EPA
TVA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
EPA-600/7-79-114
May 1979
Tennessee Valley
Authority
Office o'f Power
Emission Control
Development Projects
Muscle Shoals Al 3566O
ECDP B-2
Computerized FGD
Byproduct Production and
Marketing System:
Users Manual
Interagency
Energy/Environment
R&D Program Report
-------
EPA-600/7-79-114
ECDP B-2
May 1979
Computerized FGD Byproduct Production
and Marketing System: Users Manual
by
W. L. Anders
Tennessee Valley Authority
Office of Power
Emission Control Development Projects
Muscle Shoals Alabama 35660
EPA Interagency Agreement 09-E72I - BH
Program Element No. INE-624A
EPA Project Officer: Charles J. Chatlynne
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
DISCLAIMER
This report was prepared by the Tennessee Valley Authority and has
been reviewed by the Office of Energy, Minerals, and Industry, U.S.
Environmental Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the views and
policies of the Tennessee Valley Authority or the U.S. Environmental
Protection Agency, nor does mention of trade names or commercial
products constitute endorsement or recommendation for use.
11
-------
ABSTRACT
This users manual describes a computerized system consisting of a
number of integrated programs, models, and data bases that have been
developed to make cost comparisons among power plant strategies designed
to meet clean air regulations. The data bases, programs, and the procedures
and requirements that are necessary for data base access and program
execution are described. The power plant data base contains actual and
projected information for all U.S. fossil-fuel power plants. A scrubbing
cost model allows cost comparisons between any two of five compliance
strategies: limestone scrubbing with sludge waste disposal, limestone
scrubbing with gypsum production, sodium sulfite scrubbing with sulfur
production, magnesia scrubbing with sulfuric acid production, and the
use of clean fuel with no scrubbing. For salable flue gas desulfurization
(FGD) byproducts, potential marketing revenues are included in cost
comparisons. The sulfur and sulfuric acid data base contains actual and
projected information for all U.S. sulfur-burning acid plants. The
transportation data base contains legal rail mileages between all rail
rate basing points in the 37 eastern states (Docket 28300) and also
contains location-related data for every named U.S. location. Each of
the data bases and programs can generally be used independently of the
other parts of the system.
iii
-------
CONTENTS
Figures vii
Tables viii
Abbreviations and Glossary xi
I. Introduction I- 1
II. System Description II- 1
Data Sources II- 3
Premises II- 3
System Organization II- 4
Supply Subsystem II- 9
Supply Data Base II- 9
Supply Programs 11-11
Demand Subsystem 11-11
Demand Data Base 11-13
Demand Programs 11-13
Transportation Subsystem 11-14
Nonintegrated Transportation Data Base 11-16
Transportation Programs 11-16
Linear Programing Model Subsystem 11-16
Summary 11-18
III. Program Descriptions Ill- 1
PROJECT Program Ill- 1
ADDLIME Program Ill- 4
STMCAP Program Ill- 6
Input Ill- 6
General Processing Ill- 9
Options 111-10
Overrides for Internal Program Values 111-14
ACDUPDT Program 111-18
GENACD Program 111-20
Manual Transportation Procedure 111-22
TRNCOST Program 111-25
GENPGM Program 111-28
GENSORT Procedure 111-32
APEX System 111-34
REPTSOL Program 111-34
Final Strategy Selection Procedure 111-37
v
-------
IV. File and Record Descriptions IV- 1
Supply Subsystem IV- 1
Power Plant Data Base IV- 1
Other Supply Subsystem File and Record Descriptions . . . IV- 9
Demand Subsystem IV-31
Transportation Subsystem IV-42
Linear Programing Subsystem IV-47
V. System Usage V- 1
General v~ 1
Execution Modes V- 2
Time-Sharing Execution V~ ^
Batch Execution V-lb
VI. References
Appendices
A. Integrated Transportation Data Base A- 1
B. Related CYBERNET Manuals and Publications B- 1
C. Rail Mileage File Description and Usage C- 1
VI
-------
FIGURES
Number
II- 1 Byproduct marketing system II- 2
II- 2 Supply subsystem II- 5
II- 3 Demand subsystem II- 6
II- 4 Transportation subsystem II- 7
II- 5 Linear programing marketing model subsystem II- 8
II- 6 Supply subsystem flow chart 11-10
II- 7 Demand subsystem flow chart 11-12
II- 8 Transportation subsystem flow chart 11-15
II- 9 Linear programing marketing model subsystem flow chart . . 11-17
III- 1 Block diagram for the calculation of power plant
projection data (program PROJECT) Ill- 2
III- 2 Block diagram for the addition of delivered limestone
costs to power plants (program ADDLIME) Ill- 5
III- 3 Block diagram for the calculation of power plant
scrubbing costs as a part of the overall system
(program STMCAP) Ill- 7
III- 4 Block diagram for the calculation of power plant
scrubbing costs independently of the overall system
(program STMCAP) Ill- 8
III- 5 Block diagram for the calculation of delivered sulfur
costs to acid plants (program ACDUPDT) 111-19
III- 6 Block diagram for the calculation of acid plant
avoidable production costs (program GENACD) 111-21
III- 7 Block diagram for the selection of rate basing data
for smelters, power plants, and acid plants (manual
procedure) 111-23
III- 8 Block diagram for calculation of transportation rates
from power plants and smelters to acid plants
(program TRNCOST) 111-26
III- 9 Block diagram for the generation of the linear
programing marketing model (program GENPGM) 111-29
111-10 Block diagram for sorting the power plant data from
the linear programing model generation (program
GENSORT) 111-33
III-ll Block diagram for solving the linear programing
marketing model (APEX linear programing system) 111-35
111-12 Block diagram for generating a report of the
equilibrium solution (program REPTSOL) 111-36
IV- 1 Power plant data base structure IV- 3
IV- 2 Railroad rate territories IV-46
vii
-------
TABLES
Number Page
III- 1 Scrubbing Cost Generator (STMCAP) Program Execution
Options IZI~ 2
III- 2 Scrubbing Cost Generator (STMCAP) Internal Program
Value Overrides 111-15
IV- 1 Files Used in the Supply Subsystem IV~ 2
IV- 2 Power Plant Data Base Plant Data IV~ *
IV- 3 Power Plant Data Base Regulation Data IV~ 5
IV- 4 Power Plant Data Base Boiler Data (General) IV~ 6
IV- 5 Power Plant Data Base Boiler Data (By Year) IV- 7
IV- 6 File Format for Scrubber Investment and Operating
Factors (SASDAT6) IV- 9
IV- 7 File Format for Scrubber Investment and Operating
Costs (SCRPRC) IV-10
IV- 8 File Format for Delivered Costs of Limestone to
Power Plants (LIMEEST) IV-11
IV- 9 File Format for Power Plant Site-Specific
Adjustments (SCRSIT) IV-12
IV-10 File Format for Projected Plant-Level Data (PLAS) .... IV-13
IV-11 File Format for Projected Boiler-Level Data (BLAS). . . . IV-14
IV-12 File Format for Projected Plant-Level Data with
Delivered Limestone Costs Included (UPLIME) IV-15
IV-13 File Format for Calculated Power Plant Scrubbing
Costs and Related Quantities (SCRCST) IV-16
IV-14 Report Format for Options and Overrides IV-17
IV-15 Report Format for Plant-Level Scrubbing Costs IV-19
IV-16 Report Format for the Edit of User-Supplied Power
Plant Data IV-20
IV-17 Report Format for the Edit of Data Base Power Plant
Data IV-21
IV-18 Report Format for Boiler Scrubbing Costs IV-22
IV-19 Report Format for the Plant Scan Summary IV-23
IV-20 Report Format for Emissions and Compliance IV-24
IV-21 Report Format for the Emissions Summary IV-25
IV-22 Report Format for the Edit of Scrubber Investment
and Operating Factors IV-26
IV-23 Report Format for Detailed Scrubbing Costs IV-27
IV-24 Files Used in Demand Subsystem IV-31
IV-25 File Format for Sulfuric Acid Avoidable Production
Cost Factors (ACDPAR) IV-32
IV-26 File Format for Acid Plant and Smelter Data
(SACDSML) IV-3 3
viii
-------
TABLES (continued)
Number
Page
IV-27 File Format for Sulfur Terminal Data (SULTER) IV-34
IV-28 File Format for Acid Plant and Smelter Data with
Delivered Costs of Molten Sulfur Included (ACDSML) IV-35
IV-29 Report Format for Delivered Sulfur Costs to Acid
Plants IV-36
IV-30 File Format for Sulfuric Acid Avoidable Production
Costs and Related Quantities (ACDCST) IV-37
IV-31 File Format for Smelter Production Costs and Related
Quantities (SMLCST) IV-37
IV-32 File Format for Rail Rate Basing Point Location
Data (RBNSORT) IV-38
IV-33 Report Formats for Sulfuric Acid Avoidable Production
Costs IV-39
IV-34 Files Used in Transportation Subsystem IV-42
IV-35 File Format for Transportation Points to be Used for
Shipping Rate Calculations (TRNPTS) IV-43
IV-36 Sample Tariff Table Definitions for the ICC Docket
28300 Territory . IV-45
IV-37 Files Used in the Linear Programing Subsystem IV-47
IV-38 File Format for Power Plant Strategy Preselection
Results (GENDATA/GENREPT) IV-48
IV-39 Report Format for Power Plant Strategy Preselection
Results IV-49
IV-40 Report Format for Model Solution IV-50
V- 1 Sample Listing of the Permanent File Catalog to Verify
that the Necessary Files Are Present for System Usage,
and a Sample Listing of the Catalog after All Programs
Have Been Executed V- 5
V- 2 Sample Procedure File to Interactively Execute the
Program that Adds Delivered Limestone Costs (ADDLIME) ... V- 7
V- 3 Sample Usage of a Procedure File to Interactively
Execute the ADDLIME Program V- 7
V- 4 Sample Procedure File to Interactively Execute the
Scrubbing Cost Generator Program (STMCAP) V- 8
V- 5 Sample Usage of a Procedure File to Interactively
Execute the STMCAP Program V- 9
V- 6 Sample Scrubbing Cost Generator (STMCAP) Input Options,
Overrides, and User Data for Interactive Execution .... V-10
V- 7 Sample Procedure File to Interactively Execute the
Program that Calculates Delivered Molten Sulfur
Costs to All Sulfuric Acid Plants (ACDUPDT) V-ll
V- 8 Sample Usage of a Procedure File to Interactively
Execute the ACDUPDT Program V-ll
V- 9 Sample Procedure File to Interactively Execute the
Program that Calculates Sulfuric Acid Avoidable
Production Costs (GENACD) V-12
ix
-------
TABLES (continued)
Number
V-10 Sample Usage of a Procedure File to Interactively
Execute the GENACD Program V-12
V-ll Sample Terminal Session to Add a Power Plant to
the Transportation File (TRNPTS) V-13
V-12 Sample Procedure File to Interactively Execute the
Program that Builds the Linear Programing Marketing
Model (GENPGM) V-15
V-13 Sample Usage of a Procedure File to Interactively
Execute the GENPGM Program V-16
V-14 Sample Job Stream to Execute the Power Plant Projection
Program in Batch Mode (PROJECT) V-17
V-15 Example of Submitting a Batch Run to Execute the
PROJECT Program; Checking the Job Status; and Checking
for Correct Execution V-19
V-16 Sample Job Stream to Execute the Scrubbing Cost
Generator Program in Batch Mode (STMCAP) V-20
V-17 Example of Submitting a Batch Run to Execute the
STMCAP Program; Checking the Job Status; and Checking
for Correct Execution V-21
V-18 Sample Job Stream to Execute the Transportation Cost
Generator Program in Batch Mode (TRNCOST) V-22
V-19 Example of Submitting a Batch Run to Execute the
TRNCOST Program; Checking the Job Status; and Checking
for Correct Execution V-23
V-20 Sample Job Stream to Execute the APEX Linear
Programing Package and the Report Generator Program
(REPTSOL) in Deferred Batch Mode (Overnight) to Generate
an Equilibrium Model Solution and a Report of the
Solution V-24
V-21 Example of Submitting a Deferred Batch Run to Execute
the APEX Package and REPTSOL Program to Solve the
Model, Generate an Equilibrium Solution, and Prepare
a Report of the Solution V-25
V-22 Checking the Status of the Run Shown in Figure V-19
and Submitted in Figure V-20; then Verifying the
Results V-26
-------
ABBREVIATIONS AND GLOSSARY
ABBREVIATIONS
ACFL
CDC
CDS
CENTRE
EDS
EPA
FGD
FIPS
FPC
k
M
NEDS
NRBT
PEDCo
PLI
RJE
SPLC
SRI
Alternative clean fuel level
Control Data Corporation
Compliance Data System
Centre Mark Company
Energy Data System
U.S. Environmental Protection Agency
Flue gas desulfurization
Federal Information Processing Standard
Federal Power Commission
Thousand (103)
Million (106)
National Emissions Data System
National Rate Basis Tariff
PEDCo Environmental, Inc.
Procedural Language Interface
Remote Job Entry
Standard Point Location Code
Stanford Research Institute
xi
-------
GLOSSARY
Alternative clean fuel level: The value assigned to premium price for
fuel that will meet the sulfur oxide emission standard.
APEX: A comprehensive mathematical programing system including linear
and mixed integer programing, matrix reduction, and parametric
capabilities.
Capacity factor: Capacity factor is calculated as the ratio of the
annual quantity of heat consumed in the boiler in comparison to the
quantity that would have been consumed if the boiler had operated
at rated capacity (full load) for the entire year (8760 hr).
Centre Mark Company: Source of geographic information on locations in
the United States, including latitudes, longitudes, county data,
and various other information related to over 100,000 locations.
Compliance Data System: A data base containing compliance information
and status for all emission sources in the United States as they
relate to clean air requirements.
CYBER 76 BATCH SERVICE: A computing service owned and operated by
Control Data Corporation that provides greater computational speeds
and capacities than those provided by the SCOPE 3.4 BATCH SERVICE.
CYBERLINK: A communications interchange to transmit information between
the SCOPE 3.4 BATCH SERVICE, the CYBER 76 BATCH SERVICE, and the
NOS INTERACTIVE AND BATCH SERVICE (registered trademark of CDC).
CYBERNET: A worldwide data processing and communications network owned
and operated by Control Data Corporation (registered trademark of CDC).
Docket 28300: A general investigation by the Interstate Commerce Commission
of the reasonableness of class rates in the United States (except
in the mountain Pacific and transcontinental territories) that
resulted in the class rates and tariffs in use today.
Energy Data System: A data base containing fuel quality and consumption
data, plant design and operating data, emission regulations, com-
pliance information, future megawatt capacities, and air quality data.
FPC Form 67: Federal Power Commission form used to report annual steam-
electric plant and water quality control data.
Interactive time-sharing (also known as conversational time-sharing): A
mode of operation of a data processing system in which a series of
user instructions and computer replies are exchanged on a one-
by-one basis, effectively constituting a conversation or interaction
that takes place between the user and the computer.
xii
-------
Limestone slurry scrubbing: A process for removing sulfur oxides from
flue gases by scrubbing the gases with a limestone slurry. In the
case of sludge waste, the resulting slurry of calcium sulfites,
sulfates, and unreacted limestone is discarded in a disposal pond.
In the case of gypsum production, the slurry is oxidized to gypsum,
dewatered, and stored.
Magnesia slurry scrubbing: A regenerative process for the removal of
sulfur oxides from flue gases by scrubbing the gases with a
magnesium oxide slurry. The magnesium sulfite formed in the slurry
is removed and thermally decomposed into magnesium oxide and a
stream of concentrated sulfur dioxide gases. The regenerated
magnesium oxide is returned to the scrubbing tower and the concen-
trated sulfur dioxide stream is fed to a conventional contact
sulfuric acid plant for the production of commercial (98%) sulfuric
acid.
NAMELIST: A FORTRAN programing specification that permits input and
output of groups of variables and arrays with an identifying name.
National Emissions Data System: A computer-based EPA emission inventory
system for storing and retrieving estimates of the criteria
pollutants from both point and area sources.
National Rate Basis Tariff: Tariff containing alphabetical lists of all
rail stations with rate basis applicable.
NOS INTERACTIVE AND BATCH SERVICE: A computing service owned and
operated by Control Data Corporation that provides both interactive
time-sharing and small-to-medium batch processing capabilities.
PEDCo Environmental, Inc.: The company that gathers information under
contract to EPA on FGD by direct interviews with and surveys of
utilities in the United States.
Procedural Language Interface: A SYSTEM 2000 Procedural Language
Interface establishes communications between a FORTRAN or COBOL
program and a SYSTEM 2000 data base.
Remote batch processing: A mode of operation of a data processing
system in which a series of user instructions are submitted to a
computer as a single entity (called a run or job) using a remote
job entry terminal and a data link (such as a telephone line),
rather than using the input facilities of the central computer site
(called on-line or over-the-counter batch processing). Once the
computer begins performing the series of tasks it continues until
processing is completed, and the results are returned to the user
as a single entity using the remote terminal and data link.
Remote job entry: See remote batch processing above.
xiii
-------
SCOPE 3.4 BATCH SERVICE: A computing service owned and operated by
Control Data Corporation that provides medium- to large-scale batch
processing capabilities.
Sodium scrubbing (Wellman-Lord/Allied Chemical process): A regenerative
process for the removal of sulfur oxides from flue gases by scrubbing
the gases with a solution of sodium sulfite. The sodium bisulfite
formed is thermally decomposed to sodium sulfite and sulfur dioxide
gas. The regenerated sodium sulfite is returned to the scrubbing
tower and the sulfur dioxide is reduced with natural gas to form
molten sulfur.
Standard Point Location Code: A transportation-oriented 6-digit number
prescribed by the National Motor Freight Association under the
guidance of the SPLC policy committee. It is used as a logistical
linkage between all possible shipping origins and destinations for
truck and/or rail.
SYSTEM 2000: A hierarchical data management system for creating, main-
taining, and retrieving information from large data bases (registered
trademark of MRI Systems Corporation).
xiv
-------
COMPUTERIZED FGD
BYPRODUCT PRODUCTION AND MARKETING SYSTEM:
USERS MANUAL
I. INTRODUCTION
The U.S. Environmental Protection Agency (EPA) - Tennessee Valley
Authority (TVA) studies of byproduct marketing for flue gas desulfuri-
zation (FGD) processes have been in progress since the early 1970's.
During these studies a computer model consisting of a system of programs
and data bases was developed for the numerous computation processes
involved. As the system was refined and expanded to meet increasingly
complex requirements, it became evident that it could also be generally
useful as a tool in decision making involving a wide range of emission
abatement problems.
In general, power plants out of compliance with clean air standards
have two' options. They can reduce the pollutants in the fuel burned, a
clean fuel strategy, or they can remove pollutants after the fuel is
burned, a scrubbing strategy. Scrubbing strategies are varied but they
can be grouped into two basic categories: those that produce a waste or
"throwaway" byproduct and those that produce a marketable byproduct.
A scrubbing system that produces a marketable byproduct such as
sulfuric acid is not a practical strategy for all power plants because
markets would quickly become saturated and revenues would be adversely
affected. A scrubbing system that produces waste byproducts is not
satisfactory in every case because of disposal problems, delivered raw
material costs, and plant operating characteristics. Neither will the
exclusive use of a clean fuel strategy solve the problem for all plants.
The supply of naturally occurring clean fuel is limited and significantly
increased demands would result in fuel price increases. The costs of
cleaning fuel, such as coal washing, may also be prohibitive. Considera-
tion must, therefore, be given to engineering, economic, transportation,
and marketing factors to allow in-depth analyses that will result in
selecting a suitable strategy for particular power plants. Since many
strategies may be selected that result in compliance with clean air
requirements, information must be available to make decisions that
result in the least cost of compliance in the long run.
1-1
-------
To provide an additional basis for cost analyses and comparisons in
decisions involving alternate emission control strategies it was felt
that the computer system developed for byproduct marketing studies
should be made available for access on a commercial, nationwide, inter-
active time-sharing and remote batch data-processing network.
This manual provides the information and procedures necessary to
use the system. It is primarily intended for the analyst-programer but
it should also be useful to those requiring detailed information about
the system who do not have extensive systems or programing experience.
This manual is primarily a users manual, however, and does not
provide the concepts and background information necessary for use of
the system. Several publications prepared during the course of by-
product marketing studies are essential in the use of this manual and
should be used as references. Marketing Sulfuric Acid from SOg Abate-
ment Sources—The TVA Hypothesis (1), Potential Abatement Production and
Marketing of Byproduct Sulfurio Acid in the U.S. (2), and Potential Pro-
duction and Marketing of FGD Byproduct Sulfur and Sulfuric Acid in the
U.S. - 1983 Projection (3), provide background information and data.
Detailed Cost Estimates for Advanced Effluent Desulfurization Processes
(4) defines the cost estimates upon which the system is based.
The byproduct marketing system is available through TVA under a
technology transfer agreement with EPA. Procedures for releasing the
system are initiated upon receipt of a written request. At the present
time, under the same technology transfer agreement, selected runs of the
system based on user-supplied data can be made by TVA. The section on
system usage provides the necessary details. All inquiries concerning
the byproduct marketing system and this manual should be directed to
Emission Control Development Projects, Tennessee Valley Authority,
Muscle Shoals, Alabama 35660, telephone No. (205) 383-4631, extension
2516.
1-2
-------
II. SYSTEM DESCRIPTION
The byproduct marketing system (Figure II-l) consists of a number
of integrated computer programs, models, and data bases which can be
used to make cost comparisons of FGD strategies designed to meet clean
air regulations. For strategies which produce a salable byproduct the
marketability of the byproduct is determined and its effect on FGD costs
is included in the cost comparisons. The power plant data base contains
actual and projected information on over 900 U.S. power plants. The
system can use this data or user-supplied data to develop situations for
comparison of alternate FGD strategies. Any two of five strategies can
be compared in the current system: the use of clean fuel without FGD,
limestone scrubbing with sludge waste disposal, limestone scrubbing with
gypsum production, sodium scrubbing with sulfur production, and magnesia
scrubbing with sulfuric acid production. For comparisons based on the
use of clean fuel without FGD, an alternative clean fuel level (ACFL) is
used to reflect the cost differential between a complying fuel and a
noncomplying fuel. In the case of magnesia scrubbing with sulfuric acid
production, the system determines the marketability of the acid produced
at a price sufficient to recover the incremental cost of its production.
For this capability, data bases on sulfur and sulfuric acid transportation
costs and the'U.S. sulfuric acid manufacturing industry are included in
the system.
The entire system can be used to compare the magnesia scrubbing
strategy with either the limestone scrubbing or clean fuel strategies
for any combination of geographic and power plant situations. The
numerous subsystems, programs, and data bases can be used separately or
in various combinations to provide information on a wide range of related
FGD processes and on sulfuric acid manufacture, transportation, and
marketing. Provisions for sulfur marketing are being incorporated into
the system. When this is completed, the system capabilities that are
described for sulfuric acid throughout this manual will also apply to
sulfur.
The system is in continual use and is updated and refined to reflect
the best current information and technology. It is developed on the
Control Data Corporation CYBERNET system with provisions for interactive
time-sharing services and remote batch processing.
II-l
-------
SUPPLY
SUBSYSTEM
TRANSPORTATION
SUBSYSTEM
POWER PLANT
DATA
RATE
DATA
DEMAND
SUBSYSTEM
ACID
PLANT
DATA
SCRUBBING
COST
GENERATOR
TRANSPORTATION
COST
GENERATOR
PRODUCTION
COST
GENERATOR
MARKET SIMULATION
LINEAR
PROGRAMING
MODEL
f
\
EQUILIBRIUM
SOLUTION
RESULTS
/\
\/
Figure II-l. Byproduct marketing systei
II-2
-------
DATA SOURCES
Much of the data used in this system exists in other automated
systems and published reports. In this system, however, it is organized
as an integrated structure of power plant data, emission regulations,
emissions data, compliance strategies, compliance costs, byproduct
production, potential markets for salable byproducts, transportation
costs for salable byproducts, and projections of net revenues resulting
from the sale of byproducts. The primary data sources include Federal
Power Commission (FPC) Form 67 data; the EPA Energy Data System (EDS);
the Compliance Data System (CDS); a special report from the U.S. Bureau
of Mines on limestone sources and costs; EPA's Utility FGD Survey
prepared by PEDCo Environmental, Inc., (PEDCo); TVA-EPA engineering
cost estimates of FGD processes; Centre Mark Company (CENTRE) geographic
data for latitudes and longitudes of shipping points; Standard'Point
Locator Codes (SPLC); Federal Information Processing Standard (FIPS)
codes; National Rate Basis Tariff (NRBT) data; Docket 28300 transportation-
related data; TVA's Worldwide Fertilizer-Related Data Base (sulfuric
acid data); elemental sulfur source and transportation data; National
Emissions Data System (NEDS) data; and Stanford Research Institute (SRI)
reports on sulfuric acid plants.
PREMISES
The system is based on an extensive set of premises which is fully
covered in the referenced publications (1-4). Those most directly
related to the development of the byproduct marketing system are discussed
below.
The determination of costs for FGD systems is based on economic
premises developed jointly by EPA and TVA which permit comparison of
different systems using a common basis for developing capital costs and
annual revenue requirements. These costs can be further developed to a
single yearly cost which is comparable between FGD systems evaluated on
the same premise set. The incremental cost difference between systems
can thus be directly related to other costs or revenues—in this case
revenue from the sale of sulfuric acid.
In most cases FGD systems producing a salable byproduct are more
expensive than either the use of clean fuel or the use of a scrubbing
system producing a waste byproduct. The higher cost of the marketable-
byproduct-producing system is, however, reduced to some degree by sale
of the byproduct. The byproduct sales price used to analyze marketing
potential is a minimum price rather than a projection of an actual
price. The minimum price used is determined by the total additional
(incremental) cost associated with producing and marketing a salable
byproduct, compared to the costs of producing a waste byproduct. The
total incremental cost consists of the actual production-related
investment and operating costs and transportation costs to the demand
points. If the incremental costs can be recovered, the two systems
compared will be economically equivalent and a marketable byproduct
strategy will be selected.
II-3
-------
The model does not consider either the cost benefits which might be
derived from net sales revenue in excess of the incremental costs or the
distribution of these benefits between the power plant producer and the
acid plant consumer. It determines if the quantity of acid produced
under the power plant conditions and locations specified can be disposed
of at the price fixed by the incremental cost between the FGD systems
compared.
Another important premise is that power plants will not establish
extensive marketing systems but will sell to sulfuric acid producers
manufacturing acid from sulfur. It is assumed that the producers will
reduce their own production accordingly if purchase of power plant acid
is economically attractive. An important portion of the byproduct
marketing system is therefore the determination of manufacturing costs
for acid producers as compared to purchase of the power plant acid.
The byproduct marketing system does not calculate compliance costs
in terms of total dollars. Far too many variables are involved for a
single total dollar value alone to be meaningful; instead, many factors
such as equipment costs, capital and investment costs, and operating
costs are integrated and reduced to a projection of costs in terms of
cents/MBtu. This figure is more useful in the analysis of the various
alternatives available because it permits comparisons of predictable and
controllable plant and boiler operating factors and fuel characteristics.
Neither does the system attempt to maximize benefits for a single power
plant or acid plant; instead, a solution is provided that results in
maximum cost savings for both industries combined, considering the
interaction of all power plants and acid plants at the same time. The
number of power plants and acid plants considered in the model can be
controlled so that subsets of either or both industries can be considered.
A marketing model can also be constructed for any subset. However,
equilibrium solutions generated for the current studies address both
industries as a whole.
SYSTEM ORGANIZATION
The overall system was shown diagrammatically in Figure II-l. It
can be divided into four subsystems: the supply subsystem (Figure II-2)
in which data related to FGD systems and byproduct production are generated
the demand subsystem (Figure II-3) in which data related to acid producers
are generated, the transportation subsystem (Figure II-4) in which
byproduct-related transportation costs are generated, and the linear
programing subsystem (Figure II-5). The linear programing subsystem
uses data from the other subsystems to develop a model and produce an
optimum equilibrium solution. The system develops solutions on a case-
by-case basis that will result in maximum benefit to both industries
when considered as a whole.
Execution of the total system is a complex and costly task. Signifi-
cant savings can be realized by proper planning and analysis; runs of
selected parts of the system as opposed to running the total system may
II-4
-------
FPC
FORM 67
DESIGN
POWER
PLANT
PROJECTIONS
EPA
ENERGY
DATA
SYSTEMS
DATA BASE
FPC
FORM 67
OPERATIONAL
DATA
COMPLIANCE
DATA
SYSTEMS
DATA
BASE
CENTRE
GEOGRAPHIC
DATA
SUPPLY
SUBSYSTEM
DATA
U.S. BUREAU
OF MINES
SPECIAL REPORT
PEDCO
REPORTS
ENGINEERING
COST
ESTIMATES
COST FACTORS
PROJECTED
SUPPLY
DATA
INTERACTIVE
TERMINAL
INQUIRIES
AND
ANALYSIS
SCRUBBING
COST
GENERATOR
Figure [1-2. Supply subsystem.
JI-5
-------
TVA
WORLDWIDE
FERTILIZER
RELATED
DATA
BASE
COMPLIANCE
DATA
SYSTEMS
DATA
BASE
CENTRE
GEOGRAPHIC
DATA
SULFUR
TRANSPORTATION
DATA
DEMAND
SUBSYSTEM
DATA
SULFUR
TERMINALS
PRODUCTION
COST FACTORS
INTERACTIVE
TERMINAL
INQUIRIES
AND
ANALYSIS
NATIONAL
EMISSIONS
DATA
SYSTEMS
DATA
BASE
STANFORD
RESEARCH
INSTITUTE
REPORTS
SULFUR
TARIFFS
PROJECTED
DEMAND
DATA
ACID PRODUCTION
COST
GENERATOR
Figure 11-3. Demand subsystem.
11-6
-------
DEMAND
POINTS
INTERACTIVE
TERMINAL
INQUIRIES
AND
ANALYSIS
TRANSPORTATION
SUBSYSTEM
DATA
CENTRE
GEOGRAPHIC
DATA
TRANSPORTATION
COST
GENERATOR
BARGE
TRANSPORTATION
DATA
TARIFFS
I
Figure FI-4. Transportation subsystem.
I [-7
-------
SUPPLY
DATA AND
COSTS
f TRANSPORTATION
I DATA AND
V COSTS
LINEAR PROGRAMING MODE!
EQUILIBRIUM
SOLUTION
REPORT
DEMAND
DATA AND
COSTS
INTERACTIVE
TERMINAL
INQUIRIES
AND
ANALYSIS
(\J
-,
EQUILIBRIUM
SOLUTION
FOR COMPUTER ANALYSIS
Figure 31-5. Linear programing marketing model subsystem.
II-8
-------
satisfy many requirements. All of the files, whether provided independ-
ently or generated within the system, are available for listing and
analysis. In some cases an examination of the detailed results from the
runs made for earlier studies may be adequate.
The execution of the system can be described in its simplest form
as:
1. Execute the supply subsystem to project costs and quantities
of supply.
2. Execute the demand subsystem to estimate the market (price and
quantity) for the byproduct.
3. Execute the transportation subsystem to calculate the trans-
portation costs for all possible combinations of supply points
and demand points.
4. Execute the linear programing marketing model to determine the
optimum combination of supply points, demand points, and trans-
portation between them.
The preceding discussion and block diagrams (Figures II-l through
H-5) are useful for conceptual purposes only. Greater detail is required
to show the actual interface of the data bases and programs that make up
the system.
SUPPLY SUBSYSTEM
The supply subsystem as shown in Figure II-6 generally includes all
of the data and programs that are related to power plants. Smelters are
a special case of supply and for reasons described later are discussed
as a part of the demand subsystem.
Supply Data Base
Four of the data files and the power plant data base used in the
supply subsystem were developed independently of the normal system flow
and are provided directly with no programs or procedures to generate
them. Where modifications to these files are necessary they can be
edited as needed. Modification capabilities for the power plant data
base are not expected to be needed by users because the necessary data
base values are extracted within the system and placed into standard
sequential data files. The data files provided in addition to the power
plant data base are (1) the scrubbing investment and operating factors
based on the general cases developed in Detailed Cost Estimates, (2)
scrubbing cost data by year which are time and area adjustments for the
investment and operating factors in step 1, (3) site-factor adjustments
for individual plants, which are modifications to the general case
estimates of investment and operating costs based on known plant-specific
II-9
-------
SCKUBBKR
INVESTMENT
AND OPERATING
FACTORS
SASDATfi
SCRUBBER
INVESTMENT
AND OPERATING
COST DATA
N.^ SCRPRC
DELIVERED
COSTS OF
LIMESTONE
TO
POWKR PLANTS
I.IMEKST
POWKK
PLANT
DATA
BASE
1
CAI.I
i
UI.ATION
OF
POWKR I'l.ANT
PROJECTION
DATA
PRO.IICCT
POWKR PLANT
SITK SPECIFIC
ADJUSTMENTS
SCRSIT
rl'RO.IF.CTED
M.ANT I.KVF.l,
DATA
\
I'RO.IKCTED
BOILER LEVEL
DATA
ADDITION OF
DKI.IVKKKD
I.IMKSTONK
COSTS
AUDI.INK
TROJECTED
PLANT LEVEL
DATA AND
LIMESTONE
COSTS
IIPI.IMK
I
SCKDHKINi;
COST
CKNRRATOR
S1MCAI'
POWER PLANT
SCRUBBING
COSTS AND
ill'ANTITIKS
SCRCST
Figure II-6. Supply subsystem flow v-hart,
11-10
-------
exceptions to the values normally used, and (4) a delivered cost of
limestone for each power plant. All of these are described in section IV.
Supply Programs
The supply subsystem is made up of two major programs and a small
data merge program. The major programs are the power plant data extract-
projection program and the scrubbing cost generator program. They are
both described in detail in section III. Even though the projection
program is a major program, frequency of use is expected to be low
because the output sequential files that it generates can be modified
directly without executing the program. The supply subsystem as typi-
cally run consists of the following steps: (1) necessary corrections or
additions are made to the data base and the four independent input
files; (2) the projection program is modified as required and executed
to extract the necessary data from the data base, and plant-level and
boiler-level files of projected power plant data are created; (3)
the delivered costs of limestone are added to the projected plant data
file generated in step 2; and (4) the scrubbing cost generator program
is executed. The scrubbing cost generator program can be executed in
different modes depending upon specific requirements.
In addition to the mode described above in the typical execution of
the system as a whole, the scrubbing cost generator program can be
executed independently. Instead of plant and boiler input from the data
base, specific input data can be prepared for plants, boilers, fuel
characteristics, fuel consumption, compliance regulations, etc. Scrub-
bing costs can be developed based on these values completely independent
of any data base values.
Several options are provided to control the level of output reports
produced by the program. Reported results can vary, depending upon
specifications provided, from very general, summary-level reports to
very detailed reports showing many of the intermediate results in the
calculations. Section III provides detailed information of these
caoabilities.
DEMAND SUBSYSTEM
The demand subsystem as shown in Figure II-7 consists of programs
and data related to sulfuric acid production and the cost factors
associated with it. All acid producers are grouped together in a single
file with a code that indicates the feedstock used. The demand subsystem
analyzes the feedstock code and selects only those producers burning
elemental sulfur (referred to as acid plants for convenience) as potential
buyers of power plant byproduct sulfuric acid. The acid producers using
smelter off-gas as the feedstock (referred to as smelters for convenience)
are not potential buyers of power plant byproduct acid; they are in fact
competitors of the power plants for the sale of byproduct acid.
11-11
-------
ACID PLANT
AVOIDABI.K
PRODUCTION
COST FACTORS
AC III PLANTS
AND
SMKI.TKRS
M01.TKN SULFUR
TARIFF
DATA
CALCULATION OK
DKI.IVKRKI)
SULFUR COSTS
TO ACID
.ANTS
IKI.IVKRKI)
COS!' RKPORT
ACID PLANTS,
SMKI.TKRS,
AND DKI.IVKRKI)
SULFUR COSTS
OPTIONS AND
VARIABLKS
AVOIDABLE
ACID PRODUCTION
COST CF.NF.RATOK
ACID PLANT
AVOIDABI.K
PRODUCTION
COSTS AND
OIJANTITIF.S
SHKI.TF.R
PRODUCTION
COSTS AND
JUANTITIES
RKPORTS OF
AVOI DABI.F. COS'I
CALCULATIONS
Figure II-7. Demand subsystem flow chart.
