United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S8-88/064 May 1 988
Project Summary
Development of the Fuel
Choice Module in the Industrial
Combustion Emissions Model
Tim Hogan, Joel L Horowitz, and Thomas Cook
The Industrial Combustion Emissions
(ICE) Model is one of four stationary
source emission and control cost
forecasting models developed by EPA
for the National Acid Precipitation
Assessment Program. The ICE Model
projects air pollution emissions (sulfur
dioxide, sulfates, nitrogen oxides, and
particulate matter), costs, and fuel mix
for industrial fossil-fuel-fired (natural
gas, distillate and residual fuel oil, and
coal) boilers by state and year (1980
baseline, 1985, 1990, 1995, 2000,
2010, 2020, and 2030).
The ICE Model was originally devel-
oped from the Industrial Fuel Choice
Analysis Model (IFCAM), which relies
on a life-cycle cost-of-fuel logic. Two
reports describe the development of an
updated forecast model (i.e., ICE)
which relies on a broader range of
factors shown to be relevant to the
industrial boiler fuel choice decision.
These reports describe the develop-
ment and basis for the improved fuel
choice decision logic used in the ICE
Model (Version 6.0).
This Project Summary was devel-
oped by EPA's Air and Energy Engi-
neering Research Laboratory, Research
Triangle Park, NC, to announce key
findings of the research project that is
fully documented in two separate
reports of the same title (see Project
Report ordering information at back).
Introduction
The Industrial Combustion Emissions
(ICE) Model is one of several emission
forecasting models developed by EPA for
use by the National Acid Precipitation
Assessment Program (NAPAP). The ICE
Model (Version 6.0) projects air pollution
emissions (sulfur dioxide, sulfates, and
nitrogen oxides), costs, and fuel mix for
industrial fossil-fuel-fired (natural gas,
distillate and residual fuel oil, and coal)
boilers by state (excluding Alaska and
Hawaii) and year (1980 base year, 1985,
1990, 1995, 2000, 2010, 2020, and
2030). The ICE Model does not include
projections related to the combustion of
LPG, municipal or agricultural solid
waste, or non-purchased, self-generated
by-product fuels (i.e., wood, black liquor,
coke oven gas, blast furnace gas, refinery
off-gas, and refinery still gas).
Background
The ICE Model is a disaggregated
process engineering model. Models of
this type simulate the effects of specific
policies on technical alternatives for new
and existing equipment. The ICE Model
is designed to assess the impact of
several factors on industrial boiler fuel
choice decisions and air pollution emis-
sions, including local and Federal air
pollution emissions regulations, fuel
price forecasts, and capital and annual
operating and maintenance (O&M) costs
of firing alternative fuels or retrofitting
pollution control equipment.
The ICE Model projects the distribution
of industrial boiler characteristics (e.g.,
new versus existing, boiler size and
capacity utilization rate) and selects the
fuel type and pollution control com-
pliance strategy for each unit. An impor-
tant model feature is the approach
chosen in the ICE Model to select the
fuel type for new industrial boilers.
One option is to estimate the life cycle
costs for each fuel type (including boiler
-------
and pollution control equipment capital,
O&M, and fuel expenses) and select the
low-cost alternative. This cost compar-
ison can be performed on an after-tax
basis because, in the past. Federal
income tax laws treated investments in
coal and alternative fuel-firing boilers
differently than investments in oil- and
gas-fired boilers. Specifically, the regular
investment tax credit was denied for
investments in industrial oil- and gas-
fired boilers.
Many analysts recognize that actual
decisions by industrial firms consider a
broader range of factors than just capital,
O&M, and expected fuel costs. These
other considerations include: reliability
of fuel supplies, risks of operating
disruptions, uncertainty regarding future
fuel prices, capital budgeting constraints,
and whether purchasers have past ex-
perience with coal.
Some analysts believe that a decision
framework which considers only ex-
pected costs is an unreliable predictor of
fuel choice decisions for new industrial
boilers. They are further concerned that
the use of this narrow approach in in-
dustrial energy demand analyses con-
ducted in the past has resulted in
unreasonably high or lowforecasts of the
market share for coal in new industrial
boilers and, as a result, this procedure
is biased.
EPA decided that a more comprehen-
sive evaluation of the factors affecting
the boiler fuel choice decision was
required to eliminate this bias in the
forecasted results. A qualitative review
identified several important factors in
addition to life-cycle costs of capital
investment and annual operating
expenses. Recent new industrial boiler
sales data showed that the fuel choice
decision was not based solely on the
comparison of readily quantifiable life-
cycle costs for alternative fuels.
