers (packed column)
/ Regenerative thermal oxidizers
/ Refrigeration systems
/ Thermal incinerators (recuperative)
/ Wet scrubbers (particulate control)
Note that all but one of the devices control gases (primarily
volatile organic compounds). The sole exception is the wet
scrubbers category, which includes Venturis, wet impingement
scrubbers, and similar devices.
Before describing the various ECI in detail, we need to
discuss the methodology that was employed to structure the
indexes, develop survey forms, and collect and analyze the input
data.
Structuring and Surveying
In structuring the ECI, we first selected the major
categories of add-on control devices used in controlling air
pollution from point sources. Why was the selection limited to
add-ons? For one thing, despite the introduction of various
process modifications—material substitutions, equipment changes,
chemical injections, and the like—add-ons are still the most
commonly used control methods and the methods with which
technical personnel are most familiar. Secondly, due to time and
resource constraints, we were unable to survey, let alone analyze
the costs of, the dozens of process modifications, fugitive
emission controls, and sundry other control technologies being
used today. This is not to say that indexes should not be
developed for these technologies. As we discuss in Section V,
there is considerable merit for doing so eventually. Rather, we
simply had to "draw the line" somewhere. And that line was drawn
around the add-ons listed above.
-------
14
Our next step was to develop a survey form to send to
control equipment vendors. Because we wanted to compare past
("historical") price changes to future changes predicted by the
ECI, we devised a survey form that requested both kinds of
information. Also, because each type of control device has
unique features, we crafted a different form for each of the
eight categories. Figure 4 depicts an example, the survey form
for flares. Part I of the form requests historical list prices
by quarter from 1989 to 1994. Why did we select 1989 as the
starting date? We did so because, according to a well-accepted
cost engineering rule-of-thumb, costs should not be escalated
over periods longer than five years. Note that the form asks for
prices to be expressed relative to the price in first quarter
1989, which is arbitrarily set at 100.0. This not only made it
easier for vendors to respond to the survey, but it also
facilitated analysis of the results. Finally, prices were
requested for all devices manufactured by the vendor surveyed.
We did not ask for price history breakdowns according to
different device sizes or designs.
Not so in Part II of the form. Notice that it requests data
for three device sizes: "small," "medium," and "large". Because
different control devices have differing applications, these
sizes vary according to device type. Based on our experience, we
chose three size ranges for each device—a small, a medium, and a
large—to reflect the capacities typically installed. In Figure
4, for instance, the small size range corresponds to a flare tip
diameter of < 12 inches, while the medium and large ranges
correspond to 12 - 48 inches and > 48 inches, respectively.
For each size range, the form asks for price breakdowns
("percent of list price") according to device component. These
are the primary components that make up the device. Some of
these components, such as fans and instrumentation, are common to
many devices, while others (e.g., flare burner tips) are found
with only specific types. As explained below, these price
breakdowns were used to compute the the equipment cost indexes.
The Survey Responses
Survey responses were simply averaged and tabulated. No
effort was made to "weight" the data before averaging, as each
response was given equal credence.
Historical price data: Some vendors reported annual, rather than
quarterly prices. To make them compatible with other responses,
we assigned the same price to each quarter of a given year.
Also, a few vendors were unable or unwilling to report prices for
one or more quarters during the five-year period. Rather than
entering zeroes or other assumed numbers for these quarters, we
just reported "no data" and did not average them with the other
responses.
-------
15
Figure 4. Example Vendor Survey Form
Part I. Historical Price Data
For each of the dates below, please indicate the overall
list prices of your flares relative to the price in first quarter
1989. If your prices changed less (or more) often, please so
indicate.
Year
1989
1990
1991
1992
1993
1994
Quarter
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Relative Price
100.0
Did your price changes vary according to equipment size,
style, etc.? If so, please indicate the extent to which they
varied.
-------
16
Figure 4 (cont'd.)
Part II. Component Price Data
Please estimate how much (%) each of the following
components has contributed to the total list price of your
flares. Include labor, materials, and profit. If you cannot
give an exact percentage for a component, please list a range,
Component
Flare burner tip
Pilot burners
Flare tower (stack) & support
Utility piping (from base)
Liquid & gas seals
Ladders & platforms
Dampers & valves
Knockout drum
System fan
Fan motor & drive
Instrumentation & controls
Other (specify)
Totals:
Percent of List Price
Small11
100.0
Mediumb
100.0
Large"
100.0
" Flare tip diameter: < 12 inches.
b Flare tip diameter: 12 to 48 inches
0 Flare tip diameter: > 48 inches.
-------
17
Component price data: There were some blanks in these responses
as well. For example, a few vendors did not report data for one
or more size categories, because they had not built units of
these sizes. Here again, these blanks were reported as "no data"
and omitted from the data averaging. While some vendors reported
zeroes for the "Other" component, a few completed that line on
the form. Most of the "Other" costs consisted of engineering,
project management, startup, and the like. However, one vendor
included "structural supports" in this category. Finally, some
vendors indicated that the size categories on the survey form did
not correspond to the sizes of the units they manufactured.
Instead of reporting no data for these size ranges, they entered
results, but noted that they applied to different sizes.
Results: Historical Price Data
We received enough vendor responses to present historical
price data for seven control device categories. These results
are shown in Tables 1 and 2. Note that, over this period, the
prices of six of the seven device categories increased, by
amounts ranging from 6.7% (catalytic incinerators) to 20.5%
(thermal incinerators). The exception is the refrigeration
systems category, whose prices decreased by 0.7%, relative to
first quarter 1989 prices. The average change for the seven
devices was +10.4%.
It is very difficult to account for these price changes, as
many factors—the costs of raw materials, fabrication labor
costs, and market forces—affect pricing. However, in two
cases—catalytic incinerators and refrigeration systems—the
changes can be at least partly explained. With catalytic
incinerators, a drop in catalyst prices led vendors to lower
their prices by approximately 7% between fourth quarter 1991 and
first quarter 1992." (See Table 1.) Although catalytic
incinerator prices crept upward during the next two years, they
did not return to the third quarter 1991 peak level of 107.1.
Similarly, a design innovation marketed by a leading vendor of
refrigeration systems prompted him to lower his prices in mid-
1993.
How do these price changes compare to changes reflected by
published indexes? In Table 3 are listed the Marshall and Swift
(M&SI) and the Chemical Engineering Plant (CEPI) Indexes for
1989-1st quarter 1994. The values shown are the M&SI composite
and the "Equipment" component of the CEPI. The CEPI/Equipment
index values have been converted from monthly to quarterly
figures by simply averaging the index values for the months of a
• Telephone conversation between William M. Vatavuk (U.S.
Environmental Protection Agency, Research Triangle Park, NC) and
Darrell Bump (ABB Air Preheater, Wellsville, NY), August 1994.
-------
18
Table 1. Control Device Equipment Relative Prices: 1989-94
(Catalytic Incinerators through Refrigeration Systems)
Year
1989
1990
1991
1992
1993
1994
Quarter
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Overall Change (%) :
Relative Price (1st quarter 1989 = 100)
Catalytic
Inciner.
100.0
100.3
100.4
100.9
102.7
103.0
103.8
104.2
106.8
107.0
107.1
107.1
100.2
100.2
100.2
100.2
104.2
104.2
104.2
104.2
106.7
+ 6.7
Flares
100.0
100.0
101.5
101.5
104.0
104.0
110.0
110.0
110.0
110.0
110.0
110.0
113.0
113.0
113.0
113.0
113.0
113.0
113.0
113.0
113.5
+ 13.5
Gas
Absorbers
100.0
100.5
101.0
101.5
103.4
103.8
104.2
104.6
106.0
106.4
106.8
107.2
107.6
108.0
108.4
108.9
109.0
109.1
109.3
109.4
112.2
+ 12.1
Refrig.
Systems
100.0
100.0
100.0
100.0
103.3
103.3
103.3
105.5
107.3
107.3
107.3
107.3
109.6
109.6
109.6
109.6
110.8
110.8
99.3
99.3
99.3
- 0.7
-------
19
Table 2. Control Device Equipment Relative Prices: 1989-94
(Regenerative Thermal Oxidizers through Wet Scrubbers)
Year
1989
1990
1991
1992
1993
1994
Quarter
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Overall Change (%) :
Relative Price (1st quarter 1989 = 100)
Regenerative
Thermal
Oxidizers
100.0
100.2
100.6
101.2
101.6
103.9
104.3
104.6
105.8
105.9
107.2
107.9
108.0
108.1
109.3
109.6
114.7
114.8
115.0
109.7
108.9
+ 8.9
Thermal
Incinerators
100.0
100.7
101.0
101.8
106.9
108.1
108.2
109.3
112.6
112.7
112.8
113.3
115.3
115.6
115.7
115.8
117.9
118.3
118.3
118.5
120.5
+ 20.5
Wet
Scrubbers
100.0
100.5
102.0
102.5
102.9
103.3
103.7
104.1
106.0
106.4
106.8
107.2
107.6
108.0
109.4
109.8
110.0
110.1
110.3
110.4
112.1
+ 12.1
-------
20
Table 3. Marshall and Swift Equipment and
Chemical Engineering Plant Indexes: 1989 - 1st Qtr.19946'7
Year
1989
1990
1991
1992
1993
1994
Quarter
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Second
Third
Fourth
First
Overall Change (%) :
Index Value
M & S
884.7
894.7
897.0
903.9
905.8
912.2
917.9
924.5
925.9
928.6
935.1
932.9
932.9
943.5
949.7
946.1
952.4
966.6
966.9
970.8
980.3
+ 10.8
CEPI /Equipment
388.1
391.8
392.4
391.8
388.9
391.3
393.6
395.1
396.0
397.5
397.8
396.5
394.7
393.1
390.2
390.8
392.1
393.9
394.9
396.1
400.4
+ 3.2
-------
21
given quarter. Over this five-year period, the M&SI increased by
10.8%. This increase is higher than the price increases for
three of the seven devices. More importantly, the M&SI increase
is quite close to the average price increase, 10.4%. Good as
this agreement is, the fact remains that the M&SI increase does
not reflect the wide (-0.7 to +20.5%) range in the device price
changes.
The "Equipment" component of the CEPI grew by a much smaller
amount (3.2%) than did the M&SI. This change in the CEPI was
much lower than the cost increases for seven of the eight
devices, and higher than the change registered by the eighth
(refrigeration systems, - 0.7%). Clearly, the CEPI was a poor
barometer of control device price changes, at least during this
five-year period.
What can we learn from these comparisons? If these two
indexes are representative, we can conclude that a published
index might track the average price of control equipment
reasonably well. However, because device price histories are so
diverse, no published index alone can provide an acceptable
indicator of the price changes for a given type of air pollution
control equipment.
Results: Component Price Data
As mentioned above, we received enough data from vendors to
develop equipment cost indexes (ECI) for eight control device
categories. These included the seven for which we have compiled
historical price data plus carbon adsorbers. The averages of the
reported component price percentages (factors) for these devices
are listed in Appendix A. Here, we will focus on one of the
categories—gas absorbers—as a "case study" and examine it in
detail. We picked gas absorbers for three reasons: (l) they are
widely used for gaseous emissions control, (2) they are complex
devices with many components, and (3) by virtue of the OAQPS
Control Cost Manual8 chapter on gas absorbers, there is thorough,
timely background material available for sizing and costing them.