11-12
-------
Even though the market is assumed to be in equilibrium between
supply and demand at some point in time, future production increases by
the smelters will result in additional quantities of byproduct acid.
Any market analysis and marketing model runs must take into account the
increased production of smelter acid as well as power plant acid. A
projected estimate of future production increases for all smelters is
calculated based on their current capacity and compliance status with
respect to clean air regulations. An additional production increase is
projected for smelters out of compliance; the increased removal of
sulfur dioxide required for compliance will result in a corresponding
increase in abatement acid production. Because smelters are committed
to the production of byproduct acid, whereas power plants have a wider
choice of emission control alternatives, it is expected that they will
be able to market their additional production before any power plant
byproduct acid can be marketed unless transportation costs are prohibitive.
For this reason the smelters are considered as having a zero unit produc-
tion cost for the increased production which gives them an advantage
over the power plants in the marketing model. Power plants will generally
have a greater-than-zero unit production (incremental) cost. The smelters
are separated from the acid plants that were also in the original input
data and combined with the potential power plant byproduct acid producers
to make up the total supply considered in the linear programing marketing
model. The acid plants that remain in the original input data make up
the total demand considered in the model.
Demand Data Base
Five of the data files used in the demand subsystem are provided as
independent input files with no programs to generate them. These
independent files were either generated manually from various sources or
were already developed as a part of other studies. They are described
in detail in section IV. As in the supply subsystem, the data can be
changed as required or new records can be added if necessary. The
independent data files provided are (1) values for calculating avoidable
costs of acid production (investment costs, maintenance costs, fixed and
variable production costs, delivered cost of sulfur, etc.), (2) tariff
rates for sulfur, (3) rail mileages for all possible rate basing points,
(4) acid plants and smelters—their location, capacity, etc., and (5)
molten sulfur terminals.
Demand Programs
The demand subsystem is made up of two programs and various data
files that determine potential byproduct acid consumers. The first
program is used to calculate the delivered costs of elemental sulfur to
each acid plant using the cheapest combination of barge and rail trans-
portation from Port Sulphur, Louisiana, via the various molten sulfur
terminals. The second program is used to calculate an estimated avoidable
cost of acid production based on the results of the first program and
plant capacity, plant efficiency, age, compliance status, etc. The
smelters are processed by the programs and placed in a separate file as
mentioned previously.
11-13
-------
The calculation of the delivered cost of sulfur to each acid plant
uses transportation-related data which are discussed in the transporta-
tion subsystem narrative. The transportation of sulfur by barge, however,
applies only to the development of avoidable acid production costs. The
documentation for the program that calculates the delivered cost of
sulfur describes the procedure in detail. An NRBT rate basing point
must be determined for any acid plants or sulfur terminals before they
can be added to runs of the demand subsystem or of the system as a
whole. More details are provided in the description of the transporta-
tion subsystem.
The demand subsystem is typically run in the sequence just presented.
The second program may require several executions, varying the cost
factors in each run to determine the effects of changes (both individually
and in combination with other changes) on the calculated avoidable costs
of sulfuric acid production. For more detailed information, refer to
sections III and IV.
TRANSPORTATION SUBSYSTEM
The transportation subsystem (Figure II-8) consists of the programs
and data required to calculate rail transportation costs from all poten-
tial suppliers to all potential consumers. Some of the data files are
also used in the demand subsystem to calculate the contribution of
sulfur transportation costs to avoidable production costs.
The transportation subsystem was designed on the basis that a
totally automated transportation data base integrating all rail, barge,
and truck data would be developed, along with the programs necessary for
interface. The data base was created and the total size was over 25
million characters. It contains every named location in the 48 contiguous
states for shipping by rail, barge, or truck. There are over 100,000
truck points, over 20,000 rail stations associated with the various rail
carriers, and almost 2,000 barge points on the navigable inland waterway
system.
The data base was designed to provide the information necessary to
determine the least-cost method of shipment between any two locations
via rail, barge, truck, or any combination of the three modes. Once the
data base was created an analysis was made, based on the data base
structure, of the computer costs required to link all truck points to
the nearest rail and barge points and to link all rail points to the
nearest barge point where applicable. In both cases the appropriate
mileage between the points would also be included during the linking
process. On the basis of potential benefits to the byproduct marketing
system alone the cost of linking all points did not appear to be justified.
As a result, the transportation subsystem is currently limited to rail
except for the molten sulfur barge terminals.
More details on the integrated transportation data base are provided
in Appendix A. There may be some potential for its use in other applications
11-14
-------
;OWER PLANT
SCRUBBING
COSTS AND
UANT I TIES
ACID PLANT
AVOIDABLE
PRODUCTION
COSTS AND
QUANT ITIES
ACDCST
SMELTER
PRODUCTION
COSTS AND
QUANT IT IKS
SMLCST
TRANSPORTATION
RATE BASING
DATA
RBNSORT
SELECTION OE RATE
BASING DATA FOR
SMELTERS, POWER PLANTS,
AND ACID PLANTS
MANUAL PROCEDURE
SULFUR 1C
ACID
TARIFF
DATA
X'31 3H2S
I
POWER PLANTS,
ACID PLANTS, AND
SMELTERS WITH
RATE BASING DATA]
TRNPTS
I
RANSPOKTATION
COST
GENERATOR
I'RNCOST
TRANSPORTATION
TES FROM POWER
PLANTS AND
SMELTERS TO
ACID PLANTS
. TRNCST
DIAGNOSTICS
AND
PROGRAM EXECUTION
REPORT
-'ic-jre II-3. Tra:-.s-portatio:i subsystem flow chart
-------
or systems. An additional reason for including the data base is that
the final step of linking the locations may be completed in the future.
If so, it should then be made an integral part of the byproduct marketing
system.
Nonintegrated Transportation Data Base
Three of the data files used in the transportation subsystem are
provided independently with no automated generation procedure. One of
these files, the rail mileage file, is used in the demand subsystem.
The rate basing point file is also required in the demand subsystem if
new sulfur terminals or acid plants are added. The remaining file
contains the tariff rates. Three other files (power plants, smelters,
and acid plants) are also used for input data for the transportation
subsystem but they are generated as an output of the two previous sub-
systems.
Transportation Programs
The transportation subsystem contains a manual procedure and a
transportation cost generation program. A manual procedure is required
because the fully integrated transportation data base is not currently
used.
The manual procedure determines the rate basing point for each
potential supplier, consumer, and sulfur terminal in the system. A part
of this manual procedure must be completed before the demand subsystem
discussed previously can be run when new locations are added. All
mileage and tariff considerations rely on these rate basing points to
determine shipping charges.
The transportation cost generation program calculates shipping
costs for all possible combinations of potential suppliers and potential
consumers by using the results of the manual procedure, the rail mileage
file, and the applicable tariffs. The output file of transportation
costs and the output data from the two previous subsystems complete the
requirements for running the linear programing marketing model subsystem.
LINEAR PROGRAMING MODEL SUBSYSTEM
After the supply, demand, and transportation subsystems have been
completed, the linear programing subsystem (Figure II-9) can be run.
The subsystem consists of three main programs and a procedure to sort
the data to be reported. The power plant, smelter, acid plant, and
transportation data are inputs for the subsystem. The first program of
the subsystem generates the linear programing model for APEX. APEX is
an optimization system which produces the optimum result by maximizing
gain or minimizing losses for the situation being considered. All of
the data items in the input files are not usable in the model itself.
They must be separated into another file with the necessary identifica-
tion, sorted, and combined with the APEX results in the third program in
order to provide a more descriptive report.
II-16
-------
TRANSPORTATION
RATES FROM
POWER PLANTS
AND SMELTERS
TO ACID PLANTS
POWER PLANT
SCRUBBING
COSTS AND
QUANTITIES
ACID P1.ANI
AVOIDABLE
PRODUCTION
COSTS AND
QUANTITIES
SMELTER
PRODUCTION
COSTS AND
QUANTITIES
MODEL
GENERATION
OPTIONS AND
VARIABLES
GENERATION
Of
LINEAR
PROGRAMING
MODEL
MODEL GENERATION
REPORT
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
(UNSORTED)
.1NEAR
PROGRAMING
MODEL
SORTING OF
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
APEX LINEAR
1'ROGRAM ING
SYSTEM
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
(SORTED)
SOLUTION
FILE FOR
USER RKPORI
GENERATION
STANDARD
APEX REPORTS
REPORT
GENERATION
OF
E()IIILI BRHIM
SOLUTION
REPORT
GENERATION
VARIABLES
EQUILIBRIUM
SOLUTION FOK
AUTOMATED
ANALYSIS
EQUILIBRIUM
SOLUTION REPORT
(SUPPLIERS,
CONSUMERS, (1UAN
TIT IKS, AND
COSTS
FiTura II-9. Linear programing marketing nodel subsystem flow chart.
II-• 7
-------
The third and final major program in the subsystem is the report
generator which builds the final output report format. This program is
required even though a standard output report is provided by the APEX
system. The APEX report is very general and inflexible for descriptive
presentation of results; therefore, another file is also provided to
allow customized reporting of results. An on-line storage copy of the
output produced by the report generator is written to allow further use
if required. The subsystem is typically run in the sequence described
above. A more detailed description is provided in sections III and IV.
A greater variation in run cost is more likely in this subsystem
than in the previous ones so careful planning and evaluation of require-
ments are necessary. The use of batch versus interactive runs and
avoidance of large model runs that may not be any more meaningful than a
smaller scale run are very important in controlling costs.
SUMMARY
The system description section has been as general and basic as
possible because the following program and data description sections are
of necessity very specific and involved. Even though the overall system
can be described conceptually so that it is relatively easy to understand
it is very complex at the operational level. Because of this complexity--
and so that the functional aspects of each program and data file can be
better analyzed using documentation in subsequent detailed sections—a
final step-by-step summary may be useful. The preceding flow charts and
narratives have shown the various relationships between the programs and
data used in the system.
There are 12 basic steps in a typical system execution:
General
1. The power plant data base, the independent files, the programs
and the user procedures are loaded from magnetic tape to on-line
storage.
Supply
2. The projection program executes, extracts the power plant-related
data from the data base, and generates projected plant-level and
boiler-level data files.
3. The limestone delivered cost program executes, adds from an
independent file the delivered cost of limestone for each power
plant to the projected plant data file created in step 2, and
generates a combined projected plant data and limestone cost
file.
11-18
-------
4. The scrubbing cost generator program executes; uses the independ-
ent files of investment and operating data, cost data, and site-
specific location factors; uses the projected plant-level data
file created in step 2 and updated in step 3; and finally uses the
projected boiler-level data file created in step 2; and generates
a file of power plant scrubbing costs and quantities as well as
various optional printed reports that have been selected.
Demand
5. The program to calculate delivered sulfur costs to acid plants
executes; uses the independent files of acid plants and smelters,
sulfur tariff rates, rail mileages, and sulfur terminals; and
generates a file containing the delivered cost of sulfur to each
acid plant and an optional printed report of the delivered costs.
6. The program to generate the avoidable production costs of sulfuric
acid for each acid plant executes, uses the independent file of
avoidable production cost factors and the acid plant file with
delivered costs of sulfur generated in step 5, and generates a
file of acid plants with their avoidable costs and quantities,
a file of smelters with their production costs and quantities,
and optional printed reports.
Transportation
7. The acid plant file and smelter file from step 6 and the power
plant file from step 4 are examined manually and, where necessary,
,the independent file of rate basing points is used to select the
appropriate rate basing point for each power plant, acid plant,
and smelter. A transportation file based upon locations in the
three input files is generated manually.
8. The transportation cost generator program executes; uses the
independent files of rail mileages and sulfuric acid tariffs and
the transportation file for power plants, smelters, and acid
plants from step 7; and generates a file of transportation costs
for every possible combination of shipments from power plants and
smelters to acid plants and also a printed report related to the
costs.
Linear Programing Marketing Model
9. The program to generate the linear programing model executes;
uses the power plant file from step 4, the smelter file and acid
plant file from step 6, and the transportation cost file from
step 8; and generates the linear programing model based upon a
user-provided ACFL value. It also generates a file of information
that is not input to the model but is required for reporting
results.
II-19
-------
10. The APEX linear programing system executes, solves the linear
programing model generated in step 9, and generates a file con-
taining the solution that will be used in the report program. It
also generates a standard report describing the solution.
11. The report generator program executes, uses the file containing
the model solution generated in step 10 and the file of information
for reporting created in step 9, and generates a printed report
of the solution showing potential sales by power plants and smelters
and potential purchases by acid plants (including the price
received, price paid, and transportation costs for each trans-
action where applicable) as well as an on-line file of the report
for further analysis.
12. The report from step 11 is analyzed. Based on the results of the
market potential shown in the model solution and the ACFL used
final strategy selections are made for all power plants.
Because the independent files are not generated by the programs in
the system and therefore have no detailed description of their creation
the data they contain should be analyzed carefully (refer to section IV).
The resources expended in the research and development of these data
files and data bases were at least as great as the resources required
for developing the remainder of the system. The independent data include
the power plant data base, delivered costs of limestone to each power
plant, scrubber investment and operating factors, scrubber cost data
power plant site-specific adjustments associated with the various scrubbing
processes, acid plant and smelter data, acid plant avoidable production
cost factors, molten sulfur terminals, rate basing point data, rail
mileage data, and sulfur and sulfuric acid tariffs. External files can
generally be modified as required. The internal files can be modified
either directly or by reexecution of the appropriate program.
11-20
-------
III. PROGRAM DESCRIPTIONS
This section provides a general narrative and block diagram for all
programs and procedures that make up the system. Each of the block
diagrams is a subset of the detailed subsystem flow charts shown in
section II (Figures II-6 through II-9). The narratives are intended to
aid in usage, not to document all of the internal program details. A
current source listing of the appropriate program must be referred to
for exact coding details. This should not be necessary except for the
most rigorous analysis and is not typically required for system execution.
This section and the following sections on file and record descrip-
tions and systems usage require a greater understanding of the concepts
of programing and data processing than did the preceding sections. The
descriptions should still be of some value to those with no specialized
background or training but the system is designed to execute on modern
high-speed computers and it is impossible to avoid the terminology and
concepts required for actual usage.
PROJECT PROGRAM
The projection program (Figure III-l) was designed to extract
information from the power plant data base and generate the data projec-
tions required by the scrubbing cost generator. Because data must be
projected for the industry as a whole, some generalizations, assumptions,
and data default values must be used where individual data items are
either not reported or are inconsistent with other values. When this
occurs the values used are based upon extensive analyses of both current
and historical data.
Users can enter specific data values into the generated plant and
boiler records and override program projections if necessary. This is
obviously not practical on a plant-by-plant basis for the whole industry
which is made up of more than 900 power plants and 3500 boilers that
have over 5 million characters of data associated with them in the data
base.
Familiarity with the power plant data base is required during
subsequent program documentation and section IV should be referred
to as required. The projection program is the only interface between
the byproduct marketing system and the power plant data base. It extracts
all of the necessary information and writes the data to sequential files
for use throughout the rest of the system. Based upon the data base
description, tha data categories required from the data base are (1)
III-l
-------
POWER
PLANT
DATA
BASE
PLANTS
CALCULATION
OF
POWER PLANT
PROJECTION
DATA
PROJECT
PROJECTED
PLANT LEVEL
DATA
PLAS
f PROJECTED
/ BOILER LEVEL
1 DATA
BLAS
Figure III-l.
Block diagram for the calculation of power plant
projection data (program PROJECT).
Ill-2
-------
general plant data, (2) plant data for both the base year and the year
to be projected (1973 and 1978 in the original studies), (3) regulations,
(4) regulations by pollutant, fuel type, and applicability, (5) general
boiler data, (6) boiler/stack configuration data, and (7) boiler data by
year (available for the base year, but not for the year to be projected
because of the method of reporting on FPC Form 67).
Projections of future fossil-fuel usage (both quantities and character-
istics) are reported by the utilities on FPC Form 67 and entered in the
data base at the plant level, not the boiler level. Most of the clean
air regulations apply at the boiler level and all scrubbing cost generator
calculations related to compliance strategies and alternatives are made
at the boiler level. An allocation method based upon historical relation-
ships between plant-level and boiler-level operating characteristics and
fuel consumption is used to analyze the plant-level data and derive data
at a boiler level for the year to be projected. Although program projections
for plants with no base-year boiler data could be made directly at the
boiler level, all program projection estimates for plants with incomplete
or no base-year boiler-level data are made at the plant level. As a
result, the same procedure is used for all plants to derive data at the
boiler level. Six separate cases of data conditions are identified for
projection purposes.
Case 1 identifies plants for which both base-year boiler data and
projection-year plant data are available.
Case 2 identifies plants for which base-year boiler data are
available, projection-year plant data are not available, and plant
capacity for the projection year is no more than 100 MW greater than the
base-year plant capacity.
Case 3 is the same as case 2 except that it identifies plants for
which the plant capacity in the projected year is over 100 MW greater
than the base-year plant capacity.
Case A identifies plants for which neither base-year boiler data
nor projection-year plant data are available. This case is included
primarily for programing considerations.
Case 5 is currently not used.
Case 6 identifies plants that do not have base-year boiler data
available but have projection-year plant data available.
The key factor in allocating fuel to the boiler level from the
plant level is the boiler capacity factor. An analysis of historical
boiler capacity factors using FPC data from 1969 through 1973 and Detailed
Cost Estimates resulted in an algorithm based primarily on boiler age to
estimate capacity factors for the projected year. It is recognized that
there have been some major changes in fossil-fuel usage patterns along
with related cost implications. These patterns should be reflected in
new data as it become available and appropriate adjustments can be made
if required.
III-3
-------
Regulations may apply at the plant, stack, and boiler levels. When
regulations apply at the plant level the appropriate regulations are
included in the plant data record. When regulations apply at the boiler
level, in most cases they apply for all boilers at a given plant and
they are included in each boiler data record for that plant. For the
cases where individual boilers have a specific regulation, the regulation
that applies is included in the appropriate boiler data record.
If stack-level emission regulations apply, some problems exist.
Power plants are designed with a wide variety of boiler - stack configura-
tion options and these options are reflected in the reporting of data on
FPC Form 67, which generally lists the options rather than actual or
customary use. A procedure was developed with EPA concurrence to apply
stack regulations at the boiler or total plant level. When a plant has
only one stack all boilers obviously feed the same stack and stack-level
regulations are treated as a plant-level regulation. When one-to-one
boiler - stack configurations exist the stack regulations are applied as
a boiler regulation. When multiple stack - boiler combinations exist
and all stacks have the same regulation, each boiler is assigned the
stack regulation value on a boiler-by-boiler basis. If stacks within a
plant have different regulations each boiler reported to feed multiple
stacks is assigned the most rigorous regulation of the stacks fed.
In actual usage plant- and stack-level regulations are expected to
have little if any effect. The best candidates for byproduct marketing
are relatively new boilers that typically have to meet Federal boiler-
level emission standards.
The above procedures apply for each power plant in the data base.
The results are written to two data files—plant data (PLAS) and boiler
data (BLAS).
ADDLIME PROGRAM
The ADDLIME program (Figure III-2) is a merge program that reads
the PLAS file created by the projection program and the independent
LIMEEST file of delivered costs of limestone to each power plant and
creates a combined output file. The output file, UPLIME, is saved so
that the projection program will have to be run only when program
changes are made or the power plant data base is updated. If a projec-
tion program run is made and the ADDLIME program is not run, the limestone
cost field will be zero and the scrubbing cost generator program will
use a constant default value for every plant instead of the value for
each plant in the LIMEEST file.
III-4
-------
DELIVERED
COSTS OF
LIMESTONE TO
POWER PLANTS
LIMEEST
PROJECTED
PLANT LEVEL
DATA
PLAS
ADDITION OF
DELIVERED
LIMESTONE
COSTS
ADDLIME
PROJECTED
PLANT LEVEL
DATA AND
LIMESTONE
COSTS
UPLIME
Figure III-2. Block diagram for the addition of delivered
limestone costs to power plants (program ADDLIME).
111-5
-------
STMCAP PROGRAM
The scrubbing cost generator program (Figures III-3 and III-4) was
designed to provide a consistent method for projecting comparative costs
of installing FGD systems on power plants. Design capabilities include
both cost and byproduct quantity projections for up to six FGD processes.
Engineering data now included are limited to the use of clean fuel and
the limestone, magnesia, sodium, and gypsum scrubbing processes; the
remainder of the system relating to the marketing and transportation of
abatement byproducts is now limited to sulfur and sulfuric acid, but the
design capabilities of the program allow addition of other byproducts.
When the overall program potential and the possible variations in its
use are considered (such as the execution time options, input-data-
modification options, override conditions, selection of output reports,
and print options for displaying intermediate results of some of the
more complex calculations) the scrubbing cost generator is by far the
most complex program in the system.
The results of the scrubbing cost generator have a critical influence
on the remainder of the system. Selection of a strategy that will lead
to compliance (clean fuel, a disposable byproduct process, or a marketable
byproduct process) and the potential sales when a marketable process is
indicated are both totally dependent upon the projected costs and quantitie
calculated for each compliance strategy by the scrubbing cost generator.
To minimize modifications to both the program and to the standard
input data, program execution was designed for usage flexibility.
However, this flexibility increases the procedural complexity associated
with program execution. Because the internal program coding is also
complex the overall requirements for usage are relatively difficult. A
general discussion of the input and output should help to clarify details
of the options and overrides that will be covered later.
The primary source of input data used in the program (other than
the power plant data) is the investment and operating data for the
various scrubbing processes in Detailed Cost Estimates. Even a cursory
analysis of this program and its results requires familiarity with that
publication.
Program processing, whether accomplished using the supply subsystem
(Figure III-3) or as an independent execution with user-supplied power
plant data (Figure III-4), will always require the following input
files: the independent file of investment and operating factors, the
independent file of cost data, and the options and overrides to be used.
The block diagram for this program when executed using the supply sub-
system (Figure III-3) also shows an independent file of site-factor
adjustments. If these are required this file should obviously be used
but it is optional as far as uninterrupted program processing is concerned
In any case the site-factor file will normally contain records for a
limited number of plants. Specific site-factor adjustments are required
III-6
-------
ROJECTED
PLANT LEVEL
DATA AND
LIMESTONE
COSTS
SCRUBBER
INVESTMENT
AND
OPERATING
COST DATA
SCRUBBER
INVESTMENT
AND
OPERATINC
FACTORS
PROJECTED
BOILER LEVEL
DATA
POWER PLANT
SITE SPECIFIC
ADJUSTMENTS
OPTIONS
AND
OVERRIDES
SCRUBBING
COST
GENERATOR
DETAILED COST
CALCULATION
RRPOKT
(OPTIONAL)
EMISSIONS
REPORT
(OPTIONAL)
EMISSION
SUMMARY
REPORT
(OPTIONAL)
BOILER COST
REPORT
(OPTIONAL)
PLANT SCAN
REPORT
(OPTIONAL)
POWER PLANT
SCRUBBING
COSTS AMIS
QUANTITIES
(OPTIONAL)
Figure III-3.
Block diagram for the calculation of power plant scrubbing costs
as a part of the overall system (program STMCAP).
I].!-/
-------
OPTIONS,
OVERRIDES, AND
USER SUPPLIED
PLANT AND
BOILER DATA
SCRUBBER
INVESTMENT
AND
OPERATING
FACTORS
SCRUBBING
COST
'.ENF.RATOR
OPTIONS AND
OVERRIDES
REPORT
DETAILED COST
CALCULATION
REPORT
(OPTIONAL)
INPUT EDIT
REPORT
(OPTIONAL)
EMISSIONS
REPORT
(OPTIONAL)
PLANT SCAN
REPORT
(OPTIONAL)
BOII.KR COS!
REPORT
(OPTIONAL
POWER PLAN!
SCRIIBBINC
COSTS AND
QUANTITIES
(OPTIONAL)
Figure III-4. Block diagram for the calculation of power plant scrubbing cost
independently of the overall system (program STMCAP).
-------
only when unusual plant conditions would cause significant deviations
from the base-case calculations. Site-factor adjustments are normally
not used for user-supplied power plant data because the input data will
usually reflect any unusual plant characteristics.
Power plant data can be provided in either of two ways. If program
execution is a part of the supply subsystem (Figure III-3) the input
power plant data will typically be the output from the projection program
(which uses the power plant data base) and the program that adds limestone
costs. If execution is independent of the supply subsystem (Figure III-4)
the input power plant data will be provided by the user.
General Processing
The minimum output from the program is the options and override
report; any other output must be specified using the option specifica-
tions. If no other output is specified none is generated so program
execution without at least one optional output specification is
meaningless.
The options are the first data processed by the program, immediately
followed by the processing of some of the overrides. The overrides can
be functionally grouped into three categories. The first category is
made up of overrides for internal program constants and tables; the
second category consists of overrides to the independent file of scrubbing
cost data; and the third category consists of overrides for the independent
investment and operating factors. Although all of the overrides are
optional, there is such a wide variation in the requirements for individual
program execution that some of the internal program values almost always
require overrides. The scrubbing investment and operating factors and
cost overrides are not necessary in many cases.
The conceptualized program flow can be generally described as
follows: program control options are processed; any specified overrides
to the internal program constants and tables are processed; the independent
cost data along with any specified overrides to the cost data are processed;
the independent investment and operating factors along with any overrides
specified are processed and an internal file for using this data later
in the program is generated; and finally the power plant data and any
site-specific data values are processed (site-specific data do not
apply if power plant data are user supplied). Concurrently with all of
the above, specified output data files and reports are generated.
For input processing all of the input data are assigned to a fixed
unit number (or file) except for the cost overrides, investment and
operating factor overrides, and the power plant data (if user supplied).
The unit number (TAPES for example) to be read for these three data
inputs may be controlled by the user by means of the options. For
output data the only fixed unit (or file) assignments are the scrubbing
cost output file that will typically be used as input to the transporta-
tion and linear programing subsystems and the option and override report
(overrides to internal program constants and table values only). All
111-9
-------
other output, including the reports resulting from the other categories
of overrides, may be controlled by the user by means of the options
specified. The only limitation on output report control is the number
of units provided in the current version of the program. Four separate
unit numbers for reports are provided, when a greater number of reports
is required some of them must be combined on the same unit. Program
changes can be made in special cases to increase the number of output
units but this will increase program size and processing costs.
The following detailed description of the options is based upon the
current version of the scrubbing cost generator program and, as is the
case when general documentation cannot be kept independent of detailed
program coding, program modifications may cause exceptions to the docu-
mented procedures. Although program changes that alter the execution
procedures are not now planned they could be required sometime in
the future. If this occurs, any changes to the documented instructions
for program processing will be listed when the program is executed so
that required adjustments to previous usage procedures can be made.
Options
The options to control the scrubbing cost generator program are
always read from the standard input file and must be provided in NAMELIST
format. (NAMELIST is described in the FORTRAN manual listed in
Appendix B.) The internal program variable name of each option an
abbreviated description of each, the default value, and the user-supplied
values are always written immediately to the standard output print file
After reading and listing the options, the overrides of the internal
program constants and table values are read from the standard input
file. The variable names, along with their default values and the user-
supplied values, are immediately written to the standard output print
file. The above steps are independent of any options specified.
After the steps just described are executed the specified options
control program processing. As shown in Table III-l, up to 16 options
are provided. Some of these options are mutually exclusive while others
are mutually inclusive. Except for the IN, 10, and KSTOP options they
are all initially set to zero and the particular processing step controlled
by each option is omitted unless an appropriate nonzero value is specified
The usage sections and the sample output reports should be referred to
for more details. The option names, descriptions, and acceptable values
of each are as follows:
The IN option specifies the unit number that cost data overrides
(KPRICE option), investment and operating data overrides (KSAS option),
and user-supplied power plant data (KSIM option) will be read from. If
none of these three options are provided, the unit specified by the IN
option will not be read. The default value is the standard input file
and no other unit number is provided. If the standard input file is
unsatisfactory a program change must be made to specify the desired unit
number.
111-10
-------
TABLE III-l. SCRUBBING COST GENERATOR (STMCAP) PROGRAM EXECUTION OPTIONS
Option name
Description
IN Controls the unit number that is read for overrides and user-
supplied power plant data
10 Controls the unit number used for reporting overrides, investment
and operating data, detailed cost calculations, and power plant
data editing results
KSIM Controls the source of power plant data to be used (either user
supplied or from the data base)
KCHECK Controls the generation of a detailed scrubbing cost report based
on user-supplied power plant data
KOFF Forces offsite ponding calculations (not used during normal
execution)
KSCAN Controls the generation of a one-line summary for each plant
KEMISS Controls the generation of an emissions report for each plant
KCOSTLP Controls the generation of both the scrubbing cost file used in
the linear programing subsystem and a compliance cost summary
report for each plant
KCOSTP Controls the generation of a compliance cost summary report for
each plant
KPRICE Controls the reading of overrides for the independent scrubbing
cost data
KSAS Controls the reading of overrides for the scrubbing investment
and operating factors
LSAS Controls the generation of a report of the complete files of
investment and operating factors and investment and operating
cost data
KEDIT Controls continued program execution after the LSAS option above
has been processed
LCP Controls the generation of a detailed scrubbing cost report based
on data base power plant data
KSTART Specifies the beginning number of power plants that will be
processed from the data base
KSTOP Specifies the ending number of power plants that will be processed
from the data base
1 11-1 1.
-------
The 10 option specifies the unit number that a listing of cost data
overrides (KPRICE option), investment and operating data overrides (KSAS
option), the investment and operating data (LSAS option), detailed cost
calculation results (KCHECK option), and the power plant data edit
listing (KSIM option) will be written to. The default value is the
standard print file. Unless the indicated options are nonzero, no
output will be generated on the unit specified by the 10 option except
for the power plant data which are listed regardless of the KSIM option.
The allowable units are 6, 20, 30, or 40; a unit of zero can be supplied
if none of the listings are required.
The KSIM option controls the source of power plant data to be used
by the program. If KSIM is set to zero the output from the projection
and add limestone cost programs will be read, the site-factor file (if
provided) will be read, and the KCHECK option will be ignored. If KSIM
is not zero, user-supplied power plant input data will be read from the
unit specified by the IN option, the site-factor file is ignored, and
the KCHECK option will be processed if specified.
The KCHECK option is used only when KSIM is not equal to zero;
i.e., user-supplied input power plant data are processed from the unit
specified by the IN option. Selection of this option by specifying a
nonzero value results in the generation of a detailed cost report written
to the unit specified by the 10 option. (This option is the counterpart
of the LCP option for the plant and boiler data from previous subsystem
program execution and is described later.) The IPROC variable specifies
the process code for which the detailed report will be written and must
be supplied along with other user data if the KCHECK option is used. In
general, the KCHECK option is only useful during program and cost data
modification analysis and must be used carefully; it can result in large
volumes of output data for printing.
The KOFF option forces a program analysis of offsite ponding and is
not generally used. In normal usage, the distance to the pond (either
user supplied or from the data base) automatically controls the calcula-
tion of offsite ponding costs.
The KSCAN option is used to provide a quick-look summary report
consisting of a one-line entry for each plant processed by the program.
The one-line entry contains information such as plant number; plant
name; SPLC; number of boilers; whether or not the plant was calculated
to be in or out of compliance; a flag for any type of data exception
found during processing—such as regulation, fuel, or operating data; an
offsite ponding indicator; and the number of boilers scrubbed to meet
emission limits (if out of compliance). The allowable values for KSCAN
are zero if no report is required or the unit number that the report
will be written to, which may be 6, 20, 30, or 40. A one-page emission
and abatement summary for all plants combined is also generated as a
result of the KSCAN option.
111-12
-------
The KEMISS option controls the writing of the emission report for
each power plant processed. The allowable values for KEMISS are zero if
no report is required or the unit number that the report is to be
written to, which can be 6, 20, 30, or 40.
The KCOSTLP option controls the writing of both the scrubbing cost
output file that is typically input to the transportation and linear
programing subsystems and a report of the scrubbing costs on a boiler-
by-boiler basis. If the KCOSTLP option is not specified then the
KCOSTP option (see later options) is ignored. The generation of a
scrubbing cost data file and the scrubbing cost report cannot be specified
independently. The allowable values are zero if neither the scrubbing
cost file nor scrubbing cost report is required or the unit number that
the report is to be written to, which can be 6, 20, 30, or 40. The unit
that the scrubbing cost file is written to is fixed.
The KCOSTP option controls the writing of the compliance cost
summary report for each power plant calculated to be out of compliance
including costs and quantities of the byproduct produced for each process.
In order for this option to apply the KCOSTLP option above must be
nonzero or the KCOSTP option is ignored. The allowable values are zero
if no compliance cost summary report by plant is required or the unit
number that the report is to be written to, which can be 6, 20, 30,
or 40.
The KPRICE option controls the reading of overrides to the independ-
ent scrubbing cost data file. If the KPRICE option is nonzero, then
after the cost data for the specified year is read, user overrides are
read from the unit specified by the IN option. They are applied to the
appropriate cost values and are listed on the unit specified by the 10
option. Allowable values are zero if no overrides to the cost data file
are required or a nonzero value if overrides have been provided on
unit IN.
The KSAS option controls the reading of overrides to the external
file of investment and operating factors. If the KSAS option is nonzero,
then as the investment and operating factors file is read the file
specified by the IN option above is also read and the overrides found
are applied to the appropriate data records and listed on the unit
specified by the 10 option. Allowable values are zero if no overrides
to the investment and operating factors are required or a nonzero value
if overrides have been provided on unit IN.
The LSAS option controls the writing of a report of the complete
files of investment and operating factors and costs and is independent
of the KSAS option above. The file is listed on the unit specified by
the 10 option. This option is not normally selected since the KSAS
option provides for listing modifications and a complete listing of all
the factors is not a usual requirement. Allowable values are zero if a
complete file listing is not required or a nonzero value if a report of
all factors is to be written to the unit specified by the 10 option.
111-13
-------
The KEDIT option is usually used in conjunction with either the
KSAS or LSAS option above. When the KEDIT option is specified, a
program stop occurs immediately after processing the investment and
operating factors. The option is useful primarily for analysis and
editing because the format used in the program to write the file is more
descriptive than a direct copy of the data file and the cost data are
also included. When this option is specified the only other options
honored are IN, 10, KSAS, LSAS, and KPRICE, but the overrides specified
by KPRICE will simply be read and listed; no calculations will be made.
Allowable values are zero if no program stop is required or a nonzero
value if a program stop is to occur after listing the data; i.e., the
program execution is merely to list part (KSAS) or all (LSAS) of the
investment and operating data and no other processing is to be done.
The LCP option controls the writing of a detailed cost report when
power plant input data are from the projection and add limestone cost
programs; i.e., KSIM above equals zero. It is the counterpart of the
KCHECK option discussed previously, and the same precautions discussed
for the KCHECK option apply to the LCP option. Allowable values are
zero if no report is required or a nonzero value (typically 6) if a
detailed cost report is to be written to the unit specified by the 10
option.
The KSTART option is used to limit the processing of power plants
from the projection and add limestone cost programs (KSIM equal zero).
The value of KSTART should be the FPC number of the plant with which
processing is to begin. All plants with FPC numbers less than KSTART
will not be considered by the program. Allowable values are zero if no
limits are to be placed on the plants or if limited processing is
required the appropriate FPC number is specified.
The KSTOP option is the ending FPC number for the same requirement
as KSTART. Allowable values are 9999999999 if no limits are to be
placed on the plants or the appropriate FPC number when limited processing
is required. The plant number given for KSTOP will be included in the
processing.
This completes the documentation of the options. As previously
discussed the default options and an abbreviated description of each
will be written to the standard print file, immediately followed by a
listing of the user options that were specified.
Overrides for Internal Program Values
After the options have been processed the next step is the processin
of the overrides for the internal program constants and table values
shown in Table III-2. Reference to Detailed Cost Estimates is mandatory
for the use of this category of overrides. In the same manner as the
options, the default values of the program constants and table values
will be written to the standard print file, immediately followed by a
listing of the user overrides that were specified. These overrides are
in NAMELIST format and will be read from the standard input file.