Fuel Choice Module
Development
Data on new industrial boiler sales
were evaluated as a function of expected
cost (boiler and pollution control equip-
ment capital, O&M, and fuel) differences
and other factors in three statistical
analyses (Phases I, II, and III). The initial
and final statistical analyses (Phases I
and III) were performed by EEA. Addi-
tional insights were gained from an
additional statistical analysis (Phase II)
by Joel L. Horowitz and Thomas Cook.
The data base analyzed includes over
400 orders for new industrial coal, oil,
and gas boilers between 1977 and 1983.
The probability of selecting coal was
estimated from this data base as a
function of boiler size, region, previous
experience with firing coal on-site in
existing boilers (yes or no), and expected
cost differences.
Prior to the late 1970s, the market
share of coal in new boilers since World
War II was so low(approximately 5%)that
there was no real experience to analyze
which showed any variability in fuel
choice. In general, natural gas was
underpricing other fuel sources due to
price controls; therefore, that premium
fuel was the overwhelmingly dominant
fuel choice.
However, in the late 1970s and early
1980s the coal market share (as a
percent of fossil fuel capacity) for new
boilers rose to 15-30%. Even though this
period was characterized by sudden
shifts in incentives (e.g., oil embargo,
changes in tax laws), at least the coal
versus oil/gas market's share changed
substantially. This study was initiated
with the hope that analysis of new boiler
orders during that period would shed
new light on the determinants of fuel
choice decisions.
The market share of coal in new boiler
orders was found to be a strong function
of boiler size. Almost 75% of the large
boilers were ordered with coal-firing
capability, in comparison with approxi-
mately 50% of the medium-sized boilers
and 11 % of the small boilers.
The impact of fuel choice decisions of
having previous coal experience also was
apparent in the data. For both medium
and large boilers, the very large majority
of decisions were made for coal when
previous coal experience was a factor.
A much lower percentage of small boiler
orders chose coal with prior coal exper-
ience, but on a relative basis this lower
market share still greatly exceeded that
observed for small units without
experience.
Plant location was also important. New
industrial boilers built at plants without
previous coal experience in Federal
Regions 4 (South Atlantic) and 5 (Mid-
west) are, on the average, more likely to
choose coal than plants without coal
experience located elsewhere. More
than half of U.S. coal consumption is
accounted for by these two regions.
Apparently close proximity to coal sup-
plies and the demonstrated reliability of
coal boilers in other plants in the same
area may also be important considj
erations.
In addition to boiler size, location, anc
previous coal experience, there are othei
important factors. However, the date
base on new industrial boiler orders doe;
not include information on plant-specifk
equipment costs, fuel price expectations
perceptions of fuel supply reliability
equipment reliability, capital budge'
constraints, or the costs of lost produc
tion due to steam supply disruption
Therefore, this study summarizes th«
distribution of fuel choice decisions aj
a function of the available data (fulh
recognizing the data limitations) tc
capture the effects of these less readih
quantifiable effects on the fuel choice
decision by industrial plant managers.
-------
-------
T. Hogan is with Energy and Environmental Analysis, Inc., Arlington, VA 22209;
J. L. Horowitz is with the University of Iowa, Iowa City. IA 52442; and T.
Cook is with the University of Denver, Denver, CO 80210.
Larry G. Jones is the EPA Project Officer (see below).
The complete report consists of two volumes, entitled "Development of the
Fuel Choice Module in the Industrial Combustion Emissions Model:"
"Volume 1. Phases I and III." (Order No. PB 88-198 577/AS; Cost: $14.95)
"Volume 2. Phase II," (Order No. PB 88-198 585/AS; Cost: $14.95)
The above reports will be available only from: (costs subject to change)
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Officer can be contacted at:
Air and Energy Environmental Research Laboratory
U.S. Environmental Protection Agency
Reserach Triangle Park, NC27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
•x"r\\hT/f
'1- "
U.S. OFFICIAL MAIL
A OEMALTY
— « I
JUN3Q'8
U.S.POSIAG^
HO .25 =
Official Business
Penalty for Private Use $300
EPA/600/S8-88/064
Vr^
0000329 PS
H S 6lt«R PROTECTION HGENCT
REGION 5 LIBRART
230 S DEARBORN STREET
CHICAGO IL 60604
GOVERNMENT PRINTING OFFICE. 1988—548-013/87(
------- |