We will continue to focus on gas absorbers as we progress through
this report.
First, consider the component price data, as shown in Table
4. Of the dozen components listed, the absorber column accounts
for the largest fraction (0.17 to 0.257) of the equipment price
for all three sizes. At approximately 0.11 and 0.10, the next
largest contributors are the "instrumentation & controls" and the
"system fan," respectively. The remainder of the price is more
or less equally divided among the other nine components. Note,
however, that these fractions vary among the small, medium, and
large sizes. This is especially true for the "absorber column,"
"packing," and "other" components. Similar variations occur for
the other seven device categories. (See Appendix A.)
-------
22
Table 4. Average Reported Gas Absorber Component Price Factors
Component
Absorber columnd
Packing
Platforms and ladders
Internal piping
Recirculation tank
Pumps
Dampers & valves
Mist eliminator
System fan
Fan motor & drive
Instrumentation & controls
Other6
Totals:
Fraction of List Price
Small11
0.235
0.090
0.040
0.060
0.085
0.090
0.060
0.045
0.100
0.070
0.110
0.015
1.000
Mediumb
0.257
0.067
0.043
0.050
0.100
0.087
0.043
0.043
0.093
0.053
0.107
0.057
1.000
Largec
0.170
0.065
0.065
0.065
0.080
0.085
0.050
0.065
0.115
0.070
0.110
0.060
1.000
a Capacity: < 2,000 acfm. However, one vendor specified an
upper limit of 10,000 acfm.
b Capacity: 2,000 - 40,000 acfm. However, one vendor
specified a range of 10,000 - 40,000 acfm.
c Capacity: > 40,000 acfm.
d Including: gas inlet and outlet ports; liquid inlet port
and outlet port/drain; liquid distributor/redistributor; sump
space; and, as applicable, trays or packing support plates.
e Includes "engineering".
-------
23
Constructing the Equipment Cost Indexes
The component price factors comprise one of the two sets of
inputs needed to construct the equipment cost indexes (ECI). The
second set are the appropriate Producer Price Indexes (PPI). As
explained in Section I, the PPI are compiled and published
monthly by the Bureau of Labor Statistics. Of the hundreds of
PPI, certain ones correspond to the components comprising the
control devices. How were the "right" PPI selected? First, we
reviewed the lists of control device components and attempted to
find matches among the PPI. For example, the "pumps" component,
which pertains to gas absorbers and wet scrubbers, corresponds to
PPI #3561-13, "Centrifugal pumps". We were able to find PPI for
most of the components in this way. However, there were no
published PPI for several components. For the latter, we used
surrogates.
For instance, as a surrogate for the column component in the
gas absorbers list, we used the PPI for "Laminated plastic plate,
sheet, and profile shapes" (PPI # 3083-1). Our reasoning was
that, as most absorber columns are fabricated of FRP (fiberglass-
reinforced plastic), the price of these columns would follow
changes in the price of laminated plastic plate. This is a
reasonable assumption, given that absorber vendors typically
fabricate their vessels, mist eliminators, recirculation tanks,
and other major components, while they usually purchase their
pumps, fans, piping, and other auxiliaries. Although there are
exceptions, most control device vendors operate in such a
fashion. That is, they fabricate the major equipment items,
purchase the smaller components from other vendors, and assemble
all of them into a finished control system.
Table 5, which lists the PPI and other data used to
construct the gas absorbers ECI, reflects this reasoning. Notice
that Table 5 also contains the various index values for the first
and second quarters of 1994. In the case of the PPI, these
values have been obtained by simply averaging the indexes for the
months of each quarter. (However, the quarterly BLS Employment
Cost Index values are listed as published.) Of the indexes
listed, all have increased, except for "Centrifugal pumps," which
dropped, and "Unsupported plastic profile shapes, rods, &
tubes/polyethylene," which did not change.
The equipment cost index for gas absorbers was constructed
by "marrying" the data in Tables 4 and 5, in the following
manner. First, we selected special quantities—the equipment
cost weighting factors. Each equipment cost weighting factor
denotes the fraction of the total equipment cost that a given
component contributes. Thus, the equipment cost weighting
factors are identical to the component price factors listed in
Table 4 and Appendix A. Referring again to Table 4, the
weighting factors for the absorber column would be 0.235, 0.257,
-------
24
Table 5. Indexes Used In Contracting the
Gas Absorbers Equipment Cost Index
Component
Absorber column
Packing
Platforms &
ladders
Internal piping
Recirculation tank
Pumps
Dampers & valves
Mist eliminatorb
System fan
Fan motor & drive
Instrumentation &
controls
Other
(engineering)
Producer Price Index
(number)
Laminated plastic plate,
sheet, and profile shapes
(3083-1)
Unsupported plastic
profile shapes, rods, &
tubes /polypropyl ene
(3082-105)
Metal ladders, including
ladder accessories
(1089-0557)
Plastic pipe (3084)
(Same as absorber column)
Centrifugal pumps
(3561-13)
Air-conditioning ducts,
including dust -collecting
(3444-637)
(Same as absorber column)
Centrifugal blowers &
fans (3564-3)
Universal motors, A.C.&
B.C. (3621-12)
Process control
instrumentation (1182)
Employment Cost Index
(professional specialty &
technical occupations)
Index Value
First8
129.4
97.5
98.0
109.2
129.4
146.4
125.7
129.4
130.4
100.3
139.2
124.6
Second
130.5
97.5
98.8
106.5
130.5
147.3
127.1
130.5
130.0
102.6
139.9
125.3
1994.
These are the index values for first and second quarter
b Mist eliminators are installed by a vendor and sold with
the FRP columns. They also are fabricated from FRP. Other mist
eliminator designs employ stainless steel and other materials and
may be sold as stand-alone units. Prices of the latter would be
escalated via PPI different from those shown in this table.
-------
25
and 0.170, respectively, for the "small," "medium," and "large"
units.
Next, we multiplied each weighting factor by the ratio of
the PPI or Employment Cost Index for the most recent quarter by
the corresponding index value for the base quarter (first quarter
1994). Based on the Table 5 data, this ratio for absorber
columns would be 130.6/129.4 = 1.009, or a 0.9% increase. The
products of the weighting factors and index ratio for absorber
columns would be 0.237 (small), 0.259 (medium), and 0.172
(large).
Lastly, we summed these weighting factor-index ratio
products and multiplied the sum by 100 to obtain the ECI for the
period in question (second quarter 1994). For gas absorbers, the
ECI were 100.51 (small), 100.53 (medium), and 100.48 (large). In
this instance, the ECI values were quite close.
Tables 6 to 9 show the ECI for the eight control devices for
second, third, and fourth quarter 1994 and first quarter 1995,
respectively. Except for first quarter 1995 (which are
preliminary), all the ECI's are final values. In each table are
listed the ECI's for the "small," "medium," and "large" device
size ranges, as well as the "averages" (arithmetic means) of
these three. Also, to give an indication of how the ECI varies
according to device size category, each table includes a column
labeled "Range/Average x 100%". This column lists the ratio of
the ECI range (highest - lowest) and the average ECI (minus 100,
the baseline ECI), all times 100%. For example, in Table 9, the
highest, lowest, and average ECI's for carbon adsorbers are
106.72, 104.87, and 106.02, respectively. The range/average
(R/A) ratio is, therefore:
R/A ratio = [(106.72 - 104.87)7(106.02 - 100)] x 100%
= 30.7%.
The next highest R/A ratio in Table 9 is 26.8% (wet
scrubbers), though the others are somewhat lower, ranging from
1.5 to 9.0%. The R/A ratios in Tables 6 to 8 behave similarly.
That is, one or two are large (> 20%), while the others are
lower. There is no discernible pattern to this behavior,
however. The ratios do not increase with time (i.e., the further
we get from the first quarter 1994 baseline). Nor are the ratios
consistently larger for some devices and smaller for others. The
only conclusion that we can draw is that for most (i.e., 6 of the
8) devices in a given quarter the R/A ratios are relatively
low—typically, < 20%. On the whole, this indicates a somewhat
c This is not a Producer Price Index, although it also is
compiled and published by the Bureau of Labor Statistics.
-------
26
Table 6. Control Device Equipment Cost Indexes:
Second Quarter 1994
Control Device
Carbon adsorbers
Catalytic
incinerators
Flares
Gas absorbers
Refrigeration
systems
Regenerative
thermal oxidizers
Thermal
incinerators
Wet scrubbers
Equipment Cost Index
(First Quarter 1994 = 100)
Small
100.95
100.96
100.69
100.51
100.41
100.84
100.76
100.59
Medium
101.21
100.89
100.74
100.53
100.41
100.86
100.77
100.58
Large
101.28
100.86
100.76
100.48
100.41
100.85
100.76
100.61
Average"
101.15
100.90
100.73
100.51
100.41
100.85
100.76
100.59
Range/
Average
x 100%b
28.7
11.1
9.6
9.8
0
2.4
1.3
5.1
• Arithmetic averages of the three ECI values.
b Calculated as follows:
R/A = {(High Index-Low Index)/jAverage index-100|} x 100%.
-------
27
Table 7. Control Device Equipment Cost Indexes:
Third Quarter 1994
Control Device
Carbon adsorbers
Catalytic
incinerators
Flares
Gas absorbers
Refrigeration
systems
Regenerative
thermal
oxidizers
Thermal
incinerators
Wet scrubbers
Equipment Cost Index
(First Quarter 1994 = 100)
Small
100.79
103.04
99.93
100.90
100.49
101.92
101.81
101.26
Medium
101.38
103.02
99.71
100.94
100.50
102.16
101.86
101.32
Large
101.43
103.03
99.62
100.96
100.45
102.16
101.82
101.43
Average'
101.20
103.03
99.75
100.93
100.48
102.08
101.83
101.34
Range/
Average
x 100%b
53.3
0.7
124.
6.5
10.4
11.5
2.7
12.7
11 Arithmetic averages of the three ECI values.
b Calculated as follows:
R/A = {{High Index-Low Index)/|Average index-100)} x 100%.
-------
28
Table 8. Control Device Equipment Cost Indexes:
Fourth Quarter 1994
Control Device
Carbon adsorbers
Catalytic
incinerators
Flares
Gas absorbers
Refrigeration
systems
Regenerative
thermal oxidizers
Thermal
incinerators
Wet scrubbers
Equipment Cost Index
(First Quarter 1994 = 100)
Small
101.80
103.99
101.12
100.87
100.83
102.04
102.63
101.96
Medium
102.84
104.02
101.08
100.94
100.85
102.38
102.68
102.18
Large
102.94
104.07
101.14
101.00
100.84
102.35
102.64
102.33
Average*
102.53
103.99
101.12
100.94
100.84
102.26
102.65
102.15
Range/
Average
x 100%b
45.1
2.0
5.4
13.8
2.4
15.0
1.9
17.2
* Arithmetic averages of the three ECI values.
b Calculated as follows:
R/A = {(High Index-Low Index)/(Average index-100|} x 100%.