111-14
-------
TABLE IiT-2. SCRUBBING COST GENERATOR (STMCAP)
INTERNAL PROGRAM VALUE OVERRIDES
Override name
Description
OPYEAR Specifies the year for which calculations will be made
CEINDX Chemiaal Engineern-ng index for scaling investment costs
j^TE Used for calculating the capital charges portion of indirect
costs
START Allowance for startup, modifications, and interest during
construction
INTYPE Plant overhead portion of indirect costs
MAIN Maintenance rate
INDR Indirect investment costs
NACOST Sodium sulfate credit
RETRO Retrofit difficulty factor
FLOCAT Location factor adjustment
CONS Capital investment related to construction costs not covered in
other categories
PREMIS Technology relativity factor
GRACE Compliance allowance factor
SITEFAC Site-factor adjustment
I LI-15
-------
The OPYEAR override specifies the year for which calculations will
be made. Specification of a given year implies that costs for the
various areas of scrubbing have been provided in the independent cost
file. The value of OPYEAR is used to search the cost file for a matching
value. If no cost record is found for the specified OPYEAR value a
program stop occurs.
The CEINDX override is the Chemical Engineering index used to scale
investment costs associated with the various processes.
The RATE override is used in computations related to the capital
charges portion of indirect costs and has two values associated with it.
The first value is for new installations and the second value is for
existing installations. Tables 59 and 60 on pages 134 and 135 of Detailed
Cost Estimates show sample average capital charges as a part of the area
contribution analysis if the base-case summary includes data for new
installations.
The START override specifies the allowance for startup, modifica-
tions, and interest during construction. It has six values, one for
each of the six processes included in the program design. Pages 27, 88,
and 89 of the Detailed Cost Estimates should be referred to for sample
values.
The INTYPE override specifies the plant overhead portion of indirect
costs. Just as for the START override there are six values associated
with it, one for each of the six processes. Table 59 on page 134 of the
Detailed Cost Estimates should be referred to for sample values.
The MAIN override specifies the maintenance rate. One value for
each of the six processes is provided. Table 13 on page 29 and Table 59
on page 134 of the Detailed Cost Estimates should be referred to for
sample values.
The INDR override specifies the indirect investment costs. This
override has 12 values. The first six values apply to new installations
for each of the six processes and the other six apply to existing installa-
tions for each of the six processes. Table 10 on page 27 and Tables 35
and 36 on pages 88 and 89 of the Detailed Cost Estimates should be
referred to for sample values.
The NACOST override specifies the rate per ton of sodium sulfate
that will be applied as a credit against the costs of the sodium process.
The RETRO override specifies the retrofit difficulty factor that
will be applied. Power plants in the data base have a PEDCo retrofit
difficulty factor specified where available. The primary reason for the
RETRO override is for application to user-supplied data.
The FLOCAT override specifies any location factors that are signifi~
cantly different from the base cases. It is normally set to one, particu~
larly when processing all of the plants from the data base. It is most
111-16
-------
typically used when user-provided power plant data are processed and
known location factors show that adjustments to the base-case data
should be made. Functionally it can be related to the site-specific
factors that can be applied to power plant data from the projection
program except that the FLOCAT factor can be applied regardless of the
source of power plant data.
The CONS override specifies the percentage of total capital invest-
ment that is related to construction costs not specifically covered
in the other capital investment categories. Table 35 on page 88 of the
Detailed Cost Estimates should be referred to for sample values.
The PREMIS override is a technology relativity factor. It allows
technology changes to be taken into account in the year for which calcula-
tions will be made, relative to the base year used in Detailed Cost
Estimates (the base case is assumed to have a process premise factor of
1.0). There are six PREMIS values, one for each of the processes. They
are applied uniformly to all plants independently of the scrubbing
investment and operating cost data.
The GRACE override is a compliance allowance factor. During the
conversion of regulations to a Ib SC>2/MBtu basis and the calculation of
both actual and allowed emissions on a tons of sulfur per year basis,
various rounding differences occur. The GRACE override allows these
differences to be taken into account rather than considering a plant out
of compliance on the basis of small differences between actual and
allowed emission levels.
The SITEFAC override provides for site-specific factors that differ
from the base-case assumptions. It is applied uniformly to all plants
rather than to specific plants, which is the case of site factors
provided in the site-factor adjustment file. It is provided primarily
for user-supplied data, but it is not typically used because the actual
data normally reflect any exceptions to the base case. Up to six site
factors may be provided, one for each process.
The overrides of the internal program constants and table values
complete the basic requirements for running the scrubbing cost generator
program. All of the options and overrides just presented will not be
required for each execution of the program. Most runs will probably be
relatively straightforward and can use the results of the base-case
calculations and the associated scaling factors. The options and over-
rides are provided because data file updating and program modification
for only slightly different run requirements can consume more resources
than actual program execution. Dynamic run time modification allows a
wide variety of processing requirements to be honored and at the same
time preserves the base-case input data and a standard version of the
program. This reduces the indirect costs associated with program usage
and allows many program executions to be made in a shorter time. This
can be significant when several runs are made to analyze the cost difference
that results from varying the possible combinations of fuel and operating
characteristics for a power plant (such as heat and sulfur content of
111-17
-------
the fuel, boiler capacity factors and heat rates, overrides to the base-
case calculations and the associated scaling factors, etc.).
The preceding program documentation, the section on program usage
where sample setups for execution are described, and the file and records
sections which describe input data formats and sample output listings of
results should be examined in detail before actual program execution is
attempted. If the program processing procedures, input requirements,
and possible output results are not clear after a review of all of the
sections described previously, a source program listing can be referred
to or trial and error procedures may be useful if problems are very
minor.
ACDUPDT PROGRAM
The purpose of the ACDUPDT program (Figure III-5) is to determine
transportation rates for the shipment of molten sulfur from Port Sulphur
to each sulfuric acid plant. These rates are used in a subsequent
program as a part of the data required to calculate avoidable acid
production costs. Four input files are required for program execution—
all provided independently rather than being generated by a program in
the system. These files are the acid plant and smelter data file that
contains information about plant location and operating characteristics
the sulfur terminal data file that contains information about the location
and rates associated with each barge terminal used in the water trans-
portation of molten sulfur from Port Sulphur, the rail mileage file that
contains the distance between any given source and destination rate
basing points, and the sulfur tariff file that contains the rates for
shipping sulfur by rail. Only acid plants east of the Rocky Mountains
have sulfur transportation rates calculated. Rail rates used in the
system are limited to the Docket 28300 area. Transportation rate calcu-
lations are further limited to acid plants that burn elemental sulfur
as the feedstock. The plant status code allows the determination of
both of these qualifications (see section IV).
To determine the transportation rates for molten sulfur from Port
Sulphur to a given acid plant the appropriate barge terminal must be
found. The distance from the acid plant to all barge terminals is
determined by accessing the rail mileage file using the rate basing
point of the acid plant and the rate basing points of the terminals.
These distances are used to access the sulfur tariff file and the rail
shipping rates are determined. The terminal is selected that results in
the lowest total transportation costs for each acid plant. Total rates
from Port Sulphur to each acid plant through any terminal can be calcu-
lated because each terminal record contains the necessary barge rate
information for that terminal. The total rate is the sum of the barge
rate from Port Sulphur to the terminal, the terminal handling charges
associated with the transfer of the molten sulfur from barge to rail
(including intermediate storage charges where applicable), and the rail
rate from the terminal to the acid plant (if applicable).
111-18
-------
MOLTEN SULFUR
TARIFF
DATA
ACID PLANTS
AND SMELTERS
CALCULATION OF
DELIVERED
SULFUR COSTS
TO AC 1D
PLANTS
ACDUPDT
;iCID PLANTS,
SHELTERS,
iND DELIVERED
1ULFUR COSTS
ACDSML
DELIVERED
COST REPORT
Figure III-5. Block diagram for the calculation of delivered
sulfur costs to acid plants (program ACDUPDT).
IM-19
-------
The output from the program is an updated acid plant file with the
sulfur transportation data added and a report that includes the terminal
selected and the total transportation rate for each acid plant. The
updated acid plant file is in the same format as the input file; new
information is inserted into fields that were already provided. The
added data include the terminal selected, the barge rate from Port
Sulphur to the terminal, terminal handling charges, rail rates from the
terminal to the acid plant, and the total rate. The output report that
is generated is written to the standard print file.
GENAGO PROGRAM
The GENACD program (Figure III-6) calculates an avoidable unit
production cost for each of the sulfuric acid plants to be considered as
a potential demand point for byproduct acid. The input to the program
consists of the acid plant data file (output from the ACDUPDT program) ,
the independent avoidable production cost file, and output report selec-
tion specifications. Output from the program is a smelter file, an acid
plant file that includes avoidable production cost estimates, and up to
three reports depending upon the report specifications provided. The
input file of acid producers contains two types of plants—those burning
elemental sulfur as a feedstock as well as those using smelter off-gas.
Because the smelters are not potential demand points of abatement acid
but are actually potential suppliers, these two types of producers must
be considered separately.
The smelters are processed by calculating a projected estimate of
future production increases based upon current capacity and compliance
status. Plants out of compliance are projected to have a greater produc-
tion increase as a result of increased sulfur dioxide removal to meet
clean air standards. These projected increases are assigned a zero unit
production cost; therefore when they are later combined with the power
plants to form the total supply input to the marketing model they will
have an advantage over the power plants (which typically have a greater-
than-zero incremental production cost). After the projected production
increases and unit production costs for the smelters are completed they
are written to an output file. The remaining acid plants are then used
in avoidable production cost calculations and will be considered as
consumers of byproduct acid from both smelters and power plants. To
provide maximum flexibility and to minimize program modification and
recompilation, the factors used in the calculations are in a sequential
data file rather than coded within the program (see section IV).
Avoidable production costs are calculated for the acid plants
remaining in the original input file and an acid plant avoidable produc-
tion cost output file is written. Up to three reports may be selected
as a result of program execution: (1) a report of the acid plants
considered, (2) a report of the sulfur transportation charges for each
acid plant, and (3) an avoidable production cost report. When the
program is executed interactively, the various report options are printed
and a user response is made to select the desired reports. All three
reports are written to the standard print file.
111-20
-------
ACID PLANT
AVOIDABLE
PRODUCTION
COST FACTORS
ACDPAR
ACID PLANTS,
SMELTERS,
AND DELIVERED
SULFUR COSTS
ACDSML
OPTIONS
AND
VARIABLES
AVOIDABLE ACID
PRODUCTION
COST GENERATOR
GENACD
ACID PLANT
AVOIDABLE
PRODUCTION
COSTS AND
QUANTITIES
SMELTER
PRODUCTION
COSTS AND
QUANTITIES
ACID PLANTS
SELECTED
REPORT
DELIVERED
SULFUR COST
REPORT
AVOIDABLE
PRODUCTION
COST
REPORT
Figure 1II-6. Block diagram for the calculation of acid
plant avoidable production costs (program GENACD).
ITT-?!
-------
The GENACD program documentation and procedures just described
apply only to the automated capabilities of the demand subsystem (those
locations in the Docket 28300 area east of the Rocky Mountains) which
limit the acid plants and smelters considered to that same area. There
is no direct equivalent to the Docket 28300 concept for the western and
transcontinental railway systems. A separate run of the program is
required for western acid plants and smelters and the data must be
handled manually; it cannot be combined directly with the data for the
Docket 28300 area.
MANUAL TRANSPORTATION PROCEDURE
As previously discussed, an integrated transportation data base was
to have been developed so that the transportation cost generation process
could be completely automated. This is reflected in the format and
description for the TRNPTS file (see section IV and Appendix A) . When
the final step of linking all rail, barge, and truck points was not
completed, the manual procedure shown in Figure HI-7 was substituted
and transportation rates in the linear programing model were limited to
rail shipments within the Docket 28300 area.
Before the linear programing model can be generated, shipping rates
must be calculated for every possible combination of supply and demand
points that will be considered in the model. This includes power plants
smelters, acid plants, and transshipment terminals for the Western
States and Canada. For every possible supply and demand point combina-
tion, the location of each point and the distance between them are
required. The location is required to select the appropriate tariff
table from the tariff file; the distance is required to select the
applicable tariff rate from the tariff table (tariff files are described
in section IV).
There are four sources of locations that must be used to build the
TRNPTS file; three are data files from the supply and demand subsystems.
The data files are the power plant scrubbing cost file (SCRCST), the
smelter cost file (SMLCST), and the acid plant avoidable production cost
file (ACDCST). The fourth source of location data is for the transship-
ment terminals. The transshipment terminal data are found in the linear
programing model generator program (GENPGM) rather than in a previously
created data file. A source program listing can be referred to if
necessary.
All rail mileages used to determine the applicable rates within the
Docket 28300 area are derived using 2632 rate basing points. The
required data for these rate basing points are in the RBNSORT file. if
the actual rate basing point is not known (which is normally the case) ,
then maps, comparisons between the location SPLC and rate basing point
SPLC, and other geographic- and transportation-related data must be used
to estimate the applicable rate basing point.
111-22
-------
ACID PLANT
AVOIDABLE
PRODUCTION
COSTS AND
QUANT IT IKS
SMELTER
PRODUCTION
COSTS AND
QUANTITIES
POWER PLAN'I
SCRUBBING
COSTS AND
QUANT ITIES
TRANSPORTATION
RATE BASINC;
DATA
SELECTION OF RATE
BASINC DATA FOR
SHELTERS, POWER
PLANTS, AND
ACID PLANTS
MANUAL PROCEDURE
1
POWER PLANTS,
ACID PLANTS,
AND SMELTERS
WITH RATE
BASING DATA
TKNPTS
Figure II1-7. Block diagram for the selection of rate basing data
for smelters, power plants, and acid plants (manual procedure).
TV! -23
-------
The actual tariff table and rate selection process is already
automated in the TRNCOST program that follows this manual procedure.
The TRNPTS file provides all of the necessary data. However, the building
of the TRNPTS file is not automated. An automated procedure for building
the TRNPTS file could have been developed if the integrated transportation
data base had been completed. The locations to be modeled could be
determined by examining the four location sources; the rate basing data
for any location could be extracted from the data base (the data base
contains every U.S. location); and the TRNPTS records could be auto-
matically generated. Instead, the four location sources and the RBNSORT
file must be examined manually and the TRNPTS file built manually,
record by record, for each location.
Only 5 of the 10 fields in the TRNPTS file (section IV) are actually
used: the location identification code; the location SPLC; the rate
basing point SPLC; the rate basing point number (rail mileage index) •
and the tariff table index. Once the rate basing point number is found
for a location, a TRNPTS record can be built. The identification code
and SPLC from the appropriate location source file are combined with the
rate basing point SPLC, the rate basing point number, and the tariff
table index from the RBNSORT file to form a TRNPTS record.
The TRNPTS file is divided into two sections. The first section
contains all demand locations, the second section contains all supply
locations. A special record with an all-zero identification code is
used to separate the sections. Transshipment terminals are included in
the supply section because previous analyses of the western and Canadian
supply-demand have indicated a potential excess supply from these loca-
tions, not a demand. In the demand section the records should be in
sequence by the acid plant identification code. In the supply section
the records should be grouped by power plant, smelter, and transshipment
terminals, and within each group the records should be in sequence by
the respective identification codes. (The codes were originally assigned
so that the correct grouping will occur in the supply section if the
records are in sequence by the identification codes.)
As the TRNPTS file is processed by the TRNCOST program, rates are
calculated for every possible combination of supply and demand locations
in the file. There is no edit check to determine whether the locations
are necessary for the model or if the identification codes are correct
so the identification codes must be entered exactly as found in the four
location sources. In the TRNCOST program these codes are used as keys
to build the transportation cost file (TRNCST). At model generation
time (program GENPGM) the identification codes of the locations required
for the model are used as access keys into the TRNCST file. Any unnec-
essary rates are ignored but a missing rate resulting from either an
omitted location or an incorrect identification code in the original
TRNPTS file will cause model generation to abort.
All acid plants, smelters, and transshipment terminals are typical!
included in every model (presently about 100 acid plants, 15 smelters
and 10 transshipment terminals). The number of power plants included i
111-24
-------
a function of the ACFL and the number of plants in the SCRCST file (the
SCRCST file contains only the plants for which scrubbing costs were
calculated). The SCRCST file never contains all plants in the data base
and in general all plants in the SCRCST file are not included in a
single model (presently the data base contains about 1000 power plants;
the SCRCST file has not exceeded 200 plants; the number of plants included
in a model has not exceeded 60).
In a totally automated transportation system, it is conceivable
that all power plants might be included in the TRNPTS file even though
all plants would never be included in a single model. The advantage
would be that once rates are calculated for all possible combinations,
no further runs of the transportation cost generator would be required
until new locations are added or tariffs are updated. Model runs could
include any power plant in the data base and the necessary rates would
already be in the TRNCST file. The controlling factor for including
potentially unnecessary locations in the TRNPTS file is the cost of
calculating rates for these locations in the TRNCOST program.
The transportation system is not completely automated, however, and
a controlling factor is also the time required to manually examine the
locations, determine the rate basing points, and build the TRNPTS records.
To determine the number of power plants that will be included in a given
model and reduce the TRNPTS power plant records to that number, the ACFL
value to be used can be compared to the limestone scrubbing costs
(cents/MBtu) in the scrubbing cost file (SCRCST).
In normal use of the system, the TRNPTS file is rarely created in
its entirety; it is merely updated from one model run to the next. The
only changes typically required are to add and remove power plant records.
All of the preceding procedures still apply except that the new records
must be inserted among the existing records based on identification code
so that the correct sequence is maintained.
TRNCOST PROGRAM
The TRNCOST program (Figure III-8) calculates rail freight rates
for all possible combinations of suppliers (eastern power plants and
smelters, western and Canadian producers through transshipment terminals)
and consumers (eastern acid plants). Program inputs consist of the
TRNPTS file of rate basing data (created by the manual procedure) for
all suppliers and consumers to be considered in the marketing model, the
independent rail mileage data file, and the independent file of tariff
tables. Program output includes a data file of rail rates for all
possible combinations of suppliers and consumers, and a report of informa-
tion related to the calculation of the rail rates.
Determination of rates for every possible combination of supply and
demand points is done on a supply point basis. The TRNPTS file contains
all locations to be considered, separated into supply and demand. For
each potential supply location rail rates are developed from that supply
111-25
-------
POWER PLANTS,
ACID PLANTS, AND
SMELTERS WITH
RATE BASING
DATA
TRNPTS
SULFURIC
ACID TARIFF
DATA
X3V3H2S
TRANSPORTATION
COST
GENERATOR
TRNCOST
1
^TRANSPORTATION
RATES FROM POWER
PLANTS AND
SMELTERS TO
ACID PLANTS
TRNCST
DIAGNOSTICS
AND PROGRAM
EXECUTION
REPORT
Figure III-8. Block diagram for calculation of transportation rates
from power plants and smelters to acid plants (program TRNCOST).
1 Ii-26
-------
point to all demand points. The current program version provides rate
entries for 200 demand points per supply point in the TRNCST output file
used in the marketing model. The procedure for creating the TRNCST file
can be described as follows:
1. The rate basing points and tariff table indices of all demand
locations are taken from the demand section of the TRNPTS file.
2. The rate basing point and tariff table index of the first supply
location is taken from the supply section of the TRNPTS file.
3. The rail mileage between the supply location from step 2 and each
demand location from step 1 is obtained using each pair of locations
as source and destination points to access the rail mileage file.
4. The rail mileage values from step 3 and the tariff table indices
from steps 1 and 2 are used to select the appropriate tariff
table (using the indices) and the rate entry within the table
(using the mileage).
5. For each pair of supply-demand points in steps 1-4, an entry in
the TRNCST record is built that contains the tariff table index
of the supply point, the tariff table index of the demand point,
the tariff table number used, the rail mileage between the
supply and demand points, and the rail rate to be used in the
model. For each supply point, up to 200 of these entries per
record are built, one entry for each demand point (the TRNCST
record contains 200 entries but the number of entries with valid
rate data depends on the number of demand points in the TRNPTS
file).
6. Steps 2-5 are repeated for each supply record in the TRNPTS file.
The sulfuric acid tariff file contains tables that are based upon
the various tariff rates that apply depending upon the source and desti-
nations of shipments. The RBNSORT file that was used to build TRNPTS
records has tariff table indices that correspond to the various rate
basing points and were a part of the data included in the TRNPTS file.
A tariff table selection matrix is coded as a part of the TRNCOST program
so that the supply point tariff index and demand point tariff index can
be used as subscripts into the tariff table selection matrix (or array)
where the tariff table number that applies is stored. When the table
number is known the rate within that table depends only upon the distance
to be shipped.
The TRNCST file is written as a random file. The supply point
identification code is used as the key at file creation time. The data
are retrieved in the model generation program (GENPGM) by a random read
of the file with the appropriate supply point identification code as a
key.
111-27
-------
The TRNCOST program execution can be generally summarized by the
following: For each supply point - demand point combination of locations
in the TRNPTS file, the rail mileage between each pair of points is
found from the RAILWA file using the rate basing numbers of each point-
the applicable rail rate for potential shipments between each pair of
points is found from the tariff file using the tariff table indices from
the TRNPTS file and the rail mileage; and finally, a random data file is
written, one record per supply point with rate entries for up to 200
demand points per record.
GENPGM PROGRAM
The GENPGM program (Figure III-9), the first program in the linear
programing marketing model subsystem, generates the model to be solved
by the APEX linear programing package. All preparation and processing
of data to be used in the model must either have been completed before
the execution of this program or must be done during program execution
before the model is generated. There are no provisions for changing the
model once it is built. The data from the other subsystems (SCRCST
SMLCST, ACDCST, and TRNCST files) discussed previously should have '
already been prepared and ready for use in the model.
As previously discussed in the system and subsystem sections DOW
plant byproduct acid is assumed to replace existing supply, not add t
it. Based upon this assumption, total demand will be unaffected by
power plant compliance strategies, so the total potential demand is
included in all models. In addition, potential production increases b
current suppliers of byproduct acid (eastern smelters and the net suppl
from the West and Canada)are assumed to be unaffected by power plant
compliance strategies, so projected production increases by current
suppliers are also included in all models. The only data that remain
be included in the model are the power plant data, costs for the altern°
tive clean fuel level (ACFL) that will be used, and the option that 9~
determines how the ACFL will be used.
There are many assessments and interpretations of the conditions
that influence or directly contribute to the total cost of a clean fuel
strategy. In many cases projections of future ACFL costs may be more
appropriate than current costs and several separate model runs may be
required to analyze the overall effects that result from variations in
ACFL. Because of this, and to provide maximum user flexibility, the
ACFL used in the strategy selection process, including the generation of
a model, is a user-supplied condition. Because related cost calculati
in the system are on a cents/MBtu basis, the ACFL must also be in
cents/MBtu. The ACFL is not limited to any particular clean fuel strate
it may be total use of naturally occurring lower sulfur fuel, coal
washing, fuel blending, etc., but in any case the increased costs that
would result from implementing a clean fuel strategy can be calculated
in terms of cents/MBtu and provided as the ACFL value.
111-28
-------
"ACTDPLANT
AVOIDABLE
PRODUCTION
COSTS AND
QUANTITIES
SMELTER
PRODUCTION
COSTS AND
QUANTITIES
POWER PLANT
SCRUBBING
COSTS AND
QUANTITIES
r /^TRANSPORTATION,
/ RATES FROM
/ POWER PLANTSi
I AND SMELTERS
TO ACID PLANTS
GENERATION
OF
LINEAR
PROGRAMING
MODEL
GENPGM
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
(UNSORTED)
GENDATA
MODEL GENERATION
OPTIONS AND
VARIABLES
LINEAR
PROGRAMING
MODEL
MODLIN
MODEL
GENERATION
REPORT
Figure III-9. Block diagram for the generation of the linear
programing marketing model (program GENPGM).
T.7'f-29
-------
Two options are provided to control the method by which the ACFL
will be used in the selection process. In both cases the first step i
the program is a comparison of the clean fuel and limestone scrubbing °
strategies in which the lowest cost strategy is selected. Then, depe d
ing on the option, the model will be built based on different compariso"
If the first option is chosen, all plants in which limestone scrubK-i
is the lowest cost strategy will go into the model as a comparison of 8
the limestone scrubbing strategy with the byproduct strategy. All
plants in which clean fuel is the lower cost strategy will be further
compared to determine the incremental cost of the byproduct strategy
compared to the clean fuel strategy in terms of acid cost. If the a id
cost is less than $30/ton, the plant will go into the model as a comoa i
of clean fuel strategy with the byproduct strategy. Those plants with
an incremental acid production cost of over $30/ton will not go into tH
model. ne
The second option, in effect, imposes a clean fuel bias. After
comparison of limestone scrubbing and clean fuel, if clean fuel is f
to be the lowest cost strategy the plant will not go into the model anri
no further comparisons of strategies are made. If limestone scrubbi
is found to be the lowest cost option the plant will go into the mod"!
as a comparison of limestone scrubbing and byproduct strategies. Us^
this option a direct comparison of the clean fuel strategy and the lng
byproduct strategy is not made.
Both the ACFL cost and the option to control its use have importa t-
effects on the type of model generated and thus on the costs of its
execution and the nature of its results. In the case of both ootions
the ACFL is used in one sense as a screening factor which determines fK
selection of plants to be included in the model. The nature of thi
screening factor is further determined by the choice of options Th
model structures defined by these decisions can differ in the number^f
plants included and the types of strategies directly compared This i
turn determines the quantities of acid considered for marketing and th
transportation network used and therefore the outcome of the eniHUK <
HuiJ-ibrium
solution.
An additional assumption is used in the power plant selection
process. Power plants with a sulfuric acid producing potential of
than 66,000 tons/yr are assumed too small to be competitive in the
market and the only options considered are either a clean fuel or a
disposable byproduct strategy.
The preceding description of the power plant selection process ha*
been on a plant basis to avoid complexity but the actual strategy sel*
tion procedure is on a boiler-by-boiler basis. The only difference is
that combined strategies may result for a single plant. Because it •
not probable that a power plant will install two different scrubbing
processes, the only strategy combinations allowed are either a clean
fuel strategy and a disposable byproduct strategy, or a clean fuel
111-30
-------
strategy and a marketable byproduct strategy. Where mixed strategies
are indicated the division between strategies is always on a discrete
boiler-by-boiler basis—a boiler is not split between strategies.
Although the strategy selection process for all plants cannot be
completed until the model is solved, plants that would select a clean
fuel strategy in any case based upon the ACFL to be used can be pre-
selected and excluded from the model. Plants that would select a
disposable byproduct strategy because of size (potential acid production
less than 66,000 tons/yr) when a clean fuel strategy is not indicated
can also be preselected and excluded from the model. These preselected
plants are written to a report file.
There are several problems with modeling the 11 Western States of
the transcontinental and mountain Pacific areas and Canada. In the
present model these locations cannot be treated on the same basis as the
37 Eastern States of the Docket 28300 area. Shipping rates between
points within these areas and between these areas and the locations that
must be considered within the Docket 28300 area have been impractical to
automate and must be done manually on a case-by-case basis. Because of
this a separate manual analysis of the western and Canadian supply-
demand is done independently of the transportation cost generator. The
results of previous analyses have indicated a potential net supply from
these areas.
However, the number of shipping rates required, from net supply
locations indicated by the analyses to eastern demand points, has still
been excessive (over 1000 rates would have to be developed by hand).
Two steps are taken to reduce the number of rates required to consider
the effects of a potential supply from the West and Canada in the model
solution. First, the potential supply is summarized for each Western
State and Canada. Next, assumed transshipment terminal locations are
assigned within the 37 Eastern States. This reduces the number of
required shipping rates to a more practical level.
Transshipment terminals have been defined and, as a result of
including them in the TRNPTS file, shipping rates have been calculated
to all potential eastern demand points. Potential western and Canadian
supply can now be considered on the same basis as eastern supply points
except for one difference. For eastern supply locations, the point of
production and the point considered for marketing are the same. The
unit production cost is the only factor involved in marketing from that
point. For the transshipment terminals there is a second factor, the
cost of shipping from the point of production (the Western States and
Canada) to the point considered for marketing (the terminals). So that
all potential supply can be processed equally, western and Canadian
production costs are combined with shipping rates to the terminals. The
sum is used as a total production cost at the terminals, the point
considered for marketing in the model.
All data for the net western and Canadian supply are coded inter-
nally in the model generator program (except for the transportation
111-31
-------
rates calculated from the TRNPTS file) and included in every model. The
transshipment terminal identification code, the terminal "production
cost" (actual production costs plus shipping costs to the terminal), the
potential supply quantities, and a terminal usage handling charge (also
added to terminal production costs) make up the internal program data.
During program execution the internal program data and data from the
SCRCST, SMLCST, TRNCST, and ACDCST files are combined to build the
required model.
The strategy selection process can be summarized by the following
statements. For each plant in the SCRCST file, records are read on a
boiler-by-boiler basis and the disposable byproduct scrubbing costs, the
marketable byproduct scrubbing costs, and the user-supplied ACFL are
compared. If the disposable byproduct costs are lowest, the incremental
costs of the marketable byproduct strategy are calculated based on a
comparison with the disposable byproduct strategy and the potential
production and incremental cost are included in the model (if the poten-
tial production is less than 66,000 tons/yr a disposable byproduct
strategy is preselected). If the ACFL is lowest, the incremental costs
of the marketable byproduct strategy are calculated based on a comparison
with the ACFL and the potential production and incremental cost are
included in the model (if the incremental cost is greater than $30/ton
or the potential production is less than 66,000 tons/yr, a clean fuel
strategy is preselected). Boiler data are always combined into plant
totals for the strategy selected. When a split strategy is indicated,
the plant total is divided into two separate totals for the boiler and
strategy combination indicated. In all cases where potential suppliers
are included in a model the appropriate transportation data from the
TRNCST file are also included. When all boilers for all plants have been
processed and the model has been built, the next step in the subsystem
is the solution of the model using the APEX linear programing package.
There are no further user options or modifications in the linear pro-
graming marketing model subsystem. The only choice available is either
to solve the model and report the results or discard the model generated
by the GENPGM program.
The purpose of this manual is to provide the information necessary
to use the byproduct marketing system. It is impractical to try to also
include all of the linear programing concepts and implications required
to create the marketing models. Users who desire detailed information
should refer to a source program listing of the GENPGM program or the
APEX linear programing manual (see Appendix B).
GENSORT PROCEDURE
The GENSORT procedure (Figure 111-10) sorts the power plant data
from the GENPGM program so that it can be used in the REPTSOL program to
build a report of strategy selection for all plants. Two types of
records must be sorted. The first type consists of data for all power
plants that were included in the model as potential candidates for a
marketable byproduct strategy, the second type is a strategy status
111-32
-------
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
GENDATA
SORTING OF
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
GENSORT
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
(SORTED)
GENREPT
Figure 111-10. Block diagram for sorting the power plant data
from the linear programing model generation (program GENSORT)
111-33
-------
record for every plant considered in the strategy selection process by
program GENPGM. The record types are separated by a special record of
zeroes. The GENDATA and GENREPT record formats in section IV should be
referred to if necessary.
APEX SYSTEM
The APEX system (Figure III-ll) is basically an optimization system.
Its main function is to optimize linear models to either maximize gains
or minimize losses. The four major activities involved in handling all
linear programing problems have either been previously completed during
execution of the programs and procedures in the various subsystems or
will be completed when the model is solved. These activities are summa-
rized by the following steps:
1. All of the data that are related to solving the problem is collected
and organized into a data base (the data base may simply consist
of some logical grouping of the data into one or more data files) .
Typical information usually includes costs, capacity relationships
transfer combinations, etc.
2. The data from step 1 are transformed into a format that a linear
programing system can analyze. The new format is usually referred
to as a linear model or an LP model, and a matrix generator
program is used to build the model.
3. Once the model is generated it is solved by the linear programing
system to produce an answer that is both feasible and optimal.
4. APEX produces a standard set of printed reports to describe the
solution but in most cases special customized reports are required
Therefore a binary solution file is also provided to allow maximum
flexibility in developing solution results by using a report
generator program. The REPTSOL program is the report generator in
the byproduct marketing system.
As previously discussed, linear programing knowledge is not required
to use the byproduct marketing system. Users who require detailed
information should refer to the source program listings of GENPGM,
REPTSOL, and an APEX manual (see Appendix B).
Once the model is built by program GENPGM no further options are
provided to control the model. The procedure to solve the model and
report the results must either be executed or the model must be discarded
REPTSOL PROGRAM
The REPTSOL program (Figure 111-12) generates a report analysis of
the model solution. Input to the program consists of the SMLCST, ACDCST
GENREPT (contains required data from SCRCST), and MODLOUT files. The '
111-34
-------
LINEAR
PROGRAMING
MODEL
MODLIN
APEX LINEAR
PROGRAMING
SYSTEM
APEX
I
SOLUTION
FILE FOR
USER REPORT
GENERATION
MODLOUT
STANDARD
APEX
REPORTS
Figure III-ll. Block diagram for solving the linear programing
marketing model (APEX linear programing system).
111-35
-------
:LUTION FILE
FOR USER
REPORT
EN E RAT I ON
PLANT
AVOIDABLE
PRODUCTION
COSTS AND
QUANTITIES
SMELTER
PRODUCTION
COSTS AND
QUANTITIES
POWER PLANT
REPORT DATA
FROM MODEL
GENERATION
(SORTED)
REPORT
GENERATION
VARIABLES
REPORT
GENERATION
OF
EQUILIBRIUM
SOLUTION
REPTSOL
S EQUILIBRIUM
GLUT ION FOR
AUTOMATED
ANALYSIS
FINRPT
EQUILIBRIUM
SOLUTION REPORT
(SUPPLIERS,
CONSUMERS,
QUANTITIES, AND
COSTS)
Figure 111-12. Block diagram for generating a report of
the equilibrium solution (program REPTSOL).
II1-36
-------
GENREPT file contains power plant data required in the report that could
not be included in the model itself and the MODLOUT file contains the
solution of the model from APEX. The title that will be used to identify
the overall report is provided by the user. The report is currently
divided into six sections as follows:
1. Cost and sales summary
2. Projected acid plant purchases
3. Projected sales by eastern smelters
4. Projected sales from the West (via transshipment terminals)
5. Projected sales from Canada (via transshipment terminals)
6. Projected power plant sales
FINAL STRATEGY SELECTION PROCEDURE
Even after the final report is produced the implications of the model
solution must be analyzed before the strategy selection process is
complete. If the model solution indicates a potential market for the
total projected production considered in the model from a given power
plant, a marketable byproduct strategy is selected. If the model solu-
tion indicates no potential market for any part of the projected
production considered in the model from a given power plant, a marketable
byproduct strategy is not indicated and the otherwise-best strategy must
be selected.
The otherwise-best strategy depends on the option that was specified
in the GENPGM program to control how the ACFL value was to be used. If
the option was specified to allow the system to choose the best alterna-
tive, the costs of a disposable byproduct scrubbing strategy are compared
to the costs of a clean fuel strategy (based upon the ACFL value used)
and the least-cost strategy of the two is selected. If the option
specified in the GENPGM program was that a clean fuel preference was to
be used, the otherwise-best strategy compared to a byproduct marketing
strategy will be a disposable byproduct strategy because a clean fuel
strategy would have been selected in the first place if it was less
costly at the ACFL value used.
Strategy selection is not as obvious when the model solution
indicates a potential market for only a part of the projected production
considered in the model from a given power plant. The plants where this
occurs must be analyzed on a case-by-case basis. The problem is further
compounded if the plant was divided by the GENPGM program into a partial
clean fuel - partial scrubbing strategy, with the model solution used to
select the final strategy. The ACFL value, the SCRCST data for the
plant, the GENPGM preselection results (GENREPT), and the model solution
must all be analyzed to determine what part of the total quantity
considered for marketing was produced from each boiler. These quantities
must be manually compared to the potential market quantity indicated in
111-37
-------
the model solution. Even though the GENPGM program did not allow a
boiler to be divided or split between strategies, the quantities considered
for marketing in the model are not on a boiler basis and a market potential
may be indicated for only a fraction of the marketable byproduct produc-
tion from a given boiler. As previously discussed a combination of two
different scrubbing strategies at a single plant is improbable so the
production from each boiler (and the associated costs) must be compared
to the potential market quantity and grouped into either a mixed clean
fuel - marketable byproduct strategy, a mixed clean fuel - disposable
byproduct strategy, or a total clean fuel strategy. Based on the results
of previous models, even when the market potential is very close to
projected production quantities a split strategy probably cannot be
avoided. Final strategy selection for plants with a market potential
for only part of the projected production can be difficult. When this
occurs, in order to avoid a very arbitrary and subjective strategy selec-
tion, further analysis including additional model runs may be required.
