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)

-------