-------
29
Table 9. Control Device Equipment Cost Indexes:
First Quarter 1995 (Preliminary)*
Control Device
Carbon adsorbers
Catalytic
incinerators
Flares
Gas absorbers
Refrigeration
systems
Regenerative
thermal
oxidizers
Thermal
incinerators
Wet scrubbers
Equipment Cost Index
(First Quarter 1994 = 100)
Small
104.87
105.35
105.53
103.62
102.21
103.25
104.06
106.37
Medium
106.48
105.45
105.45
103.85
102.23
103.56
104.07
107.45
Large
106.72
105.47
105.45
103.66
102.11
103.49
104.00
108.35
Averageb
106.02
105.42
105.48
103.71
102.18
103.44
104.04
107.39
Range/
Average
x 100%c
30.7
2.2
1.5
6.2
5.5
9.0
1.7
26.8
* Because the PPI inputs to the ECI have been denoted as
"unrevised" by the Bureau of Labor Statistics, these ECI values
should be considered preliminary, as well. Final ECI values will
be disseminated once the final first quarter 1995 PPI's become
available.
b Arithmetic averages of the three ECI values.
c Calculated as follows:
R/A = {(High Index-Low Index)/(Average index-lOOj} x 100%.
-------
30
Finally, it is interesting to compare the ECI values with
the Chemical Engineering (CEPI) and Marshall & Swift Indexes
(M&SI). The CEPI and M&SI values for 1994 are listed below,
along with the preliminary CEPI for first quarter 1995.
Date
1st Q '94
2nd Q '94
3rd Q '94
4th Q '94
1st Q '95
(prelim. )
Index Value
CEPI ("Equipment")
400.5
403.5
408.5
415.1
424.8
100.0
100.8
102.0
103.6
106.1
M&SI (Composite)
980.3
990.8
998.1
1004.4
Una vail .
100.0
101.1
101.8
102.5
Unavail .
The right-hand column under each index lists the ratio of
the index for each quarter and the first quarter 1994 index,
times 100. We can directly compare these values to the ECI
"averages" in Tables 6 to 9. Over these four quarters, the
CEPI/Equipment ratios compare fairly well with the average ECI's,
falling approximately in the upper-middle of the ranges. For the
second quarter 1994 (Table 6), the CEPI exceeds four of the
average ECI's and falls below the other four. In third quarter
1994, however, the CEPI is higher than six of the eight average
ECI's (Table 7). In Table 8 (fourth quarter 1994), only one of
the ECI (catalytic incinerators) exceeds the CEPI. Finally,
Table 9 (first quarter 1995) shows the CEPI again to exceed the
average ECI's of six of the eight devices. This agreement may be
due, in part, to the fact that both the CEPI and the ECI's are
based upon some of the same Producer Price Indexes (e.g.,
stainless steel).
The M&SI ratios, which have different bases, also agree
fairly well with the ECI's in the second, third, and fourth
quarters of 1994. Like the CEPI, the M&SI fall within the upper-
middle of the ECI ranges. The M&SI exceed seven of the ECI in
the second and third quarters, and five ECI in fourth quarter
1994. Again, because the M&SI was unavailable for first quarter
1995, no comparison could be made for that period.
Admittedly, the differences among the ECI and CEPI/M&SI are
not large, absolutely or relatively. Keep in mind, however, that
these comparisons are being made over just a single year (four
quarters). Longer-term comparisons could produce larger
discrepancies in some cases, smaller in others. For that reason,
we should resist the temptation to brand the CEPI or M&SI as a
"good" or "bad" surrogate for the ECI's, in general.
-------
31
Reflections on the ECI
One might question how the ECI, being constructed from
weighting factors and other published indexes, could accurately
track control device equipment prices. Suppose, for instance,
that the price of a thermal incinerator component—say,
centrifugal fans—increases by 23% during a given quarter, while
the prices of all other components do not change. Further
suppose that this component makes up (a representative) 10% of
the price of the incinerator. (See Appendix A.) Does this mean
that the incinerator vendor would raise his prices by precisely
2.3% (10% of 23%) to recover his costs? Not necessarily.
Depending on the current market, he could increase his prices by
this amount, a different amount, or not at all. If the thermal
incinerator market were especially "bearish," the vendor might
choose to absorb part or all of this boost in fan prices.
Conversely, if the demand for thermal incinerators were high, he
might increase his prices by more than 2.3%. In such a case, an
equipment cost index might provide less than an ideal escalator
of incinerator price changes.
On the other hand, rarely do prices change for one
component, and not for the others. As we saw in Table 5, some
prices increase, others decrease, and still others stay
unchanged. If a device is made of of several components, each
with a unique price history as chronicled by a PPI or another
index, there is a higher probability—a "central tendency," if
you will—that such a weighted ECI would track price changes more
closely than would another type of index. Interestingly, other
published indexes are based on such a weighting scheme. For
instance, as indicated above, the CEPI rests heavily upon a PPI
foundation. And these indexes are widely used by analysts and
cited in the literature.
What do equipment vendors think of the ECI methodology?
Some insist that the ECI cannot accurately escalate price
changes, for the reasons stated above. However, one wet scrubber
manufacturer supports the ECI methodology. He wrote: "It should,
therefore, be possible to arrive at historic market price data,
if the official price indexes were applied for the main
components of the system (manufactured stainless steel, motors,
pumps, instruments, etc.) and the weighted percentages in your
Table II be applied for each component."9 Clearly, there is
ample room for argument regarding the ECI.
Applying the Historical and Equipment Cost Indexes
How can the historical and equipment cost index data be used
to escalate control device costs? Actually, they are quite easy
to use, either separately or together, as the following example
illustrates. Consider a gas absorber with a first quarter 1989
equipment cost of $100,000. What would its cost be, if escalated
-------
32
to (1) third quarter 1991 and (2) second quarter 1994 dollars,
respectively?"
First, refer to Table 1, where a relative price of 106.8 is
given. Because the original cost is in first quarter 1989
dollars, the base value is 100.0. After substituting these
values into equation (1), we obtain:
Cost (3rd qtr.'91) = $100,000 x (106.8/100.0) = $106,800.
However, escalating the original cost to second quarter 1994
dollars requires two steps. First, escalate the equipment cost
to first quarter 1994 dollars via the historical price data in
Table 1. The corresponding index value is 112.2. Second,
escalate this cost from first to second quarter 1994 using the
ECI for gas absorbers, as given in Table 6. Here, we find the
average index value of 100.51, where the base value is 100.0
(first quarter 1994). We can combine both steps into one, as
follows:
Cost (2nd qtr.'94) = $100,000 x (112.2/100.0) x (100.51/100.0)
= $112,772 ~ $112,800.
In this calculation, note that the base value "100.0"
appears twice. This happens because, for convenience, 100.0 was
chosen as the base for both the historical and equipment cost
indexes.
" Escalation beyond second quarter 1994 would be ill-advised
in this case, as costs should not be escalated over periods
exceeding five years. (See previous discussion.) Instead,
current prices for the equipment should be obtained from vendors
or other equally reliable sources.
-------
33
. Total Annual Cost Indexes
Having developed escalation indexes for control device
equipment costs, the next logical step is to devise similar
indexes for the device total annual costs (TAG). However, there
are several reasons why we should not do so. First of all,
there is little need to escalate electricity, labor, and other
direct annual ("O&M") costs, as current values for these costs
are readily available from public and private sources.
Secondly, direct annual costs include a variety of
components. The costs of these components typically exhibit
vastly different escalation rates. For instance, the price of
some fuels, such as natural gas, have fluctuated considerably
over the years, when compared to, say, operating labor. No index
could begin to track price changes in so many different
components. Thirdly, unlike equipment costs, the prices of
utilities, labor, and other direct annual costs may exhibit
considerable regional variation. This would militate against
using a single national index for tracking prices. Instead,
regional indexes would be needed or, at least, adjustment factors
to account for these variations.
Finally, the composition of the annual costs also varies
according to the type of control device being costed. A thermal
incinerator, for instance, would have the following direct annual
costs: fuel (typically, natural gas), operating and supervisory
labor, maintenance labor and materials, and electricity. Of
these, the fuel cost would predominate. By contrast, the O&M
cost for a carbon adsorber would cover labor, maintenance, and
electricity, as well as steam, cooling water, and carbon
replacement. To escalate these O&M costs, we would need to
develop and use a different annual cost index for each type of
control device. The effort required to do that might be better
spent by obtaining current prices for the various O&M costs.
Despite all this, there may be cases where there is not
enough time or information to collect current annual costs. To
meet these needs, we have developed total annual cost indexes
(TACI) to complement the equipment indexes presented in Section
II. In this section, we will present a methodology for
developing the TACI and apply this methodology to their creation.
Nevertheless, the TACI are to be used only when no other reliable
annual cost data are available.
Control Costs: A Refresher
First, let's review cost terminology. As Chapter 2 of the
Manual explains, the total annual cost is comprised of three
major elements:
-------
34
TAG = DC + 1C - RC
(2)
where: DC = sum of direct annual costs ($/year)
1C = sum of indirect annual costs ($/year)
RC = sum of recovery credits ($/year)
The table below shows the composition of these elements
TAG Element
Direct Annual Costs (DC)
Indirect Annual Costs (1C)
Recovery Credits (RC)
Costs/Credit Included
Labor (operating, supervisory)
Maintenance (labor, materials)
Utilities (electricity, etc.)
Operating materials
Replacement parts
Waste treatment/disposal
Overhead
Capital recovery
Taxes, insurance, admin, chgs.
Recovered materials
Most of the costs in this table are functions of one or more
stream variables, facility parameters (e.g., annual operating
hours), financial parameters (annual interest rate, etc.), and/or
other costs, such as the total capital investment (TCI) of the
control system. From a cost-escalating standpoint, the single
most important parameter is the gas volumetric flowrate
(ftVmin)f as it is the primary determinant of the control device
size and, in turn, its equipment cost and total capital
investment. Although other stream parameters can also be
important—stream inlet temperature, waste gas heat content,
pollutant loading, etc.—it is sufficient to focus on just the
gas flowrate.
The second critical parameter is the operating factor
(annual operating hours). All of the direct annual costs plus
the recovery credits vary linearly with the operating factor.
The overhead does as well, because it is factored from the total
labor and maintenance cost.
Lastly, some costs—such as the operating, supervisory, and
maintenance labor and maintenance materials—are not functions of
any stream or facility parameters or the TCI, though they depend
upon the operating factor. Rather, these costs are more or less
constant, regardless of the system size. The table below lists
-------
35
the annual costs and the variables they primarily depend upon.
Main Determinant Variables of Annual Cost Elements
Annual Cost
Operating labor
Supervisory labor
Maintenance labor & materials
Utilities (electricity, etc.)
Operating materials
Waste treatment/disposal
Replacement parts
Overhead
Taxes , insurance , administrative
Capital recovery
Recovery credits
Main Determining Variable (s)
Operating factor
Operating factor
Operating factor
Gas flowrate, operating factor
Gas flowrate, operating factor
Gas flowrate, operating factor
Part cost, life, interest rate
Total labor + maintenance cost
Total capital investment
Total capital investment
Gas flowrate, operating factor
Note in this table that the "replacement parts" and all of
the indirect costs except overhead depend on variables other than
the gas flowrate or the operating factor. However, it can be
shown that even these costs depend (albeit indirectly) on the gas
flowrate, as well. The TCI, for instance, is normally calculated
by multiplying a factor by the purchased equipment cost (PEC).