If the results of a single model based on only one ACFL value are
inconclusive, or if more information is required for other reasons,
additional models can be built using different ACFL values so that
strategy selection is based on trends rather than a single set of con-
ditions. Using different ACFL values serves two purposes. In addition
to allowing an evaluation of the effects on strategy selection as clean
fuel strategy costs vary, the ACFL value used directly affects the total
potential supply quantities that are considered in the model and there-
fore the amount of competition present for the available market. As the
ACFL increases, a marketable byproduct strategy becomes an option for
more and more power plants. At higher ACFL values, power plants that
continue to have a market potential in the presence of greater competi-
tion resulting from the additional supply considered are those plants
with relatively low incremental costs, geographic advantages for the
available market, or both.
Another factor that can significantly affect strategy selection, in
addition to the changes in supply that were just discussed, is changes
in demand. The avoidable production costs used to determine potential
demand are very dependent on sulfur costs (sulfur-related costs are
typically the most significant portion of avoidable production costs).
In order to model the results of potential reductions in sulfur costs
(which would reduce avoidable production costs and in turn reduce the
potential demand for power plant byproduct acid), the sulfur cost in the
ACDPAR file can be modified and new avoidable production costs generated
by executing the GENACD program. Additional models can then be built to
determine effects on strategy selection as a result of changes in
demand.
Typically, several models are developed by varying both the poten-
tial supply (by using different ACFL values) and the potential demand
(by using different sulfur cost values). Based on the results from all
of the models, the power plants selected for marketable byproduct strate-
gies are those where the model solutions indicate a continuing market
potential over a range of increasing levels of supply and decreasing
levels of demand.
111-38
-------
IV. FILE AND RECORD DESCRIPTIONS
This section contains the file and record descriptions and output
report formats for the byproduct marketing system. For the binary files
a general narrative is provided because a listing of the actual data
must be decoded and is not directly useful.
SUPPLY SUBSYSTEM
Data files used in the supply subsystem are listed in Table IV-1.
Power Plant Data Base
A graphic representation of the power plant data base is shown in
Figure IV-1. A System 2000 data base description of the power plant
data used in the byproduct marketing systen is shown in Tables IV-2
through IV-5.
The graphic representation shows three levels in the data base.
The highest level contains general power plant data such as location,
name, and FPC number. The second level contains three different cate-
gories of data, referred to as "repeating groups" in the terminology of
System 2000. These repeating groups contain plant data by year,
emission regulations, and general boiler data. As shown in the graphic
representation they are subsets of the general plant level and are
relatively independent of each other. The third level repeating groups
contain regulations by pollutant, fuel type, and equipment identification;
boiler and stack configurations; and boiler data by year. The regula-
tions by type are a subset of the regulations repeating group; the
remaining groups at the third level are subsets of the general boiler
data repeating group.
The initial data base was created from the FPC Form 67 data for the
years 1969-1973 including projected data for 1978 and 1983. Additional
data are included from the EDS data base, the CDS data base, and PEDCo
reports; from various other sources to allow the addition of SPLC and
FIPS state and county codes; and from sources containing projections of
boilers scheduled to come on-line that were not included in the Form 67
data. The Form 67 data are primarily from pages 2, 5, 9, and 14; a
pseudo page of 00 is used by FPC to include data items not actually
filed by the utilities on Form 67. The units of measure are the same as
reported on Form 67, used in the EDS and CDS systems, or reported by
PEDCo.
IV-1
-------
TABLE IV-1. FILES USED IN THE SUPPLY SUBSYSTEM
File name
Description^
PLANTS Data base for power plants, System 2000 format (input to
PROJECT)
SASDAT6 Investments and operating factors for the various scrubbing
processes (input to STMCAP)
SCRPRC Cost data that correspond to the investment and operating
factors by year (input to STMCAP)
LIMEEST Estimated delivered cost of limestone to each power plant
(input to ADDLIME)
SCRSIT Site-factor adjustments for specific plants (input to STMCAP)
PLAS Projected plant-level data (output from PROJECT, input to
ADDLIME and optionally to STMCAP)
BLAS Projected boiler-level data (output from PROJECT,
input to STMCAP)
UPLIME PLAS file merged with LIMEEST (output from ADDLIME, input
to STMCAP)
SCRCST Power plant scrubbing costs and related quantities (output
from STMCAP, input to manual procedure in transportation
subsystem, input to GENPGM in linear programing subsystem)
IV-2
-------
POWER
PLANTS
<
i
PLANT LEVEL
DATA BY YEAR
REGULATIONS
BOILERS
REGULATIONS
BY POLLUTANT,
FUEL TYPE,
AND APPLICABILITY
BOILER AND
STACK
CONFIGURATION
BOILER LEVEL
DATA BY YEAR
Figure IV-1. Power plant data base structure.
-------
TABLE IV-2. POWER PLANT DATA BASE PLANT DATA
9999,9)
9 WITH MANY
FUTURE ADDITIONS)
DATA BASE NAME: IS PWRPLTS
100* PLANT-NMBR (NAME X(10>)
no* PLANT-NAME: (NON-KEY NAME" x<3o>>
120* UTILITY-NMBR (NAME X(6»
130* UTILITY-NAME (NON-KEY NAME X(30»
140* FIPS-REGION (NON-KEY INTEGER NUMBER 99)
150* FIPS-STATE (INTEGER NUMBER 99)
160* FIPS-COUNTY (NON-KEY INTEGER NUMBER 999)
170* AQCR (NON-KEY INTEGER NUMBER 999)
180* NEDS-STATE (NAME XX)
190* NEDS-COUNTY (NAME XXXX)
200* NEDS-PLANT (NAME XXXX)
2.1.0* SPLC (INTEGER NUMBER 9(6))
220* PEDC0~MILES--TO-POND (DECIMAL. NUMBER
340* CDS-COMPLIANCE-CODE (INTEGER NUMBER
500* PLANT-DATA-BY-YEAR (RG)
510* PLANT-COAL.-CONSUMP (DECIMAL NUMBER 9(9),99 IN 500)
520* PLANT-COAL-BTU (DECIMAL NUMBER 9(9),99 IN 500)
5:50* PL ANT-COAL.-SULFUR (NON-KEY DECIMAL. NUMBER 9,9999 IN 500)
540* PLANT-COAL-ASH (NON-KEY DECIMAL NUMBER 9,9999 IN 500)
550* PLANT-COAL-H20 (NON-KEY DECIMAL NUMBER 9,9999 IN 500)
560* PLANT-OIL-CONSUMP (DECIMAL NUMBER 9(9),99 IN 500)
570* PLANT-OIL-BTU (DECIMAL NUMBER 9(9).99 IN 500)
580* PLANT-OIL-SULFUR (NON-KEY DECIMAL. NUMBER 9,9999 IN 500)
590* PLANT-GAS-CONSUMP (DECIMAL. NUMBER 9(9).99 IN 500)
600* PLANT-GAS-BTU (DECIMAL. NUMBER 9(9),99 IN 500)
610* PLANT-FUEL-BASIS (NON-KEY NAME X IN 500)
620* PLANT-NET-MW-GEN (DECIMAL NUMBER 9(9),99 IN 500)
630* PLANT-HEAT-RATE (DECIMAL NUMBER 9(9).99 IN 300)
640* PLANT-CAPACITY (DECIMAL NUMBER 9(9).99 IN 500)
650* PLANT-DATA-YEAR (INTEGER NUMBER 9999 IN 500 WITH MANY FUTURE
AUDITIONS)
-------
TABLE IV-3. POWER PLANT DATA BASE REGULATION DATA
1000* REGULATIONS (RG)
1010* REG-POLLUTANT-NAME (NAME
1020* REG-FUEL-TYPE (NAME X IN
1030* REG-SOURCE (NAME X(6) IN
1040* REG-TYPE-CQDE (NAME XXXX
1500* REGS-BY-POLL-FUEL-EGUIP
FUTURE ADDITIONS)
ADDITIONS)
XXX IN 1000 WITH MANY
1000 WITH MANY FUTURE
1000)
IN 1000 WITH MANY FUTURE ADDITIONS)
ID (RG IN 1000)
IN 1500)
IN 1500)
1500 WITH
MANY FUTURE ADD
IN 1500)
99.99 IN
1500)
1510* REG-EQUIP-ID (NAME X(10) IN 1500)
1520* REG-VALUE (DECIMAL NUMBER 9 (5). 99
1530* REG-UNITS-CODE (NAME XX IN 1500)
1540* REG-UNITS-TEXT (NON-KEY NAME X(20)
1550* REG-NEW-SOURCE-CODE (NAME X(5) IN
ITIONS)
1560* REGAPPLICABILITY-CODE (NAME X IN 1500 WITH MANY FUTURE ADDI
TIONS)
1570* REG-HEAT-INPUT-BASIS-CODE (NAME X
1580* REG-CONV-TO-PCT-S (DECIMAL NUMBER
1590* REG-CONV-TO-LB-S02-MMBTU (DECIMAL
1600* REG-CONV-TO-LB-TSP-MMBTU (DECIMAL
1610* REG-CONV-TO-LB-NOX-MMBTU (DECIMAL
1620* REG-EXCESS-AIR (INTEGER NUMBER 99
ADDITIONS)
1630* REG-DATE-APPROVED (DATE IN 1500)
1640* REG-AVG-TIME-CODE (NAME XXX IN 1500)
1650* REG-COMPLIANCE-DATE (DATE IN 1500)
1660* REG-VALUE-STATUS (NAME X(7) IN 1500)
1670* REG-VALUE-MAX-OP (DECIMAL NUMBER 9(5)
FUTURE ADDITIONS)
1680* REG--COMPUTED-FIELD (DECIMAL NUMBER 9 (5), 99 IN 1500)
1690* REG-HEAT (NAME X IN 1500 WITH MANY FUTURE ADDITIONS)
NUMBER 99.99
NUMBER 99.99
NUMBER 99.99
IN 1500 WITH
IN 1500)
IN 1500)
IN 1500)
MANY FUTURE
99 IN 1500 WITH MANY
-------
TABLE IV-4. POWER PLANT DATA BASE BOILER DATA (GENERAL)
2000* BOILERS (RG)
2010* BOILER-ID (NAME XXX IN 2000 WITH MANY FUTURE ADDITIONS)
2020* NEDS-POINT-]: D (NON-KEY NAME XX IN 2000)
TOTAL-STACKS-SERVED (NON-KEY INTEGER NUMBER 9(7) IN 2000)
TOTAL-GENERATORS-SERVED (NON-KEY INTEGER NUMBER 9(7) IN 2000)
BOILER-MANUFACTURER (NON-KEY NAME XXXX IN 2000)
BOILER-YEAR-MADE (NAME XXXX IN 2000 WITH MANY FUTURE ADDITIONS)
ASSOC-GEN-CAPACITY (DECIMAL NUMBER 9 (9), 99 IN 2000)
MAX-STEAM-CAPACITY (NON-KEY DECIMAL NUMBER 9(9). 99 IN 2000)
COAL-FEED-RATE (NON-KEY DECIMAL NUMBER 9 (9). 99 IN 2000)
OIL-FEED-RATE (NON-KEY DECIMAL NUMBER 9 (9), 99 IN 2000)
GAS-FEED-RATE (NON-KEY DECIMAL NUMBER 9(9), 99 IN 2000)
BOILER-EFF-100-PCT (NON-KEY DECIMAL NUMBER 9.9999 IN 2000)
BOILER-EFF-75-PCT (NON-KEY DECIMAL NUMBER 9.9999 IN 2000)
BOILER-EFF-50-PCT (NON-KEY DECIMAL NUMBER 9.9999 IN 2000)
TOTAL-AIR (NON-KEY DECIMAL. NUMBER 9(9). 99 IN 2000)
EXCESS-AIR (NON-KEY DECIMAL NUMBER 9.9999 IN 2000)
WET-DRY-BOTTOM (NON-KEY NAME X IN 2000)
FLY-ASH (NON-KEY NAME X IN 2000)
FIRE-TYPE-1 (NON-KEY NAME XXXX IN 2000)
FIRE-TYPE-2 (NON-KEY NAME XXXX IN 2000)
FIRE-TYPE-3 (NON-KEY NAME XXXX IN 2000)
FIRE-TYPE-4 (NON-KEY NAME XXXX IN 2000)
PEDCO-RETROFIT-DFFCLTY-FCTR (DECIMAL NUMBER 9,999 IN 2000)
PEDCO-START-UP-DATE (NAME X<7> IN 2000)
PEDCO-REG-STATUS (NAME X IN 2000 WITH MANY FUTURE ADDITIONS)
PEDCO-NEW-RETROFIT (NAME X IN 2000)
PEDCO-SIZE-FLU-GAS-DESULFURING (INTEGER NUMBER 9999 IN 2000)
PE DC 0- PROCESS (NAME X<15) IN 2000)
PEDCO-VENDOR (NAME X(15) IN 2000)
STACKS-SERVED (RG IN 2000)
STACK-ID (NAME XXX IN 2600)
2030*
2040*
2050*
2060*
2070*
2080*
2090*
2100*
2110*
2120*
2130*
2140*
2150*
2160*
2170*
2180*
2210*
2220*
2230*
2240*
2250*
2260*
2270*
2280*
2290*
2600*
2610*
-------
TABLE IV-5. POWER PLANT DATA BASE BOILER DATA (BY YEAR)
3000* BOILER-DATA-BY-YEAR ( RG IN 2000)
3010* BOILER-COAL-CONSUMP (NON-KEY DECIMAL NUMBER 9<9).99 IN 3000)
3020* BOILER-OIL-CONSUMP (NON-KEY DECIMAL NUMBER 9(9),99 IN 3000)
3030* BOILER-GAS-CONSUMP (NON-KEY DECIMAL NUMBER 9 (9), 99 IN 3000)
3040* UINTER-UKDAY-HIGH (NON-KEY NAME X IN 3000)
3050* WINTER-WKDAY-LOW (NON-KEY NAME X IN 3000)
3060* WINTER-WKEND-HIGH (NON-KEY NAME X IN 3000)
3070* WINTER-WKEND-LOW (NON-KEY NAME X IN 3000)
3080* SUMMER-WKDAY-HIGH (NON-KEY NAME X IN 3000)
3090* SUMMER-WKDAY-LOW (NON-KEY NAME X IN 3000)
3100* SUMMER-WKEND-HIGH (NON-KEY NAME X IN 3000)
3110* SUMMER-WKEND-LOW (NON-KEY NAME X IN 3000)
3120* LOW-PERIOD-WKDAY-HIGH (NON-KEY NAME X IN 3000)
3130* LOU-PERIOD-WKDAY-LOW (NON-KEY NAME X IN 3000)
3140* LOW-PERIOD-UKEND-HIGH (NON-KEY NAME X IN 3000)
3150* LOU-PERIOD-UKEND-LOU (NON-KEY NAME X IN 3000)
3160* BOILER-HOURS-OPERATED (INTEGER NUMBER 9999 IN 3000)
3170* BOILER-CAPACITY-FACTOR (DECIMAL. NUMBER 9,9999 IN 3000 WITH
MANY FUTURE ADDITIONS)
3180* BOILER-COAL-PARTICULATE (NON-KEY DECIMAL NUMBER 9(9).99 IN
3000)
3190* BOILER-OIL-PARTICULATE (NON-KEY DECIMAL NUMBER 9(9)*99 IN
3000)
3200* BOILER-COAL-SOX (NON-KEY DECIMAL NUMBER 9(9).99 IN 3000)
3210* BOILER-OIL-SOX (NON-KEY DECIMAL. NUMBER 9(9).99 IN 3000)
3220* BOILER-COAL-NOX (NON-KEY DECIMAL NUMBER 9(9).99 IN 3000)
3230* BOILER-OIL-NOX (NON-KEY DECIMAL NUMBER 9(9).99 IN 3000)
3240* BOILER-6AS-NOX (NON-KEY DECIMAL NUMBER 9(9).99 IN 3000)
3250* SPLIT-FUEL-CONSUMP-INDEX (INTEGER NUMBER 9 IN 3000 WITH MAN
Y FUTURE ADDITIONS)
3260* SPLIT-FUEL-CONSUMP-CNTR (INTEGER NUMBER 99 IN 3000 WITH MAN
Y FUTURE ADDITIONS)
3270* BOILER-DATA-YEAR (NAME XXXX IN 3000 WITH MANY FUTURE ADDIT I
ONS)
-------
The repeating groups developed from the FPC Form 67 data utilize
the plant-level and boiler-level data for the years that actual data
were reported plus the 5- and 10-year plant-level data projections.
The plant-level repeating group occurs one time for each power
plant. It is based on either Form 67 data for the latest year reported
or new plants that are scheduled in the future.
The plant-data-by-year repeating group occurs one time for each
year that plant-level data were either reported or projected on Form 67.
A plant with complete data for all years plus 5- and 10-year projections
has seven occurrences at this level in the current data base, 1969-1973
1978, and 1983. As the data base is updated three additional occurrences
can be added for each year. For example, if the data reported for 1974
are added, the 1974 actual plant-level data as well as the 5- and 10-
year plant-level projections for 1979 and 1984 can be added (assuming
the data are reported as in previous years).
The regulation repeating group is built based on data extracted
from the EDS data base. As the data base is updated these values are
modified, added, or deleted as required.
The boiler-level repeating group corresponds to the plant-level
repeating group. Because boiler data are not provided on Form 67 beyond
the year that actual consumption information is reported, other sources
are used to identify boilers projected tocome on-line between the last
year reported and the year necessary for projections. This allows
boiler-level fuel allocation estimates to be calculated for projected
years. The repeating group occurs one time for each boiler, either
reported on Form 67 or added based on other sources.
The stack repeating group is based on Form 67 data. There is one
occurrence per boiler for each stack serving the boiler.
The boiler-data-by-year repeating group corresponds to the plant-
data-by-year repeating group except that occurrences are based on actual
consumption data only, not projected data. A boiler with complete data
for all years will have five occurrences at this level in the current
data base, one for each of the years 1969-1973. As the data base is
updated, one occurrence can be added for each year.
Neither detailed knowledge of the data base nor direct access into
it is required to use the system. The extract-projection program is
the procedural language interface (PLI) between the data base and the
remainder of the system. Output from the program consists of standard
sequential data files which are much simpler and less costly to access
Any user modifications required should typically be made to the sequen-
tial files, not directly to the data base. If the data base must be
directly accessed the appropriate System 2000 manual must be used (see
Appendix B).
IV-8
-------
Because the power plant data, base was designed to satisfy several
requirements rather than being limited to a single system, the repeating
groups contain components that are not used in the current byproduct
marketing system. Many or all of these components or repeating groups
may be deleted and other components or repeating groups may be added as
continuing requirements dictate. No assumptions should be made as to
the continuing availability or validity of any component or repeating
group not used directly in the byproduct marketing system. The PLI
projection program (PROJECT) source listing must be examined to determine
the components and repeating groups being used at any given time. As
modifications are made that affect the components and repeating groups
used (other than merely adding or updating data values), new data base
descriptions can be made available. However, because direct data base
access is unnecessary to use the system, if a modified PLI projection
program is provided that corresponds to the modified data base, system
usage as described will be unaffected.
The current data base contains about 6 million characters of data
so careless or unnecessary use should be avoided. As an example, depend-
ing on the particular System 2000 command, costs can vary from a few
dollars to hundreds of dollars for a single command. This precaution
applies particularly to commands in update mode.
Other Supply Subsystem File and Record Descriptions
Examples of the remaining files and applicable printed reports are
shown in Tables IV-6 through IV-23. (Details and examples of NAMELIST
usage as shown in several of the tables can be found in the FORTRAN
manual listed in Appendix B.)
TABLE IV-6. FILE FORMAT FOR DELIVERED COSTS
OF LIMESTONE TO POWER PLANTS (LIMEEST)
Field
number Description Format
1 FPC number no
2 Delivered cost of limestone F6.2
IV-9
-------
TABLE IV-7. FILE FORMAT FOR SCRUBBER INVESTMENT AND OPERATING FACTORS
(SASDAT6)
Field
number
1
2
3
4
5
6
7
8
9
10
11
12
13
Description
Process index (1-6)
Boiler status (1 - new, 2 - old)
Sequence code
Not used
Area index (1-13)
Not used
Price index (1-24)
Not used
Base scale or quantity
Not used
Sulfur coefficient
Not used
Air coefficient
Format
11
11
13
IX
12
IX
12
IX
F9.0
IX
F5.3
IX
F5.3
SASIN NAMELIST variable
names for user override
IU
JU
ISEQU
-
KU
-
LU
-
AU
-
A1U
-
A2U
IV-10
-------
TABLE IV-8. FILE FORMAT FOR SCRUBBER INVESTMENT
AND OPERATING COSTS (SCRPRC)
Field
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Description
Capital
Lais or
No. 2 fuel oil
No. 6 fuel oil
Natural gas
Steam (500 psi)
Process water
Electricity
Heat credit
Limestone
Lime (limestone)
Lime (magnesium oxide)
Lime (sodium)
Magnesium oxide
Coke
Vanadium catalyst
Vanadium catalyst
Sodium carbonate
Antioxidant
Catalyst
Analyses
Sulfuric acid
Water treatment cost
Offsite disposal
transport rate
Applicable year
Unit
0
$/hr
$/gal
$/gal-
$/kftJ
$/klb
$/kgal
$/kWh
$/MBtu
$/ton
$/ton
$/ton
$/ton
$/ton
$/ton
$/liter
$/liter
$/ton
$/lb
$/lb
$/hr
$/ton
$/kgal
$/ton
Format
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
Note: The PRICIN NAMELIST has a variable name of
PRICE that must be used as well as the
appropriate subscript (field number above)
when user overrides are used.
IV-11
-------
TABLE IV-9. FILE FORMAT FOR POWER PLANT
SITE-SPECIFIC ADJUSTMENTS (SCRSIT)
Field
number
1
2
3
4
5
6
Description
FPC number
Site factor for process No. 1
(currently limestone)
Site factor for process No. 2
(designed for lime, not used currently)
Site factor for process No. 3
(currently magnesia)
Site factor for process No. 4
(currently sodium solution)
Site factor for process No. 5
Format
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
NAMELIST
(designed for catalytic oxidation,
not used currently)
Site factor for process No. 6
(currently gypsum)
NAMELIST
IV-12
-------
TABLE IV-10. FILE FORMAT FOR PROJECTED PLANT-LEVEL DATA (PLAS)
Field
number
Record 1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Record 2
1
2
3
4
Record 3
1
2
3
4
5
6
7
8
9
10
Description
FPC number
Not used
SPLC
Coal heat content
Coal sulfur content
Oil heat content
Oil sulfur content
Gas heat content
Plant heat rate
Test field 1
Test field 2
Test field 3
Test field 4
Test field 5
Test field 6
Plant capacity
Not used
Data case identifier
Plant name
Distance to pond
Not used
Not used
SIP type (P or B)
Not used
Coal SIP code
Not used
Coal regulation value
Test field 1
Not used
Oil SIP code (if
different from coal)
Oil regulation value
Test field 2
Format
110
IX
16
F9.1
F5.3
F10.1
F5.3
F10.1
F10.1
F7.0
F7.0
F7.0
F7.0
F6.0
F4.2
F6.0
3X
IX
3A10
F10.2
3X
6X
Al
3X
12
3X
F10.2
F10.2
3X
12
F10.2
F10.2
PLANTIN NAMELIST variable
names for user data
IFPC
-
ISPLC
BTUCOL
SULCOL
BTUOIL
SULOIL
B TUG AS
HTRATE
-
-
-
-
-
-
-
-
—
_a
PEDCO
-
—
ISIPlb
-
ISIP2
-
REGC
-
-
ISIPO
REGO
_
a. Plant name is not included in the NAMELIST data; it must be provided
as a separate record following all PLANTIN NAMELIST data; FORMAT is
still 3A10.
b. Neither a 'B' for a boiler-level regulation nor a
level regulation is an allowable NAMELIST value.
must be used to overcome the limitation.
For a boiler regulation - ..., ISIP1 = 2, .
For a plant regulation - ..., ISIP1 = 16, .
'P' for a plant-
A coded equivalent
IV-13
-------
TABLE IV-11. FILE FORMAT FOR PROJECTGED BOILER-LEVEL DATA (BLAS)
Field
number
Record 1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Description
Boiler identification
Not used
Boiler capacity
Boiler air rate
Coal consumption, ktons
Oil consumption, kbbl
Gas consumption, kft^
Boiler capacity factor
Not used
Boiler startup year
Not used
Number of stacks served
Stack identification
FPC number
PEDCo scrubber status
Retrofit difficulty
factor
Format
A3
3X
F11.2
F11.2
F11.2
F11.2
F11.2
F9.4
3X
F4.0
2X
17
5A3
110
11
F5.3
BLRIN NAMELIST variable
names for user data
IB
_
GENCAP
AIRI
TOTCOLI
TOTOILI
TOTGASI
CAPFPC
^
STRTYRI
_
_
_
IFPCB
_-i
RETRO
Record 2
1
2
3
4
5
6
7
8
9
10
Boiler identification
Not used
Coal SIP code
Not used
Coal regulation value
Test field 1
Not used
Oil SIP code (if
different from coal)
Oil regulation value
Test field 2
A3
3X
12
3X
F10.2
F10.2
3X
12
F10.2
F10.2
_
ISIP2
_
REGC
__
_
ISIPO
REGO
—
Note:
The following variable names may also be specified on a
boiler basis for user-provided BLRIN NAMELIST data:
Coal heat content - HRCOAL, coal sulfur content - SULCOL oil
heat content - BTUOIL, oil sulfur content - SULOIL, gas heat
content - BTUGAS, and process code for detailed analysis of
costs - IPROC (1 - limestone, 2 - not used, 3 - magnesia,
4 - sodium, 5 - not used, 6 - gypsum).
IV-14
-------
TABLE IV-12. FILE FORMAT FOR PROJECTED PLANT-LEVEL DATA
WITH DELIVERED LIMESTONE COSTS INCLUDED (UPLIME)
Field
number
Record 1
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
. 17
18
Record 2
1
2
3
4
.
Record 3
1
2
3
4
5
6
7
8
9
10
Description
FPC number
Not used
SPLC
Coal heat content
Coal sulfur content
Oil heat content
Oil sulfur content
Gas heat content
Plant heat rate
Test field 1
Test field 2
Test field 3
Test field 4
Test field 5
Test field 6
Plant capacity
Not used
Data case identifier
Plant name
Distance to pond (miles)
Not used
Limestone cost
(delivered, cents/ton)
SIP type (P or B)
Not used
Coal SIP code
Not used
Coal regulation value
Test field 1
Not used
Oil SIP code (if
different from coal)
Oil regulation value
Test field 2
Format
110
IX
16
F9.1
F5.3
F10.1
F5.3
F10.1
F10.1
F7.0
F7.0
F7.0
F7.0
F6.0
F4.2
F6.0
3X
IX
3A10
F10.2
3X
F6.2
Al
3X
12
3X
F10.2
F10.2
3X
12
F10.2
F10.2
PLANTIN NAMELIST variable
names for user data
IFPC
-
I SPLC
BTUCOL
SULCOL
BTUOIL
SULOIL
BTUGAS
HTRATE
-
-
-
-
-
-
-
-
-
_a
PEDCO
-
VLIME
ISIPlb
-
ISIP2
-
REGC
-
-
ISIPO
REGO
"
Plant name is not included in the NAMELIST data; it must be provided
as a separate record following all PLANTIN NAMELIST data; FORMAT is
still 3A10.
Neither a 'Bf for a boiler-level regulation nor a 'P1
level regulation is an allowable NAMELIST value
must be used to overcome the limitation.
For a boiler regulation - ..., ISIP1 = 2,
For a plant regulation - ..., ISIP1 = 16,
for a plant-
A coded equivalent
IV-15
-------
TABLE IV-13. FILE FORMAT FOR CALCULATED POWER PLANT SCRUBBING COSTS
AND RELATED QUANTITIES (SCRCST)
Field
number Description Forma t
1 FPC number of power plant HO
2 Power plant SPLC 16
3 SIP indicator (1 for plant, 2 for boiler) H
4 Number of boilers (total) J2
5 Number of boilers with adequate information for
calculations 12
6 Number of boilers that must be scrubbed to meet
compliance 12
7 Boiler identification A3
8 Test field F10.0
9 Heat input, MBtu (cumulative) Flo!o
10 Emissions removed by scrubbing (cumulative total
including this boiler, tons sulfur) F^Q Q
11 Total dollar cost, limestone scrubbing (cumulative) F10.0
12 Cents/MBtu cost of limestone scrubbing
(cumulative) ¥1Q 4
13 Total dollar cost, magnesium oxide scrubbing (cumulative) F10.0
14 Total dollar cost, gypsum scrubbing (cumulative) FIQ Q
15 Total dollar cost, sodium scrubbing (cumulative) '
16 Emissions removed by scrubbing, cumulative total
including this boiler, but limited to the point that
compliance is reached, tons sulfur (compare to
field 10 above)
17 Plant name 3A10
IV-16
-------
TABLE IV-14. REPORT FORMAT FOR OPTIONS AND OVERRIDES
AVAILABLE OPTIONS AND DEFAULTS
IN- 5 INPUT FILE FOR OVERIDE AND SIMULATED DATA
10= 6 OUTPUT FILE FOR LISTING SUIITCHABLE 6/20/30/40/
KSIM = 0 READ NAMELIST PLANT/BOILER RECORDS FROM IN
KCHECK= 0 DETAIL OUTPUT FROM COSTNEU ON FILE 10
KOFF= 0 FORCES OFF-SITE POND CALCULATIONS CHECKOUT
KSCAN= 0 WRITE A 1-LINE RECORD FOR EACH PLANT ON IO=KSCAN
KEMISS= 0 PLANT'BOILER EMISSION DATA ON IO=KEMISS
KCOSTLP= 0 WRITES LP TAPElOr REPORT ON KCOSTLP
KCOSTP= 0 PLANT SUMMARY ON KCOSTP (KCOSTLP ALSO ACTIVE
KPRICE= 0 NON-ZERO CHANGES TO PRICES FOLLOW OVERIDE DATA
KSAS= 0 NON-ZERO CHANGES TO SASDATA FOLLOW OVERIDE
LSAS= 0 NON-ZERO LISTS CURRENT SASDATA ' PRICES ON 10
KEDIT= 0 NONZERO STOP AFTER LISTING SASDATA
LCP= 0 ALLOWS DETAIL PRINT FROM COSTNEU (CHECKOUT)
KSTART 0 STARTING F PC NUMBER FOR SELECTED PLANTS
KSTOP=9999999999 ENDING FPC NUMBER SELECTED
NAMELIST OPTIONS SELECTED FOR THIS RUN APPEAR ON 10
*OPTIONS
KSAS
LSAS
LCP
KPRICE =
KED1T
KOFF
KSIM
IN
10
KSCAN
KEMISS =
KCOSTLP -
KCOSTP =
KCHECK =
KSTART =
KSTOP
*END
Of
Or
Or
Of
Of
Ov
Of
5,
20 f
30 f
40 f
20 f
6,
Of
0,
9999999999
(continued)
-------
TABLE IV-14 (continued)
<
._!
oo
PRESET VALUES OF CONSTANTS AND INTERNAL TABLES
WHICH USER MAY OVERIDE BY NAMELIST INPUT
CONSTANTS
OPYEAR 2000. CEINDX 160.2
CONS 1.05 RETRO 1.0 FLOCAT 1.0
NACOST20.0 GRACE 0.000
RATE .149
PROCESS
LIMESTONE
LIME
MAGNESIA
SODIUM SOL
CATL-OXDAT
GYPSUM
1.2
1.2
1.2
1.2
1.2
1.2
T
1
1
1
1
1
1
Y
»
•
*
»
*
»
PE
2
2
2
2
2
2
MAIN
.08
.08
.07
.06
.04
.08
INDR
1
1
1
1
1
1
.35
.35
.37
.37
.37
.35
1
1
1
1
1
1
PREMIS
,41
.41
.43
.43
.43
,41
1
1
1
1
1
1
.2
.2
.2
1 2
,2
.2
SITEFAC
1
1
1
1
00
00
1.00
00
00
1.00
USER OVERIDE SELECTIONS FOLLOW
$OVERIDE
OP YEAR = .1983E+04r
CEINDX = .3421E+03r
RATE:
START
INTYPE
MAIN
INDR
NACOST
RETRO
FLOCAT
CONS
PREMIS
GRACE
SJTEFAC
= .149E+00
= .116E+01
= .12E+01.
» .153E+OOv
f .116E+01r
.12E+01f .
.118E4-01r .118E4-01
12E+01f . 12Et01» .1
= .SE-Olr .8E-01? ,7E-01r .6E-01* .4E-01
= . 135E+01
.141E4-01
= .2E+02»
= ,!E4-01f
= .lE+Olr
= .105E+01
f . 135E+01f
f .141E+01f
f
.137E+01» .137E+01
,143E-f01r .143E+01
f .USE+Oli
,116E+01»
2E+01r . 12E+01*
, .8E-01f
y . 137E+01r
r .143E-f-01f
.135E+01*
.141E+Of
= .12E4-01? .12E4-01> ,13E+01r .12Ef01f .12E+01r . 12E+01 r
- .lEfOOf
* .lEfOlf .lEtOlf «1E^
\-Oli ,lEt01> .lEfOli
r .lEfOlf
-------
TABLE IV-15. REPORT FORMAT FOR PLANT-LEVEL SCRUBBING COSTS
vO
COMPLIANCE COSTS FROM 1983 PROJECTIONS
PLANT CODE AND NAME 9990009910 FIRST PLANT
CAPACITY FACTOR COAL .024 OIL. 0,000 GAS 0,000
SULFUR CONTENT COAL .0360 OIL .0010
TOTAL CAPACITY 1482.MW TOTAL BOILERS 10
SCRUBBED 1482.MW
PROCESS LIMESTONE
INVESTMENTS
* 162824358.
*/KW 109.9
FIRST YEAR COSTS
BYPRODUCT REVENUES
EXCLUDED
MIL/KWH
CENTS/MMBTU
BYPRODUCT
TONS/YR
COST */TON
INCREMENTAL COSTS
IN COMPARISON TO
LIMESTONE PROCESS
*
*/TON
10
MAGNESIA
171467201.
115.7
SODIUM
187184073.
126.3
GYPSUM
177983095.
120,1
139917800,
25.84
245,8
SLUDGE
442226,
316.4
0,
0.
147558693,
27,25
259,2
H2S04
236820.
623.1
7640894.
32.
167351951 .
30,90
294.0
S
70988,
2357,5
27434151.
386.
156772359,
28,95
275.4
CAS04.2H20
519555,
301 ,7
16854559,
32,
-------
TABLE IV-16. REPORT FORMAT FOR THE EDIT OF
USER-SUPPLIED POWER PLANT DATA
$PLANTIN
IF PC
BTUCOL =
BTUOIL =
BTUGAS =
SULCOL =
SULOIL =
HTRATE =
ISPLC
ISIF'l
ISIP2
REGC
ISIPO
RE: GO
PEDCO
VLIME
SITEFAC =
*END
4770004100,
.105358E+05T
,1375885E4-06f
,1E+04»
.2E-02r
.98600128537E+04r
436270,
2,
IT
0.0»
Or
,442E+03f
, lE+Olr . J.E+01r . lE+Olt . lE-f-Olt
lE4-0.tr .lE-fOlr
BOILER DATA
IB YR TOTCOL TOTOIL
1 72 308V,3 0.0
2 73 3010.7 0.0
TOTGAS AIR
0.0 2234600.
0.0 2234600.