But the PEC, in turn, is primarily a function of the gas
flowrate. Moreover, the initial costs of parts—fabric filter
bags, carbon adsorber carbon, etc.—are also functions of the gas
flowrate. Again, the gas volumetric flowrate is the main
variable of interest.
However, we must not forget that the labor, maintenance,
utilities, and other direct annual costs also depend upon the
unit costs for these expenditures (i.e., $/hour, $/kWhr, etc.).
Although these unit costs vary from site to site, based on
geography, industry, and other factors, they are not nearly as
variable as the gas flowrate or the operating factor. The
operating factor might range over an order of magnitude, while
the gas flowrate may vary by a factor of 100 or more. Rarely do
the unit costs exhibit such variability.
Because the various annual costs depend on the gas flowrate
and/or the operating factor to different extents, the proportion
that each cost contributes to the total annual cost will also
depend on these two parameters. The costs of utilities,
-------
36
operating materials, waste treatment/disposal, and the recovery
credits all vary linearly with both the gas flowrate and the
operating factor. For instance, if either parameter increased by
50%, each of these costs would rise by 50%, assuming that all
other parameters remained constant. On the other hand, a 50%
increase in the gas flowrate might only cause a 25 to 30% jump in
the total capital investment, as the TCI's dependence on the
flowrate is weaker. By the same token, the capital recovery and
taxes, insurance, and administrative costs—both of which are
factored from the TCI—also would increase by 25 to 30%.
Total Annual Cost "Weighting Factors"
How does this discussion relate to the escalation of the
total annual cost? First, it shows that the TAG cannot be
escalated using a single index, as is the equipment cost (and, in
turn, the TCI). We can use a single escalator with the TCI
because, as we have shown in Section II, the price of a
refrigeration system, wet scrubber, or other device increases (or
decreases) by about the same percentage over a quarter regardless
of the size of the unit. Not so with the TAG, because: (1) each
of the costs comprising the TAG has an individual price history
and (2) the contribution of each cost to the TAG will vary
according to the gas flowrate and/or the operating factor.
To account for both the "cost mix" and the "cost
contributions" we could derive complex equations that incorporate
the gas flowrate, operating factor, and other parameters;
calculate the TAG for any given set of parameters; and present a
TAG escalation index for every flowrate from 1 to 1,000,000
ft3/min. and operating factor from 8 to 8000 hours/year. But
doing so would be not only cumbersome, but misleading, as it
would imply that cost escalation is a very precise procedure,
when, in reality, it is as much art as science. On the other
hand, to present a single escalation index for the TAG would be
equally misleading, as the TAG components do vary appreciably
according to the gas flowrate and operating factor.
There is a middle ground, however. We can compute and
present six TAG escalation indexes for each control device
type—one each for "small," "medium," and "large" gas flowrates
at both "low" and "high" operating factors. As we might expect,
both the flowrate and operating factor ranges would vary
according to the device. For instance, the gas flowrates used
with thermal incinerators were 250 (small), 25,250 (medium), and
100,000 standard ft3/minute (large). (Corresponding flowrates
for gas absorbers were 1000, 22,300, and 80,000 actual
ft3/minute.) Also for each flowrate, cost indexes were
calculated for 800 and 8,000 hours/year, the low and high
operating factors for thermal incinerators. As was done with the
equipment cost indexes, the TAG indexes for the small and medium
-------
37
cases were calculated using the midpoint of the size (flowrate)
ranges, while for the large case, twice the lower end of the
range was used.
In calculating the TACI, we used special quantities—the TAC
weighting factors, which are analogous to the equipment cost
weighting factors presented in Section II. Like their ECI
counterparts, each TAC weighting factor denotes the fraction of
the total annual cost that a given cost component contributes.
If the natural gas cost for a thermal incinerator comprises 75.4%
of the TAC at a given gas flowrate and operating factor, the
natural gas weighting factor would be 0.754. However, for a
different flowrate and/or operating factor, the weighting factor
could be much different. Clearly, the weighting factors for the
other annual costs also would change with changes in these two
parameters.
How were the TAC weighting factors used in computing the TAC
indexes? Before we answer that question, recall that each of the
annual costs changes in a different way. That is, the price of
electricity might increase by 4% in a given year, while the
natural gas price might rise by 2%. And in certain situations,
the price may actually decline." As discussed in Section I,
price changes are commonly reported in terms of indexes, of which
the BLS Producer Price Indexes are among the most familiar.
Recall that each of the thousands of PPI's has a "base
period" (month-and-year) for which the index value is arbitrarily
set at "100". Prices in future periods are expressed relative to
this base of 100. For example, if the "fabricated metals" PPI
were, say, 107.3 for January 1994, the price in that month would
be 7.3% higher (i.e., {[107.3/100]-!} x 100%) than it was in the
base year and month. Now, as explained in Section II, it is
often convenient to express price changes between two periods as
the ratio of the indexes for the two periods of interest. This
allows us to make price comparisons without knowing either the
index base period or base value. If the "fabricated metals" PPI
for November 1993 were 106.8, for example, the relative change in
the index would have been 0.0047 ([107.3/106.8]-!), or 0.47%,
between that month and January 1994. The following equation
expresses the relationship between the indexes for the two
periods and the base period index:
' This is hardly unprecedented. During 1993, the U.S.
Bureau of Labor Statistics reported price declines for several
commodities, including fabricated steel plate, tranformers, and
sodium hydroxide (caustic soda). The last registered a 23% drop.
-------
38
(W/tW = Ijj/Ij, (3)
where: IJ2 = cost "j" index value for period 2.
Ij! = cost "j" index value for period 1.
Ijb = cost "j" index value for base period.
Notice that Ijb cancels out in this equation, proving that the
change in an index is independent of the base index.
Furthermore, we can show that the change in a total annual
cost index, consisting of "n" cost components, can be expressed
in terms of (1) the weighting factors and (2) the changes in the
corresponding indexes of the component costs. That is:
TACIf/TACIj = E (W.F.Jjdjf/Iji) (4)
= (W.F.) !(!,,/!„) + (W.F.)2(I2f/I2i) + ...
where: TACI = total annual cost index for final (f) and
initial (i) periods, respectively
W.F. = cost weighting factor for cost l,2,...n
A TACI Illustration
To illustrate, again consider gas absorbers. Table 10 lists
weighting factors for the "medium" capacity gas absorber (22,280
scfm) operated at the "high" operating factor (8,000 hours/year).
These weighting factors were computed via the cost spreadsheet
programs to be described in Section IV. Also shown in Table 10
are the Employment Cost Index and Producer Price Indexes for the
respective total annual cost components corresponding to the
first and second quarters of 1994, all obtained from the BLS.a
Note that five of the six costs increased during this three-
month period, while one, "chemicals" (caustic soda) actually
decreased. The index ratio for the "capital recovery; taxes,
insurance, & administrative" component is actually the ratio of
the equipment cost index for medium-capacity gas absorbers from
Table 6. Because the base period for the ECI is first quarter
1994, the index value is 100.0. (See "Equipment Cost Index"
discussion in Section II.) If we substitute these index ratios
and corresponding weighting factors into equation 4, we obtain
the total annual cost index ratio:
* A quarterly index value is simply the arithmetic average
of the monthly PPI's for that quarter.
-------
39
Table 10. TACI Illustration: Gas Absorbers
Cost Component
Labor, maintenance,
& overhead
Electricity
Water
Chemicals
Wastewater
treatment
Capital recovery;
taxes, insurance, &
administrative
Index Value
1st Q '94
120.2
126.0
126.0
92.9
126.0
100.0
2nd Q '94
121.2
128.0
128.0
83.0
128.0
100.51
Index
Ratio
(^oAuo)
1.008
1.016
1.016
0.893
1.016
1.0051
Weight .
Factor
(W.F.j)
0.101
0.041
0.002
0.723
0.031
0.102
-------
40
TACI2Q/TACI1Q = (0.101) (1.008) + (0 . 041) (1. 016) + (0 . 002) (1. 016) +
(0.723) (0.893) + (0.031) (1.016) + (0.102) (1.0051)
= 0.9255
In other words, based on the component index ratios and
weighting factors, the (predicted) decrease in the total annual
cost index for this device would be 7.45%! Even though five of
the six cost components increased—three (electricity, water, and
wastewater treatment) by nearly 2% each—the total annual cost
index dropped precipitously. Why? Simply because the chemicals
cost—which comprised 72.3% of the TAG—fell, by nearly 11%.
This decrease more than offsets the increases in the other
components.
Suppose that, for expediency's sake, we had just averaged
the changes in the cost components without first weighting them?
(In effect, this would be assigning each component an equal
weighting factor of 1/6 = 0.167.) How much would the TACI have
changed then? We can get the answer simply by averaging the
index ratios (column four in Table 10). The result would have
been 0.9924—somewhat higher than the calculated result above.
To say the least, the cost weighting factors are important to the
escalation of total annual costs.
This example shows how sensitive the total annual cost (and
TACI) can be to component unit prices. Normally, chemical prices
do not change by so much over such a relatively short period.
But during 1994, caustic soda (NaOH) prices have not been
behaving normally. In fact, the NaOH market has been very
unsettled and, consequently, hard to gauge. First, as the table
above shows, the PPI for NaOH dropped from 92.9 to 83.0 from
first to second quarter 1994, a 10.6% decrease.
However, the weekly Chemical Marketing Reporter (CMR), an
oft-cited source of chemical prices, has been telling quite a
different tale. A May 28, 1994, article entitled "Caustic Surge
Leads to Potent Price Hike" reads: "Strong demand and a tighter-
than-anticipated supply of caustic soda have led U.S. producers
to push prices up even further...Spot prices for Gulf Coast
caustic are pegged at $120 to $130 per ton...In January, caustic
sold for $20 to $25 per ton...."10 In other words, according to
the CMR, caustic prices have not been dropping, as the BLS
claims, but skyrocketing by factors of 5 to 6. But before we
dismiss the BLS data, we should examine the price listing for
NaOH in the CMR. For the week ending January 7, 1994, the CMR
reported a price of $300 to $330/ton for "76% Na20, F.O.B. Gulf
Coast". For the week ending January 14, the price remained the
same—and stayed the same for each and every week through the end
of June (the end of the second quarter). Thus, if one did not
read the CMR article, one might conclude that caustic prices were
constant throughout both quarters of 1994.
-------
41
With this kind of price confusion, it is no wonder that cost
escalation—especially TAG escalation—is often uncertain. To
quantify the uncertainty in this illustration, we twice reran the
gas absorber total annual cost program to generate two new sets
of TAG weighting factors. The first time we assumed no change to
the NaOH price (i.e., we took the CMR price listing at face
value). For the second rerun, we deleted the NaOH (chemicals)
cost from the list of TAG components, assuming, in effect, that
no caustic would be needed. The results for the gas absorber
described above were as follows:
Gas Absorber TACI for Different NaOH Price Scenarios
NaOH Price Basis:
Producer Price Index
Chemical Marketing Reporter
No NaOH
Total Annual Cost Index x
100
92.55
100.26
100.92
These differences are significant, to be sure, and require
no comment.
Keep in mind, of course, that these TACI values would
pertain only to a medium-capacity gas absorber operating at 8,000
hours/year. As we will see in Section IV, the weighting factors
and resulting TACI would be significantly different for gas
absorbers of other capacities and operating factors.
TACI for All the Control Devices
Tables 11 to 14 list the total annual cost indexes for all
eight control devices, pertaining to second, third, and fourth
quarter 1994 and first quarter 1995, respectively. With the
exception of the first quarter 1995 indexes (which are
preliminary), all of the TACI are final values. For each device,
the TACI are given for three sizes ("small," "medium," and
"large") and two operating factors ("high" and "low"). The sizes
are listed in Tables A-l to A-8 of Appendix A. For all but three
devices, the low and high operating factors are 800 and 8,000
hours/year, respectively. The exceptions are carbon adsorbers,
flares, and refrigeration systems. Their operating factors are
listed in a footnote in each of these tables.
As we might expect, the TACI vary noticeably among the three
unit sizes and two operating factors. The differences, however,
are larger for some devices than for others. This is especially
true for gas absorbers. For example, for second quarter 1994,
the TACI ranges from 91.65 (large unit, high operating factor) to
99.29 (small, low). Contrast these values with the third quarter
-------
42
Table 11. Total Annual Cost Indexes: Second Quarter 1994
Control
Device
Carbon
adsorbers
Catalytic
incinerators
Flares
Gas
absorbers0
Refrigeration
systems
Regenerative
thermal
oxidizers
Thermal
incinerators
Wet scrubbers
Total Annual Cost Indexes
(First Quarter 1994 = 100)
Small"
Low
O.F.b
100.98
100.86
100.02
99.29
100.49
100.84
100.72
100.76
High
O.F.
100.95
100.74
99.51
97.60
100.63
100.85
100.55
100.83
Medium
Low
O.F.
101.38
100.65
99.40
96.27
100.59
100.78
99.92
100.84
High
O.F.
100.84
100.01
98.84
92.55
100.56
101.09
98.51
101.05
Large
Low
O.F.
101.47
100.56
99.15
95.57
100.51
100.87
99.45
100.90
High
O.F.
100.73
99.79
98.68
91.65
100.71
101.02
98.18
101.24
* These capacities are listed in Tables A-l through A-8
(Appendix A).
b Operating factors (O.F.) are 800 (low) and 8,000 (high)
hours/year, with the following exceptions:
Carbon adsorbers—864, 8,640;
Flares 876, 8,640;
Refrigeration systems—520, 2,080.
c TACI are based on assumption of NaOH (chemicals) price
based on PPI data. (See discussion above.)
-------
43
Table 12. Total Annual Cost Indexes: Third Quarter 1994
Control
Device
Carbon
adsorbers
Catalytic
incinerators
Flares
Gas
absorbers0
Refrigeration
systems
Regenerative
thermal
oxidizers
Thermal
incinerators
Wet scrubbers
Total Annual Cost Indexes
(First Quarter 1994 = 100)
Small"
Low
O.F.b
101.10
102.78
101.94
103.49
100.82
101.93
101.70
101.74
High
O.F.
101.78
101.97
102.85
105.77
101.38
102.01
101.24
101.96
Medium
Low
O.F.
102.15
102.61
102.76
106.64
101.01
102.16
99.67
102.48
High
O.F.
102.97
100.76
103.66
112.08
101.63
103.12
96.04
103.40
Large
Low
O.F.
102.28
102.52
103.01
108.00
100.96
102.30
98.61
102.82
High
O.F.
103.14
100.34
103.65
113.22
102.03
103.25
95.19
104.10
a These capacities are listed in Tables A-l through A-8
(Appendix A).
b Operating factors (O.F.) are 800 (low) and 8,000 (high)
hours/year, with the following exceptions:
Carbon adsorbers—864, 8,640;
Flares—876, 8,640;
Refrigeration systems—520, 2,080.
c TACI are based on assumption of NaOH (chemicals) price
based on PPI data. (See discussion above.)
-------
44
Table 13. Total Annual Cost Indexes: Fourth Quarter 1994
Control
Device
Carbon
adsorbers
Catalytic
incinerators
Flares
Gas
absorbers6
Refrigeration
systems
Regenerative
thermal
oxidizers
Thermal
incinerators
Wet scrubbers
Total Annual Cost Indexes
(First Quarter 1994 = 100)
Small"
Low
O.F.b
101.99
103.43
101.83
107.70
100.99
102.04
102.40
101.92
High
O.F.
102.06
101.99
102.05
114.41
101.27
102.03
101.49
101.91
Medium
Low
O.F.
103.58
103.31
102.14
118.29
101.03
102.25
99.82
101.96
High
O.F.
102.96
100.14
102.36
134.40
101.00
102.28
95.09
101.79
Large
Low
O.F.
103.48
103.19
102.17
122.09
100.90
102.32
98.51
102.01
High
O.F.
102.88
99.45
102.29
137.99
101.04
102.13
94.03
101.84
' These capacities are listed in Tables A-l through A-8
(Appendix A).
b Operating factors (O.F.) are 800 (low) and 8,000 (high)
hours/year, with the following exceptions:
Carbon adsorbers—864, 8,640;
Flares 876, 8,640;
Refrigeration systems—520, 2,080.
c TACI are based on assumption of NaOH (chemicals) price
based on PPI data. (See discussion above.)
-------
45
Table 14. Total Annual Cost Indexes: First Quarter 1995
(Preliminary)'
Control
Device
Carbon
adsorbers
Catalytic
incinerators
Flares
Gas
absorbers'1
Refrigeration
systems
Regenerative
thermal
oxidizers
Thermal
incinerators
Wet scrubbers
Total Annual Cost Indexes
(First Quarter 1994 = 100)
Smallb
Low
O.F.C
104.69
104.70
102.21
115.21
102.26
103.23
103.74
103.93
High
O.F.
103.50
102.87
100.29
126.90
102.35
103.21
102.45
102.83
Medium
Low
O.F.
106.76
104.36
100.67
135.67
102.32
103.40
100.72
104.85
High
O.F.
103.93
100.71
98.98
165.07
102.10
103.26
95.16
102.78
Large
Low
O.F.
107.09
104.11
100.05
142.44
102.09
103.43
99.19
105.34
High
O.F.
103.94
99.88
98.75
171.84
102.04
103.05
93.95
102.87
a Because the PPI inputs to the TACI have been denoted as
"unrevised" by the Bureau of Labor Statistics, these TACI should
be considered preliminary, as well. Final TACI will be
disseminated once the final first quarter 1995 PPI's become
available.
b These capacities are listed in Tables A-l through A-8
(Appendix A).
c Operating factors (O.F.) are 800 (low) and 8,000 (high)
hours/year, with the following exceptions:
Carbon adsorbers—864, 8,640;
Flares 876, 8,640;
Refrigeration systems—520, 2,080.
d TACI are based on assumption of NaOH (chemicals) price
based on PPI data. (See discussion above.)
-------
46
1994 TACI (Table 12). Here, the range is 103.49 (small unit, low
operating factor) to 113.22 (large, high). This steep increase
in the TACI—an increase that continues through the next two
quarters—is almost solely due to the price of caustic soda,
which, according to the Producer Price Index, jumped 110% from
second quarter 1994 to first quarter 1995. Except for second
quarter 1994, the gas absorbers TACI is seen to increase with
increasing size and operating factor.
The indexes for the other devices do not increase as much as
the gas absorbers TACI. Still, for two devices, patterns do
emerge. For catalytic and thermal incinerators, the TACI
generally decrease with increasing size and operating factor.
This reflects the influence of the fuel (natural gas) cost
contribution. As both the incinerator size and operating factor
increase, the natural gas contribution to the total annual cost
also increases. According to the BLS, the natural gas price
steadily decreased from first quarter 1994 to first quarter 1995.
The price drop during this twelve-month period was 8.9%.
Accordingly, the TACI for the largest incinerators at the higher
operating factor decreased. For example, the thermal
incinerators low-end TACI dropped from 98.18 (second quarter
1994) to 93.95 (first quarter 1995). Conversely, the TACI for
the smallest incinerators at the lower operating factor
increased, reflecting the greater influence of the equipment
cost. And the equipment costs grew during the past year,
primarily because of increases in steel prices.
For the other five devices, the TACI exhibit no consistent
pattern, although for some there are significant differences
among sizes and operating factors. The TACI differences are
largest for carbon adsorbers, flares, and wet scrubbers. This is
especially noticeable for first quarter 1995 (Table 14). Here,
the TACI ranges for carbon adsorbers, flares, and wet scrubbers
are 3.6%, 3.5%, and 2.6%, respectively. Finally, the TACI for
refrigeration systems and regenerative thermal oxidizers are
essentially equal for all sizes and operating factors.
-------
47
IV. Total Annual Cost Spreadsheet Programs
As we discussed in Section III, control costs depend upon a
variety of emission stream, control device, and financial
parameters. Often this dependency is quite complex, as in the
sizing and costing of gas absorbers. Clearly, it would be
cumbersome to make these calculations by hand, especially if
costs were needed for a range of input parameters (e.g., waste
gas flowrate). To make the calculation of the capital and annual
costs and the TAG weighting factors easier, we developed 12
spreadsheet programs, one for each of the eight devices covered
in this report, plus four others that we thought would be helpful
to cost analysts. We wrote these programs in Eight-in-One™, a
spreadsheet program that is compatible with Lotus 1-2-3™
(versions 1A and 2), is structured similarly, and utilizes many
of the same commands. Each program yields itemized total capital
investment and total annual costs and TAG weighting factors for a
given set of input parameters. We wrote programs for the
following control devices:
• Particulate emission controls:
/ Electrostatic precipitators
/ Fabric filters
/ Mechanical collectors (cyclones)
/ Venturi scrubbers8
/ Wet impingement scrubbers
• Gaseous emission controls:
/ Carbon adsorbers (fixed-bed regenerative)
/ Catalytic incinerators (fixed-bed)
/ Gas absorbers (packed-bed)
/ Flares
/ Refrigeration systems
/ Regenerative thermal oxidizers
/ Thermal incinerators (recuperative)
Most of the programs were based on design and cost data and
procedures in the OAQPS Control Cost Manual (Fourth Edition,
1990) (Manual). The exceptions were the programs for mechanical
collectors, venturi scrubbers, and wet impingement scrubbers.
These three were based on information in Estimating Costs of Air
Pollution Control (Lewis Publishers/CRC Press, 1990)." Those
interested in obtaining copies of the programs may contact the
author, William M. Vatavuk, at (919)-541-5309 (fax: 919/541-
0839) .
a Basis of the "wet scrubbers" device discussed in Sections
I to III of this report.