CAFF GENCAP PSCR RETRO COLSIP COLVAL OILSIP OIL.VAL
.580 1300. 1.140 7 500.00 0 4.00
.565 1300. 1.140 3 4.00 0 4.00
-------
TABLE IV-17. REPORT FORMAT FOR THE EDIT OF DATA BASE POWER PLANT DATA
*****9990009910****« CAP .570**** POWER PLANT A ****»
SULCOL BTUCOL SULOIL BTUOIL BTUGAS PEDCO VLT.ME SITEFAC COL.SIP COLVAL OILSIP OILVAL
.042 10268. ,002 136600. 1000. 0.00 0.00 1.00 1,00 1.00 1.00 1.00 1.00 0 0.00 0 0,00
BOILER DATA
IB YR TOTCOL. TO TO It..
1 63 1514.4 0,0
2 63 1514.4 0,0
3 69 2871.3 0,0
TOTGAS AIR
0.0 1166120.
0.0 1166120.
0.0 1829000.
CAPF GENCAP PSCR RETRO COLSIP COLVAL OILSIP OILVAL
.531 704. 1.158 4 2.00 4 1.50
.331 704. 1.158 4 2.00 4 1.50
,617 1.150. 1.158 4 2.00 4 1.50
***##9990009920***** CAP .570**** POWER PLANT B ****B
SULCOL BTUCOL SULOIL BTUOIL BTUGAS PEDCO VLIME SITEFAC COLSIP COLVAL OILSIP OILVAL
.037 10536. .002 137589. 1000. 0.00 0.00 1.00 1.00 1,00 1.00 1.00 1.00 0 0.00 0 0.00
BOILER DATA
IB YR TOTCOL TOTOIL.
1 72 3050.0 0.0
2 73 3050.0 0.0
TOTGAB AIR
0,0 2234600.
0,0 2234600.
CAPF GENCAP PSCR RETRO COLSIP COLVAL. OILSIP OILVAL
.572 1300. 1.140 3 4.00 3 4.00
.572 1300. 1.140 3 4.00 3 4.00
-------
TABLE IV-18. REPORT FORMAT FOR BOILER SCRUBBING COSTS
LPFPCN SPLC
105000100 536460
BID STARTUP
8 1659371.
LPFPCN SPLC
220000100 199526
BID STARTUP
2 3528489.
r-i LPFPCN SPLC
< 265000050 236800
1
£5 BID STARTUP
1 6616406.
LPFPCN SPLC
410000050 296118
BID STARTUP
2 3537997.
1 3595420.
3 3755089.
LPFPCN SPLC
41O000100 296920
BID STARTUP
1 2617317.
S Nft NGB NBS
2211
ACCHET
1851816.
S NB NGB NBS
1111
ACCHET
827721',=;.
S NB NCiB NBS
2111
ACCHET
24767520.
S NB NGB NBS
1333
ACCHET
9791033.
19582066.
29545908 .
S NB NGB NBS
1311
ACCHET
4811954.
YEAR SULCOL
1983 .0250
BTUCOL
9300.
ACCSUL LIMESTONE
2290.
YEAR SULCOL.
1983 .0280
ACCSUL
5387.
YEAR SULCOL
1983 .0350
ACCSUL
33230.
YEAR SULCOL
2776929.
BTUCOL
12800.
LIMESTONE
6037693.
BTUCOL
12000.
LIMESTONE
15227548.
BTUCOL
1983 .O350 10550.
ACCSUL
14942.
29884.
45089.
YEAR SULCOL
LIMESTONE
8311295.
15538735.
23012269.
BTUCOL
1983 .0370 10300.
ACCSUL
7951.
LIMESTONE
5340665.
SULOIL
.0050
MAGNESIA
3314939,
SULOIL
.01OO
MAGNESIA
8096447.
SULOIL
.0050
MAGNESIA
16752575.
SULOIL
.025O
MAGNESIA
11373632.
20241354.
28781268.
SULOIL
.0010
MAGNESIA
8160460.
BTUOIL
136600.
GYPSUM
2521776.
BTUOIL
140000.
GYPSUM
6397774.
BTUOIL
144745.
GYPSUM
15619522.
BTUOIL
149000.
GYPSUM
8900322.
17100976.
25459457.
BTUOIL
136000.
GYPSUM
5793009.
BTUGAS
985.
CFUEL
1.500
BTUGAS
1000.
CFUEL
.729
BTUGAS
1000.
CFUEL
.615
BTUGAS
1000.
CFUEL
.849
.794
.779
BTUGAS
1000.
CFUEL
1.110
CAP DUCT
.238 2.
CAP DUCT
.578 2.
CAP DUCT
.515 4.
CAP DUCT
.650 2.
. 650 2 .
.650 2.
CAP DUCT
.596 2.
AIR
175379.
AIR
281225.
AIR
1083360.
AIR
319222.
319222.
365333.
AIR
169666.
TOTCOL
99.6
TOTCOL
163.5
TOTCOL
1O32.0
TOTCOL
464.0
464.0
472.2
TOTCOL
233.6
TOTOIL
0.0
TOTOIL
695.9
TOTOIL
0.0
TOTOIL
0.0
0.0
0.0
TOTOIL
0.0
YEAR
1981
YEAR
1964
YEAR
1982
YEAR
1970
1969
1972
YEAR
1965
-------
TABLE IV-19. REPORT FORMAT FOR THE PLANT SCAN SUMMARY
i — i
<3
i
N3
LO
40000500
45000200
45000600
450O08OO
105000100
140000800
140009910
165000100
170000200
170000300
175000200
175000300
185000500
185000600
185000800
220000100
265000050
265000100
26500030O
310000100
410000050
41OOOO1OO
485000400
52O00020O
627500100
TOMBI6BEE
BARRY
G OR GAS
GREENE COUNTY
AMES (IOWA)
AMOS
MOUNTAINEER
APACHE
CHOLLA
FOUR CORNERS
BAILEY
MCCL.ELLAN
LAKE CATHERINE
RITCHIE
WHITE BLUFF
ENGLAND
BRANDON SHORES
CRANE
UAGNER
LELAND OLDS
COLEMAN
REID
NEW BOSTON
MILLER-
BIG CAJUN
NB=
NB=
NB=
NB=
NB=
NB=
NB=
NB=
NB=
NB=
NB=
NB=
NB=
NB^
NB='
NB =
NB~
NB==
NB =
NB =
NB =
NB=
NB =
NB =
NB^
3
2
1
2
2
3
1
3
4
5
1
1
1
1
2
1
1
1
2
1
:;
3
2
2
•;>
NGB =
NGB=
NGB=
NGB=
NGB=
NGB=
NGB=
NGB =
NGB=
NGB=
NGB-=
NGB =
NGB=
NGB-
NGB^=
NGB=
NGB==
NGB =
NGB=
NCiB==
NCiB=
NGB =
NGB==
NGB=
NGB =
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
}.
0
0
0
?!
1
0
0
0
NBS=
NBS =
NBS =
NBS =
NBS =
NBS=
NBS =
NBS=
NBS =
NBS =
NB3 =
NBS=
NBS =
NBS =
NBS=
NBS =
NBS =
NBS=
NBS =
NBS-
NBS =
MS--'
NBS =
NBS =
NHS==
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
3
I
0
0
0
SPLC=479442
SPLC=47991A
SPLC=473763
SPLC=47A917
SPLC=536460
SPLC=277620
SPLC=277310
SPLC=795741
SPLC=791168
SPLC=783133
SPLC--607150
SPLC=617450
SPLC=613150
SPLC=A 10430
SPLC=611826
SPLC=1 99526
SPL.C=236BOO
SPLC-232314
SPLC=234000
SPLC=516924
SP 1C =2 961 18
SPl.C=29A920
SPLO142000
SPLC=6A8450
SPLC-644450
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
COMP
NEW
NEU
NEU
NEU
NEW
NEU
NEU
GRAC
NEU
PSCR
SIP
PSCR GAS
PSCR
PSCR SIP
SIP
SIP
.SIP
SIP
KOFF
PSCR
GAS
GAS
SCRB
SCRB
SCRB
SCRB
SCRB
-------
TABLE IV-2C, REPORT FORMAT FOR EMISSIONS AND COMPLIANCE
***** EMISSION AND COMPLIANCE r FOR 45000800
PLANT SIP NB = 2 NBS= 0
REG-CD= 3 LBS02/MMBTU COAL 4,00 OIL 4.00
TONS S/YR ALLOWED 28036. EMITTED 26473.
MMBTUS 28036058. 0. 0.
IB N/0 SUL FrRAC MMBTUCOL OIL GAS
1 0 13737.-2.000 14547816. 0. 0.
2 0 12736.-2.000 13488243. 0. 0.
XSEMR CD
0.
0.
RE6C
REGO ALLOWED
***** EMISSION AND COMPLIANCE
PLANT SIP NB = 2 NBS= 1
FOR 105000100
REG-CD= 4 LBS02/MMBTU COAL. 5.00 OIL 2.50
TONS S/YR ALLOWED 556. EMITTED 2290.
MMBTUS 1851816. 0. 0.
IB N/0 SUL FRAC MMBTUCOL OIL GAS
7 0 1620. -2. 000 1310184. 0. 0,
8 N 2290. .842 1851816. 0. 0,
XSEMR CD
0.
327. 4
REGC
1.20
REGO ALLOWED
.80
556.
***** EMISSION AND COMPLIANCE r FOR 220000100
BOILER SIP N»= 1 NBS= 1
IB N/0 SUL FRAC MMBTUCOL OIL GAS
2 0 5387. .707 4185088. 4092127.
XSEMR CD REGC REGO ALLOWED
0. 1422. 2 1.56 .32 1960.
***** EMISSION AND COMPLIANCE f FOR 410000050
BOILER SIP NB= 3 NBS= 3
IB N/0 SUL FRAC MMBTUCOL
1 0 14942. .747 9791033.
2 0 14942. .747 9791033.
3 0 15205, ,747 9963842.
OIL GAS XSEMR CD
0. 0. 3401. 4
0. 0. 3401. 4
0. 0. 3461, 4
REGC REGO ALLOWED
2,00 1,50 4896.
2.00 1.50 4896.
2,00 1,50 4982,
-------
TABLE IV-21. REPORT FORMAT FOR THE EMISSIONS SUMMARY
EMISSION AND ABATEMENT SUMMARY
REGION NPI..T NBLR EMISSIONS NPC NPS NBS REMOVED EXCESS*
EASTERN 22 38 309939. 12 5 7 84552* 17730,
WESTERN 3 12 30060, 200 0, 0,
TOTAL 25 50 339999, 14 5 7 34552. 17730,
-------
TABLE IV-22. REPORT FORMAT FOR THE EDIT OF SCRUBBER
INVESTMENT AND OPERATING FACTORS
EDIT OF CURRENT SASDATA AND PRICES
i
S3
LIMESTONE PROCESS FOR NEW
FOR YEAR 1978. CEIND.X 214,7
BOILER SUL ==39848. AIR = 888000.
ISEQ AREA
10
20
30
40
50
60
70
80
130
140
150
160
170
180
190
200
210
220
230
240
250
260
270
280
281
282
283
284
290
300
310
320
1 RAW MATL
1
1
1
1
2 FEED PREP
2
o
4 S02 SCRUB
4
4
4
4
5 REHEAT
5
5
6 FANS
6
6
7 CAL SOLIDS
7
7
7
7
8 OFFSITE CH
8
8
8
12 UTILITIES
12
13 SERVICES
13
ITEM
1 CAPITAL
10 LIMESTONE
2 LABOR
8 ELECTRICTY
21 ANALYSES
1 CAPITAL
2 LABOR
8 ELECTRICTY
1 CAPITAL
2 LABOR
7 PROC WATER
8 ELECTRICTY
21 ANALYSES
1 CAPITAL
2 LABOR
6 STEAM 500
1 CAPITAL
2 LABOR
8 ELECTRICTY
1 CAPITAL
2 LABOR
7 PROC WATER
8 ELECTRICTY
21 ANALYSES
1 CAPITAL
7 PROC WATER
8 ELECTRICTY
24 DISPOSAL
1 CAPITAL
8 ELECTRICTY
1 CAPITAL
8 ELECTRICTY
QTY/AMT
419000.
175000.
41.70,
39OOOO.
15 20,
899000,
6470.
5150000,
4655000.
4900.
17470O.
22530000.
1 1 40 .
585000.
1250.
557800.
544000,
1250.
26511000.
3923000 .
3340,
75600,
700000,
380,
4133700.
49600,
888000 .
412000,
67000.
230000 .
638000 .
2? 00 00 ,
RATE
0.00
6.00
10.00
.03
1.5.00
0.00
10.00
.03
0.00
10.00
.06
.03
15.00
0.00
10,00
1 ,40
0,00
10.00
.03
0.00
10,00
.06
.03
15,00
0.00
.06
,03
:l. . 00
0,00
.03
0.00
,03
COST
561544.
1050000.
41.700.
1.0530.
22800,
1204840,
64700,
139050.
6238630.
49000.
1O482,
608310.
1.71.00.
78401.7,
12500.
780920,
729069,
11? 500,
715797,
5257604.
33400,
4536,
18900,
5700,
5539984,
2976,
23976,
412000,
89793,
6210.
855047,
5940.
**SUL
.650
1 , 000
.436
1 ,OOO
, 73 1.
.720
,436
1 , 000
0.000
,436
0,000
0.000
,731
0.000
.436
0,000
0,000
, 436
0.000
,6:1.0
:436
J , 000
1 - 000
, 73 1
,610
1 : 000
.1 ,000
1,000
0,000
0.000
0,000
0.000
**AIR
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0,000
.740
0.000
1 .000
1,000
0,000
.680
0 . 000
1 .000
.630
0,000
1 ,000
0.000
0,000
0.000
0,000
0.000
0.000
0,000
0.000
0,000
,430
.1. .000
.370
1.000
-------
TABLE IV-23. REPORT FORMAT FOR DETAILED SCRUBBING COSTS
DETAILED COSTS FOR BOILER 1,
INPUT DATA SUL= 105159. AIR= 2234600. DUC=4.0 CAP= .580
SUMSUL= 105.1.59, STRTYR=1972.
LIMESTONE P
SCALED RAW
SCALED FE:ED
SCALED SO2
SCALED RE HE:
SCALED FANS
SCALED CAL
SCALED UTIL
SCALED SERV
*UNADJUSTED
ROCESS
MATL
PREP
SCRUB
AT
SOLIDS
ITIES
ICES
TOTAL
CAPITAL
INVESTMENT
INVESTMENT
INVESTMENT
INVESTMENT
INVESTMENT
INVESTMENT
INVESTMENT
INVESTMENT
INVESTMENT
INVESTMENT-COSTS
787317
1807990
9215140
:l. 09 56 98
972959
6705226
99635
897660
21581625
65061015
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
TINVEST:=--UAJ*CONS*CEINDX/160,2
COSTS FOR LABOR
SCALED RAW MATL COST= 54193.ACCUM 54193.
SCALED FEED PREP COST= 84083.ACCUM 138276.
SCALED S02 SCRUB COST= 63680.ACCUM 201955.
SCALED REHEAT COST= 16245.ACCUM 218200.
SCALED FANS COST^ 16245.ACCUM 234445.
SCALED CAL SOLIDS COST= 43406.ACCUM 277851.
OVERHEAD AIU 1.3 ADJCST 361206.
,COST= TINVEST*.3363=
787317,
1807990,
0,
0.
0,
6705226.
0,
0,
9300533*
=13973112,
TOT-CST 14334318,
COSTS FOR STEAM 500
SCALED REHEAT CO^T
OVERHEAD AIU 1,2
1423989.ACCUM
ADJCST
1423989.
1708787.
TOT-CST 16043105.
COSTS FOR PROC WAT^R
SCALED S02 SCRUB msT =
SCALED CAL SnMDS COST =
OVE
ADJ 1
19114.ACCUM
8674,ACCUM
ADJCST
19114,
27788,
33345,
TOT-CST 'I 6076451,
(continued)
-------
TABLE IV-23 (continued)
10
COSTS FOR ELECTRICTY
SCALED RAW MATL CC)ST =
SCALED FEED PREP COST=
SCALED S02 SCRUB COST=
SCALED FANS COST==
SCALED CAL SOLIDS COST=
SCALED UTILITIES COS>T==
SCALED SERVICES COST=
OVERHEAD ADJ 1.2
COSTS FOR LIMESTONE
SCALED RAW MATL COST=
OVERHEAD ADJ 1.0
COSTS FOR ANALYSES
SCALED RAW MATL COST=
SCALED S02 SCRUB COST=
SCALED CAL SOLIDS COST=
OVERHEAD ADJ 1.2
20136. ACCUM
265903. ACCUM
1109239. ACCUM
1305239. ACCUM
36142. ACCUM
11 324. ACCUM
10831 .ACCUM
ADJCST
1.479153. ACCUM
ADJCST
33583. ACCUM
251 87. ACCUM
8396.ACCUM
ADJCST
20136.
286040.
1395279.
2700517.
2736660.
2747983.
2758815.
3310578.
1479153.
1479153.
33583.
58770.
67166.
80599.
TOT-CST 19387028.
TOT-CST 20866181.
TOT-CST 20946780,
DETAILED COSTS FOR BOILER 2
INPUT DATA SUL= 102485. AIR= 2234600. DUC=4.0 CAP= .565
SUMSUL= 102485. STRTYR=1973.
SODIUM SOL. PROCESS CAPITAL
SCALED RAW MATL INVESTMENT
SCALED S02 SCRUB INVESTMENT
SCALED REHEAT INVESTMENT
SCALED FANS INVESTMENT
SCALED PURGE INVESTMENT
SCALED S02 REGEN INVESTMENT
SCALED S02 REDUC INVESTMENT
SCALED SULFUR STR INVESTMENT
SCALED UTILITIES INVESTMENT
SCALED SERVICES INVESTMENT
*UNADJUSTED TOTAL INVESTMENT
TJNVEST=UAJ*CONS*CEINDX/160.2
INVESTMENT-COSTS
404148.
8874612,
1260525.
1046289.
2670936.
526342?:.
49.1.1022.
431518.
309334.
940063.
26111871.
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
COMMON
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
FACILITIES
81261624.COST= TINVEST*,3185=
404148.
0.
0.
0,
2670936.
5263423.
4911022.
431518.
0.
0.
13681048.
16011621.
(continued)
-------
TABLE IV-23 (continued)
COSTS FOR LABOR
SCALED RAM MAIL GUST-
SCALED S02 SCRUB COST =
SCALED REHEAT COST-
SCALED FANS COST-
SCALED PURGE COST™
SCALED s02 REGFN COST™
SCALED S02 PL' DUG COST-
SCALED SULFUR STR COST-
OVERHEAD ADJ 1.3
COSTS EOR NATURE GAS
SCALED S02 REDUC COST =
OVERHEAD ADJ 1.2
COSTS FOR STEAM 500
SCALED REHEAT COST-
SCALED PURGE COST-
SCALED S02 REGEN COST--'
OVERHEAD ADJ 1,2
COSTS EOR PROC WATER
SCAL.ED RAW MATT... COST-
SCALED PURGE COST-
SCALED S02 RE-bEN COST =
OVERHEAP ADJ 1 2
COSTS FOR E::I.F:CTR:I:CTY
SCALED RAW MAIL COST^
SCALED S02 SCRUR COST--
SCALED FANS COST=
SCALED PURGE COST™
SCALED S02 REGFN COST-
SCAL.ED SO 2 RKDUC COST-
SCALE !0 SULFUR STR COST-"
SCAL.ED I.I I II. 1 TIES COST-
SCALED SERVICES COST-
OVERHEAD AD .1 1 .2
^
33957, ACCUM
51408.ACCUM
:l.0782,ACCUh
1. 0782.ACCUM
:I0820A,ACCUM
108206,ACCUM
1.08206. AC GUM
33901 .ACCOM
ADJCST
231.36:51 .ACCUM
ADJCST
1160962, ACCUM
363136, ACCLJM
40051 82. ACCUM
ADJCST
9 70. ACCUM
3?.5.)2- ACCUM
1029740 .ACCOM
ADJCST
2 20 69, ACCUM
107007, ACCUM
1351&53.ACGUM
31f!7f!0. ACCUM
573804, ACCUM
10 IS IV. ACCUM
87V;, ACCUM
:l. 1.037. ACCUM
11 99 A, ACCUM
ADJCST
3X957,
85364,
96146.
106928.
215133,
323339,
431545.
465446,
605080,
2313651,
2776381 ,
1 1 60 96 2 ,
1524099.
5529281 .
6635137.
970,
36532.
1066272,
1279526.
22069,
129077.
1480729.
1799509.
2373313,
2474832,
24B3169.
2494206,
2506202,
3007443,
TOT--CST 16616702
TOT-CST 19393083.
TOT-CST 2602B219.
TOT-CST 27307746<
TOT-CST 30315188,
(continued)
-------
TABLE 1V-23 (continued)
u>
o
COSTS FOR HT CREDIT
SCALED 802 REDUG COST=
OVERHEAD ADJ 1,2
COSTS FOR LIME (NA )
SCALED RAW MAIL COST-
OVERHEAD ADJ 1,0
COSTS FOR VAPNT CAT-
SCALED RAW MAIL COST=
OVERHEAD ADJ 1.0
COSTS FOR SODIUM CBN
SCALED RAW MAT!... COST =
OVERHEAD ADJ 1.0
COSTS FOR ANTI--OX-NA
SCALED RAW MAIL COST=
OVER HE;: AD ADJ i.o
COSTS FOR ANALYSES
SCALED RAW MATL. COST =
SCALED S02 SCRUB COST==
PURGE COST=
SO2 REGEN COST=
S02 REDUC COST=
SULFUR STR
•131390,ACCUM -131390.
ADJCST -157668, TOT-CST 30157520,
SCALED
SCALED
SCALED
SCALED
OVERHEAD ADJ 1.2
10223.ACCUM
ADJCST
29060.ACCUM
ADJCST
1317623. ACCUM
ADJCST
ISO3954.ACCUM
ADJCST
13563.ACCUM
24159,ACCUM
42384.ACCUM
26914,ACCUM
58279,ACCUM
12715.ACCUM
ADJCST
1.0223.
10223. TOT-CST 30:1.67743,
29060,
29060. TOT-CST 30196803.
13J.7623.
1317623, TOT-CST 31514426,
1583954,
1583954. TOT-CST 33098380.
13563,
37722,
80107.
107021,
165300,
178015.
2.13618. TOT-CST 33311998.
-------
DEMAND SUBSYSTEM
Data files used in the demand subsystem are listed in Table IV-24,
Examples of the files not previously described and applicable printed
reports are shown in Tables IV-25 through IV-33. Binary files are
described.
TABLE IV-24. FILES USED IN DEMAND SUBSYSTEM
File name
Description
ACDPAR Acid plant avoidable production cost factors
(input to GENACD)
SACDSML Acid plant and smelter data (input to ACDUPDT)
SULTER Molten sulfur terminals (input to ACDUPDT)
ACDSML Same as SACDSML above, with delivered sulfur costs
added (output from ACDUPDT, input to GENACD)
ACDCST Calculated avoidable production costs for sulfuric
acid (output from GENACD, input to manual procedure
in transportation subsystem, input to GENPGM and
REPTSOL in linear programing subsystem)
SMLCST Calculated production costs for smelters (output
from GENACD, input to manual procedure in trans-
portation subsystem, input to GENPGM and REPTSOL
in linear programing subsystem)
RBNSORT NRBT rate basing point data (required to add new
sulfur terminals or acid plants, input to the
manual procedure in the transportation subsystem)
RAILWA Rail mileages between all possible combinations of
NRBT rate basing points (input to ACDUPDT, input
to TRNCOST in transportation subsystem)
X313SUL Molten sulfur tariff data (input to ACDUPDT)
IV-31
-------
TABLE IV-25. FILE FORMAT FOR SULFURIC ACID AVOIDABLE
PRODUCTION COST FACTORS (ACDPAR)
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Description
Tons of sulfur per ton sulfuric acid (before year 60)
Tons of sulfur per ton sulfuric acid (after year 60)
Year of technology change
Sulfuric acid plant investment ($/ton/yr)
Capacity for base case plant (Mtons/yr)
Scale factor for determining investment for other
sized plants
Variable conversion cost per ton ($/ton)
Fixed annual conversion cost ($/yr)
Taxes and insurance rate
Time preference rate for money
Compound maintenance rate
Economic useful life
Percent sulfuric acid concentration
Port Sulphur price ($/short ton)
TVA sulfuric acid price ($/ton)
Proportion of 330 TPD capacity estimate
Not used
Not used
Number of years considered
Years considered
Unit cost inflation factor
Transportation cost inflation factor for sulfur
. _ __ — ^ — ^
Format
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
Free form
IV-32
-------
TABLE
Field
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
IV-26. FILE FORMAT FOR ACID PLANT AND SMELTER DATA
Description
Acid plant sequence number
Not used
Acid plant name
Not used
Acid plant location
SPLC
SPLC for rate basis location
Rate basis location
Rate basis index
Tariff index (MES)
Year plant built
Plant capacity (1000 tons/year)
Days operated
Not used
Latitude
Longitude
Not used
Not used
Not used
Not used
Percent sludge acid used
CDS compliance code
TVA plant status code
Not used
(SACDSML)
Format
13
IX
2A10
IX
A10,A5
15
A6
A10, A5
14
12
A2
A4
A3
Al
A5
A5
14
14
14
13
A4
A2
13
13
IV-33
-------
TABLE IV-27. FILE FORMAT FOR SULFUR TERMINAL DATA (SULTER)
Field
number Description Format
1 Sulfur terminal SPLC 16
2 Rail miles index 14
3 Barge mode index A10
4 Rail tariff index (MES) 12
5 Barge rate F5.2
6 Terminal type A2
7 Rate estimation flag Al
8 Statute miles from Port Sulphur to sulfur terminal A4
9 Sulfur terminal location 2A10,A1
10 Rate basis location A10.A5
11 Sulfur terminal handling charge F5.2
IV-34
-------
TABLE IV-28. FILE FORMAT FOR ACID PLANT AND SMELTER DATA WITH
DELIVERED COSTS OF MOLTEN SULFUR INCLUDED (ACDSML)
. —
Field
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
1.
Description
Acid plant sequence number
Not used
Acid plant name
Not used
Acid plant location
SPLC
SPLC for rate basis location
Rate basis location
Rate basis index
Tariff index (MES)
Year plant built
Plant capacity (1000 tons/year)
Days operated
Not used
Latitude
Longitude
Unit cost of delivered sulfur
Unit Port Sulphur barge rate to sulfur terminal
Unit rail rate for sulfur terminal to acid plant
Unit handling charge for sulfur terminal
Percent sludge acid used
CDS compliance code
TVA plant status code
Sulfur terminal index
Fo rma t
13
IX
2A10
IX
A10.A5
15
A6
A10.A5
14
12
A2
A4
A3
Al
A5
A5
14
14
14
13
A4
A2
13
13
IV-35
-------
TABLE IV-29. REPORT FORMAT FOR DELIVERED SULFUR COSTS TO ACID PLANTS
ACID PLANT 1AGRICO CHEM-WILLIAMS
OLD PORT SULFUR RATE - 10.21 NEW RATE
ACID PLANT 2AGRICO CHEM-WILL IAMS
OLD PORT SULFUR RATE - 10.26 NEW RATE
ACID PLANT 7ALLIED CHEMICAL CORF-
OLD PORT SULFUR RATE = 9.66 NEW RATE
ACID PLANT 10 ALL I ED CHEMICAL CORF-
OLD PORT SULFUR RATE = 22,16 NEW RATE
ACID PLANT 11 ALLIED CHEMICAL CORF-
OLD PORT SULFUR RATE - 13.44 NEW RATE
ACID PLANT 13ALLIED CHEMICAL CORF-
OLD PORT SULFUR RATE." = 16,34 NEW RATE
ACID PLANT 15AMERICAN CYANAMID
OLD PORT SULFUR RATE = 13,79 NEW RATE
ACID PLANT 16AMERICAN CYANAMID
OLD PORT SULFUR RATE = 9,85 NEW RATE
SULFUR TERMINAL NO. 5 FLTAMPA
= 10,22 ( 5,80+ 3.35+ 1.07)
SULFUR TERMINAL. NO. 14 LAPORT SULPHUR
= 9.61 ( 7.40+ 0.00+ 2,21)
SULFUR TERMINAL NO, 11 LABATON ROUGE
= 9.41 ( 5.20+ 2,00+ 2,21)
SULFUR TERMINAL. NO, 40 VARICHMQND
= 20,65 ( 12,80+ 6,78+ 1,07)
SULFUR TERMINAL NO, 40 YARICHMOND
= 13.45 ( 5.60+ 6,78+ 1.07)
SULFUR TERMINAL NO. 17 MDBALTIMORE
= 16.35 < 8.40+ 6.88+ 1.07)
SULFUR TERMINAL NO. 26 NJSAYREVILLE
= 13,73 ( 5,60+ 7.06+ 1,07)
SULFUR TERMINAL NO, .1.9 MSPASCABOULA
= 9.59 ( 5,60+ 1.78+ 2.21)
ACID PLANT 17AMERICAN CYANAMID
OLD PORT SULFUR RATE = 22.41 NEW RATE =
SULFUR TERMINAL NO, 8 ILJOLIET
22,16 ( 5,60+ 14.35+ 2,21)
ACID PLANT 18AMERICAN CYANAMID
OLD PORT SULFUR RATE = 11,62 NEW RATE =
SULFUR TERMINAL NO, 6 GASAVANNAH
11,63 ( 5.20+ 5.36+ 1.07)
-------
TABLE IV-30. FILE FORMAT FOR SULFURIC ACID AVOIDABLE
PRODUCTION COSTS AND RELATED QUANTITIES (ACDCST)
Field
number
1
2
3
4
5
6
Description
Acid plant sequence number
Acid plant name
Acid plant location
SPLC
Unit production cost of sulfuric acid
using elemental sulfur (Port Sulphur)
Sulfuric acid plant production
Format
13
2A10
A10,A6
16
F10.2
F10
TABLE IV-31. FILE FORMAT FOR SMELTER PRODUCTION
COSTS AND RELATED QUANTITIES (SMLCST)
Field
number
1
2
3
A
5
6
Description
Pseudo FPC code
Smelter name
Smelter location
SPLC
Unit production cost of sulfuric acid
Sulfuric acid production
Format
110
2A10
A10,A6
16
F10.0
F10.0
.
IV-37
-------
TABLE IV-32. FILE FORMAT FOR RAIL RATE BASING
POINT LOCATION DATA (RBNSORT)
Field
number Description Format
1 State abbreviation A2
2 Rate basing point name 2A10.A7
3 Rate basis SPLC 16
4 Rail mileage index 14
5 Tariff table index 12
IV-38
-------
TABLE IV-33. REPORT FORMATS FOR SULFURIC ACID
AVOIDABLE PRODUCTION COSTS
<
u-
REPORT 1
SULFURIC ACID PLANTS CONSIDERED
* NAME LOCATION
1. AGRICO CHEM-WILLIAMS
2. AGRICO CHEM-UILLIAMS
7. ALLIED CHEMICAL CORP
10. ALLIEK CHEMICAL CORP
11. ALLIED CHEMICAL CORP
13. ALLIED CHEMICAL CORP
15. AMERICAN CYANAM.TD
16. AMERICAN CYANAMID
17. AMERICAN CYANAMID
18. AMERICAN CYANAMID
TOTAL ACID CAPACITY IS 2811;
ANNUAL
YEAR CAPACITY
PIERCE
DONALDSQNVL
GEISMAR
NITRO
HOPEWELL
FRONT ROYAL
BOUND BROOK
MOBILE
JOLIET
SAVANNAH
FL
LA
LA
UV
VA
VA
NJ
AL
IL.
GA
1975
1975
1968
1940
1965
1945
1945
1967
1937
1967
900
900
375
101
150
120
48
19
37
162
REPORT 2
SULFUR FREIGHT RATES
+LOCATION
PORT SULFUR RATES
TOTAL BARG RAIL
HANDLING
J. .
9 ^
7.
10.
11 .
13.
15.
16.
17.
18.
PIERCE
DONALDSONVL
GEISMAR
NITRO
HOPEWELL
FRONT ROYAL
BOUND BROOK
MOBILE
JOLIET
SAVANNAH
FL
LA
LA
WV
UA
UA
NJ
AL
IL
GA
22
20
20
44
29
35
29
20
47
25
.05
.74
.30
.58
.03
.29
.66
.69
.84
.10
5
7
5
12
5
8
5
«::;
5
£*
.79
.40
.19
,80
,59
. 39
.59
,59
. 59
.19
3
0
':>
6
6
A
7
:l
14
5
.34
.00
.00
.77
. 77
.87
.06
.78
,15
.35
1
^>
o
1
1
1
.1.
2
0
1
,07
.20
.20
.07
.0;'
,07
.07
- 20
,20
.07
(continued)
-------
TABLE IV-33 (continued)
REPORT 3
SULFUR PRICE F.O.B,
BASE YEAR
INVESTMENT RATE
MAINTENANCE RATE
UNIT COST INFLATION FACTOR-NEW
UNIT COST INFLATION FACTOR-OLD
TRANSPORTATION INFLATION FACTOR
62,50
1983
.10
,05
1.00
1.00
2.16
•c-
o
SULFUR1C ACID PRODUCTION COSTS
t LOCATION
:u PIERCE:
2, DONALDSOMUL
7, GEISMAR
10, NITRO
11. HOPE:: WELL
13, FRONT ROYAL
15, BOUND BROOK
16, MOBILE
17, JOLIET
18, SAVANNAH
19< KALAMAZOO
FL
LA
LA
WV
MA
MA
NJ
AL
IL
CiA
MI
SULFUR
FACTOR
,3282
,3282
,3282
,3282
,3282
.3282
,3282
. 3282
,3282
,3282
» 3282
CONVERSION
- RETRO
-- TOT
1983
-3. 18
-3. 18
-1, 10
3,6?
1 ,42
3.11
6,61
8,76
7,86
.86
8,97
0.00
3,04
0.00
0.00
5.10
0.00
7.06
0.00
0.00
0,00
0,00
24,57
27, 18
26,07
38,83
36,56
35,21
43,91
36,06
44.07
29,61
47,06
-------
Rail Mileage File (RAILWA)
The RAILWA file contains rail mileages between 2632 rate basis
locations published for the various tariff associations. The file is
binary so a specific file format is not provided; listing of the data as
stored is not meaningful.
A detailed narrative of the design and development of the rail
mileage file is provided in Appendix C because of its potential use in
applications other than the byproduct marketing system. The file can be
used to determine rail mileages for various purposes such as finding all
rate basing locations within a specified distance by rail from a given
location, the distance from several rate basing points to other rate
basing points including all other points if required, etc.
Sulfur Tariff File (X313SUL)
The X313SUL file contains tariff rates for shipping sulfur by rail
with individual rates dependent upon the distance to be shipped. The
file as used in the system is binary and because of the coding technique
used a simple listing of the file is not particularly useful. A binary
data file was used because the tariff rates for sulfur were taken from a
generalized structure of rail rates (both class rates and commodity
rates) where the conservation of memory and storage as well as efficient
data access were necessary (see also the sulfuric acid tariff file).
processing of the data requires a program such as ACDUPDT (described in
section III). The source program listing can be used to examine the
exact procedures but the coding details are not necessary for use of the
system.
Conceptually, the tariff structure can be thought of as a table or
array with the columns representing the various commodities shipped and
the rows representing the exact shipping rate per commodity, one row for
each distance specified in the tariff. When several tariffs are applicable,
a, separate table or array can be used. Using the concept of tariff
tables, a knowledge of the source, destination, commodity, and distance
to be shipped allows the selection of the appropriate table (array),
column, and row, respectively, that contains the applicable rate value.
Because the sulfur tariff is a class tariff the process is greatly
gimplified. Only one tariff applies; therefore only a single table is
required for sulfur shipping rates.
IV-41
-------
TRANSPORTATION SUBSYSTEM
Data files used in the transportation subsystem are shown in
Table IV-34. The file format for TRNPTS (the only nonbinary file not
previously described) is shown in Table IV-35.
TABLE IV-34. FILES USED IN TRANSPORTATION SUBSYSTEM
File name
Description
RAILWAa Rail mileages between all possible combinations of NRBT rate
basing points (input to ACDUPDT in the demand subsystem, input
to TRNCOST)
SCRCST Boiler scrubbing costs and emissions data (output from STMCAP
in the supply subsystem, input to the manual procedure, input
to GENPGM in the linear programing subsystem)
ACDCST3 Calculated avoidable production costs for sulfuric acid
(output from GENACD in the demand subsystem, input to the
manual procedure, input to GENPGM and REPTSOL in the linear
programing subsystem)
SMLCST3 Calculated production costs for smelters (output from GENACD
in the demand subsystem, input to the manual procedure, input
to GENPGM and REPTSOL in the linear programing subsystem)
RBNSORTa NRBT rate basing point data (input to the manual procedure,
required in the demand subsystem to add sulfur terminals)
TRNPTS Power plants, acid plants, smelters, and transshipment
terminals with the appropriate location data (output from the
manual procedure, input to TRNCOST)
X313H2S Sulfuric acid tariff data (input to TRNCOST)
TRNCST Transportation costs from potential supply points to
potential demand points (output from TRNCOST, input to GENPGM
in the linear programing subsystem)
Described in the preceding demand subsystem file descriptions.
Described in the preceding supply subsystem file descriptions.
a.
b.
IV-42
-------
TABLE IV-35. FILE FORMAT FOR TRANSPORTATION POINTS
TO BE USED FOR SHIPPING RATE CALCULATIONS (TRNPTS)
Field
number
Demand Section
1
2
3
4
5
6
7
8
9
10
Description
(acid plants)
Identification code
Location
Miles to rail
Rate basis SPLC
Rail mileage index (rate basing point number)
Tariff table index
Miles to barge
Barge node index
Miles to Cairo
Miles to Harvey Locks
Format
13
16
15
16
14
12
15
A10
15
15
gygply__Sec_tion (power plants, smelters, and transshipment terminals)
1 Identification code HO
2-10 As shown above
IV-43
-------
Sulfurlc Acid Tariff File (X313H2S)
The X313H2S file contains tariff rates for shipping sulfuric acid
by rail with individual rates dependent on the source and destination
points and the distance to be shipped. The file as used in the system
is binary, and because of the coding techniques used a simple listing of
the file is not particularly useful. A binary data file was used
because the tariff rates for sulfuric acid were taken from a generalized
structure of rail rates (both class rates and commodity rates) where the
conservation of memory and storage as well as efficient data access were
necessary (see also the sulfur tariff file). Processing of the data
requires a program such as TRNCOST (described in section III) . The
source program listing can be used to examine the exact procedures, but
the coding details are not necessary for use of the system.
Conceptually, the tariff structure can be thought of as a table or
array with the columns representing the various commodities shipped and
the rows representing the exact shipping rate per commodity, one row for
each distance specified in the tariff. When several tariffs are appli-
cable, a separate table for each must be used.
Using the concept of tariff tables, knowledge of the tariff that
applies, the commodity, and the distance to be shipped allows the
selection of the appropriate array, column, and row, respectively, in
which the rate value that applies is entered. Because the sulfuric acid
tariff is a commodity tariff, the tariff rate that applies not only
depends on the distance to be shipped but also on the source and destina-
tion location of the shipment; therefore multiple tables are required
for various source and destination combinations. The sample tariff
table numbers shown in Table IV-36 illustrate the organization to allow
determination of the appropriate table (array) based upon source and
destination locations within the Docket 28300 territory. A map is also
provided in Figure IV-2.
Transportation Cost File (TRNCST)
The TRNCST file is generated by the TRNCOST program for locations
in the TRNPTS file and is used in the GENPGM program. The file is word
addressable and randomly indexed. The index key is the supply point
identification code. The file contains one record per potential supply
point and each record contains 200 entries, 1 entry for each potential
demand point to be considered. The file is binary so a direct listing of
the file is not meaningful. The source program listings of the creating
program (TRNCOST) or the accessing program (GENPGM) can be referenced if
required but is not necessary to use the system. The actual values were
built during the rate selection procedure. Each entry contains the
following data:
1. The supply point index to the tariff tables
2. The demand point index to the tariff tables
3. The tariff table used
4. The rail mileage between the supply point and demand point
5. The rail rate between the supply point and demand point
IV-44
-------
TABLE IV-36. SAMPLE TARIFF TABLE DEFINITIONS
FOR THE ICC DOCKET 28300 TERRITORY
Tariff
table
number
Tariff
series
Application
1 WTL W-1000
2 WTL W/S-1001
3 IFA 1-1002
4 IFA I/S-1003
5 SWL SW-1004
6 SWL SW/E-1005
7 SWL SW/W-1006
8 SWL SW/S-1007
9 TL E/S-1008
10 TL-CTR E-1009
11 CTR E/W-1010
12 SFTB S-1011
Between stations within WTL territory
Between stations in WTL and Southern
territory
Between stations within IFA territory
Between stations in IFA and Southern
territory
Between stations with Southwestern
territory
Between stations in Southwestern and
Official territory
Between stations in Southwestern and
WTL territory
Between stations in Southwestern and
Southern territory
Between stations in Official (except
Illinois) and Southern territory
Between stations within New England,
Trunkline, and CTR and between those
stations and IFA territory
Between stations in Official and WTL
territory
Between stations within Southern
territory
IV-45
-------
WESTERN TRUNK
TERRITORY
GENERAL FREIGHT
COMMITTEE TERR
TRANS-CONTINENTAL
TERRITORY
SOUTHER
I ASSN
SOUTHWESTERN
TERRITORY
Figure IV-2. Railroad rate territories.
-------
LINEAR PROGRAMING SUBSYSTEM
Data files used in the linear programing subsystem are listed in
Table IV-37. The only nonbinary file required for system executions and
not previously described is shown in Table IV-38. Table IV-39 illus-
trates a printed report from the model generator program. Table IV-40
illustrates a final report of the model solution. Binary files are
described.
TABLE IV-37. FILES USED IN THE LINEAR PROGRAMING SUBSYSTEM
File name
Description
SCRCST3 Boiler scrubbing costs and emissions data (output from STMCAP
in the supply subsystem, input to the manual procedure in the
transportation subsystem, input to GENPGM)
Calculated production costs for smelters (output from GENACD
in the demand subsystem, input to the manual procedure in the
transportation subsystem, input to GENPGM and REPTSOL)
Calculated avoidable production costs for sulfuric acid
(output from GENACD in the demand subsystem, input to the
manual procedure in the transportation subsystem, input to
GENPGM and REPTSOL)
TRNCSTC Transportation costs from potential supply points to
potential demand points (output from TRNCOST in the trans-
portation subsystem, input to GENPGM)
GENDATA Report data (output from GENPGM, input to sort procedure)
GENREPT Same as GENDATA except sorted (output from sort procedure,
input to REPTSOL)
HODLIN Linear programing model (output from GENPGM, input
to APEX)
HODLOUT Model solution (output from APEX, input to REPTSOL)
FINRPT On-line copy of final report (output from REPTSOL)
Described in supply subsystem file descriptions.
Described in demand subsystem file descriptions.
Described in transportation subsystem file descriptions.
b.
c.
IV-47
-------
TABLE IV-38. FILE FORMAT FOR POWER PLANT STRATEGY
PRESELECTION RESULTS (GENDATA/GENREPT)
Field
number
Power
1
2
3
4
5
6
7
8
Description
Plants Considered in Model Report Record
Not used
Relative sequence number of plants
to be considered in the model
SPLC
FPC number
Unit production costs (incremental)
Production capacity to be considered
in model
Total heat input for capacity to be
considered
Plant name
Format
10X
14
16
A10
F10.2
F10.0
F10.0
3A10
9 Total disposable byproduct strategy
costs F10.0
10 Total marketable byproduct strategy
costs F10.0
Power Plant Status Report Record
la Strategy status from preselection
procedure prior to model 4A10
2 FPC number A10
3 Emissions to be removed F10.0
a. Possible status values: all clean fuel, all
scrubbing, part clean fuel, part scrubbing, total
emissions removed by clean fuel (summary), total
emissions removed by scrubbing (summary).
IV-48
-------
TABLE IV-39. REPORT FORMAT FOR POWER PLANT
STRATEGY PRESELECTION RESULTS
i
j>
VO
NO. SF'LC FPC COPE
1.411440 720000900
21176711145001900
34957851655000300
46864632185000900
52836252455000250
COST TONS-H2S04 HEAT
PLANT NAME
0,00 67592, 66640000,ROXBORO
9,41 75726. 33768383.CAMPBELL
25.23 196573. 53997072,CRYSTAL RIVER
O.OO 101605. 67650014.PARISH
22.88 33B976.127866860.GHENT
LMSTN CSTS
31639237.
16815410.
29567219.
33537684.
69514691.
MGO CSTS
30183021.
17528253.
31956257.
32936912.
71695468.
PLANT STATUS
ALL CLEAN FUEL
ALL CLEAN FUEL
ALL CLEAN FUEL
ALL CLEAN FUEL
ALL SCRUBBING
ALL SCRUBBING
ALL SCRUBBING
ALL SCRUBBING
ALL SCRUBBING
PART CLEAN FUEL
PART SCRUBBING
TOTAL EMISSIONS REMOVED CLEAN FUEL
TOTAL EMISSIONS REMOVED SCRUBBING
WESTERN POWER PLANT
WESTERN POWER PLANT
FPC COUE
105000100
220000100
410000100
720000100
720000900
2345000200
3945000600
4740000100
4785009910
1145001900
114500.1.900
(TONS S)
(TONS Si)
3705002700
4035000350
TONS SULFUR
2061.
4E!48.
7156.
4667.
22071.
5982.
19677,
15803.
12259,
13786.
24727.
2113473.
3B5593.
731146
783120
-------
TABLE IVr-40. REPORT FORMAT FOR MODEL SOLUTION
PROBLEM NAME $.50 ACFL
TOTAL INDUSTRY COST 649486.18
TOTAL ACID TRANSPORT COST 12740.67
TOTAL EASTERN SMELTER ACID PRODUCTION COST 0.00
< TOTAL WESTERN SMELTER ACID PRODUCTIN COST 0,00
g TOTAL CANADIAN ACID PRODUCTION COST 3002.00
TOTAL. EASTERN STEAM PLANT PRODUCTION COST 2337.06
TOTAL MARKET INCENTIVE FOR STEAM PLANT ACID 0.00
TOTAL ACID PLANT PRODUCTION COSKPORT SULFUR) 631406.44
TOTAL ACID PLANT DEMAND (000 TONS) 23255,
(continued)
-------
TABLE 1V-AO (continued)
EASTERN SMELTER SALES
WESTERN SMELTER SALES
ARIZONA CHICAGO
ARIZONA ST.LOUIS
ARIZONA MEMPHIS
ARIZONA BATONROUGE
ARIZONA HOUSTON
ARIZONA
TOTAL
NEW MEXICO CHICAGO
NEW MEXICO ST.LOUIS
NEW MEXICO MEMPHIS
NEW MEXICO BATONROUGE
NEW MEXICO HOUSTON
NEW MEXICO TOTAL
UTAH
UTAH
UTAH
UTAH
UTAH
UTAH
CHICAGO
ST.LOUIS
MEMPHIS
BATONROUGE
HOUSTON
TOTAL
MONTANA CHICAGO
MONTANA ST.LOUIS
MONTANA MEMPHIS
MONTANA BATONROUGE
MONTANA HOUSTON
MONTANA TOTAL
CANADIAN ACID SALES
BUFFALO
DETROIT
CANADA TOTAL
EASTERN STEAM PLANT SALES
PORT SULFUR SALES
CAPACITY
419.
118.00
380.00
96.00
144.00
SALES
419.
0.00
0.00
0.00
0.00
118.00
118.00
50.00
166.00
0.00
50.00
9.00
275.00
0.00
96.00
0.00
0.00
0.00
96.00
0.00
144.00
0.00
0.00
O.OO
144.00
200.00
0.00
200.00 200.00
780.472 315.922
22320.077
(continued)
-------
TABLE IV-40 (continued)
Ul
M
ACID PLANT PURCHASE SUMMARY
1AGRICO CHEM-WILLIAMS PIERCE
PORT SULFUR PURCHASES
FL
2AGRICO CHEM-UILLIAMS DONALDSONVL LA
PORT SULFUR PURCHASES
7ALLIED CHEMICAL CORP GEISMAR
PORT SULFUR PURCHASES
LA
10ALLIED CHEMICAL CORP NITRO WV
PORT SULFUR PURCHASES
11ALLIED CHEMICAL CORP HOPEWELL VA
PORT SULFUR PURCHASES
13ALLIED CHEMICAL CORP FRONT ROYAL MA
PORT SULFUR PURCHASES
15AMERICAN CYANAMID BOUND BROOK NJ
PORT SULFUR PURCHASES
CANADA-BUFFALO
16AMERICAN CYANAMID MOBILE AL
PORT SULFUR PURCHASES
17AMERICAN CYANAMID JOLIET IL
PORT SULFUR PURCHASES
STEAM PLANT 2 1145001900 CAMPBELL
18AMERICAN CYANAMID SAVANNAH GA
PORT SULFUR PURCHASES
DEMAND
900.
900.000
900.
900.000
375.
375.000
101,
101.000
150.
150.000
1.20.
120.000
49.
0.000
49.000
20.
20.000
38.
11.274
26,726
162,
162.000
COST
24.57
27.18
26.07
38.83
36.56
35.21
43,91
24.79
36.06
44.07
30.23
29.61
(continued)
-------
TABLE IV-40 (continued)
EASTERN SMELTER SALES SUMMARY
EASTERN SMELTER NEW JERSEY ZINC CO
133 E I DUPONT DE NEM GIBBSTOUN
EASTERN SMELTER ST JOE MINERALS
75 MONSANTO COMPANY E ST LOUIS
M EASTERN SMELTER AMER METAL (AMAX)
< 75 MONSANTO COMPANY E ST LOUIS
Ol
LJ EASTERN SMELTER AMAX LEAD COMPANY
75 MONSANTO COMPANY E ST LOUIS
EASTERN SMELTER ENGELHARD-NAT*L ZINC
91 PENNSALT CHEMICALS TULSA
EASTERN SMELTER ST JOE MINERAI S
72 MINN MIN + SMELT COPLEY
137 ALLIED CHEMICAL CORP CLEVELAND
EASTERN SMELTER CITIES SERVICE 01 .
27 ARMY AMMUNITION PL.T 7 YNER
28 ARMY AMMUNITION PLT RADFORD
68 KERR-MCGEE COT TO ND ALE
95 REICHHOLD CHEMICALS TUSCALOORA
128 HOME GUANO COMPANY DOTHAN
130 COLUMBIA NITROGEN MOULTRIE
EASTERN SMELTER AMER SMELT ( ASARCC) )
138 EL PASO PRODUCTS EL PASO
PALMERTON PA
NJ
HERCULANEUM MO
IL
MONSANTO IL
IL
SALEM (BUICH) MO
IL.
BARTLESVILLE OK
OK
JOSEPHTOWN PA
OH
OH
COPPERHILL TN
TN
VA
Fl.
Al...
AL
GA
EL PASO TX
TX
CAPACITY SALES
26.000 26.000
26.000
14,000 14.000
14.000
12.000 12.000
12.000
7.000 7.000
7.000
9.000 9.000
9.000
f54.000 54.000
41,000
13.000
200.000 200.000
99.000
23, 000
11,000
41,000
8.000
18.000
16.000 16.000
16.000
SALES
PRICE
12.79
11.66
8.92
17.71
8.29
15.96
18.36
14.08
21 .58
23.41
21 .04
22 .81
21.58
6.18
(continued)
-------
TABLE IV-40 (continued)
CANADA SALES SUMMARY
SALES
SALES
PRICE
i
l_n
-p-
BUFFALO
15 AMERICAN CYANAMID
51 E I DUPONT DE NEM
53 ESSEEX CHEMICAL CO
133 E I DUPONT DE NEM
135 CITIES SERVICE OIL
DETROIT
BOUND BROOK NJ
CORNWELLS HTSPA
NEWARK NJ
GIBBSTOWN NJ
MONMOUTH JCT NJ
49,000
56.000
12.000
57.000
26.000
24.79
26.03
23.48
29.37
24.79
(continued)
-------
TABLE IV-40 (continued)
WESTERN SALES SUMMARY
SALES
SALES
PRICE
ARIZONA
HOUSTON 88 OLIN CORPORATION
PASADENA
TX
118.000
27.56
NEW MEXICO
CHICAGO 17 AMERICAN CYANAMID
ST.LOUIS 79 NL INDUSTRIES INC
BATONROUGE 126 AMERICAN CYANAMID
HOUSTON 88 OLIN CORPORATION
JOLIET
ST LOUIS
FOR TIER-
PASADENA
IL
MO
LA
TX
50.000
166.000
50.000
9.000
25.54
28.16
29,56
27,56
UTAH
ST.LOUIS 75 MONSANTO COMPANY
E ST LOUIS IL
96.000
27.000
MONTANA
ST.LOUIS 79 NL INDUSTRIES INC
ST LOUIS
MO
144.000
28.79
(continued)
-------
TABLE IV-40 (continued)
STEAM PLANT SALES SUMMARY
INCREMENTAL
CAPACITY COST SALES
SALES
PRICE
1 720000900 ROXBORO
98 ROYSTER COMPANY
108 SWIFT CHEM CO
109 SWIFT CHEM CO
119 WEAVER FERTILIZER
67.592
0*000
NORFOLK VA
WILMINGTON NC
NORFOLK VA
NORFOLK VA
2 1145001900 CAMPBELL
17 AMERICAN CYANAMID JOLIET IL
19 AMERICAN CYANAMID KALAMAZOO MI
107 SWIFT CHEM CO CALUMET CITY IL
3 1655000300 CRYSTAL RIVER
4 2185000900 PARISH
88 OLIN CORPORATION PASADENA TX
5 2455000250 GHENT
20 AMERICAN CYANAMID HAMILTON OH
75.726
9*410
196.573 25*230
101.605 0*000
338*976 22.880
15.00
.59
26.00
26.00
26.73
19.00
30.00
17.110
17.110
17*110
17.110
30*230
17.490
30,230
101.61
71.00
6*180
34.030
(continued)
-------
Linear Programing Model Problem File (MODLIN)
The MODLIN file contains the marketing model in the format required
by the APEX linear programing system, and is generated by the GENPGM
orogram. Users requiring exact details should refer to a source program
!j j Sting of the GENPGM program and an APEX users manual (see Appendix B).
A detailed understanding of the file format is not necessary to use the
system.
Linear Programing Model Solution File (MODLOUT)
The MODLOUT file is produced by the APEX linear programing system.
It contains the solution to the model and allows customized reports to
ue developed in addition to the standard APEX reports. The file is used
•tn the REPTSOL program to generate a final report analysis of the strategy
selection process, including the model solution. Users requiring exact
details should refer to a source program listing of the REPTSOL program
and an APEX users manual (see Appendix B). A detailed understanding of
the file format is not necessary to use the system.
IV-57
-------
V. SYSTEM USAGE
GENERAL
The byproduct marketing system was developed on the Control Data
Corporation (CDC) CYBERNET system which provides both interactive (con-
versational) time-sharing services and remote batch processing by means
of a. nationwide commercial data processing network. A list of manuals
related to the use of the byproduct marketing system on the CYBERNET
system is provided in Appendix B. The Interactive Service Reference
Set Volume 1, Tutorial, provides comprehensive descriptions and illus-
trations (including logging into the system, user input, and terminal
responses) of the procedures and examples shown in this section.
Initially the system will not be kept on-line for independent, full-
time* interactive-batch access by all users. Many potential users may
not have a contractual agreement with CDC to access the CYBERNET system.
Even where agreements do exist, the background experience required to
fully utilize the capabilities of the byproduct marketing system itself
aa well as the capabilities related to its use on CYBERNET may be
unavailable or too costly to develop during a preliminary investigation
Of potential benefits. Another consideration is the actual cost of
keeping the system on-line full time. The storage charges that would be
incurred, especially during the initial stages of system availability,
would probably be excessive when compared to the level of usage.
Under an information exchange agreement between EPA and TVA, capabili-
ties will be provided by TVA to process user requests (subject to limita-
tions based upon available funding, the costs that would have to be
absorbed, and the source of and justification for the requests). This
will allow users to analyze some of the general system capabilities with
a minimum amount of investigation and investment. After analyzing the
results obtained and the associated costs that were incurred to provide
the results (approximate costs that had to be absorbed to satisfy a
given request will be included with the results) each user should be
able to determine if potential benefits justify independent usage and
Billing. This will also allow a more cost-effective use of the funds
for information exchange since several user requests can probably be
processed together to reduce both the man-hours and computer resources
required.
V-l
-------
The system will be kept on-line for independent full-time access by
all users only when this is justified by level of usage, the availability
of funding, and the number of users with the requirements and capability
to use the system independently.
Even when the system is available on a full-time on-line basis,
users desiring to access the system independently must have the necessary
contractual agreement for direct billing by CDC since neither TVA nor
EPA can be responsible for computer charges that are not directly under
their control. (This would not necessarily apply to the storage charges
for keeping the system on-line for all users.)
It is the intent of the information exchange agreement between EPA
and TVA to give potential users the opportunity to investigate possible
benefits from the byproduct marketing system. The capability provided
by TVA under the information exchange agreement may be simply providing
clarification and additional information about the system as well as
providing computer runs. In either case when the requirements of a
specific user affect the service that can be provided to all users, the
user will be required to establish procedures for independent access and
direct billing.
For those users who require complete freedom to access and modify
the system to their own specifications, special arrangements can probably
be made to provide individual copies of the system by means of magnetic
tape or by copying the system from on-line files, either into the users'
own catalog of files or onto off-line storage facilities. These users
would then have total freedom of use of the system but would have to
obtain updated versions of the system as they become available and would
be responsible for charges associated with their individual copies of
the data and programs.
Those users whose requirements can be satisfied as a result of the
information exchange agreement will find the remainder of the sections
on actual usage unnecessary except as an aid to further understanding of
the internal relationships between the various parts of the system.
Other users, who prefer to access the system independently, should
analyze the remainder of the sections on usage very carefully; they will
have an effect on both the results and the cost of obtaining the results.
EXECUTION MODES
The CDC CYBERNET system provides distinctly different operating
environments that may be used for processing such applications as the
byproduct marketing system. If interactive time-sharing capabilities
are unnecessary and remote job entry (RJE) facilities are available, the
SCOPE 3.4 BATCH SERVICE and CYBER 76 BATCH SERVICE offer priority
schedules that may result in significant savings. If interactive time-
sharing is required, the NOS INTERACTIVE AND BATCH SERVICE is the logical
choice. In addition to interactive time-sharing, NOS also provides two
optional types of batch capabilities. The first batch option is a
V-2
-------
standard RJE capability with the only difference from the SCOPE 3.4 and
CYBER 76 services being that NOS assigns time-sharing tasks a higher
priority than remote batch tasks. The second NOS batch option allows
interactive users to mix either single batch commands or groups of batch
commands with normal conversational commands and execute both types
during a single time-sharing session. It also allows interactive users
to create a complete batch run at the terminal and submit it to the
system to be executed completely independent of the time-sharing session
in the same way that a card deck would be prepared and submitted to the
system through an RJE terminal. Significantly reduced costs may be
realized by the use of NOS deferred batch processing which is the same
as standard RJE batch except that runs are deferred until after normal
working hours and lower rates apply. In order to take advantage of the
specific capabilities of each of the operating environments, users of
either interactive or remote batch terminals can transmit jobs to any of
the three systems through the CYBERLINK Communications Interchange.
Because NOS provides capabilities for both interactive time-sharing
and remote batch processing as well as some combined capabilities of
both, it was selected as the primary system for the byproduct marketing
programs and data bases. While some of the tasks required for previous
models were not processed on NOS (because of core requirements, storage
requirements, or processing time requirements) the added complexity that
would result for users from splitting the data bases, files, and programs
over more than one operating system is not justified on a potential
cost-saving basis. It is not possible to provide enough information
here to allow users to take advantage of all the CYBERNET network capa-
bilities and the usage sections, including examples, will be limited to
NOS. Users with the expertise to utilize the other operating systems
should have no problem adapting to the operating system or combination
of systems that will best meet their own requirements.
The text editor should provide the necessary capability for exami-
nation and modification of the independently provided files, as well as
the intermediate files generated within the system. The only exceptions
are the power plant System 2000 data base and the binary files, but
modifications to these are not necessary during normal use of the system.
The text editor will also allow examination and modification of the
procedure files and source programs where necessary without any compila-
tions or program executions required.
The mode of execution of the various programs is a function of the
system resources that will be required and their associated costs. All
of the programs can be executed in either interactive, batch, or combined
interactive-batch mode but in most cases the individual program character-
istics make one mode of execution preferable to another.
For the ADDLIME, ACDUPDT, GENACD, GENPGM, and REPTSOL programs and
the transportation manual procedure with its text editing requirements,
execution during a time-sharing session offers the greatest amount of
flexibility without significantly increasing costs. The scrubbing cost
V-3
-------
generator can also be executed in a time-sharing session if the execu-
tion is on a stand-alone basis with user-supplied power plant data.
The PROJECT, TRNCOST, and APEX programs, as well as the scrubbing
cost generator program when executed as a part of the supply subsystem
using the power plant data from the data base (after appropriate
processing by the PROJECT program) are more economically executed in a
deferred batch mode. The batch run may be created in a time-sharing
mode and submitted independently of the time-sharing session or it may
be submitted through an RJE terminal, but in general these programs are
not executed interactively on a routine basis.
The above guidelines for program execution will obviously depend on
individual requirements; there will be exceptions, such as a very small
model, a large quantity of user-supplied data to the scrubbing cost
generator program, etc., but users should make certain that they under-
stand all of the potential cost implications before deviating from the
suggested modes of execution. Those users accessing the system independ-
ently should be careful if they choose to directly access or modify the
System 2000 data base. As mentioned previously there is a potential for
significant charges to be incurred.
In most of the usage examples (both batch and interactive) the
programs are executed from relocatable program modules generated during
previous program compilations. This saves the resources required to
recompile each time the program is executed.
Before beginning any system usage, verification that the necessary
files are present can be made using the CATLIST command. An example is
shown in Table V-l. Note that the System 2000 power plant data base
consists of six different files instead of a single file. These six
files have "TA" through "TF" as the first two characters of the file
name; the last five characters are a hash code derived from the data
base name. The CATLIST command should also be used during the various
stages of system execution to verify that the appropriate files have
been created and catalogued.
Time-Sharing Execution
The text editor is the most convenient method for the examination
and modification of the data files, programs, and procedure files used
in the system. Individual fields or entire files may be edited, complete
records may be added or deleted, and new files may be created using the
text editor. Most users should find it helpful to examine some of the
actual data and program files (referring as necessary to section IV on
file and record descriptions and section III on program documentation)
before attempting any program executions. It is expected that the text
editor will also be used to create or modify any batch runs that will be
executed independently of a time-sharing session.
The first program typically executed in a time-sharing session is
the ADDLIME program. This program takes the plant data file from the
V-4
-------
I
U:
TABLE V-l. SAMPLE LISTING 0¥ THE PERMANENT HLE CATALOG TO VERIFY THAT
THE NECESSARY FILES ARE PRESENT FOR SYSTEM USAGE AND A SAMPLE
LISTING OF THE CATALOG AFTER ALL PROGRAMS HAVE BEEN EXECUTED
/CATLIST
CATALOG OF USERNUM
FILE NAMES(S)
78/06/01. 15.14.27
TA2SHUO
PRJLGO
SASDAT6
X313SUL
TRNCOST
BYP
TB2SHUO
ADDLIME
SCRPRC
GENACD
TRNLGO
BPRJ
TC2SHUO
LIMEEST
SCRSIT
GENALGO
GENPGM
BSCR
TD2SHUO
STMCAP
ACDUPDT
ACDPAR
GENPLGO
BTRN
TE2SHUO
SCRLGO
RAILUA
RBNSORT
REPTSOL
BLPRPG
TF2SHUO
BSCROPT
SACDSML
TRNPTS
RPGLGO
PROJECT
ISCROPT
SULTER
X313H2S
RAILCAL
40 FILE(S)
/CATLIST
CATALOG OF USERNUM
FILE NAMEf(S)
78/06/08. 12.15.02
TA2SHUO
PRJLGO
SASDAT6
X313SUJ.
TRNCOST
BYP
PR J OUT
SCR30
TRNCST
MODLOUT
TB2SHUO
ADDLIME
SCRPRC
GENACD
TRNLGO
BPRJ
UPLIME
SCR 40
TRNOUT
FINRPT
TC2SHUO
LIMEEST
SCRSIT
GENALGO
GENPGM
BSCR
ADLOUT
ACmJOUT
TRNDAY
BYPDAY
TD2SHUO
STMCAP
ACDUPDT
ACDPAR
GENPLGO
BTRN
SCRCST
ACDSML
GENDATA
TE2SHUO
SCRLGO
RAILUA
RBNSORT
REPTSOL
BLPRPG
SCROUT
ACDRPT
GENREPT
TF2SHUO
BSCROPT
SACDSML
TRNPTS
RPGLGO
PLAS
SCR DAY
ACDCST
MODLIN
PROJECT
ISCROPT
SULTER
X313H2S
RAILCAL
BLAS
SCR20
SMLCST
APXRPT
66 FILE(S)
-------
PROJECT program and adds the delivered cost of limestone to each power
plant record (based on the FPC number) from the independent LIMEEST
file. The UPLIME file is output from the program. This program execu-
tion is necessary only if new plant and boiler data files are created by
execution of the PROJECT program (see section III). Sample usage of the
program in a time-sharing session is shown in Tables V-2 and V-3.
The second program typically executed in a time-sharing session is
the stand-alone application of the scrubbing cost generator to make
calculations based on user-supplied power plant, boiler, and regulation
data. As shown in the program description in section III, the invest-
ment and operating factors file (SASDAT6); the cost data (SCRPRC); and
the file of options, overrides, plant data, and boiler data are required
for program execution. The output is dependent on the options and
overrides used. Sample usage of the program in a time-sharing session
is shown in Tables V-4 and V-5. Table V-6 shows a sample input file for
interactive use containing options, overrides, and user data.
The ACDUPDT program is typically executed in a time-sharing session.
As shown in section III, the sulfur tariff data file (X313SUL), the rail'
mileage file (RAILWA), the acid plant and smelter file (SACDSML), and
the sulfur terminal file (SUITER) are required. Output is a report of
delivered sulfur costs and an acid plant and smelter file (ACDSML) con-
taining delivered costs of sulfur. Sample program usage is shown in
Tables V-7 and V-8.
The GENACD program is typically executed in a time-sharing session.
The data for avoidable production cost generation (ACDPAR), the acid
plant and smelter file with updated costs (ACDSML), the file names to be
used, and the report selection options are required during execution.
Output is an acid cost file, a smelter cost file, and selected reports.
Sample program usage is shown in Tables V-9 and V-10.
The transportation manual procedure to provide transportation
locations and the associated data in the TRNPTS file is typically done
in a time-sharing session using the text editor. The output file from
the scrubbing cost generator (SCRCST), the output files from the avoid-
able production cost generator (ACDCST and SMLCST), the transshipment
terminal data in GENPGM and the external file of rail rate basing point
data (RBNSORT) must be referred to for the data. As new power plants
acid plants, smelters, sulfur terminals, or transshipment terminals are
considered, the appropriate records must be added to the TRNPTS file in
correct sequence by means of the text editor. A sample time-sharing
session to add a power plant location-related record to the TRNPTS file
is shown in Table V-ll.
GENPGM, the first program in the linear programing model subsystem
is typically executed in a time-sharing session even though the subsequent
solution of the model by the APEX program is done in batch mode. The
output files from the other subsystems must be provided as input and
user options, including the ACFL, must be provided during program execu-
tion. Once the model is created the only remaining option is to either
V-6
-------
TABLE V-2. SAMPLE PROCEDURE FILE TO INTERACTIVELY EXECUTE THE
PROGRAM THAT ADDS DELIVERED LIMESTONE COSTS (ADDLIME)
01740
01760 MESSAGE. EXECUTE ADD LIMESTONE COSTS (ADDLIME)
01780 GET»ADIiLIME.
01800 GET»TAPE1=BPLAS.
01820 GETfTAPE2=LIMEEST.
01840 PURGE* T.ADLOUT/NA.
01860 DEFINE* IADL.OUT.
01880 FTN(I=ADDLIME»L=0>
01900 LGOrrlADLOUT.
01920 REPLACEtTAPE3=IUPLIME.
01940 RETliRN»AniiLIME»M3U»TAPEl»TAPE2.
01960 MESSAGE. AM i. IME COMPLETE.
01980 RETURNrMESSAGE,
02000 GOTO,9END.
02020 EXIT.
02040 MESSAGE. ADD LIME FAILED* DAYFILE TO FOLLOW.
020AO RETURNfMESSAGE.
02080 GOTO 9EXIT.
f
—i .
TABLE V-3. SAIIPLE USAGE OF A PROCEDURE FILE TO INTERACTIVELY
EXECUTE THE ADDLIME PROGRAM
/OLDfPROCFIL
/-PROCFIL»S=1ADDLIM.
10.24.30.MESSAGE. EXECUTE ADD LIMESTONE COSTS (ADDLIME)
10.24.37.MESSAGE. Ann LIME COMPLETE.
9ENDfEXIT.
-------
TABLE V-4. SAMPLE PROCEDURE FILE TO INTERACTIVELY EXECUTE
THE SCRUBBING COST GENERATOR PROGRAM (STMCAP)
02100 1ISCR,OLD»MESSAGE/UN=LIBRARY.
02120 MESSAGE. EXECUTE SCRUBBING COST GENERATOR (STMCAP)
02140 GET,TAPE1=SASDAT6.
02160 GET,TAPE4=SCRPRC.
O218O RETURN,TAPE2,TAPES.
02200 GETfSCRLGO.
02220 GETrINPUT=ISCROPT.
02222 PURGE,ISCROUT/NA.
02224 RETURN,ISCROUT.
02226 DEFINErISCROUT.
02240 SCRLGO,,ISCROUT.
02260 RETURN,TAPE2,TAPE3,INPUT.
02280 IFCFILE(TAPE10,AS))REPLACE,TAPE10=ISCRCST.
02300 IF(FILE(TAPE20,AS>>GOTO,1T20.
02320 1T30A,IFGOTO,1T3O.
02340 1T40A,IF
-------
TABLE V-5. SAMPLE USAGE OF A PROCEDURE FILE TO
INTERACTIVELY EXECUTE THE STMCAP PROGRAM
/-PROOFILrS=lISCR,
12.43,17.MESSAGE, EXECUTE SCRUBBING COST GENERATOR (STMCAP)
12.43.32,MESSAGE, TAPE20 CREATED AND SAVED AS ISCR20.
12.43,35.MESSAGE, TAPE30 CREATED AND SAVED AS ISCR30,
12,43,40.MESSAGE, TAPE40 CREATED AND SAVED AS ISCR40.
12,43,43,MESSAGE, SCRUBBING COST GENERATION COMPLETE.
< 9END.EXIT,
VO
-------
TABLE V-6. SAMPLE SCRUBBING COST GENERATOR (STMCAP) INPUT OPTIONS,
OVERRIDES, AND USER DATA FOR INTERACTIVE EXECUTION
$OPTIONS KSAS=5,LSAS=0»KPRICE=5»KOFF=0»KSIM=1»IN=5,IO=40»
KSCAN=30r KEMISS=30rKCOSTLP=20 rKCQSTP*6rLCP=Or KCHECK=1r
KEHIT=--0*
fOVERIDE OPYEAR=197B.rPREMIS=l.2»l*2rl.3rl.2rl.2rl.2i>
CEINDX=214«7rGRACED.1*
*PRICIN PRICE(2)=.lE-f02*
*SASIN IU=6»ISEQU=10fJU=1»LU=1»KU=1»AU=419000.*A1U=.650rA2U=0.*
*SASIN IU=44»ISEQU=0»JU=0»KU=0»LU=0*
*PLANTIN IFPC=9990009910FBTUCOL=10535*8rBTUOIL = 137588*5»BTUGAS
SULCOL=.037 »SULOIL=.002iHTRATE=9860.0128537»ISPLC=436270»
ISI PI=2»
PEDCO=0.0 f VLIME=442.tSITEFAC=1*0»1.0»1.0rl*Orl«0»ltO*
SAMPLE PLANT A
*BLRIN IFPCB=9990009910fTOTCOLI=3089,27324»TOTOILI=0*0»TOTGASI=0«0»
AIR1=2234600.»STRTVRI=1972fCAPFPC=.5797»GENCAP=1300.0»IB=lr
ISIP2=07»R£GC=500.0TREGO=4.00»RETRO=1.14»IPROC=1*
$BLRIN IFPCB=9990009910»TOTCOLI=3010,732455»TOTOILI=0,OfTOTGASI=0.0>
AIRI=2234600.,STRTYRI=1973»CAPFPC=.565>GENCAP=1300.»IB=2»
ISIP2=03»REGC=4.0»REGO=4.0,RETRO=1.14fIPROC=4*
$BLRIN IFPCB=00000*
-------
TKKLR M-l. SHffilA TOC^m Y11E TO 1TO1ACTIWW TOCTO TH£ ?HOGMM. THkT CflLCOIATES
DELIVERED MOLTEN SULFUR COSTS TO ALL SULFURIC ACID PLANTS (ACDUPDT)
03340 lIACDUPFOLIiFMESSAGE;:/UN=:LIBRARY.