-------
48
A sample printout of the gas absorbers program appears in
Figure 5. Like the others, this program consists of six
sections: (1)"Cost Reference Date," (2) "Input Parameters," (3)
"Design Parameters," (4) "Capital Costs," (5) "Annual Cost
Inputs," and (6) "Annual Costs". As the name implies, the "Cost
Reference Date" section simply lists the date to which the costs
in the program pertain. This date ranges from fourth quarter
1986 (fabric filters) to third quarter 1991 (gas absorbers). If
the costs are updated, either directly or via the VAPCCI, this
date would change.
The "Input Parameters" section contains technical data that
must be entered by the user. The program needs these data to
compute the design parameters, the costs, or both. Because these
input parameters vary so much according to control device designs
and applications, there are no "default" values for them. In
other words, the user must provide this information.
Input parameters include such standard stream parameters as
the waste gas volumetric flowrate, temperature, and pollutant
loading—data that are common to all control devices. Another
standard parameter, the control device "pressure drop," also may
be a required input. Along with the gas flowrate, the pressure
drop is used to calculate the control device fan horsepower
requirement and, in turn, most (if not all) of the control system
electricity requirement. In many cases, the user must provide
the pressure drop. However, with other devices (e.g., thermal
and catalytic incinerators) the pressure drop is calculated from
one or more of the input parameters. With the latter devices,
the pressure drop would not be an input parameter.
The input parameters section also lists data specific to a
certain type of device. In the case of gas absorbers, the user
must enter various solvent data (molecular weight, density,
diffusivity, and viscosity) and packing parameters—data that can
be found in Chapter 9 of the Manual.
The next section, "Design Parameters," lists data that are,
for the most part, calculated by the program based on the input
parameters. For gas absorbers, these include operating line
data, pollutant-air equilibrium information, and such key
absorber parameters as the absorption factor, column height, and
pressure drop.
The "Capital Costs" section displays the control device
equipment cost (itemized), the purchased equipment cost, and the
total capital investment (TCI). No costs for auxiliary equipment
(e.g., ductwork and fans), are included, although some_"packaged"
devices, such as thermal incinerators, come equipped with some
auxiliaries. Both the purchased equipment cost (PEC) and TCI are
"factored" from the total equipment cost (TEC), according to the
procedure described in Chapter 2 of the Manual. The PEC is
-------
49
Figure 5. Sample Printout of Gas Absorbers Cost Program
TOTAL ANNUAL COST SPREADSHEET PROGRAM—GAS ABSORBERS C13
COST REFERENCE DATE: 3rd quarter 1991
INPUT PARAMETERS:
Stream parameters:
— Inlet waste gas flowrate (acfm): 1000
— Inlet waste gas temperature (oF): 100
— Inlet waste gas pressure (aim.): 1
— Pollutant in waste gas: Hydrogen chloride (HC1)
— Inlet gas poll, cone., yi Cmole fraction): 0.001871
— Pollutant removal efficiency (fraction): O.99
— Solvent: Aqueous caustic soda
— Inlet pollutant cone, in solvent: 0
— Waste gas molecular weight (Ib/lb-mole): 28.85
— Solvent molecular weight (lb/Ib-mole): 18
— Inlet waste gas density (Ib/ft3): 0.0706
— Solvent density (Ib/ft3): 62.4
— Solvent specific gravity: 1
— Uaste gas viscosity <$ inlet temp. (Ib/ft-hr): 0.044
— Splvent viscosity £ inlet temp. (Ib/ft-hr): 2.16
— Minimum wetting rate (ft2/hr): 1.3
— Pollutant diffusivity in air
-------
50
Figure 5. (cont'd.)
Column surface area (ft2):
Column gas pressure drop (in w.c./ft packing
Column liquid pressure drop (ft of H2Q):
Packing volume (ft3):
120.4
0.980
60
28.4
CAPITAL COSTS:
Equipment costs (*):
— Gas absorber
— Packing
— Total
Purchased Equipment Cost
Total Capital Investment
(*)
13,
14
17
841
568
409
002
37,405
ANNUAL COST INPUTS:
Operating factor
-------
51
figured at either 1.18 times the TEC if the control device cost
does not include instrumentation and controls or 1.08 times the
TEC if it does. The TCI is then factored from the PEC. The
factor used depends on the type of control device and whether it
is a packaged unit. For instance, the factor used with thermal
and catalytic incinerators is either 1.25 (packaged units) or
1.61 (custom units). In Figure 5, the TCI factor is 2.20, a
number that applies to all gas absorbers, because most of them
are custom-built devices.
In calculating the equipment cost for some devices, the
program utilizes two powerful features of the Eight-In-One™
spreadsheet program. These features are the "IF...THEN" and
"MAX" statements. Both have been employed in writing the Thermal
Incinerators program, a sample of which is shown in Figure 6.
First, note that the "primary heat recovery" must be entered as
an input parameter. In this case, the primary heat recovery is
0.70 (70%). Under the "Equipment Costs" portion of the "Capital
Costs" section, there are four lines for the incinerator cost,
one each for primary heat recoveries of 0, 35, 50, and 70%. A
different line is needed for each heat recovery because the
incinerator equipment cost increases with increasing heat
recovery. (For more information on this, consult Chapter 3 of
the Manual.) Note that only the 70% line contains a cost, while
each of the others shows "0". To accomplish this, each equipment
cost cell was given a formula, such as:
"IF(Dll=0.70,21342*D24A0.2500,0) " .
Translation: "If the primary heat recovery equals 70%, the
incinerator equipment cost equals 21,342 times the total gas
flowrate (D24) raised to the 0.2500 power. If the heat recovery
equals some other value, the cost equals 0." The formulas for
the other three cells were written similarly.
One might ask, "If there are four values for the incinerator
equipment cost—one number and three zeroes—how will the program
know which to use to compute the purchased equipment cost?" To
solve this problem, we entered the following formula in the PEC
cell:
"1.08*MAX(D30:D33)".
Translation: "The PEC equals 1.08 times the maximum value entered
in cells D30 through D33." This feature allows the program to
ignore the cells containing zeroes.
The fourth section, "Annual Cost Inputs," lists nearly all
of the parameters needed for the program to calculate the various
annual costs. As with the "Input Parameters" discussed above,
the user must provide all of these inputs (i.e., there are no
-------
52
Figure 6. Sample Printout for Thermal Incinerators Cost Program
TOTAL ANNUAL COST SPREADSHEET PROGRAM — THERMAL INCINERATORS
COST REFERENCE DATE: April 1988
INPUT PARAMETERS
— Gas flowrate (scfm):
— Reference temperature (oF)
— Inlet gas temperature (oF)
— Inlet gas density (Ib/scf)
— Primary heat recovery (fraction)
— Waste gas heat content (BTU/scf):
— Waste gas heat content (BTU/lb):
— Gas heat capacity (BTU/lb-oF):
— Combustion temperature (oF):
— Preheat temperature (oF):
— Fuel heat of combustion (BTU/lb):
— Fuel density (Ib/ft3):
DESIGN PARAMETERS
— Auxiliary Fuel Reqrmnt (Ib/min):
(scfm):
— Total Gas Flowrate (scfm):
CAPITAL COSTS
Equipment Costs (*):
— Inc inerator :
GO* heat recovery:
-------
53
default values). Nevertheless, typical values for the parameters
are given in the Manual chapter covering the device in question.
The first, and probably most important, annual cost input is
the "operating factor"—the hours per year the control device
operates. As discussed previously, it can range from less than
800 to more than 8000 hours/year, depending upon the emission
source that the device controls. This parameter determines all
of the operating and maintenance costs—labor, utilities,
materials, and waste disposal. The operating and maintenance
labor rates also depend on the source and its location. However,
according to guidance given in the Manual, the program computes
the maintenance labor rate by factoring it from the operating
labor rate (i.e., the maintenance rate is 10% higher than the
operating) and computes the maintenance materials cost at 100% of
the maintenance labor cost.
The operating and maintenance labor factors—both given in
hours/shift—vary according to device type, as some equipment
(such as scrubbers) typically requires more operator attention
and maintenance than others (e.g., incinerators). As a rule, the
smaller devices require less attention than the larger, a fact
that has been incorporated into these programs. For instance,
the operating and maintenance labor factors for thermal and
catalytic incinerators are each 0.5 hours/shift, except for
devices treating gas flowrates less than 20,000 standard cubic
feet per minute (scfm) . The utility prices—electricity, natural
gas, process and cooling water, steam, etc.—are self-
explanatory. However, as with the other annual cost inputs, the
user should be careful that they reflect the units shown on the
spreadsheet. The natural gas price, for example, must be
expressed in "$/thousand standard cubic feet," not "$/million
BTU" .
The "annual interest rate" and "control system life" are the
two required inputs to the "capital recovery factor" (CRF). When
the CRF is multiplied by the total capital investment (TCI), the
capital recovery cost results. In accordance with Office of
Management and Budget guidance, the annual interest rate should
be 7%. However, the user can vary this parameter to determine
its effect on the total annual cost. The control system life
typically varies from 10 years (scrubbers, gas absorbers, and
carbon adsorbers) to 20 years (ESPs, fabric filters, and
mechanical collectors). But, there may be situations when the
system life falls outside this range.
The CRF is computed according to the standard formula:
CRF = i(l+i)n/[(l+i)n-l] (5)
where: i = annual interest rate (decimal)
-------
54
n = control system life (years)
Finally, the "taxes, insurance, admin, [administration]
factor" is set at 4% of the total capital investment, consistent
with Manual guidance, although the user may vary this parameter
if he/she desires.
The "Annual Costs" are displayed in the next section of the
program. Appropriately, each cost is given in "dollars/year".
The "operating labor" and "maintenance labor" costs are each
computed from the operating factor, operating/maintenance labor
rate, and operating labor factor—all of which are "Annual Cost
Inputs". However, the "supervisory labor" is factored at 15% of
the operating labor cost. The 15% supervisory labor factor and
the overhead factor (60% of the sum of all labor and maintenance
materials costs) are embedded in the program calculation
formulas, rather than being listed as annual cost inputs.
However, as both factors are standard Manual values contained in
Chapter 2, there is little incentive to vary them.
The utilities costs—natural gas, electricity, water, steam,
etc.—are calculated via the appropriate input parameters, design
parameters, and/or annual cost inputs. The equations the program
uses to compute these costs are very similar, if not identical,
to those in the Manual. However, whenever the program calculates
a negative utility parameter (e.g., natural gas usage for a
regenerative thermal oxidizer), the annual cost for this item is
set at zero.
For certain devices—namely, carbon adsorbers, catalytic
incinerators, and fabric filters—there are also annual costs for
replacement parts—carbon, catalyst, and filter bags,
respectively. These costs are calculated by multiplying the part
cost (plus taxes and freight and the part replacement labor cost)
by a capital recovery factor that accounts for the interest rate
and the part expected life. The last two costs—"taxes,
insurance, administrative" and "capital recovery"—are obtained
by multiplying the total capital investment by the "taxes,
insurance, admin, [administrative] factor" and the "capital
recovery factor," respectively. Whenever there are replacement
parts costs, the capital recovery cost is factored from the total
capital investment less the cost of the replacement parts, the
applicable sales and freight taxes, and the cost of installing
the parts in the device. (See Chapter 2 of the Manual for more
guidance on this.)