03360 MESSAGE, EXECUTING ACID UPDATE PROGRAM
O338O GET r ACDUPDT.
03400 PURGE: FAcnuouT/NA,
03420 DEFINEs-ACDUQUT,
03440 FTN(I=ACDUPDT »L=0)
0 3460 1. GO F F ACDUOU T,
03480 REPLACE r TAPE8-IACDSML..
03500 RETURNFLGOvACDUPDT.
03520 MESSAGE:. ACID UPDATE. COMPLETE.
0354O RETURNFMESSAGE.
03560 GOT0.9END,
03580 EXIT.
03600 MESSAGE. ACID UPDATE FAILED-
03620 MESSAGE. DAYi-IL.E TO FOLLOW.
03640 RETURNFMESSAGE,
03660 GOTC)»9E.::X.TT.
TABLE V-3. SAMPLE USAGE OF A PROCEDURE FILE TO
INTERACTIVELY EXECUTE THE ACDUPDT PROGRAM
OLD»PROOFIL.
READY
-PROCFILyS^lIACDIJP.
13.55.49.MESSAGE.
16.00.10.MESSAGE.
9END»EXIT,
EXECUTING ACID UPDATE PROGRAM
ACID UPDATE COMPLETE.
-------
TABLE V-9. SAMPLE PROCEDURE FILE TO INTERACTIVELY EXECUTE
THE PROGRAM THAT CALCULATES SULFURIC ACID
AVOIDABLE PRODUCTION COSTS (GENACD)
03680 lIACri.r)Ui.MESSAGE/UN=LIBRARY.
O37OO MESSAGE. EXECUTE ACID PRODUCTION COST( GENACD)
03720 PURGEr lACf'RPT/NA.
03740 PURGFrlACPCST/NA.
O37AO PURGE .ISMLCST/NA.
O3780 CSETrGENALGP.
O3800 GENALGO.
O3820 RETUKN.liENALGO.
03R40 SAVE. TAPE7^ (f ML CST.
03980 SAVE* rAPF3=IACDRPT.
O3900 MESSAGE:. ACID PRODUCTION COST
O3920 ME:SSAGF. R[:FCIRT FILE IS IACDRPT.
O3940 RETURN .MKSSAGE.
O3960 ^OT0.9t:ND.
O3V80 IrXIT.
04OOO MESSAGE. ACID COST FAILEDr DAYFIIF TO FOLLOW.
O4020 RFTURN.HFSSAGt .
O4O4O GOTO,9t:Xi:T.
TABLE V-10. SAMPLE USAGE OF A PROCEDURE FILE TO
INTERACTIVELY EXECUTE THE GENACD PROGRAM
/Ol.n.pRQCFII.
/-F-ROCFU ,S--) IACn.
14,13, 19. MtSSAGK. FXICIMt. AClU PPODUCrtON COSTf GFNACIO
EHTfR AC'ID PARAME1FF FILE NArtf
v SACT'PAR
ENTf-.R ACIfi PLANT FILE
ENTER EA?TERNi 1 • OR UF:STI- PN( ?) RltGIDiJ CODE
' 1
fS '.PKTIAL REPORT nKSLRKP (YFS OP NO >
'•• YFS
ENTER SPECIAL RFPUKT *(t-3 OR 8 jS>-=ALI. •O=f.-FPUF.i f NAMf?)
•f 9
14.16.06.MESSAC-E:, ACID PKOrHIC TION COS I
14.14.17.MESSAGE, REPORT FILE IS rrtCUK
''ENDr EXIT.
-------
TO
TO THE TRANSPORTATION FILE (TRNPTS)
i
M
u>
OLD. TRNPTS
/XEDIT
XEDIT 3,0,6
?? N90
000
?? P 10
000
10400002003422.1.20
10400003003414740
14000006002181110
17900025504569400
1.7900028004538970
35450007003472150
38000008002041380
45100001004729440
47400003004977950
?? N -7
10400003003414740
?? I 1
? 13950002504131830
?? P 5
13950002504131830
14000006002181110
17900025504569400
17900028004538970
35450007003472150
341800155502
153202
218364146102
456940240701
455370240601
153602
204210151102
472872260801
498464251601
153202
413183203901
41.3183203901
21.8364146102
456940240701
455370240601
153602
TRNPTS REPLACED
TRNPTS IS A LOCAL F-ILE
AEB r 0.428UNTS.
-------
solve it or discard it; no capability is provided for modifications
after it has been built. In addition to the model a data file to be
used in a final report is also created and must be sorted if the model
is to be solved and a report prepared of the results. The GENSORT
procedure to do this uses the ZORT control statement and is included in
the procedure for program GENPGM. Sample usage of program GENPGM and
the GENSORT procedure (ZORT) are shown in Tables V-12 and V-13.
Batch Execution
"Batch" as used in this section refers to the execution of an
independent run stream just as though it was submitted through an RJE
terminal. This same type of run can also be created and submitted
during a time-sharing session but it is executed completely independently
of the time-sharing session once it has been submitted to the system.
The use of the word batch in the above context should not be confused
with the CYBERNET NOS definition in connection with the execution of
batch commands during a time-sharing session. Usage of the batch sub-
system during a time-sharing session to execute one or more batch commands,
as opposed to submitting a batch run to be executed independently of the
time-sharing session, is a CYBERNET NOS option not related to a typical
RJE batch run.
In the batch execution examples that will be shown, the runs were
generated specifically for this manual and were limited for cost purposes.
Because of their simplicity they were submitted as normal batch (priority
P4) rather than deferred batch (priority P2). This results in a single
interactive terminal session where a run is submitted; the status is
checked to verify that the run is in the input queue; the status is
checked again after a short time within the same terminal session; the
run is found to have executed; and the DAYFILE is checked for correct
execution. Use of normal batch does not offer the cost reductions of
deferred batch. For actual system runs, batch jobs (runs) will normally
be submitted as deferred batch in a terminal session on any given day,
the job will be run overnight, and a status check and verification of
results will be made on the following day. This specific situation will
be shown in the figures related to the use of the APEX package to solve
the model and the generation of a report of the solution.
The PROJECT program is the first program in the supply subsystem
and is typically executed in batch mode. A run stream to execute the
program is prepared in a time-sharing session and submitted to the
system. The power plant data base is required to execute the program.
The output from the batch run can be saved for examination in a later
time-sharing session, it can be printed at an RJE site, or it can be
printed onsite and sent to the user. All examples shown save the results
for later examination in a time-sharing session when final disposition
can be determined. A sample job stream to execute the power plant
projection program in batch mode is shown in Table V-14.
V-14
-------
TABLE V-12. SAMPLE PROCEDURE FILE TO INTERACTIVELY EXECUTE THE PROGRAM
THAT BUILDS THE LINEAR PROGRAMING MARKETING MODEL (GENPGM)
04060 1IGENLP,OLD,MESSAGE/UN=LIBRARY,
04080 MESSAGE. EXECUTE MODEL GENERATION PROGRAM (GENPGM)
04100 PURGErMODLIN/NA.
04120 DEFINE,TAPE4=MODLIN.
04140 GETfGENPLGQ.
O4160 GENPLGO.
04180 RETURNvGENPLGO.
04200 4CONT,REWIND*TAPES.
04220 RENAMEyGENDATA=TAPE8.
04240 RFLr50000.
04260 ZORT('GENDATA,TAPE8»110..Z,ArC,IF 10)
04280 PURGErGENREPT/NA,
04300 SAVE,TAPE8=GENREPT.
04320 RETURNyTAPEX.
04340 MESSAGE. MODEL GENERATION COMPLETE
04360 MESSAGE. SUBMIT MODEL ANALYSIS AND REPORT GENERATION
04380 MESSAGE. RUN TO NOS BATCH USING BLPRPG FILE.
04400 RETURNfMESSAGE.
04420 GOTOf9END.
04440 EXIT.
04460 4EXITfMESSAGE. (GENPGM) FAILED TO EXECUTE.
04480 MESSAGE. DAYFILE TO FOLLOW
04500 RETURN»MESSAGE.
04520 GOTO*9EXIT.
-------
TABLE V--13. SAMPLE USAGE OF A PROCEDURE FILE TO INTERACTIVELY
EXECUTE THE GENPGM PROGRAM
/OLD»PROCFIL
/-PROCFIL,S=1GENLP.
13.26.23.MESSAGE. EXECUTE MODEL GENERATION PROGRAM (GENPGM)
•GIVE ACID PLANT COST FILE NAME'
? ACDCST
NUMBER OF ACID PLANTS IN MODEL IS 86
TOTAL DEMAND (1000) TONS IS 30992.000
GIVE TRANSPORTATION COST FILE NAME
? TRNCST
• GIVE SMELTER (EAST) COST FILE NAME1
? SMLCST
NUMBER OF EASTER SMELTERS IS 13
TOTAL. SMELTER CAPACITY (1000) TONS IS 419.000
TOTAL WESTERN SMELTER CAPACITY IS 738.000
f TOTAL CANADIAN IMPORT CAPACITY IS 200.000
£ SHOULD THE SYSTEM SELECT THE OPTIMUM
(LEAST COST) STRATEGY? (1=YES)
? 1
1
GIVE NAME OF SCRUBBING COST FILE
? SCRCST
ENTER CLEAN FUEL COST ALTERNATIVE (CENTS/MBTU)
? .35
ERROR - NEGATIVE INCREMENTAL COST 1395000250
ENTER TRANSPORT INFLATION FACTOR FOR 1983
? 2.16
ZERO MILES - SMELTER 9991490221 ACID PLANT 138
ZERO MILES - SEMLTER 9986848007 ACID PLANT 88
ZERO MILES - SMELTER 9986848007 ACID PLANT 88
ZERO MILES - SMELTER 9986848007 ACID PLANT 88
13,27.39.MESSAGE* MODEL GENERATION COMPLETE
13.27.40.MESSAGE* SUBMIT MODEL ANALYSIS AND REPORT GENERATION
13.27.43.MESSAGE* RUN TO NOS BATCH USING BLPRPG FILE.
9ENDrEXIT.
-------
M-\A,
301
TO
IKE SOWEB. P.UNT
PROJECTION PROGRAM IN BATCH MODE (PROJECT)
I
(—•
~J
00100 /JOB
00110 JOB»P4fT1000.
00120 ACCOUNTrUSERNUMrPASSUORrKG.
00130 CHARGErCHAFGNO**SFX*PROJECTNQ.
00140 PURGE,BBLAS/NA.
00150 GET»PRJLGO.
00160 SBULIM
00170 PRJLGO,
00180 REPLACE»TAPE9=BPLAS.
00190 REWINDrTAPElO,
00200 DEFINErBBLAS.
00210 COPYfTAPElOtBBLAS.
00220 PURGEfBPRJOUT/NA,
00230 DEFINErBPRJOUT.
00235 ENQUIRErR.
00240 REWINDrOUTPUT.
00250 COPYyOUTPUT»BPRJOUT.
00260 PURGE»BPRJDAY/NA.
00265 DEFINE»BPRJDAY«
00270 DAYFILE»BPRJDAY.
00280 EXIT.
00290 PURGEfBPRJOUT/NA.
00300 DEFINEDBPRJOUT.
00305 ENQUIRErR»
00310 REWINDfOUTPUT.
00320 COPY» OUTPUT»"BPRJOINT,
00330 PURGE»BPRJDAY/NA*
00340 DEFINErBPRJIiAY.
00350 DAYFILE»BPRJDAY,
00360 /EOR
00370 ALL
00500 /EOF
-------
A sample run submitted using a procedure file, checking the job
status after submittal, and checking of results after execution is
complete is shewn in Table V-15.
The scrubbing cost generator program (STMCAP), when executed as a
part of the supply subsystem using the files created by the PROJECT and
ADDLIME programs, is typically executed in batch mode as shown in Tables
V-16 and V-17. Based on the program description in section III, the
file of investment and operating factors (SASDAT6), the file of cost
data (SCRPRC), the plant-level data file with delivered costs of lime-
stone (UPLIME), the boiler-level data file (BLAS), a file of site-
specific adjustment factors (SCRSIT, optional), and the options and
overrides to be used are required for program execution. (If the
ADDLIME program is not executed, a default delivered cost of limestone
is assumed equally for all plants and the PLAS file can be used.) The
batch run is typically submitted during a time-sharing session and,
depending on the volume, the output is either saved for examination in a
later time-sharing session or is printed at an RJE site.
The TRNCOST program is also typically executed in batch mode. The
TRNPTS file from the manual transportation procedure, the rail mileage
file (RAILWA), and the sulfuric acid tariff table file (X313H2S) are
required. An output report and a transportation cost file are created
during program execution. Sample usage is shown in Tables V-18 and
V-19.
The linear programing marketing model is typically solved by APEX
and a report of the solution is generated by the GENPGM program in batch
mode. The report generation step could be done in a time-sharing
session from a cost basis, but because the model is solved in batch mode
and a report is almost always required, there is no point in separating
the two functions. The run is generated in a time-sharing session and
submitted to the system to be executed in deferred batch mode. Sample
usage is shown in Tables V-20 through V-22.
V-18
-------
TABLE V-15. EXAMPLE 0¥ SUBMITTING A BATCH RUTS TO EXECUTE THE PROJECT PROGRAM;
CHECKING THE JOB STATUS; AND CHECKING FOR CORRECT EXECUTION
OLD»BPROJ
/SUBMITtBPROJfN.
18.56.58,ASIOBNA
/ENQUIRE*JN=BNA.
ASIOBNA IN INPUT QUEUE.
/ENQUIRE*JN=BNA,
ASIOBNA NOT FOUND.
/ATTACH,BPRJDAY.
/LNH»FN=BPRJDAY,
18.59.58.JOBrP4fTJ.000.
18.59.58.ACCOUNT »USERNUM*,KC.
18.59.58.ABC » P4«
18.59.59.CHARGE* CHARG**» *SFX*PROJNO.
18.59.59.PURGErBBLAS/NA.
19.00.00.GETiPRJLGO.
19.00.00.SBULIM
-------
TABLE V-16. SAMPLE JOB STREAM TO EXECUTE THE SCRUBBING
COST GENERATOR PROGRAM IN BATCH MODE (STMCAP)
00100 /JOB
00110 JOB,P4rT3000.
00120 ACCOUNT »USERNUM»PASSUOR>KC.
00130 CHARGEfCHARGNOr*SFX*PROJECTNO.
00140 PURGE,BSCR20/NA.
00150 PURGE,BSCR30/NA.
00160 PURGE,BSCR40/NA.
00165 PURGEjBSCRCST/NA.
00170 DEI-INE>BSCR20.
00180 DEFINE*BSCR30.
00190 DEFINE.BSCR40.
00200 GETf1NPUT=BSCROPT.
00210 GET,TAPE1=SASCAT6.
00220 GET»TAPE4=SCRPRC.
00230 GETfSCRLGO,
00240 GET»TAPE8=IUPLT.ME.
00250 ATTACH,TAPE9=BBLAS/M=R.
00260 GET.TAPE7=SCRS1T.
00270 SBULIM
-------
TABLE V-17. EXAMPLE OF SUBMITTING A BATCH RUN TO
EXECUTE THE STMCAP PROGRAM: CHECKING THE JOB
STATUS; ANC CHECKING FOR CORRECT EXECUTION
OLDfBSCR.
10.25.5V.ASJOKSG
/ENQUIRE, JN=BSG,
ftSIOBSG IN INPUT QUEUE.
/ENQUIRE* JN=EfSC5.
ASIOBSG NOT FOUND.
/ATTACHrBSCRDAY.
/LNHjFN=BSCRI)AY.
10.27.08.JOBrF-4,T3000.
10.27, O9. ACCOUNT rUSEKMM t rKC.
10.27.09.ABB > F4.
10.27.09.CHARGE,CHAF.:G**,*SFX*PROJECTNO.
10.27. 09. PURGEfBSCRSO/NA.
10.27.09. BSCR20 NOT FOUNB» AT OOO121,
10.27. 10.PURGE»BSCR30/NA.
< 10.27.10. BBCR30 NOT FOUNI'r AT 0001.71,
I 10.27.10.PURGEr*SCR40/NA.
P 10.27,10. BSCR40 WOT FOUND r f>T 000121.
10.27.J 0.PURGE* BSCRCST/NA.
10.27. 10,tiEFINErBE;CR20.
10,27,1O.DEFINF»BSCP30,
10.27,10.DEFINEjBSCR40,
10 .27. 1.1 .GET »1NFHIT--HSCROPT .
1O.27. 11. .6ET.TAF'El=BASnftT6.
1O.27.ll.GET»TAPE4=SCRKftC,
10.27.11.GET,SCftLKO.
10.27,12.GETfTAPEB=IUPL.lME.
10.27, 13. ATTACH, TAPE9=BBLftS/f1=R.
10.27.13. BPLAS NOT FDUNTi, AT 000121-
10.27.14.EXIT.
10.27.14.f>URGErBSCROUT/NA.
10.27.14. BSCROLIT NOT FOUNIif AT 000121.
10.27.14. V\-Fl ME , BSCRCHJT .
10.27.t^.ENOUIREtR.
10.27.14. ENQUIRY COMPLETE.
10,27.14.REWIND^ OUTPUT.
10.27.14.COPY» OUTPUT tBSCRQUT.
1O.27.15. ENIJ OF INFORMATION E'NCOUNTFRF:!:!
1O.27. 15.PURGEfBSCRi:iAY/NA.
10.27. is. I:IE:F i NE , BSCK f.>A Y ,
10 .27. 1'J,. DAYF1LE . BSCPDAY .
-------
TABLE V-18. SAMPLE JOB STREAM TO EXECUTE THE TRANSPORTATION
COST GENERATOR PROGRAM IN BATCH MODE (TRNCOST)
i
NJ
ro
00000 /JOB
00100 JOB»P4rT1000.
00200 ACCOUNT »USERNUM fPASSWOR fKC.
00300 CHARGE*CHARGNOf*SFX*PROJNO.
00400 PURGErBTRNCST/NA,
00500 DEFINErTAPE4=BTRNCST.
00600 ATTACHrPSDLIB/UN=PSDRLIB.
00700 GETfTRNLGO.
00800 SBULIM(JS=500)
00900 LDSET(LIB=PSDLIB)
01000 TRNLGO.
01100 PURGEfBTRNOUT/NA.
01200 DEFINEDBTRNOUT.
01300 ENQUIREtR.
01400 REWIND^OUTPUT,
01500 COPY»OUTPUT»BTRNOUT.
01600 PURGEfBTRNDAY/NA.
01700 DEFINErBTRNDAY.
01800 DAYFlLEfBTRNDAY.
01900 EXIT.
02000 PURGE»BTRNOUT/NA.
02100 DEFINE»BTRNOUT.
02200 ENQUIRE»R*
02300 REWIND»OUTPUT.
02400 COPY»OUTPUTfBTRNOUT.
02500 PURGEfBTRNDAY/NA.
02600 DEFINEfBTRNDAY,
02700 DAYFlLEfBTRNDAY.
02800 /EOR
02900 X313H2S
03000 ITRNPTS
03100 /EOF
-------
TABLE V-19. EXAMPLE OF SUBMITTING A BATCH RUN TO EXECUTE THE TRNCOST PROGRAM;
CHECKING THE JOB STATUS; AND CHECKING FOR CORRECT EXECUTION
OLD»BTRN
/SUBMIT,BTRNrN.
13,25.17.ASrQBNA
/ENQUIRE,JN=BNA.
ASIOBNA IN INPUT QUEUE.
/ENQUIRE?JN=BNA.
ASIOBNA NOT FOUND.
/ATTACHfBTRNDAY.
/LNH»FN=BTRNDAY.
13.30.05,JOB»P4rT1000.
13.30.05.ACCOUNT,USERNUM•r KC.
13.30.05.ABG r P4.
13.30.06,CHARGErCHARG**r*SFX*PROJNO.
13.30.06.PURGEfBTRNCST/NA.
13.30.06.DEFINEr TAPE4=BTRNCS T.
13.30.06.ATTACHfPSDLIB/UN=PSDRLIB.
13,30.07.GETrTRNLGO«
13,30*07.SBULIM(JS=500)
13.30.11.LDSET(LIB=PSDLIB)
13.30.11.TRNLGO.
13.36.01. STOP
13.36.01,PURGErBTRNOUT/NA,
13.36.01,DEFINErBTRNOUT.
13.36.01.ENQUIREfR.
13.36.01. ENQUIRY COMPLETE.
13.36.01.REWINDrOUTPUT.
13,36.01.COPYf OUTPUT»BTRNOUT.
13.36.02. END OF INFORMATION ENCOUNTERED.
13.36.02.PURGE»BTRNDAY/NA.
13.36.02.DEFINE»BTRNDAY.
13.36.02.DAYFILErBTRNDAY.
-------
TABLE V-20. SAMPLE JOB STREAM TO EXECUTE THE APEX LINEAR PROGRAMING PACKAGE
AND THE REPORT GENERATOR PROGRAM (REPTSOL) IN DEFERRED BATCH MODE
(OVERNIGHT) TO GENERATE AN EQUILIBRIUM MODEL SOLUTION
AND A REPORT OF THE SOLUTION
00100 /JOB
00110 JOB,P2,T1500»CM101000.
00120 ACCOUNTiUSERNUM.PASSWOR,NC,
00130 CHARGE*CHARGNO,*SFX*PROJECTNO.
00135 SBULIMCJS=1500>
0014O ATTACH,APEX/UN=LIBRARY.
00150 ATTACH,TAPE1=MODLIN.
00160 PURGE»APXRPT/NA.
O0170 PURGE,MODLOUT/NA.
00180 DEFINE,APXRPT.
00190 DEFINE,HODLQUT.
00200 REDUCE(-)
00210 APEX
00215 RETURN,MODLOUT,
00220 GET,RPGLGO.
00230 PURGE.BFINRPT/NA,
00240 DEFINE,TAPE4=BFINRPT.
00250 RFL»101000.
00240 RPGLGO.
00270 PURGE.BLFDAr/NA.
00280 PURGE.BLPOUT/NA.
OO29O DEFINE,BLPOUT.
00300 ENQUIRE,R.
00310 REWIND,OUTPUT.
00320 COPYrOUTPUT,BLPOUT.
00330 riEFINEfBLPDAr.
00340 i:iAYFILE,BLPDAY.
00350 EXIT.
00360 PURGE,BLPDAY/NA.
00370 DEFINE,BLPDAY.
00380 DAYFILE,BLPDAY.
00390 PURGE,BLPOUT/NA.
00400 DEFINE,BLPOUT.
00410 ENQUIRE,R.
00-42O REWIND, OUTPUT.
00430 COPY,OUTPUTjBLPOUT.
00440 /EOR
00450 MODLOUT
00460 (9-29-78 $.50 ACFL)
00470 IACDCST
00480 ISMLCST
00490 GENREPT
00500 /EOF
-------
TABLE V-21. EXAMPLE OF SUBMITTING A DEFERRED BATCH RUN TO EXECUTE THE APEX PACKAGE AND
REPTSOL PROGRAM TO SOLVE THE MODEL, GENERATE AN EQUILIBRIUM SOLUTION,
AND PREPARE A REPORT OF THE SOLUTION
//
NET 051037
PLEASE SIGN ON—KCrUSERNUM
78/05/18, 12.41.32,
EASTERN CYBERNET CENTER SN147 NOS 1,0/411,477.0-0
PASSWORD
XXXXXXXX
TERMINAL: 305rTTY
f RECOVER/ CHARGEJ CHARGE»CHARGNOr*SFX*PROJECTNO
K READY.
BATCH
*RFL*20000.
/OLDfBLPRPG
/SUBMITrBLPRPGfN.
12.42.10.ASIOBNZ
/ENQUIRE,JN=BNZ.
ASIOBNZ IN INPUT QUEUE
/BYE
USERNUM LOG OFF 12.43.19.
SBU 0.667
TIO = 1779
-------
TABLE V-22. CHECKING THE STATUS OF THE RUN SHOWN IN FIGURE V-19 AND
SUBMITTED IN FIGURE V-20; THEN VERIFYING THE RESULTS
//
NET 051037
PLEASE SIGN ON—KCrUSERNUH
78/05/19. 09.41.32
EASTERN CYBERNET CENTER SN147 NOS 1.0/411.477.0-0
PASSWORD
XXXXXXXX
TERMINAL: 206»TTY
RECOVER/ CHARGE! CHARGErCHARGNO,*SFX*PROJECTNO
READY
BATCH
*RFLr200OO
/ENQUIREfJN=BNZ.
ASIOBNZ NOT FOUND.
/ATTACH,BLPDAY,
/LNH,FN=BLPDAY.
19.01.35.JOB,P2,T1500,CM101000.
19.01.36.ACCOUNT rUSERNUM,» KC.
19.01.36.ABG , P2.
19.01.36.CHARGErCHARG**,*SFX*PROJECTNO.
19.01.36.SBULIM
19.01.36.ATTACH,APEX/UN=LIBRARY.
19.01.37.ATTACH,TAPE1=MOBLIN.
19.01.37.PURGE.APXRPT/NA.
19.01.37.PURGE.MODLOUT/NA.
19.01.37. MOCLOUT NOT FOUND» AT 000121.
19.01.37.i:iEFINE,APXRPT.
19.01.38.DEFINE,MODLOUT.
19.01.38.REriUCE(->
19.01.38.APEXBLPHAY/NA.
19.57.38.PURGE.BLPOUT/NA.
19.57.38.DEFINErBLPOUT.
19.57,39.ENQUIRE.R,
19.57.40.REWIND,OUTPUT.
19.57.40.COPYrOUTPUTfBLPOUT.
19.57.41.DEFINE,BLPrJAY.
19.57.41.nAYFILErBLPDAY.
-------
VI. REFERENCES
1. Waitzman, D. A., J. L. Nevins, and G. A. Slappey. Marketing H2SO<.
from S02 Abatement Sources - The TVA Hypothesis. Bulletin Y-71,
Tennessee Valley Authority, Office of Agricultural and Chemical
Development, Muscle Shoals, Alabama. EPA-650/2-73-051, U.S.
Environmental Protection Agency, Office of Research and Development,
Washington, D.C., 1973. (NTIS PB 231 671)
2. Bucy, J. I., R. L. Torstrick, W. L. Anders, J. L. Nevins, and P. A.
Corrigan. Potential Abatement Production and Marketing of Byproduct
Sulfuric Acid in the U.S. Bulletin Y-122, Tennessee Valley Authority,
Office of Agricultural and Chemical Development, Muscle Shoals,
Alabama. EPA-600/7-78-070, U.S. Environmental Protection Agency,
Office of Research and Development, Washington, D.C., 1978.
(NTIS PB 284 200)
3. O'Brien, W. E., and W. L. Anders. Potential Production and
Marketing of FGD Byproduct Sulfur and Sulfuric Acid in the U.S. -
1983 Projection. ECDP B-l, Tennessee Valley Authority, Office of
Power, Emission Control Development Projects, Muscle Shoals, Alabama.
In press.
4. McGlamery, G. G., R. L. Torstrick, W. J. Broadfoot, J. P. Simpson,
L. J. Henson, S. V. Tomlinson, and J. F. Young. Detailed Cost
Estimates for Advanced Effluent Desulfurization Processes.
Bulletin Y-90, Tennessee Valley Authority, Office of Agricultural
and Chemical Development, Muscle Shoals, Alabama. EPA-600/2-75-006,
U.S. Environmental Protection Agency, Office of Research and
Development, Washington, D.C., 1975. (NTIS PB 242 541)
VI-1
-------
APPENDIX A
INTEGRATED TRANSPORTATION DATA BASE
The byproduct marketing system was designed to provide a framework
for modeling the market potential for FGD byproducts. Data bases and
programs are maintained to allow projections of both supply and demand,
including detailed estimates of the potential costs, quantities, and
locations involved. However, no matter how refined any projections of
potential supply and demand related to FGD byproducts may be, they are
of limited value without corresponding projections of transportation
costs.
A completely manual analysis of both the locations that must be
considered and the appropriate shipping rates is a complex and time-
consuming task. Specific supply sources, demand destinations, and the
shipping distances between them must be known before the mode of ship-
ment and the applicable rates can be determined. In the byproduct
marketing system, the number of materials that shipping rates are
required for is limited. However, the number of locations involved is
not so limited and is constantly changing. Continuing analyses of
market potential for those FGD byproducts that are currently modeled
require the addition of new locations that must be taken into account
along with locations previously considered. The analysis of market
potential for additional FGD byproducts, produced from other scrubbing
processes, can involve a completely different set of locations and
rates, for both the byproduct as well as any related materials.
A totally automated and generalized transportation system — including
rail, barge, and truck rates — would be a significant project in
itself and prohibitive for byproduct marketing purposes. However, by
limiting such a system to a location data base, with shipping rates
Determined separately and limited to those required for the byproduct
marketing system alone, the necessary resources are more practical and
could be very cost effective in the long run.
A transportation location data base was created using the System
2000 format (related manuals are shown in Appendix B) . A description is
ghown in Table A-l and should be referred to as necessary throughout the
remainder of this appendix. The data base contains all named locations
£n the 48 contiguous states that could be considered for shipping by
barge, or truck. Every location that could be identified from any
is included; no location was intentionally omitted no matter how
A-l
-------
TABLE A-l. TRANSPORTATION LOCATION DATA BASE
DATA BASE NAME IS USLOCAT
100* FIPS-STATE (INTEGER NUMBER 99)
110* STATE-ABBREVIATION (NAME XX)
120* STATE-NAME (NON-KEY NAME X(15»
200* COUNTY-RG (RG)
210* FIPS-COUNTY (INTEGER NUMBER 999 IN 200)
220* COUNTY-NAME (NON-KEY NAME X(30) IN 20O)
230* MEAN-LATITUDE (NON-KEY DECIMAL NUMBER 99.99 IN 200)
240* MEAN-LONGITUDE (NON-KEY DECIMAL NUMBER 999.99 IN 200)
250* SQ-MILE-AREA (NON-KEY INTEGER NUMBER 9(6) IN 200)
260* N-LATITUDE (NON-KEY DECIMAL NUMBER 99.99 IN 200)
270* S-LATITUDE (NON-KEY DECIMAL NUMBER 99.99 IN 200)
280* E-LON6ITUDE (NON-KEY DECIMAL NUMBER 999.99 IN 200)
290* U-LONGITUPE (NON-KEY DECIMAL NUMBER 999.99 IN 200)
310* FUTURE-GROUTH (NON-KEY NAME X(100> IN 200)
5OO* LOCATIONS-RG (RG IN 200)
510* LOCATION-SPLC (INTEGER NUMBER 9(6) IN 500)
520* LOCATION-NAME (NON-KEY NAME X(32) IN 500)
525* COUNTY-SEAT (NON-KEY NAME X IN 500)
530* ZIP-CODE (INTEGER NUMBER 9(5) IN 500)
540* LATITUDE (DECIMAL NUMBER 99.99 IN 500)
550* LONGITUDE (DECIMAL NUMBER 999.99 IN 500)
560* RAIL-SF'LC (INTEGER NUMBER 9(6) IN 5OO)
570* MILES-TO-RAIL (NON-KEY DECIMAL NUMBER 9999.9 IN 500)
580* BARGE-SPLC (INTEGER NUMBER 9(6) IN 500)
590* MILES-TO-BAR6E (NON-KEY DECIMAL NUMBER 9999.9 IN 500)
A10* RATE-BASIS-SPLC (INTEGER NUMBER 9(6) IN 500)
615* RATE-BASIS-NAME (NON-KEY NAME X(32) IN 500)
620* RAIL-MILEAGE-INDEX (NON-KEY DECIMAL NUMBER 99.99 IN 500)
630* RAIL-TARIFF-INDEX (NON-KEY INTEGER NUMBER 99 IN 500)
700* RAIL-STATIONS-RG (RG IN 500)
740* NRBT-ITEM-NUMBER (NAME X(10) IN 700)
750* CARRIER (NAME XXXX IN 700)
760* STATION (TEXT XC9) IN 700)
770* FOOTNOTE (NAME XXXX IN 700)
780* SUPPLEMENT (NAME XXXX IN 700)
790* CANCEL (NAME X IN 700)
800* BARGE-RG (RG IN 500)
810* BARGE-NODE (INTEGER NUMBER 9(10) IN 800)
820* MILES-TO-CAIRO (NON-KEY DECIMAL NUMBER 9999.9 IN 800)
830* MILES-TO-HRVY-LOCKS (NON-KEY DECIMAL NUMBER 9999.9 IN 800)
840* TVA-BARGE-NODE (NON-KEY INTEGER NUMBER 9(7) IN 800)
850* TENN-TOM (NON-KEY DECIMAL NUMBER 9999.9 IN 800)
-------
All locations are grouped by state and county. State information
is at the first level of the data base and county information is at the
second level. Location-specific information is at the third level in
the data base structure. A location repeating group was established for
every point that could be identified as being either a rail-related
location (stations, rate basing points), a barge-related location, or
simply a named geographic location. Although many locations were estab-
lished solely on the basis of rail or barge data, all locations are
assumed to be generally accessible by truck.
Rail and barge information is at the fourth (and lowest) level in
the data base structure. Each location that is a rail shipping point or
a barge shipping point has a corresponding rail station or barge repeating
group (component numbers 700 and 800 respectively). The information
necessary to determine rail mileages within the Docket 28300 area (see
Appendix C) is provided by component 620. Barge shipping mileages can
be determined directly from components 810, 820, and 830.
The transportation location data base contains over 26 million
characters. In addition to all of the states and counties, there are
over 100,000 truck points, over 20,000 rail stations that are associated
with the various rail carriers, and almost 2,000 barge terminals on the
navigable inland waterway system.
After the data base was created, for those locations that are not
rail or barge points, the nearest rail shipping point and the nearest
barge shipping point were to have been determined. The resulting data
could then have been used to update the original data base components
560-630 (locations that are rail or barge points would have zero for
components 570 or 590). However, original estimates of the computer
charges that would be incurred for such an update were based on a rela-
tively small data base. Test cases using the actual data base indicated
that computer charges for the update would be many times higher than
origlnally estimated, and prohibitive based on potential benefits to the
byproduct marketing system alone.
The final step of linking all locations by truck, rail, barge, or
any combination of the three could have significantly reduced the effort
necessary to determine required location information in the transportation
cost generator. For any combination of locations, an automated procedure
could have been developed to use the data base information and select
the mode (or modes) of shipment and the distance involved. The rail
rate selection procedure is already automated and a similar procedure
could have been developed for barge and truck rates.
Because the final step has still not been completed, usage of the
transportation location data base in the system has not been automated.
Requirements in the future may justify linking all truck, rail, and
targe points and automating the shipping point location procedure. The
primary reason for including the information in this appendix is that
benefits may be realized by using the data base in other systems in the
current form.
A-3
-------
APPENDIX B
RELATED CYBERNET MANUALS AND PUBLICATIONS
Publication
Publication No.