When applicable, "Recovery credits" are deducted from the
total annual costs for carbon adsorbers and refrigerated
condensers. A credit can be quite significant, especially if the
recovered product has a high value. In some cases, it actually
can exceed the sum of the annual costs.
-------
55
Note that the annual cost section includes two more columns:
"Wt. factor [Weighting factor]" and "W.F. (cond.) [Weighting
Factor, condensed]". The first of these lists the fractional
contribution that each annual cost makes to the total annual cost
(without recovery credits). These fractions range from 0 to
greater than 0.97, depending on the various inputs. As discussed
in Section III, the weighting factors are used in computing the
Total Annual Cost Indexes (TACI).
Lastly, the "Weighting Factor [condensed]" column lists two
numbers, each of which has been devised to streamline the
calculation of the TAG indexes. The first "condensation" is the
sum of all the labor-related weighting factors: operating,
supervisory, and maintenance labor; maintenance materials; and
overhead. The second number combines the "taxes, insurance,
administrative" and "capital recovery" costs. Both condensations
appear in the weighting factor tables.
-------
56
V. Distributing the Indexes
Because so many cost analysts would benefit from using both
the equipment and total annual cost indexes it seems only right
that the indexes be disseminated as widely as possible. With
this in mind, we have devised a strategy for distributing the
"Vatavuk Air Pollution Control Cost Indexes" (VAPCCI). The
strategy has two elements: intra-governmental dissemination and
public domain distribution. This section discusses both
elements, in terms of what is planned—and what has been done—to
implement them.
Intra-governmental dissemination: This element includes all
practical means for informing EPA, State/local, and EPA
contractor users about the indexes and supporting information
(such as the TAG spreadsheet programs). The best means for
making such distribution is through the Emission Standards
Division's Control Technology Center (CTC). The CTC has the
capability of informing literally thousands of potential users
via its computer bulletin board, newsletter (CTC News) , and other
communication tools. After announcing the availability of the
indexes, the CTC then can mail hard copies of this report and/or
computer diskettes containing the various indexes to requesters.
In addition, the CTC's Technology Transfer Network (TTN) can be
used to electronically disseminate the indexes and supporting
information. Finally, via its "Hotline," the CTC can answer
questions about the VAPCCI or refer callers to someone who can.
In these respects the CTC is a "triple threat" information bank.
Within OAQPS, we already have informed potential users about
the VAPCCI via the LAN. This notification generated some
interest. Based on this response, we gave a seminar (December
1994) to present the VAPCCI and TAG spreadsheets, and demonstrate
the use of both. As our contractors usually make more control
cost calculations than we do, we invited them also.3
Public domain distribution: As the name implies, this element of
the strategy entails distributing the VAPCCI to all users,
especially those outside the regulatory sector. The most
efficient way to do this is through technical publication.
* This seminar on the VAPCCI was an addendum to a series of
12 control cost lectures this author presented to OAQPS personnel
(primarily BSD) and BSD contractors from April 1991 to January
1993. Each of these lectures focused on a particular type of
control device (e.g., fabric filters) and provided instruction on
how to size and cost that device. The instruction closely
followed the data and procedures presented in the applicable
chapter(s) of the OAQPS Control Cost Manual.
-------
57
Several technical journals are potential outlets for the VAPCCI.
The four journals we considered were:
/ Chemical Engineering (CE)
/ Environmental Engineering World (EEW)
/ Journal of Air and Waste Management (JAWMA)
/ Pollution Engineering
Published by McGraw-Hill, CE is a high-circulation (70,000)
and readership (300,000-plus) magazine that, for over six
decades, has been dedicated to informing technical and management
personnel about developments in the chemical process industries.
In recent years, CE has devoted more coverage to environmental
issues. (For instance, the March 1995 issue contained a feature
report on the HON MACT.) During the past fifteen years, this
author has had 25 articles published in Chemical Engineering.
Environmental Engineering World, also published by McGraw-
Hill, is a recent "spin-off" from Chemical Engineering magazine.
Suzanne Shelley, the editor of EEW, has committed to publish both
a feature article about the VAPCCI and the indexes as they are
updated. She also has arranged to have the article and indexes
published in CE. When this is done, the indexes will reach the
readership of both magazines, i.e., those who are working in the
chemical processing and environmental control fields—just the
kinds of professionals that would benefit from using the VAPCCI.
The JAWMA is another periodical to which we have
contributed. This monthly periodical has been the unofficial (if
not the official) organ of the air pollution control field for
decades, both inside and outside government. The VAPCCI should
fit JAWMA's format well, and be well received by its readers.
Another monthly, Pollution Engineering is much like EEW, in
terms of format, content, and targeted audience. Older than
EEW, PE has been published since the early 1970's. The VAPCCI
would fit its format just as well as it would EEW's.
Other periodicals are possible outlets for the VAPCCI. Two
are Environment and Environmental Progress. The former,
published 10 times a year, typically publishes articles related
to environmental science and policy, though not very many on
control equipment usage, design, and cost.
The latter, Environmental Progress publishes practical
articles related to air, water, and solid waste pollution
control. It is one of several periodicals distributed by the
American Institute of Chemical Engineers to AIChE members.
However, with a quarterly publication schedule, EP would not give
-------
58
the VAPCCI nearly as much exposure as CE, EEW, JAWMA, or PE.
We should keep in mind, however, that publication of the
VAPCCI would not necessarily be limited to one or more
periodicals. As these indicies were developed on EPA time using
EPA resources, they are technically in the public domain. Hence,
no publisher or other entity can copyright the contents of the
VAPCCI. In other words, the VAPCCI can be published in several
periodicals simultaneously.
In addition to publishing the VAPCCI in journals, we plan to
send the indexes to certain control equipment vendors
periodically, to solicit their comments. The vendors solicited
will be those who responded positively to our surveys. We are
mainly interested in seeing how well the VAPCCI—especially the
equipment cost indexes—predict their price changes. If there
are large enough differences between the real and predicted price
changes, we can adjust the VAPCCI accordingly. This solicitation
would not only provide a "reality check" on the VAPCCI, but also
serve to publicize the indexes and encourage their use.
-------
59
VI. Future Work
Unlike some projects that are complete once the final
reports are written, this one will never be finished. Although
we have devised equipment and total annual cost indexes for eight
air pollution control devices, our work is far from being
complete. For one thing, the indexes will have to be—and have
been—updated quarterly. This updating process requires us to
obtain current Producer Price Indexes, as well as inputs from
other sources (e.g., the Department of Energy). These inputs
must be keyed into the VAPCCI programs to generate the updated
indexes. Finally, the updated indexes have to be disseminated to
users, periodicals, and other outlets.
The future work will not be limited to just updating the
VAPCCI. Based on feedback from vendors and other reviewers, the
indexes may have to be revised. These revisions may entail
something as easy as changing some of the index weighting
factors. But, under certain circumstances we may have to
restructure one or more of the indexes. Feedback from equipment
vendors and other users, as well as data gleaned from further
research, may point out areas where an index is deficient. As a
result, we may have to make major changes, such as adding or
deleting several components of the index. Though time- and
resource-consumptive, such revisions will be necessary to
maintain the integrity of the VAPCCI. (We do not anticipate that
such major changes will be required often, however.)
Thirdly, we expect to expand the coverage of the VAPCCI to
include additional control devices. Although those included at
present cover most of the major control device categories,
several others eventually should be added. These are the other
particulate devices (fabric filters and electrostatic
precipitators, mainly*) , the several NOX controls (selective and
nonselective catalytic reduction, combustion modifications, steam
and water injection, etc.), the many flue gas desulfurization
(FGD) processes, and the fugitive emission controls, such as wet
dust suppression.
All of this begs the question: "Who's going to do this work,
and how much in-house time and contractual resources will be
needed to accomplish it?" Although this author is not the only
person who could update, refine, and expand the VAPCCI, he is the
most likely candidate for the job. Besides which, the author's
main responsibility is the maintainance and improvement of the
• We had planned to include these two devices in this
report, but were unable to obtain the requisite data from
vendors.
-------
60
quality of OAQPS control cost estimation.
fit his job description very nicely.
Hence, this work would
The in-house time and contractual resources required will
depend on the type of VAPCCI work being done (i.e., updating,
refining, or expanding). These requirements (figured on an
annual basis) are listed in the table below. Obviously, these
estimates are quite approximate, as the number of tasks (let
alone their complexity) is still unknown.
Annual Time and Resources Needed for VAPCCI Tasks
VAPCCI Task
1. Updating1"
2. Refinement0
3 . Expansion*1
Totals:
In -House Time
(person-year) "
0.14
0.02
0.04
0.20
Contractor Resources
(hours)
None
None
240 - 400
240 - 400
Notes:
" One person-year = 1800 hours/year = 225 days/year.
b Assumes that each of eight indexes will have to be updated four
times per year (quarterly), with each index update requiring
approximately one person-day.
c Assumes that one index will need a major refinement annually,
with each refinement requiring approximately five person-days.
d Assumes that two new indexes will be developed annually, with
each development requiring approximately ten person-days of in-
house time and 240 to 400 (6 to 10 weeks) of contractor time.
In all, approximately two-tenths person and 240 to 400 hours
of contractor time would be needed to update, refine, and expand
the VAPCCI. Because this author has spent six months full-time
developing the VAPCCI (and approximately one more month updating
the indexes and revising this report), it is not too difficult to
estimate the in-house time needed to accomplish these maintenance
and development tasks. Estimating the contractual resources
(task 3) is another matter, however. Not knowing which
contractor(s) would be retained for developing new indexes, the
capability o-f the staff assigned to the project, and other
-------
61
factors, it is difficult to provide a single hours/year estimate.
Thus, a range has been listed. As note "d" indicates, this range
reflects 6 to 10 weeks of contractor time.
Note that no contractor resources have been given for tasks
1 (updating) and 2 (refinement). Although one or both of these
tasks could be assigned to contractors, they probably could be
done more efficiently in-house. The procedures for updating the
VAPCCI are easy to learn and use and are, therefore, amenable to
in-house performance. (Interestingly, for several years the
Chemical Engineering Plant Index was updated by clerical
personnel at Chemical Engineering magazine.) Refining an index
is a more complex undertaking, as it requires knowledge of how
the index was developed, along with thorough knowledge of the
control device and its components.
Clearly, index development demands technical expertise.
However, it also requires a lot of data-gathering and
analysis—the kinds of tasks that contractors are often better-
equipped to handle than are in-house staff. In-house time will
also be needed for task 3, for such jobs as writing the work
assignment, monitoring the contractor's performance, and using
the contractor's findings to devise the final index. In other
words, the contractor would perform only part of the index
development task. The rest would be done in-house.
Finally, no time has been allotted in the above table for
"information service"—answering questions about the VAPCCI,
sending copies of this report and computer diskettes, and related
chores. As explained above, the latter tasks can be handled
effectively by the Control Technology Center. Nevertheless, this
author can best handle the "Q&A" work. As no one can predict how
many calls we will get regarding the VAPCCI, we cannot estimate
the resources needed to address them. In any event, we should
hope to get many, for phone calls are an excellent barometer of
how well we've distributed the VAPCCI and how much they are being
used.
-------
62
Appendix A. Control Device Component Price Data
The tables in this appendix list the averages of the
reported component price factors for each of the eight control
devices covered in this report. These factors were used in
developing the equipment cost indexes (ECI) described in Section
II.
Table A-l. Carbon Adsorber Component Price Data
Component
Adsorber vessels
Carbon
Condenser
Internal piping
Decanter
Pump (recovered organics)
Receiver
Distillation column
Air prefilters
Dampers & valves
System & cooldown fans
Fan motors & drives
Instrumentation & controls
Other (specify) d
Totals:
Fraction of List Price
Small"
0.180
0.030
0.040
0.170
0.020
0.020
0.020
0.000
0.020
0.060
0.010
0.040
0.210
0.180
1.000
Medium1"
0.240
0.120
0.040
0.130
0.010
0.010
0.010
0.000
0.040
0.070
0.010
0.040
0.150
0.130
1.000
Large6
0.250
0.130
0.030
0.140
0.010
0.010
0.010
0.000
0.040
0.080
0.010
0.050
0.140
0.100
1.000
* Capacity: < 1,000 acfm.
b Capacity: 1,000 - 50,000 acfm.
c Capacity: > 50,000 acfm.
d Other includes "indirects" (engineering, primarily).
-------
63
Table A-2. Catalytic Incinerator Component Price Factors
Component
Catalyst
Catalyst chamber
Preheat chamber
Filter/mixer
(gas flow distribution)
Recuperative heat exchanger
Flow control dampers
System fan
System fan motor & drive
Stack
Instrumentation & controls
Other (specify) d
Totals:
Fraction of List Price
Small"
0.225
0.095
0.065
0.060
0.180
0.015
0.070
0.045
0.070
0.110
0.065
1.000
Mediumb
0.265
0.105
0.090
0.015
0.255
0.005
0.075
0.045
0.035
0.045
0.065
1.000
Large0
0.295
0.105
0.090
0.010
0.270
0.005
0.075
0.045
0.035
0.040
0.030
1.000
" Capacity: < 2,000 acfm.
b Capacity: 2,000 - 50,000 acfm.
c Capacity: > 50,000 acfm.
d Includes "engineering and startup".
-------
64
Table A-3. Flare Component Price Factors
Component
Flare burner tip
Pilot burners
Flare tower (stack) & support
Utility piping (from base)
Liquid & gas seals
Ladders & platforms
Dampers & valves
Knockout drum
System fan
Fan motor & drive
Instrumentation & controls
Other (specify)
Totals:
Fraction of List Price
Small"
0.175
0.120
0.315
0.040
0.065
0.040
0.025
0.065
0.020
0.010
0.125
0.000
1.000
Mediumb
0.125
0.075
0.390
0.055
0.080
0.080
0.020
0.070
0.020
0.010
0.075
0.000
1.000
Largec
0.095
0.035
0.425
0.070
0.080
0.105
0.015
0.085
0.020
0.010
0.060
0.000
1.000
' Flare tip diameter: < 12 inches.
b Flare tip diameter: 12 to 48 inches.
c Flare tip diameter: > 48 inches.
-------
65
Table A-4. Gas Absorber Component Price Data
Component
Absorber columnd
Packing
Platforms and ladders
Internal piping
Recirculation tank
Pumps
Dampers & valves
Mist eliminator
System fan
Fan motor & drive
Instrumentation & controls
Other6
Totals:
Fraction of List Price
Small4
0.235
0.090
0.040
0.060
0.085
0.090
0.060
0.045
0.100
0.070
0.110
0.015
1.000
Mediumb
0.257
0.067
0.043
0.050
0.100
0.087
0.043
0.043
0.093
0.053
0.107
0.057
1.000
Large0
0.170
0.065
0.065
0.065
0.080
0.085
0.050
0.065
0.115
0.070
0.110
0.060
1.000
a Capacity: < 2,000 acfm. However, one vendor specified an
upper limit of 10,000 acfm.
b Capacity: 2,000 - 40,000 acfm. However, one vendor
specified a range of 10,000 - 40,000 acfm.
0 Capacity: > 40,000 acfm.
d Including: gas inlet and outlet ports; liquid inlet port
and outlet port/drain; liquid distributor/redistributor; sump
space; and, as applicable, trays or packing support plates.
e Includes "engineering".
-------
66
Table A-5. Refrigeration System Component Price Data
Component
Refrigeration unit
Precooler
VOC condenser (s)
VOC storage/recovery tank(s)
Internal piping
Pumps
Dampers & valves
System fan
Fan motor & drive
Instrumentation & controls
Other (specif y)d
Totals:
Percent of List Price
Small"
0.230
0.050
0.165
0.025
0.085
0.025
0.030
0.040
0.050
0.225
0.075
1.000
Mediumb
0.215
0.050
0.165
0.035
0.085
0.025
0.030
0.045
0.050
0.225
0.075
1.000
Large0
0.260
0.050
0.175
0.045
0.075
0.035
0.045
0.000
0.000
0.250
0.065
1.000
* Capacity: < 10 tons.
b Capacity: 10 to 100 tons.
c Capacity: > 100 tons.
d Other includes "structural".
-------
67
Table A-6. Regenerative Thermal Oxidizer Component Price Data
Component
Combustion chambers
Burners
Heat collection material
(e.g., ceramic packing)
Insulation
Flow control dampers
System fan
System fan motor & drive
Instrumentation & controls
Other (specify) d
Totals:
Fraction of List Price
Small"
0.110
0.035
0.115
0.100
0.115
0.200
0.125
0.175
0.025
1.000
Mediumb
0.115
0.025
0.075
0.100
0.175
0.200
0.100
0.150
0.060
1.000
Large0
0.100
0.020
0.065
0.100
0.200
0.215
0.110
0.135
0.055
1.000
* Capacity: < 5,000 acfm.
b Capacity: 5,000 - 100,000 acfm.
c Capacity: > 100,000 acfm.
d Other includes "engineering, startup, project management,
and purchasing manhours".
-------
68
Table A-7. Thermal Incinerator Component Price Data
Component
Combustion chamber
Burner (s)
Insulation
Recuperative heat exchanger
Flow control dampers
System fan
System fan motor & drive
Stack
Instrumentation & controls
Other (specify)*1
Totals:
Fraction of List Price
Small"
0.100
0.050
0.100
0.350
0.080
0.100
0.050
0.040
0.080
0.050
1.000
Mediumb
0.100
0.050
0.090
0.350
0.100
0.100
0.050
0.030
0.080
0.050
1.000
Large6
0.100
0.050
0.080
0.390
0.100
0.100
0.050
0.030
0.070
0.030
1.000
* Capacity: < 500 acfm.
b Capacity: 500 - 50,000 acfm.
e Capacity: > 50,000 acfm.
d Other includes "support legs".
-------
69
Table A-8. Wet Scrubber Component Price Data
Component
Contacting unit
(e.g. , venturi)
Liquid- gas separator
Mist eliminator
Internal piping
Lining (e.g., rubber)
Recirculation liquid pump
Recirculation tank
Flow control dampers
System fan
System fan motor & drive
Instrumentation & controls
Other (specif y)d
Totals:
Fraction of List Price
Small"
0.110
0.045
0.045
0.065
0.040
0.125
0.160
0.040
0.125
0.070
0.135
0.040
1.000
Medium5
0.120
0.097
0.043
0.073
0.027
0.107
0.143
0.037
0.110
0.053
0.120
0.070
1.000
Large6
0.105
0.165
0.065
0.065
0.040
0.085
0.075
0.055
0.115
0.070
0.100
0.060
1.000
a Capacity: < 1,000 acfm. (However, one vendor specified
10,000 acfm as the upper end of this size range.)
b Capacity: 1,000 - 50,000 acfm. (However, one vendor
specified a size range of 10,000 - 50,000 acfm.)
0 Capacity: > 50,000 acfm.
d Other includes "engineering".
-------
70
References:
1. Matley, J. "CE Plant Cost Index--Revised." Chemical
Engineering, April 19, 1982, p. 153.
2. Stevens, R.W. "Equipment Cost Indexes for Process Industries,"
Chemical Engineering, November 1947, pp. 124-126.
3. "Brief Explanation of Producer Price Indexes," in: Producer
Price Indexes, November 1993. Washington: Bureau of Labor
Statistics, U.S. Department of Labor.
4. "Producer Prices," in: BLS Handbook of Methods. Washington:
Bureau of Labor Statistics, U.S. Department of Labor. 1992.
5. Ibid.
6. Telefax from Ellen Rafferty (Chemical Engineering magazine) to
William M. Vatavuk (U.S. Environmental Protection Agency,
Research Triangle Park, NC), September 26, 1994.
7. Chemical Engineering (various issues, 1989-93) .
8. Vatavuk, William M. (ed.). OAQPS Control Cost Manual (Fourth
Edition, Supplement 2). Research Triangle Park, NC: U.S.
Environmental Protection Agency. October 1992.
9. Letter from Jorgen G. Hedenhag (Air Pol, Inc., Teterboro, NJ)
to William M. Vatavuk (U.S. Environmental Protection Agency,
Research Triangle Park, NC), June 21, 1994.
10. "Caustic Surge Leads to Potent Price Hike," Chemical
Marketing Reporter, May 28, 1994, p.5.
11. Vatavuk, William M. Estimating Costs of Air Pollution
Control. Chelsea, MI: Lewis Publishers/CRC Press. 1990.
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
1. REPORT NO.
EPA-452/R-95-006
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Escalation Indexes for Air Pollution Control Costs
5. REPORT DATE
October 1995 (Date of Issue)
6. PERFORMING ORGANIZATION CODE
7. AUTHORS William M. Vatavuk
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Strategies and Standards Division
Innovative Strategies and Economics Group (MD-15)
Research Triangle Park, NC 27711
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Strategies and Standards Division
Innovative Strategies and Economics Group (MD-15)
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT This report documents the work accomplished during the FY 1993-94 "Bernard J.
Steigerwald Opportunity for Independent Study". It discusses why and how air pollution control cost
indexes were developed and presents index values for several periods (quarters). These indexes are used
to adjust ("escalate") control equipment and total annual costs from one quarter to another. In all, 16
indexes were developed and are presented herein, one Equipment Cost Index (ECI) and one Total
Annual Cost Index (TACI) for each of eight control device categories. These indexes collectively
known as the Vatavuk Air Pollution Control Cost Indexes can be used to escalate costs from the
initial (base) period (first quarter 1994) to any quarter thereafter. In this report, final ECI and TACI are
given for the second, third, and fourth quarters of 1994; and preliminary indexes, for first quarter 1995.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Stationary emission sources Gas absorbers
Costs
Control techniques
Carbon adsorbers
Catalytic incinerators
Flares
Regenerative oxidizers
Refrigeration systems
Thermal incinerators
Venturi scrubbers
Cost estimation
Cost escalation
Escalation indexes
Equipment costs
Annual costs (direct, indirect)
"Add-on" controls
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (Report)
Unclassified
21. NO. OF PAGES
72
20. SECURITY CLASS (Page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
------- |