CYBERNET Interactive Service Reference Set
Vol 1, Tutorial 84000320
Vol 2, Interactive Usage 84000360
Vol 3, Comprehensive Usage 84000370
CYBERNET Services XEDIT User Information Manual 76071000
CYBERNET Services FORTRAN Extended 4 Reference Manual 84000009
CYBERNET Interactive Service Compiler Subroutine and
Function Supplement 84002000
APEX-IH Reference Manual 76070000
CYBERNET Services SYSTEM 2000
User Information Manual 76074000
System Support Manual 76074900
Procedural Language Interface (PLI) 76075000
Immediate Access 76074400
CYBERNET Services SCOPE 3.4 Reference Manual 84000021
CYBERNET Services SCOPE 2.1 Reference Manual 84000230
(This publication covers CYBER 76 Services)
CYBERNET Services CYBERLINK Communications Inter-
change Users Guide 84000120
Note: These publications are distributed by Control Data Corporation,
Minneapolis, Minnesota.
B-l
-------
APPENDIX C
RAIL MILEAGE FILE DESCRIPTION AND USAGE
TRIANGULAR MILEAGE FILE
A primary requirement in the byproduct marketing system was the
development of rail mileages that are used to determine rail rates.
These rail mileages are used as a part of avoidable acid production cost
calculations and, more importantly, in the transportation subsystem to
determine rail costs associated with shipping from power plants and
smelters to acid plants for all of the possible combinations to be
considered in the linear programing marketing model. Rail rates are
based on 2632 rate basing points that are considered as possible source
and destination locations. In many cases, a minimum mileage for rate
determination purposes is defined even when the source and destination
rate basing points are the same. This occurs because the 2632 rate
basing points in the Docket 28300 area are not specific point locations
but represent general geographic areas which can be fairly large.
The number of mileage entries required for 2632 rate basing points
considered as either sources of shipments or destinations of shipments
Is 2632 x 2632, or 6,927,424. The costs associated with a data file
containing 6,927,424 mileage entries, both on a data storage basis and a
data access basis, require careful consideration. A common requirement
±8 the comparison of the shipping costs between various source points
and all possible destination points. Therefore the number of file
accesses for required mileages must be kept to a minimum without signifi
cantly increasing the number of mileage entries in the file.
A triangular mileage table concept was used to build a file of all
possible mileage entries required. For any number of locations, n, to
be considered, a table can be constructed such that given a source
location number, i, and a destination location, j, the mileage between
the two, M(i,j), can be read directly from the table. Table C-l is an
of such a table.
However, even though Table C-l contains the mileage from any loca-
tion to all other possible locations (including itself) the number of
entries has not been minimized. Consider the mileage from location 1 to
location 2, M(l,2), and the mileage from location 2 to location 1,
jj(2,l). The distance between the points is the same whether point 1 is
the source or destination. If a requirement is established that anytime
point 1 is to be considered in a mileage table search it will always be
C-l
-------
TABLE C-l. GENERAL MILEAGE TABLE
Destination numbers
1
2
3
* • •
n - 1
n
1
M(1,D
M(2,l)
M(3,l)
• • •
M(n - 1,1)
M(n,l)
2
M(l,2)
M(2,2)
M(3,2)
• • •
M(n - 1,2)
M(n,2)
3
M(l,3)
M(2,3)
M(3,3)
• • •
M(n - 1,3)
M(n,3)
• • •
* • •
• • •
• • •
• • •
• • •
• • •
n - 1
M(l,n - 1)
M(2,n - 1)
M(3,n - 1)
• • •
M(n-l,n-l)
M(n,n - 1)
n
M(l,n)
M(2,n)
M(3,n)
• • •
M(n- l,n)
M(n,n)
s
o
u
r
c
e
n
u
m
b
e
r
s
considered the source, then all entries in column 1 except M(l,l) can be
removed. For example, M(2,l), M(3,l), . . . , M(n,l) can also be found
at M(l,2), M(l,3), . . . , M(l,n). The entry for M(l,l) cannot be
removed because a minimum mileage may be specified even though the
source and destination rate basing points are the same. In column 2 the
same principle can be applied. The first entry, M(l,2), must be kept
because M(2,l) was removed from the first column and M(l,2) is now the
only entry for points 1 and 2. The second entry, M(2,2), is required
just as M(l,l) was required. In the case of M(3,2), the same mileage
can be found at M(2,3). The. same is true for M(4,2), M(5,2), . .
MCn,2). As in columns 1 and 2, in column 3 all entries below M(3,3) can
be removed. By examining the number of entries that could be removed in
each column, a pattern can be established. In column 1, all entries in
the column except the first one, M(l,l), can be removed. If there were
originally n entries, then n - 1 entries can be removed. In column 2
all entries except M(l,2) and M(2,2) can be removed. If there were n'
entries, then n - 2 entries can be removed. In column 3 all except the
first three entries can be removed, or n - 3. In column 4, n - 4 can be
removed, etc.
In the table of rate basing points n = 2632. Therefore n - 1
entries removed from column 1 is 2632 - 1 or 2631; in column 2, n - 2 -
2632 - 2; in column 3, n - 3 = 2632 - 3; . . . ; in column n - 1, n -
(n - 1) = 2632 - 2631; and finally, in column n, n - n = 2632 - 2632 or
zero. Examination of Table C-l will show why no entries can be removed
from the last column. M(n,n) is the last entry in the last column, and
C-2
-------
all entries above it are required just as M(l,l), M(2,2), M(3,3), etc.,
were required. Table C-2 shows all possible entries removed from the
original Table C-l. The entries in Table C-2 form a triangle and are the
basis for the term, triangular mileage table.
TABLE C-2. TRIANGULAR MILEAGE TABLE
Destination numbers
S
o
u
r
c
e
n
u
m
b
e
r
s
The number of entries that remain in each column is exactly equal
to the column number, so the total number of entries is 1+2+3+4+
. . + (n - 1) + n. A formula can be used to calculate the total number
of entries required in a triangular table such as Table C-2.
Entries required
n(n + 1)
2
In the case of the rate basing point mileages, n is 2632.
Entries required =
(2632)(2633)
All entries in Table C-2 are required; no more can be removed and
still provide all mileage possibilities. The original Table C-l contained
6,927,424 mileage entries. Table C-2 contains only 3,465,028 mileage
entries so the number of mileages required has been reduced by almost 50%.
C-3
-------
With the number of table entries minimized, the mileage entries
could be written to a sequential data file by indexing through the table
and writing a data record for each mileage entry beginning with column
1, then column 2, column 3, etc. Table C-3 shows the relative entry
number (EN) of each mileage entry written to the data file. The EN of
each mileage entry record in the data file can be calculated directly by
the following formula, where i is the row rate basing point number
(source) and j is the column rate basing point number (destination).
TABLE C-3. SAMPLE RELATIVE MILEAGE ENTRY RECORD NUMBER
FOR A TRIANGULAR MILEAGE DATA FILE
S
0 Destination rate basing points
u
r
c
e
r
a
t
e.
b
a
s
i
n
g
P
o
i
n
t
s
EN(i,j)
i +
- D/2
The following example shows how to determine which EN contains the
mileage from rate basing point 3 to rate basing point 2, or M(3,2). In
order to develop a triangular table in the first place, almost half of
the mileage entries were removed by always accessing with the lowest
rate basing point number first. Examination of the mileage M(3,2) shows
that before the mileage entry can be found, the source and destination
must be reversed, i.e., M(2,3). Now the formula can be used:
C-4
-------
EN(i,j) = i + J(J - D/2
= 2 + 3(2)/2
= 5
The mileage entry for M(2,3) is the fifth mileage entry, which can
be verified by referring to Table C-3.
If the only requirement had been the minimum number of mileage
entries, a file could be written as just described and would be the
smallest possible size. However, when rate comparisons are required and
the distance from several rate basing points to all other rate basing
points must be found, the number of file accesses would be prohibitive.
For a single rate basing point, finding the mileage records for distances
to all other points (including itself) requires 2632 file accesses. If
the mileages from 50 points are to be compared, getting all possible
distances from the 50 points to all other points requires 50 x 2632 =
131,600 file accesses. To reduce the number of file accesses, some
blocking technique is required so that a single file access will get a
block of mileage entries rather than only one entry.
In the current file, a blocking factor (BF) of 50 was chosen, and
each block contains 50 x 50 or 2500 mileage entries. Because 2632 is
not evenly divisible by 50, 2650 points were provided in the blocked
file, with the additional 18 points allowed for future expansion before
the file size will have to be increased. Instead of a file of individual
mileage records for 2632 rate basing points, there is now a file of
blocked mileage records for 2650 rate basing points and each block
contains 2500 individual mileage entries. The file size is now n(n + l)/2
2650(2651)72, or 3,512,575 entries. However, instead of requiring 2632
file accesses to find the mileage entries from one point to all other
points, only 53 (2650/50 = 53) accesses are required because each block
contains mileages from 50 points to 50 other points. The potential
reduction in file accesses is from 2632 to 53 (about 98%). The increase
in file size is from 3,465,028 entries for 2632 points to 3,512,575
entries for 2650 points (about 1.5%). The triangular blocked table and
the relative block number (BN) of each block are shown in Table C-4.
The formula to calculate the total number of blocks required is the
same as for calculating the total number of entries required except
instead of n = 2632 or n = 2650, n = 53.
Blocks required = n(n + l)/2
= 53(53 + l)/2
= 1431
The BN that contains a particular mileage entry can be found in the same
manner that an EN was found before blocking the data except that the BF
must now be considered. As an example, assume that the distance from
source rate basing point number 57 to destination rate basing point
number 102 must be found using the blocked file, i.e., find M(57,102).
C-5
-------
TABLE C-4. BLOCKED TRIANGULAR MILEAGE TABLE
Destination blocks
S
o
u
r
c
e
b
1
p
c
k
s
The first step is to determine the row block and column block that
contains the mileage to be found. Each block contains 50 x 50 mileage
entries, and based on Tables C-3 and C-4 shown previously, block 1
contains mileages from row rate basing points 1-50 to column rate basing
points 1-50; block 2 contains mileages from row rate basing points 1-50
to column rate basing points 51-100; block 3 contains mileages from row
rate basing points 51-100 to column rate basing points 51-100, etc. The
row block index (RBI) and column block index (CBI) of the block that
contains the mileage entry for source rate basing point i and destination
rate basing point j can be found by using the following formulas:
RBI(i,j) = [(i - D/BF] + 1
CBI(i.j) = [(j - D/BF] + 1
where (i - 1)/BF and (j - 1)/BF are always integer quotients; any
remainder is discarded. Table C-4 was based on a BF of 50 and the RBI
and CBI for M(57,102) are found as follows:
RBI(57,102) = [(57 - 1)/50] + 1
= (56/50) + 1
- 2
C-6
-------
The mileage entry Is in the second row of blocks.
CBI(57,102) - [(102 - 1)/50] + 1
= (101/50) + 1
= 3
The mileage entry is in the third column of blocks.
To calculate the BN based on the RBI and the CBI, the procedure is
the same as determining the EN in Table C-3. The BN is calculated using
the RBI and CBI instead of the i and j values in the unblocked table.
Instead of
the following formula is used:
BN(i,j) = RBI + CBI(CBI - l)/2
= 2 + 3(2)/2
= 5
The BN containing M(57,102) is 5. This can be verified by examina-
tion of Tables C-3 and C-4. Row block 2 contains all row source points
51-100, so it contains point 52. Column block 3 contains all column
destination points 101-150, so it contains 102. The intersection of row
block 2 and column block 3 is BN 5.
To find the particular mileage within the block once the BN is
known, a combination of the rate basing point row number, the rate
basing point column number, the RBI, and the CBI must be used to
determine the row and column within the block (each block is BF rows x
BF columns; in this case 50 rows by 50 columns). The RBI within the
block (BRBI) and the CBI within the block (BCBI) are calculated as
follows:
BRBI(i.j) = i - (RBI - 1)BF
BCBI(i,j) = j - (CBI - DBF
Using the previous example of M(57,102),
BRBI(57,102) = 57 - (2 - 1)50
BCBI(57,102) = 102 - (3 - 1)50
= 2
C-7
-------
M(57,102) is in row 7, column 2, within the block. The block entry
number (BEN) is the equivalent of the EN in the unblocked file except
that it is now limited to a single block. It can be calculated using
the BRBI, the BCBI, and the BF.
BEN = BRBI + BF(BCBI - 1)
= 7 + 50(1)
= 57
M(57,102) is the 57th mileage entry in BN 5.
When the file of mileage values is created, each block is written
so that it can be read directly using the BN as a key. By following the
steps in the preceding discussion the mileage between any two rate
basing points can be retrieved from the file. (If mileage entries to
all other rate basing points are required, all 53 blocks are processed
in the same way.) As the blocks are read from the file, 50 mileage
entries are taken from each block with only one file access per block.
The complete procedure is reviewed in the following example:
Find the mileage entry from the blocked triangular mileage file
between rate basing points 1708 and 29, M(1708,29).
1. For the mileage M(i,j) between locations i and j, i must be -j
so calculations should be for M(29,1708)
2. The RBI and CBI must be calculated
a. RBI(i.j) = [(i - D/BF] + 1
RBI(29,1708) = [(29 - 1)/50] + 1
RBI(29,1708) = 1
b. CBI(i,j) = [(j - D/BF] + 1
CBI(29,1708) = [(1708 - D/50] + 1
CBI(29,1708) = 35
M(29,1708) is in row block 1 and column block 35
3. The BN must be calculated
BN(i,j) = RBI + CBKCBI - l)/2
BN(29,1708) = 1 + 35(35 - l)/2
BN(29,1708) = 596
M(29,1708) is in BN 596
C-8
-------
4. BN 596 (2500 entries) is read from the triangular mileage file
5. The row index and column index within the block must be found
a. BRBI(i.j) = i - (RBI - 1)BF
BRBI(29,1708) = 29 - 1(1 - 1)50
BRBI(29,1708) = 29
b. BCBI(i.j) = j - (CBI - DBF
BCBI(29,1708) = 1708 - (35 - 1)50
BCBI(29,1708) = 8
M(29,1708) is in the 29th row and 8th column within block 596
6. The entry within the block must be calculated
BEN(i,j) = BRBI + BF(BCBI - 1)
BEN(29,1708) = 29 + 50(8 - 1)
BEN(29,1708) = 379
M(29,1708) is the 379th entry of block 596
Table C-4 showed the triangular blocked table format. However,
(refer to Table C-2) the blocks along the diagonal do not contain 50 x
50 or 2500 valid mileage entries. Each block on the diagonal is actually
triangular, and contains only 50(50 + l)/2 = 1275 valid mileages.
Rather than treat each block along the diagonal as a special case which
would involve variable length blocks and a more complex file generation
and accessing technique, the blocks along the diagonal are written just
as though all entries tfere valid. The same procedures still apply; the
only effect is 1225 empty mileage entries written per block on the
diagonal. Since there are 53 blocks on the diagonal, there are 53 x
1225 = 64,925 empty entries included in the file. The blocked file
contained a minimum number of 3,512,575 entries; the addition of 64,925
entries to avoid the problems described above is only a 1.9% increase in
the file size. The following paragraph summarizes the file development
process up to this point.
The original triangular mileage tables (Tables C-2 and C-3) con-
tained a minimum number of mileage entries, 3,465,028, but required
2,632 file accesses to retrieve mileage entries from one point to all
other points (including itself). A blocking technique was developed
that only increased the minimum table size by 112,472 or 3.2% (47,547
entries, or a 1.3% increase, resulted from the addition of 18 points
because of the BF of 50; and 64,925 entries, or a 1.9% increase, resulted
from including empty entries in the diagonal blocks to avoid variable
C-9
-------
length blocks). By increasing the file size 3.2%, a potential reduction
in file accesses of up to 98% resulted (from 2632 to 53), a more than
acceptable trade off. (The file size is verified by considering 1431
blocks, each containing 2500 entries. 1,431 x 2,500 = 3,577,500, and is
the same as 3,465,028 + 27,547 + 64,925.)
TRIANGULAR MILEAGE FILE (CDC FORMAT)
Four factors were taken into account to develop the rail mileage
file on the CDC system. One factor was the number of file accesses
required. As discussed previously, a blocked data file was used to
reduce the number of potential file accesses. The remaining factors
were on-line storage requirements for the total file, core storage
requirements for retrieval programs (one block of the file), and data
conversion requirements.
Using a standard decimal format for the rail mileage data results
in problems with all three of the factors not previously considered.
Each mileage entry requires up to 4 decimal characters, and based on
3,577,500 entries, 4 characters per entry, an on-line storage file size
of 14,310,000 characters would be required. On the CDC CYBERNET system
a 60-bit, 10-character word internal format is used. Even though each
mileage entry is a maximum of 4 characters, a standard FORTRAN integer
variable requires a full 10-character word. Based on 2,500 mileage
entries per block, 10 characters per entry, 25,000 characters of core
storage would be required to contain one block of data in a program.
Finally, all decimal data requires a conversion each time it is trans-
ferred between the external format of on-line storage and the internal
format required within a program. Ease of program use is the only
advantage of a decimal format. Standard data input and output routines
can be used to read or write a block of mileages and the required mileage
entry can be directly extracted from the block by subscripting.
A binary format reverses the disadvantages and advantages of a
decimal format. Although each mileage entry requires 4 decimal characters
no mileage entry is greater than 4,095. If a value never exceeds 4,095
decimal, then it can be represented by a maximum of 12 binary bits
(409510 = 7777e = 1111111111112). This is the equivalent of only 2
decimal characters, and 5 mileage entries can be packed into a single
60-bit, 10-character word. For 3,577,500 mileage entries, 2 characters
per entry, an on-line storage file size of 7,155,000 characters would be
required, only 50% of the size required for a decimal format. Core
storage requirements for a program would be 5000 characters (2500 entries
per block, 2 characters per entry), only 20% of the size required for a
decimal format. There would be no data conversion for a binary format'
binary data do not require conversion when transferred between
on-line storage and a program. The only disadvantage of using the
binary format is additional complexity for program usage. Standard read
and write routines cannot be used and a mileage entry cannot be extracted
from a block by subscripting alone. Based on the preceding comparisons
between decimal and binary format and the factors that were to be considered
a binary format was selected for the rail mileage file on the CDC system. '
C-10
-------
Packing of mileage entries into 12 bits, 5 per word, is in the same
sequence that entries would be written to a blocked file (and also to
the original sequential file), that is, in column order. Table C-4
showed the triangular blocked file. Because block 1 is on the diagonal
and contains empty entries, block 2 will be used as an example. The
same principle applies to all blocks, including those on the diagonal,
but the empty entries make the patterns more difficult to illustrate.
Table C-5 shows the mileage entries in block 2. Block 2 is in the first
row of blocks so the rate basing point rows in block 2 are 1-50. However,
block 2 is in the second column of blocks so the rate basing point
colums in block 2 are 51-100. Blocks in the second row of blocks (such
as block 3 of Table C-4) contain row rate basing points 51-100.
TABLE C-5. BLOCK 2 OF TABLE C-4, BY ENTRY NUMBER
Column rate basing points
1
2
3
4
5
6
• • •
50
51
EN1276
EN1277
EN1278
EN1279
EN1280
EN1281
* • •
EN1325
52
EN1327
EN1328
EN1329
EN1330
EN1331
EN1332
• • •
EN1376
53
EN1379
EN1380
EN1381
EN1382
EN1383
EN1384
• * •
EN1428
54
EN1432
EN1433
EN1434
EN1435
EN1436
EN1437
• • •
EN1481
55
EN1486
EN1487
EN1488
EN1489
EN1490
EN1491
• • •
EN1535
56
EN1541
EN1542
EN1543
EN1544
EN1545
EN1546
EN1590
• * *
• • •
• • •
• • •
• • •
• • •
• • •
« • *
100
EN4951
EN4952
EN4953
EN4954
EN4955
EN4956
• • •
EN5000
R
o
•c
a.
a.
s
P
o
i.
fi
t
&
C-ll
-------
Because block 1 (which Is just to the left of block 2 in Table C-4)
is a diagonal block the only valid mileage entries are EN1 through
EN1275, so the first entry in block 2 is EN1276. Block 2 is not a
diagonal block so it contains 2500 valid entries, from EN1276, M(l,51)
(the distance from rate basing point 1 to rate basing point 51) to
EN5000, M(50,100) (the distance from rate basing point 50 to rate basing
point 100). The entry numbers in the block are not EN1276 through
EN3776 (1276 + 2500) because column entries continue from block 2 into
block 3 which is just below (see Tables C-4 and C-5). The block contains
entries in column order (EN1276, EN1277, EN1278, etc.), and is the order
used to pack them into 60-bit words, 12 bits per entry. The first word
of block 2 contains EN1276, EN1277, EN1278, EN1279, EN1280. The second
word contains EN1281, EN1282, EN1283, EN1284, EN1285, etc. Word 1 of
block 2 is shown below:
Word 1
EN1276
59. ..48
EN1277
47. ..36
EN1278
35... 24
EN1279
23. ..12
EN1280
11. ..0
Entry number
Bit number
Within the word, bits are numbered in reverse order, so EN1276 is in the
leftmost 12 bits of the word that are numbered 48-59. Five entries are
packed per word and each column in a block contains 50 entries, so 10
words are required per column. There are 50 columns per block and 10
words are required per column, so 500 words are required per block.
Table C-6 shows block 2 represented by relative word number (WN) rather
than by mileage entry. Each word shown in Table C-6 contains five
separate mileage entries.
The calculations required to extract a mileage from the triangular
mileage file are still basically the same as described earlier. The
difference is that instead of finding the EN within the block that is
read and using it to directly extract the required entry from the program
array containing the mileage block, not only the EN within the block
must be found, but the word containing the entry and also the position
of the entry within the word must be found. Before using the binary
format and packing five entries per word, the EN and the word within the
block were the same because each entry occupied a full word in the
program array. The EN and the word within the block are no longer the
same and an additional mask and shift operation is required because a
standard subscript identifies an entire word, not bit positions within a
word. The calculations required to extract a mileage entry from the
blocked triangular file remain the same except that each subscript
number identifies a word with five mileage entries. The mask and shift
operation is used to identify a single mileage entry within the word.
C-12
-------
TABLE C-6. BLOCK 2 OF TABLE C-4, BY WORD NUMBER
(See also Table C-5)
Rate basing point columns
1
2
3
4
-*-.
9
10
1
WN1
WN2
WN3
WN4
WN9
WN10
2
WN11
WN12
WN13
WN14
WN19
WN20
3
WN21
WN21
WN22
WN23
WN29
WN30
4
WN31
WN32
WN33
WN34
WN39
WN40
• • •
» • »
• • •
• • •
* • •
• • •
• • •
49
WN481
WN482
WN432
WN433
WN489
WN490
50
WN491
WN492
WN493
WN494
WN499
WN500
R
a.
t
e
b
a
s
i
n
P
o
i
n
t
w
o
r
d
s
C-13
-------
In order to do the mask and shift, the WN must be calculated using
the BEN, dividing by 5 because there are 5 entries per word.
WN(i,j) = [(BEN - l)/5] + 1
where (BEN - l)/5 is an integer quotient.
As a continuation of the earlier example where the mileage entry
H(57 102) was to be found, a BN of 5 had been calculated, and within BN
5, a'fiEN in the unpacked file was found to be 57. M(57,102) is still
in BEN 57 of block 5, which is the 57th mileage entry in the block, but
in the packed file it is no longer the 57th word. The WN must be calculated
using the BEN.
WN(i,j) = [(BEN - l)/5] + 1
WN(57,102) = [(57 - l)/5] + 1
WN(57,102) = 12
The mileage entry for M(57,102) is in WN 12 of block 5. Word 12 can
now be extracted from the array using 12 as a subscript, but word
12 actually contains five mileage entries. If HILARY is the variable
name of the mileage array containing the required block, then
HILARY(WN) =
In order to determine which of the five entries within the word
contains the desired mileage, the word entry number (WEN) must be calcu-
lated, which is the relative entry position within word WN.
WEN(i,j) = [MOD(BEN - 1),5] + 1
where MOD is a modulo function with respect to the number of entries per
word, which is five. Continuing with M(57,102), which had a BEN of 57
in block 5
WEN(i,j) = [MOD(BEN - 1),5] + 1
WEN(57,102) = [MOD(57 - 1),5] + 1
WEN(57,102) = 1+1
WEN(57,102) = 2
H(57,102) is in the second 12 bits from the left of word 12 of block 5.
HILARY(12) =
M(56,102)
M(57,102)
M(58,102)
M(59,102)
M(60,102)
C-14
-------
To extract the appropriate entry from the word, a series of masks must
be defined, with the mask number used dependent upon which 12 bits
contain the desired entry.
MASK(l) = 77770000000000000000
MASK(2) = 00007777000000000000
MASK(3) - 00000000777700000000
MASK(4) = 00000000000077770000
MASK(5) = 00000000000000007777
The mask number that will be used is determined by the WEN calculated
above.
MASK(n) = MASK(WEN)
If the word within the block that contains the desired mileage, and the
appropriate mask, based upon the value of WEN, is used in a logical AND
operation, the result will be that the other four entries will be excluded
by being set to zero, leaving the desired mileage entry as the only
nonzero value within the word. A new variable (MENTRY for example) is
used to contain the results of the AND operation to avoid destroying the
other four entries in the word. These other entries may be required in
subsequent mileage searches and if they are destroyed, the complete
block must be read into the array again from storage. By placing the
results of the AND operation into a new variable, WN of the block is
unchanged and the unused entries are still available if needed in sub-
sequent mileage extractions.
MENTRY = HILARY(WN).AND.MASK(WEN)
Continuing with M(57,102)
MENTRY = MILARY(WN).AND.MASK(WEN)
MENTRY = MILARY(12).AND.MASK(2)
The operation can be illustrated as follows:
MILARY(12)
MASK(2)
M(56,102)
M(57,102)
M(58,102)
M(59,102)
M(60,102)
and
0000
mi
0000
0000
0000
MENTRY
0000
M(57,102)
0000
0000
0000
C-15
-------
The desired mileage entry is now the only value in word MENTRY, but
before it can be used it must not only be the only data in the word, it
must be the rightmost 12 bits of the word. In the CDC system, a left
end around shift can be used to move the desired 12 bits into the
rightmost part of the word, with the number of bits shifted (in incre-
ments of 12) dependent upon the original position of the entry within
the word.
For M(57,102)
M(i,j) = SHIFT [MENTRY,12(WEN)]
M(i,j) = SHIFT [MENTRY,12(WEN)]
M(57,102) = SHIFT [MENTRY,12(2)]
M(57,102) = SHIFT (MENTRY,24)
M(57,102) = MENTRY
The shift function will execute a left end around shift of 24 bits for
M(57,102). A shift of 12 bits will move the 12 bits containing the
mileage entry into the leftmost 12 bits of word MFNTRY; the remaining
12-bit shift will move the 12 bits out the left end of the word and back
into the rightmost 12 bits of the word; and after the shift is completed,
M(57,102) will be the value of MENTRY which can now be used as a normal
variable.
MENTRY
The following example illustrates the complete process required to
extract the mileage between two rate basing points from the blocked
triangular mileage file of packed binary entries. Find the mileage
between rate basing point 1793 and rate basing point 231; M(1793,231).
Because of the triangular design, instead of M(1793,231), M(231,1793)
should be found. The first step is to determine the row block and column
block by calculating the RBI and CBI.
0000
0000
0000
0000
M(57,102)
RBI(i.j) = [
-------
M(231,1793) is in row block 5 and column block 36.
BN(i,j) = RBI + CBKCBI - l)/2
BN(231,1793) = 5 + 36(35)/2
BN(231,1793) = 635
M(231,1793) is in BN 635.
BN 635 is read into the program from the rail mileage file. The
row number within block 635 and the column number within block 635 that
contains M(231,1793) must be found.
BRBI(i,j) = i - (RBI - 1)BF
BRBI(231,1793) = 231 - (5 - 1)50
BRBI(231,1793) = 31
BCBI(i.j) - j - (CBI - DBF
BCBI(231,1793) = 1793 - (36 - 1)50
BCBI(231,1793) = 43
Within block 635, M(231,1793) is in row 31 and column 43. The position
of the entry within the block can now be found.
BEN(i.j) = BRBI + BF(BCBI - 1)
BEN(231,1793) - 31 + 50(43 - 1)
BEN(231,1793) = 2131
M(231,1793) is mileage entry number 2131 within block 635. The word
that contains entry 2131 must be calculated.
WN(i,J) - [(BEN - l)/5] + 1
WN(231,1793) = [(2131 - l)/5] + 1
WN(231,1793) - 427
M(231,1793) is in word 427 of block 635. The entry number within word
427 of block 635 must be calculated.
WEN(i.j) - [MOD(BEN - 1),5] + 1
WEN(231,1793) = [MOD(2131 - 1),5] + 1
WEN(231,1793) = 1
C-17
-------
M(231,1793) is in the leftmost 12 bits (48-59) of word 427 of block 635.
A logical AND can be done between word 427 and MASK(l), with the results
in MENTRY, to remove the other four mileages.
MENTRY = HILARY(WN).AND.MASK(WEN)
MENTRY = MILARY(427).AND.MASK(1)
The final step is to shift the mileage entry into the rightmost part of
the word so that it can be used as a normal variable.
M(i,j) = SHIFT [MENTRY,12(WEN)]
M(231,1793) = SHIFT (MENTRY,12)
M(231,1793) = MENTRY
Only 12 bits had to be shifted since the entry was in the leftmost 12
bits. To avoid special cases and to make the shift procedure completely
independent of the mileage position within the word, even when the entry
is already in the rightmost 12 bits in the first place (WEN = 5), the
shift function will result in a shift of all 60 bits (12 x 5 = 60), so
MENTRY will still contain the correct value after execution of the
shift.
RAIL MILES CALCULATION PROGRAM
A utility program (RAILCAL) allows rail mileage entries to be
extracted as required from the rail mileage file. The program accepts
two different types of requests. The first type of request allows a
source rate basing point number (see the RBNSORT file format in section
IV) and a destination rate basing point number to be provided, and the
program will return the rail mileage between the two points. The second
type of request allows a source rate basing point number and a radius to
be provided, and the program will return all destination rate basing
point numbers within the specified radius, along with the rail mileage
from the specified source rate basing point.
Sample usage of the RAILCAL program during a time-sharing session
is shown in Tables C-7 and C-8.
C-18
-------
TABLE C-7. SAMPLE PROCEDURE FILE TO INTERACTIVELY EXECUTE
THE PROGRAM THAT EXTRACTS RAIL MILEAGES
o
04860 1IRAIL, OLD,MESSAGE/UN=LIBRARY,
04880 MESSAGE. EXECUTE RAIL MILEAGE CALCULATION PROGRAM
04900 GET»RAILCAL.
04920 FTN(I=RAILCALfL=0>
04940 LGO.
04960 RETURNfRAILWA.
04980 MESSAGE. RAIL CALCULATION END OF RUN.
05000 RETURN»MESSAGE.
05020 GOTOf9END.
05040 EXIT.
05060 MESSAGE. RAIL CALCULATION PROGRAM ERROR
05080 MESSAGE. DAYFILE TO FOLLOW.
05100 RETURN, MESSAGE* RAILUIA.
05120 GOTO»9EXIT.
05140 9EXIT,PURGEfBYPDAY/NA.
05160 DEFINEfBYPDAY.
05180 DAYFILEfBYPDAY.
05200 REWIND»BYPDAY.
05220 COPYBFrBYPDAY,
05240 RETURN,BYPDAY.
05260 9END»EXIT.
-------
TABLE C-3. SAMPLE USAGE OF THE PROCEDURE FILE TO EXECUTE
THE RAILCAL PROGRAM INTERACTIVELY
/OLDrPROCFIL
/-PROCFIL,S=1IRAIL.
13.12.34.MESSAGE. EXECUTE RAIL MILEAGE CALCULATION PROGRAM
IF INSTRUCTIONS ARE REQUIRED ENTER YES
ELSE ENTER NO
? YES
THIS PROGRAM UILL PROVIDE THE RAIL MILEAGE
BETWEEN ANY 2 RATE BASING POINT NUMBERS OR
IT UILL PROVIDE ALL RATE BASING POINTS WITHIN
A SPECIFIED RADIUS OF A SINGLE SOURCE RATE
BASING POINT.
IN EITHER CASE, A SOURCE NUMBER, A DESTINATION
NUMBER, AND A RADIUS MUST BE SUPPLIED. SEPARATED
BY COMMAS.
IF BOTH A SOURCE NUMBER AND A NON-ZERO
DESTINATION NUMBER ARE GIVEN, A RADIUS VALUE
OF ZERO SHOULD BE ENTERED.
IF A DESTINATION NUMBER OF ZERO IS GIVEN, ALL
POTENTIAL DESTINATION POINTS WITHIN THE
y SPECIFIED RADIUS WILL BE PROVIDED.
^ GREATER PROCESSING EFFICIENCY IS POSSIBLE IF
O SOURCE NUMBERS ARE ENTERED IN SEQUENCE.
A CARRIAGE RETURN UILL TERMINATE PROGRAM
EXECUTION WHEN ALL DATA HAS BEEN ENTERED.
ENTER SOURCE,DESTINATION,RADIUS
? 1095,1095,0
FROM 1095 TO 1095 IS 20 RAIL MILES
ENTER SOURCEtDESTINATION,RADIUS
? 1,1.0
FROM 1 TO 1 IS 20 RAIL MILES
ENTER SOURCE,DESTINATION,RADIUS
?1O95,0,20
POINTS WITHIN A 2O RAIL MILE RADIUS OF POINT 1095
DISTANCE TO 1095 IS 20 RAIL MILES
DISTANCE TO 1584 IS 0 RAIL MILES
DISTANCE TO 1967 IS 0 RAIL MILES
DISTANCE TO 2027 IS 0 RAIL MILES
DISTANCE TO 2433 IS 0 RAIL MILES
DISTANCE TO 2434 IS 0 RAIL MILES
DISTANCE TO 2635 IS 0 RAIL MILES
DISTANCE TO 2636 IS 0 RAIL MILES
DISTANCE TO 2637 IS 0 RAIL MILES
DISTANCE TO 2636 IS 0 RAIL MILES
DISTANCE TO 2639 IS 0 RAIL MILES
ENTER SOURCE,DESTINATION,RADIUS
•?
13.13.53.MESSAGE. RAIL CALCULATION END OF RUN.
9END,EXIT,
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1 REPORT NO.
EPA-600/7-79-114
3. RECIPIENT'S ACCESSIOf*NO.
4. TITLE ANDSUBTITLE
Computerized FGD Byproduct Production and Marketing
System: Users Manual
5. REPORT DATE
Mav 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
W. L. Anders
8. PERFORMING ORGANIZATION REPORT NO.
ECDP B-2
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Tennessee Valley Authority
Office of Power
Emission Control Development Projects
Muscle Shoals, Alabama 35660
10. PROGRAM ELEMENT NO.
INE-624A
11. CONTRACT/GRANT NO.
EPA IAG-D9-E721-BH
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 1/78- 1/79
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES
IERL-RTP project officer is Charles J. Chatlynne, Mail Drop 61, 919/541-2915.
16. ABSTRACT
The users manual describes a computerized system—consisting of a number of integrated programs,
models, and data bases-that has been developed to compare the costs of power plant strategies designed
to meet clean air regulations. It describes the data bases, programs, and procedures and requirements that
are necessary for data base access and program execution. The power plant data base contains actual and
projected information for all U.S. fossil-fuel power plants. A scrubbing cost model allows cost
comparisons between any two of five compliance strategies: limestone scrubbing with sludge waste
disposal, limestone scrubbing with gypsum production, sodium sulfite scrubbing with sulfur production,
magnesia scrubbing with sulfuric acid production, and the use of clean fuel with no scrubbing. For
salable flue gas desulfurization (FGD) byproducts, cost comparisons include potential marketing
revenues. The sulfur and sulfuric acid data base includes actual and projected information for all U.S.
sulfur-burning acid plants. The transportation data base contains legal rail mileages between all rail rate
basing points in the 37 Eastern States (Docket 28300) and also contains location-related data for every
named U.S. location. Each data base and program can generally be used independently of the other parts
of the system.
17.
a-
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
COS AT I Field/Group
Pollution
Flue Gases
Desulfurization
Byproducts
Marketing
Mathematical Models
Cost Comparison
Electric Utilities
Fossil Fuels
Gas Scrubbing
Pollution Control
Stationary Sources
13B 14A
21B
07A,07D 21D
14B 13H
05C
12A
^DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
187
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |