MSU

LIBRARIES
“

‘-

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES wil]
be charged if book is
returned after the date
stamped be1ow.

 

 

 

 

 

r1 :1 "5 hfi .m ‘ 1‘”; |V .
~- ‘ . , .‘- -- ': 4 .1. fi"
‘. L .h I {I :> 3“ ’(Q‘ t? T'%% Q i I“!

 

 

BIOENERGI INDUSTRIAL LOCATION DECISIONS

EMPHASIZING RAH MATERIAL TRANSPORTATION COSTS

BY

Kyle Mason Kittleson

A DISSERTATION
Submitted to
Hichigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department or Resource Development

1983

o ~<Ollmrqv"( '-

"r'w- ..

. o
. . .' can"

tw~"

O '0‘

O 4 In, _ I'Q -- .
“"“*1'v."'i'.'?‘ " o '"' "um. ‘ '
. _

”I’M.

w.

afi'.‘

l

wt?" 7
'Lg M] W

 

w "w“? ,xhu“
1J .V

'0 h,

.mfiunawuwwwrfimflflmMWmmmra

 

ABSTRACT

BIOENERGY INDUSTRIAL LOCATION DECISIONS
EMPHASIZING RAW MATERIAL TRANSPORTATION COSTS

3?
Kyle Kittleson

This dissertation describes the development of a
model for comparing alternative bioenergy production
locations based on raw material transportation costs.
Although the focus of the developmental research is wood-
energy production in Michigan's Northern Lower Peninsula,
only minor changes would be required to adapt the model to
crop-biomass production of synfuel. This research is unique
because it integrates a regional forest biomass ~inventory
based on Landsat satellite imagery and a computerized model
which forecasts network biomass transportation costs based,
in part, on this inventory. It is believed to represent the
first integrated model of this type.

The model developed here consists of four major
components: (1) a specification of the biomass supply
location and quantity, (2) estimation of the cost of
transporting biomass to each of the consumers, including
both existing consumers and the hypothetical bioenergy
plant, (3) application of a linear programming algorithm to
select, from among all the possible combinations of biomass
supply points and consumers, those combinations which
produce the lowest network transportation costs for the

entire network, and (A) a time simulation structure for

components 1.2 and 3.

Four sets of experiments were conducted using the
model: (1) a series to develop preliminary statistical
parameters for the network cost estimates, (2) a benchmark
series of model runs every five years from 1982 through 2012
which included only the existing biomass consumer network,
but no wood-energy plant, (3) a series covering the same
time period simulating a small (25 MN) wood-energy plant at
Rose City, Michigan, and (A) a series simulating the same
size .plant at Idlewild, Michigan. These experiments
indicate information produced using this model will be

valuable in bioenergy industrial location decisions.

CONTENTS

Page
CHAPTER

I. INTRODUCTION 0.0.0.0.000000000000000000000. 1
Problem Statement ..................... 1

Study Objectives ...................... 3

Research Strategy OOOOOOOOOOOOOOOIOOOOO 5

Description Of Study Area ......... 7

II. RESEARCH CONTEXT .................... 1A
Palitical context ....OIOOOOOOOOOOOOOOO 1”
Environmental Context ................. 19

General Statutory Context ........ 20

Air Pollution Context ............ 26

Harvesting Site Impacts .......... 31

Future context .....OOOOOOOOOOOOOO 32

Decisionmakers ........................ 35

III. HODBL AND DATA USED ......OOOOOOOOOOOOOOOO 38

Program I, The Biomass Location And
Supply Simulator ...................... 38 ,
Timber Type ...................... 38
Biomass Tonnage/acre ............. no
Biomass Growth ................... A1
Competition For Stumpage
In The Study Area ................ AA

Program II, Transportation Cost ....... 48
Straight-line Distance 0 e e o e e e o o o e ”9
Regional Transportation Factor ... 50

Transportation Cost Estimate ..... 52
Program III, Transportation Cost
Minimization .......................... 53
Time Simulation Framework ............. 57

IV. RESULTS OF MODELING

Results Of Modeling
Precision Of Estimates ................
Comparison Of Series 1,11 and III .....
Detailed Analysis Of Significant
Results ...............................
Analysis Of Results Affecting
Individual Pulp And Chip Consumers ....
Packaging Corporation Of America

S.D. Warren

......OOOOOCOOOOOOIOOO

M6n83h8 eeeeeeeoeeeeoeeeeeeeoeeeee
Buakil‘k eeeeeeeeooeeeeoeeeeeeeeeee
"OOd-Onel‘sy plant 0 e o e e o e e e e e e e e e e

Apparent Trends:

Results With Less Than

Eighty Percent Significance Level .....

Costs 0f Modeling

Problems Encountered In Modeling ......

V. USING THE MODEL IN DECISIONMAKING .........

Forest Biomass Inventory And Growth
Estimation O.......OOOOOOOOOOOOOOOOOIO.
Transportation Network ................

Shock Variables

Stochastic Estimation .................

Cost Of Applying The Model For Planning

Agricultural Crop Biomass .............
Urban Trash Processing And Disposal ...

VI. SUMMARY AND CONCLUSIONS

Research Results Summarized ...........
Opportunities For Further Research ....
Application Of The Model ..............

APPENDICES

A. REGIONAL FOREST BIOMASS INVENTORY AND

ANALYSIS SYSTEM

B. PROGRAM I AND II FLOW DIAGRAMS AND SOURCE

CODE LISTINGS

REFERENCES

11

59

59
59
66

7O
76
76
77
78

8O

82
85
88

9O

LIST OF FIGURE AND TABLES

FIGURE Page
1.1 Study Area Map ......... ...... .. ...... ... 7.1
TABLES
1.1 Counties In Study Area ................. 8
3.1 Area Of Major Forest Types ............. 39
3.2 Biomass Of Major Forest Types .......... NO
3.3 Weight Of Hood-energy Stands ........... nu
3P4 1981 Average Stumpage Prices ........... "5
3.5 Pulp And Chip Consumers ................ #6
”.1 Summary Of Estimate Precision Runs ..... 6O
A.2 Results Of Benchmark (I) Series ........ 63
A.3 Results Of Rose City (II) Series ....... 63
n.u Results Of Idlewild (III) Series ....... 65
h.5 Example of Benchmark Series and

Rose City for Single Firm .............. 67

h.6 Significant Differences Between
Results of Series I,II and III ......... 69

A.7 Number Of Timber Sales Simulated In
seriealand II ......OOOOOOIOOOOIOOOOOO 71

4.8 Adjusted Network Transportation Costs
For series II And III ......OOOOOOOOOOOO 73

N.9 Network Raw Material Transportation
Costs With A Hood-energy Plant At
Rose City Compared to Similar Costs
At Idlewild ............................ 7A

n.1o Mean Annual Projected Transportation
Cost Per Ton Of Biomass 1982 - 2012 .... 79

iii

CHAPTER I

INIBQDUQIIQN

Statement 21 the problem.

During the last ten years increasing energy costs
have changed the way Americans live:

- (t)he oil crisis of 1973-197N constituted a
turning point in post-war history, delivering a
powerful economic and political jolt to the
entire world. It interrupted or perhaps even
permanently slowed postwar economic growth. And
it set in motion a drastic shift in world power,
in the very substance of international
politics(Stobaugh and Yergin 1979.p.2).

One consequence of this shift has been a renewed
interest in developing sources of energy other than
petroleum. Among these alternatives is bioenergy.
Basically bioenergy is simply the conversion of
vegetation to fuel. Synthetic liquid fuel production,,
for example "gasohol", is the most highly developed
bioenergy technology. A simpler form of bioenergy
production has been used for several decades in the
forest products industry: burning wood in steam boiler
systems. The addition of turbines to these steam systems
makes electricity production possible. Institutional

heating with waste wood has become quite common in the

Pacific Northwest. It has contributed to significant

 

‘—

 

(J

(J

increases in air quality when coal systems have been
converted to wood-chip systems (Eisner 1977).
Electricity produced from burning wood (wood enerEY) is
a very new concept. As of 1982. one such plant exists in
Burlington, Vermont. Its economic and environmental
impact have been carefully monitored and extensively
documented. Whether this plant will be able to survive
economically depends on its ability to produce
electricity profitably, which in turn depends on many
factors. One of the most important is the utility's
ability to obtain fuel at a reasonable cost. Stumpage
costs are therefore important. In addition, because
forests are inherently widely dispersed, the cost of
transporting fuel from forest to plant is also important.
'Transportation considerations will play a major
role in design and locations of wood-fired power
plants. The direct costs of transporting fuel
place limits on the distance the plant can be
located away from the harvest areas;
transportation costs thus significantly affect

fuel procurement costs" (Adler,Blakey,Meyer 1978
P- 3).

Development of a wood-energy plant does not however
occur in a vacuum. In addition to the transportation
cost question faced by the wood-energy plant, a more
complex question exists: how will wood-energy development
affect transportation costs of existing wood users who
will be competing with the wood-energy plant for

stumpage?

 

 

.wu

fin

The purpose of this research is to develop an
economic modeling system, which will use the forest
biomass inventory system developed in my earlier research
(Appendix A), to predict raw material transportation
costs for the entire network of forest products
industries in a region. This forest products network
exists within a much more complex system composed of
political, social, ecological and economic subsystems,

which is discussed in the next chapter.

mommies

The objective of the study is to develop a series
of computer programs which, when applied in conjunction
with a remote sensing-based regional biomass inventory,
can be used to analyze bioenergy plant location decisions

by explicitly considering:

1. the regional resource supply

2. existing and anticipated competition
for this resource

3. cost of the resource

A. growth rate of the resource

5. a series of shock scenarios that will
permit prediction of impacts of future
structural economic change on the

plant.

Specific objectives are:

1.

For Program I, simulate random raw
material supply origins and amounts for
the entire network of resource users in
the region as realistically as
possible.

For Program II, develop a fully
automatic conversion of straight line
distances through the use of the
Regional Transportation Factor (which
is explained in Chapter IV) and-compute
transportation costs between all the
origins simulated in Program I and all
the resource users (destinations) in
the region. That is, to compute all
possible transportation costs in the
region for the period.

For Program III, develop a linear,
minimization of these transportation
costs for the regional network.
Because origins simulated in Program I
are random, this linear minimization
will also outline the supply area of
each resource user in the region.

To develop a simulation framework for

the three Pregrams that can be used to

model time. This framework should
permit:
a. stochastic annual transportation

cost minimization

b. resource growth (or shrinkage)
using generalized growth
projections

c. shocks to be introduced into the
system at specified future times
that simulate structural economic

changes in the system.

WSW
Research was done in four stages:

A. study area forest inventory and biomass
estimation

B. simulation of stumpage purchases for all
wood pulp and chip users in the area every
five years from 1982-2012 (Program I).

C. computations of transportation costs for all
possible combinations of simulated chip and
pulp origins and actual plant destinations
(Program II).

D. linear programming minimization of the
annual network transportation costs within a

time simulation framework (Program III).

MU

‘A

1“

Fe

5 1|

IL,

It was necessary to develop and test a forest
biomass inventory system based on Landsat satellite
imagery because only through use of Landsat can large
areas he inventoried in a timely, cost effective way.
Forest maps produced from Landsat images were compared to
medium scale color infrared aerial photography of the
same areas to determine whether Landsat imagery could
successfully be used for this purpose. Forest stands
were -mapped as either hardwood, conifer or mixed from
Landsat. The detailed composition of the stands was
determined from the color infrared images. Biomass (in
tons) per acre of these stands was estimated based on the
best available sources. A complete description of the
sources and why each was selected is contained in
Appendix A. The percentage of forest cover in hardwood,
conifer and mixed species and the per acre tonnage of
these types was used in the stumpage purchase simulator
by assigning random forested one-square-mile sections a..
forest type and a biomass per acre estimate based on
variable probabilities. A full report of this stage also
is contained in Appendix A.

During stage B, the stumpage purchase simulator,
random forest sections were generated and the stumpage
purchased until the pulp and chip demand of all the users
in the system was met. These sections were withdrawn

from the forest base. Computation of the cost of moving

an

a;

all purchased biomass from its location in the forest to
all the possible pulp and chip users in the region was
done in stage C. Up to and including stage C, all
computations were performed on a microcomputer.

Stage D, minimization of these biomass
transportation costs, was done using the MSU CYBER 750
mainframe computer. All origins and destinations were
input and a linear program compared alternative
combinations of these to find the least expensive set
which still met the consumer network's demands. Stages
A through D were repeated every five years for the thirty

years from 1982 through 2012.

Wofmm

The study area is the northern 33 counties in the
lower peninsula of Michigan (Table 1.1). A rectangular
grid of the area covers approximately 28,000 square
miles. Eliminating Great Lakes and inland lakes with
surface areas larger than one square-mile reduces the _
area by 28 percent to approximately 20,000 square miles
(Figure 1.1).

Individual grid cells in the modeling data base are
one square mile in size. Modeling transportation costs
is done using spatial relationships between the two
dimensional (x,y) coordinates of the individual cells
(sections). Since neither county nor township boundries

represent significant political barriers to biomass

Buskirk
S.D. Wa

Menasha

Figure 1.1.

Champion

Idlewil

Study Area

7.1

   

itibi

se City

 

transportation, there is no advantage to using them to
stratify the grid. Each section in the study area is
classified as being forested, non-forested, water or
urban. Because the purpose of this research is to
develop the analysis system rather than conduct an
inventory of the region, United States Geologic Survey
(USGS) 1:250,000 scale maps were used for this

classification, rather than Landsat satellite imagery.

Table 1.1. Michigan counties in study area
(North-to-South).

1. Emmet 18. Missaukee
2. Cheboygan 19. Roscommon
3. Presque Isle 20. Ogemaw

A. Charlevoix 21. Iosco

5. Antrim 22. Mason

6. Otsego 23. Lake

7. Montmorency 2h. Osceola
8. Alpena 25. Clare

9. Leelanau 26. Gladwin
10. Benzie 27. Arenac
11. Grand Traverse 28. Oceana
12. Kalkaska 29. Newago
13. Crawford 30. Mecosta
1N. Oscoda 31. Isabella

(continued)

31X!

SCEG

pAtte

la‘

firms

(Table 1.1 continued)

15. Alcona 32. Midland
16. Manistee 33. Muskegon
17. Wexford 25. Clare

In general, forests within this study area are a
mixture of northern hardwood and conifer. They are
somewhat unique in that they are composed of quite poor
quality stands. In their natural condition, many soils
in the study region are sandy and contain only thin
layers of organic material. They were of“ marginal
fertility and could not tolerate much farming. Problems
of three types occurred in various combinations and
differing degrees: (1) clearcutting virtually all mature
virgin white pine, (2) deliberate fires to clear slash
and open land for farming and accidential caused by
wildfire, (3) unsuccessful attempts at farming that often
removed whatever fertility survived logging. Many'.
subsequent disease and insect problems are caused by this
pattern of soil abuse.

Poor though these stands are, they represent a
major part of the raw material for 179 forest products
firms in Michigan (Michigan Department Of Natural
Resources 1977). Of these, 161 are sawmills. These are
large numbers, more than five forest products companies

per county, and would seem to suggest a thriving

 

BU

EU.

I
51
S
‘
QC
P.
8
i-
P
no
Pl;

M1,

 

1O

industry. Unfortunately, this is not the case. Most
sawmills are small and operate only part time. To a
large extent, development of a large scale forest
products industry has been limited to those using low
quality raw materials.

Six major pulp or chip users presently compete with
each other for stumpage in the study area. Location and
production capacities of these plants are discussed in
Chapter II. Even pulp buyers have limited use for the
"scrub'I oak (primarily gngzgns, ellipsoidalis) stands
which occupy many of the region's worst sites. Their
products and technology require a mixture of wood pulp
from several species. Although there is a great deal
more poor oak than anyone can use, no one knows how much
because much of it is on private land. 0.8. Forest
Service forest survey data can provide some insights into
these stands. Forest survey data is not as helpful as it
could be in this context because it is designed to ,
provide accurate estimates over relatively large
geographic areas. Most public forests have good forest
inventories which provide this point oriented data.
Private forest owners, on the other hand, as a group do
not.

During the late 1970's Consumers Power, Inc.
planned to construct a wood-energy plant at Hersey,

Michigan. One of the important factors in Consumers

11

Power's original plans for a wood energy plant was this
large area of scrub oak (Hudson and Kittleson 1978). As
a whole, given current product requirements and
technologies, most scrub oak is unmarketable for anything
except fuelwood and a wood-energy plant would be only
partially competitive with other pulp users for stumpage.
However, it is not safe to assume that this will continue
to be the case. Paper, cardboard, particle board and
plywood production technologies are developing rapidly.
It is reasonable to expect changes in these technologies
which would permit even more poorer quality stumpage to
be used. The more poorer quality stumpage which is used
by the forest products industries, the more those
industries will compete with fuelwood users for stumpage.

Forest products industries are reluctant to discuss
stumpage purchases with outsiders because they are afraid
information will be used by their competitors to adjust
stumpage bids. They regard as proprietary such
information as how much of each tree species is required'.
for a particular product, how much they might pay for
stumpage, and how far they would be willing to travel to
obtain it. When they were pressed during telephone
conversations for these details and others, particularly
their supply area, three of the six pulp users indicated
they regarded the entire study area as their supply area
and would be willing to make AOO mile round trips to

harvest stumpage. In fact, a forest products firm might

nak

VOU

bec.

haul

Unfc

in

cf
[‘1‘
(D

1

.ov
cons

StUd

two

lint:

12

make A00 mile round trips in an emergency. However, they
would not be able to continue making the trips regularly
because they will lose money on each trip, given the high
transportation costs detailed in Chapter III.

Railroads often provide a common solution to long-
haul problems in forest products industries.
Unfortunately, the railroad network is not well developed
in this region and much of what has been developed is in
the process of being dismantled at this time because of
low use levels. One reason the companies are able to even
consider these very long hauls is the high quality of the
study area road network.

Major roads throughout the area are good quality,
two lane asphalt with the exception of one four-lane
interstate highway (I-75) which essentially bisects the
area in a north-south direction. There are no natural or
man-made obstacles (e.g. toll bridges or roads) anywhere
in the area. The only real transportation constraint »
regularly encountered by forest products industries is
.winter weather. Assuming good harvest planning has left
logging sites near main roads, these delays are usually
one or two days long and occur less than a dozen times
per winter. In the past larger pulp users have stock
piled stumpage during the rest of the year for use during
the winter. More recently, having found themselves with

holding yards full of wood they couldn't use or sell

beca

buy

C03:

than

stcul

the :

llthou
GORSUH
done 9

b°13 o.

13

because of an economy in recession, they are tending to
buy on tighter schedules.

Very long distance stumpage purchases may become
cost effective for both pulp and wood-energy users
through the use of Great Lakes barges. A recent study of
transporting logs, pulp, chips and firewood by barge on
the Great Lakes predicts that, depending upon individual
circumstances, for transportation distances of longer
than approximately 100 miles the least expensive mode
should be by barge (DenUyl 1982).

The wood-energy plant in Burlington, Vermont is on
the shore of Lake Champlain. Faced with a fuel supply
problem, they decided to try having wood chips hauled to
a barge and then towing the barge to Burlington.
Although the process is more expensive and time
consuming, preliminary indications are that it can be
done economically if the facility is located near a large

body of water.

Ph-

CHAPTER II

RESEABQH.£ONIEXI

Wmmwwn
Michigan

The political context of wood-energy development in
Michigan is primarily a function of the state's present
economic problems. Michigan's economic health depends
primarily on the health of the U.S. automotive industry.
This industry has changed a great deal recently, due in
part to gasoline price changes and Japanese competition.
Michigan's twin problems of unemployment and reduced tax
revenue can be traced directly to these changes.

Considering these problems, newly elected Governor
James Blanchard's heavy emphasis on diversifying the
State's economy and creating new jobs makes a great deal
of sense. Creating jobs in the forest products industry,
in particular, has become a major objective of his new'.
administration (Johnson 1983). Because so much of
Michigan's forest is low quality timber, the emphasis has
been on attracting companies which would utilize this
resource, such as wood-energy, paper, particle board and
more recently, flake board producers. A lower level
effort has been underway for several years without much
public fanfare, so this sudden exposure to the public

spotlight has not caught foresters by surprise. However,

1”

a»
'1!

ma:

PE:

15

with the exception of a few small communities which were
being considered as locations for this type of
development, these efforts have gone largely unnoticed
outside the forest products industry prior to the recent
campaign season. Political campaigning in 1982 changed
this pattern and brought the job creating potential of
Michigan's forest resources to the attention of the
general public.

The present forest products industry employs
approximately 63,000 people in 2,300 firms (James,
Heinen, Olson and Chappelle 1982). In an effort to
quantify the general campaign theme of "forest? jobs",
many Michigan foresters have made claims that the
resource has the potential to support " 50,000 new jobs
by the year 2000' (Johnson 1983). Whether or not this
much potential actually exists remains to be
demonstrated. In any case, development of Michigan's
forest products industry is presently the subject of
widespread public interest and active political debate.”
While this interest may eventually be channeled to
produce constructive development, it has important
present implications for wood-energy research.

Commercial wood-energy is a new concept.
Therefore, it is very difficult to predict what impacts
development of it might produce. The purpose of the
research discussed in this report is to develop and test

a system that can be used to predict the impact of wood-

16

energy development on raw material transportation costs
under different sets of circumstances. While this is an
important question for wood-energy development, there are
other. equally important unknown factors which must be
addressed before the overall potential of industrial
scale wood-energy can be assessed. A detailed discussion
of these other research needs is contained in Chapter V.
However, one major source of uncertainty is influenced
heavily by political factors: stumpage availability from
private forests.

Obviously, whether a certain landowner is .willing
to sell his stumpage depends to a large extent on the
price he can get. However, there are almost certainly
many other factors which would influence his decision.
Very little research has been done on this subject,
primarily because of the physical difficulties involved
in contacting and questioning stumpage owners about
hypothetical circumstances. Although no solid ,
information is available, the availability of a
particular stand of timber is probably influenced by such
factors as: (1) the stumpage owner's need for cash, (2)
the proximity of the stand to a road, (3) the type of
timber harvesting done (selective cutting is usually less
objectionable to owners than clearcutting), (A) whether
regeneration of the stand or conversion to another type

can be assured in some way, (5) the extent to which

17

wildlife and esthetics factors will be impacted by the
harvest. There are almost certainly other factors in
addition to these. In Michigan, an interesting set of
conflicting incentives has developed to further confuse
this subject.

The Commercial Forest Reserve Act (Public Act 9A,
1925, ammended in Public Act 393, 1980) makes it possible
for a landowner to defer a portion of his state tax if he
agrees to harvest the stumpage at a later date. The
purpose of the act was to help private forest owners grow
timber more profitably by allowing them to defer their tax
payments until harvest, thereby increasing the commercial
wood supply. This act encourages forest landowners not to
harvest now. On the other hand, because of the sharp
reduction in state and federal aid available during the
recent recession, rural townships have tended to increase
the tax burden of their largest landowners. This increase
has normally taken the form of either an increase in the
reassessed value of a property or an across the board.
increase in the assessment rate for the entire township.
The increased tax burden produces a need for cash which in
turn encourages the landowner to harvest stumpage now
since timber is usually the only product which has a ready
market. In this case, a long-term state program to
increase the commercial timber supply is being subverted
by impoverished local governments.

Michigan's political and economic realities have

18

another important implication' for wood-energy
development. The research which must be done before
sound policy decisions can be reached is predisposed to
misinterpretation or misuse because the results will be
produced at a time when the entire subject is a lively
political issue. Results of wood-energy research seem
relatively straightforward compared to, for example, such
technical subjects as eugenics. Perhaps it is. However,
in terms of interpretation of research results, both
subjects share the common problem that the research
methods and technology used by scientists are so complex

that any statements of results must be carefully

qualified to preclude as much confusion as possible. In
general, if the results are not for some reason
controversial or politically interesting, these

qualifications seem to be reasonably well maintained.
However, in Michigan the subject is politically important
and precise statements of results may be subjected to_,
generalized interpretation without qualification.
Therefore, given Michigan's present politicized context,
it is very important to minimize misinterpretation of
research results when they are published. One strategy
for reducing misinterpretation or misuse is to provide as
complete a statement of the results of the research as
possible. The first component of the statement should be

a list of the assumptions made during the research, not a

19

one-sentence summary of the results themselves. Another
strategy is to carefully select report and summary
formats and use only those which contain reasonably
balanced, informative statements.

How successful these strategies will be, given the
difficult economic conditions and the broad, optimistic
assertions coming from other foresters, remains to be
seen. However inadequate these strategies may seem to be
in coping with the potential for misinterpretation and
misuse in a politically charged context, they appear to

be the best available at this time.

Enxincnmsntal Context

The following discussion of the environmental
context for wood-energy development consists of four main
sections: (1) general statutory , (2) air pollution, (3)
harvesting site impacts and (A) future considerations.
Considerable overlap exists between these categories and
other environmental impacts, notably water pollution-I
concerns, could be discussed as well. However, within
the study area for the current research these are of less
importance. In particular, given Michigan's vast fresh
water supply, high rate of flow and the small wood-energy
plant modeled in this research, the impact of the plant
on water quality should be negligible (Bechtel
1981.p.u55).
1. General SLBINI££1.£QDIBXL

C)
m
H

dew

e151

des:

20

Whether the environmental impacts considered in the
following analysis should be accepted by society depends
in part on the extent to which wood-energy development
can help meet our energy needs. If wood-energy
development occurs in Michigan, it will probably occur
elsewhere because conditions which make wood-energy
desirable here will undoubtedly exist in other locations.
In addition, Federal laws provide a common statutory
context for wood-energy development anywhere in the
nation. Therefore, it is necessary to look, at least
briefly. beyond the study region to the national wood-
energy picture.

'At present in the United States, the fuel

supplied by biomass is about ... 2 percent of

the total U.S. energy consumption. By year 2000

with appropriate research and development,

biomass has the potential of supplying the

nation with about 52 percent of projected U.S.

consumption." (Roddis 1981)

Regardless what forms it takes, bioenergy can
have a only a limited impact on our energy needs in the ,
near future. Obviously, the impact of direct conversion
of wood-energy will be even less. With this limited
potential, the environmental impacts of the bioenergy
become more important. If bioenergy had the potential to
meet a larger share of our energy needs, its negative
environmental impacts, like those of coal burning, might

be easier to overlook. As it is, whether or not the

marginal energy producing potential of bioenergy should

21

be developed will depend to a large degree on how
successfully its negative environmental impacts can be
minimized.

One of the statutory cornerstones of national
policy on this subject is the 1978 Powerplant and
Industrial Fuel Use Act. The primary purpose of the act
" is to facilitate increased energy independence by
providing for expanded use of alternative energy sources
by the nation's electrical powerplants and major
industrial installations" (Lublin and Pickholz 1980).

Specifically. the Act

' clearly sets a national policy of
requiring industrial and powerplant fuel users
to increase their use of coal and other
alternate :fuels as primary energy sources for
new and existing facilities. This requirement
is a strict one, and it places the burden on the
facilities that are capable of using an
alternate fuel to either convert or seek a
specific exemption. The statutory scheme
reverses the framework of previous fuel
conversion plans and should provide for a more
efficient and effective fuel conservation and
energy independence program" (Lubin and Pickholz
1980).

What has actually emerged from this legislation is
an extremely complex set of provisions which have not
been interpreted consistently. Many provisions are
believed by industry representatives to conflict directly
with provisions of the Clean Air Act (Dryburgh 1980,

9.779).

'PIFUA (the Act) simply reinforces the
existing market pressure to use coal instead of
gas or oil in new plants. ... The increased use
of coal, however, presents hazards beyond the

22

scope of current environmental standards, and by
establishing a national policy favoring coal
conversion, PIFUA provides more support for
industry's attempt to weaken air pollution
standards" (Dryburgh 1980. p.780).

Whether we actually have a coherent national
policy encouraging the use of coal and the development of
alternative sources of energy in general and bioenergy
specifically remains to be seen. For the purposes of
this report, PIFUA confuses the statutory context for
wood-energy development, and consequently, discourages
new investment. Unfortunately, these results are the
rule rather than the exception for recent energy-related
legislation.

To a large extent, confusion in energy and
environmental legislation is a function of a much larger
Constitutional question: how far Congress can go in
preempting or overiding what has been traditionally the
responsibility of individual states to regulate (Fisher
1980.p.786). This issue often arises in environmental
law in the form of questions concerning the extent to'.
which Congress can demand that states enforce federal
regulations. The Clean Air Act is an example of a
sweeping environmental statute where these types of
questions have produced a multitude of complex
litigation. The Priority Energy Project Act of 1980
(PEPA) added a new twist to this complexity in the

energy development arena.

The main purpose of PEPA was to provide a system

for
prcduc

I328 O.‘

‘U

Basical
PrOPell

changed

 

 

 

23

for "fast tracking" particularly desirable energy
producing development through what, by 1980. had become a
maze of legal constraints. The Energy Mobilization Board
was to be responsible for overseeing this work. By
comparison to other statutory processes, the PEPA system
seems quite straightforward and wood-energy is precisely
the kind of "geographically isolated" priority project
PEPA was to have promoted (Fisher 1980. p. 798).

_PEPA has not worked out well. What happened?
Basically three things: (1) the energy crisis which had
propelled PEPA subsided somewhat, (2) the Presidency
changed parties, (3) PEPA's regulations raised some of
the still unresolved Constitutional questions discussed
above. Ultimately, PEPA and the Energy Mobilization
Board quietly succumbed to changing political reality.
The Energy Mobilization Board still exists on paper.
However, they do not meet regularly and were never able
to achieve their objectives. The problem is that the,.
PEPA "approach is totally inconsistent with the needs of
a decisionmaking system that is asked to approve or
reject projects that have a significant effect upon the
economy and environment of discrete geographic areas"
(Fisher 1980. p. 866). The fate of PEPA is important for
wood-energy development because it is part of a larger
pattern of failed legislative efforts to establish a

consistent regulatory context. These failures have

and

his a

'1

2”

produced confusion which in turn discourages investment,
particularly in unproven technologies like wood-energy.

Another such confusing episode, at least in
Michigan, has been the Public Utilities Regulatory and
Policy Act of 1978 (PURPA). The main purpose of Section
210 of this statute was to ” encourage the development of
electric power production and energy conservation through
the use of renewable energy sources" (Hinshaw 1982,p.1).
Bioenergy is definitely a renewable source and PURPA has
passed its most difficult legal hurdle in Eeeenel Enengy
Reunion musical... Mississippi, 102 s.ct. 2126
(1982). Why wood-energy hasn't received much
'encouragement' from PURPA is difficult to say. Perhaps
because the recent oil surplus has reduced our perception
of energy scarcity. However, there is evidence that
PURPA's laudable objectives may have become bogged down
in bureaucracy.

In Michigan, the Public Service Commission is
charged with developing procedures for carrying out the =
Federal Energy Commission's guidelines. Following a
lengthy study, one independent researcher has concluded
that ' the Michigan Public Service Commission's approach
reveals a choice of confusing procedures" (Hinshaw
1982.p.ii). Hinshaw goes on to point out that there is
no evidence that these confusing procedures have
"compromised" the goals of PURPA yet.

The overall point is this -- the one

characi
so far
reason:
their

rules

1:30pm:
deci3‘10:
Federal

”Sloan

“Say (1

the? ar'

25

characteristic that all three of the statutes discussed
so far have in common is that they have, for a variety of
reasons, produced regulatory confusion in addition to
their other contributions. Because most regulations,
rules and procedures adopted along the way have added
their share, the confusion itself seems to have gradually
become the predominant factor in the general legal
context of bioenergy development. So far we seem to be
unable to do much to prevent these problems.

Ralph W. Walker II, past Chairman of the American
Bar Association's Energy Resources Commission argues that
these types of problems will continue until we stop
entrusting basic energy research and development projects
to private energy producing companies (Walker 1980.p.610).
Walker suggests that the basic problem is that we don't
have our energy facts straight. Because of the profit
incentive, private research and development concerning
energy will always produce results in their own interestr'
He continues that, since we don't have adequate
information, we continue to make patchwork legislative
decisions which guarantee confusion in the end. The
Federal Government, according to Walker, should accept its
responsibility to sponsor research on all aspects of
energy development and regulation. He concludes that this
research should be conducted in major universities because

they are the only institutions we have which are capable

26

of reasonably unbiased research (Walker 1980.p.61u).

2. A12 Pollution QQDLELL

The Environmental Protection Agency (EPA) has
recently developed a market system for trading air
pollution entitlements. The market is based on three
concepts: (1)"bubbles', (2) offsets and (3) banking to
create a central clearinghouse. The "bubble" concept
permits pollution controls to be based on single plants,
or groups of plants, rather than individual stacks within
a plant. The objective is to permit a firm to make
decisions regarding individual stacks and still -regulate
the environmental impact of the plant. The offsetting
concept means that major new industrial plants may be
constructed in areas of the country which don't presently
comply with Clean Air Act requirements. All the owner of
such a plant needs to do is obtain offsetting reductions
from existing sources of air pollution equal to the amount
projected for the new plant. The purpose of banking is to,
provide one centralized trading location for these
entitlements and permit controlled trading. Two important
questions about entitlement trading are: does it work (or,

can it be made to work) and is it legal?

The General Accounting Office recently completed a
comprehensive review of air pollution entitlement trading

(GAO 1982). As is its normal practice, the GAO concerned

11‘

A!“

the

lay

itpe]

Re"

fl

Yet a

27

itself primarily with whether or not the trading is in
the public interest. They did not address the question
of trading legality. Basically, their conclusion was
that the present problems are probably not unresolvable.
They found that most of the problems were due to the
novelty of the concept rather than serious flaws. From
their perspective, entitlement trading appeared to be a
good plan. In addition they predicted that, whether it
is desirable or not, the energy industry will probably
attempt to make widespread use of the bubble in
attainment areas (Darrell,1982.p.36). However, since
most development will probably occur in attainment areas
use of the bubble concept may never become widespread in
wood-energy. So, GAO, EPA and industry experts all
appear to favor the bubble. But, is it legal?

The most recent judicial ruling available indicates
it may not be. The U.S. Court of Appeals, District Of
Columbia, decision in NRDC v Gorsuch states that '
the EPA's regulatory change, its employment of the bubble”
concept to shrink to a relatively small size mandatory
new source review in nonattainment areas, is
impermissible" ( 17 ERC 1825). Apparently, whether or not
new wood-energy firms will find themselves dealing with
air pollution on a stack-by-stack basis or with their
entire plant under one bubble remains to be seen, adding

yet another unknown to the wood-energy legal context.

In contrast to this uncertain regulatory context,

bi

.Id

86

CE

He

at

We

Dr

5.

IV.

28

the technical aspects of air pollution control seem
relatively well known. The forest products industry has
been using wood byproduct-fired boilers for many years
and has reasonably reliable data on the air pollution
control aspects of wood burning. Every wood processing
plant generates some kind of residue. These residues
range from sawdust coming from sawmills, to chunks of
wood and peelings of bark from plywood mills. Because of
the wide range of size, moisture content and combustion
characteristics of these residues, the industry has
developed multiple fuel handling and boiler systems.
Without going into detail on how these systems work, it
seems obvious that the chemical contents of stack gases
can vary quite a bit depending on the fuel and how
efficient a given boiler is in burning a particular fuel.
Three main types of emissions-control systems are
commonly used with wood-fired boilers: dry mechanical
collectors using cyclonic separators, wet scrubbers, and .
electrostatic precipitators (Flick 1976). The dry
mechanical collectors are usually employed as the first
stage of the treatment process and are followed by either
wet scrubbing or electrostatic precipitators. Mechanical
collection followed by wet scrubbing or electrostatic
precipitators are the most effective in reducing
particulate emission. In terms of cost, the mechanical

collector and electrostatic precipitator combination is

29

the most expensive, followed closely by wet scrubbing
(Flick 1976,p.153.). Mechanical collectors alone are
roughly one third the cost of either of the two
combinations. The cost of these systems is important
because it is one of the major points at which the air
pollution regulations directly determines the economic
viability of a wood-energy plant.

Efforts to comply with air pollution standards as
inexpensively as possible has lead to considerable amount
of creative experimentation. One of the most promising
new techniques is the use of gravel bed filters for
particulate control. The gravel bed is slowly moved
through the escaping gas, trapping particles. Preliminary
indications_ are that gravel bed filters are at least as
efficient as wet scrubbing or electrostatic precipitation
and probably less expensive to build and operate (Flick
1976,p.152).

Regardless which emissions control system is used,
whether or not a plant can meet effluent quality.-
standards is primarily determined by two additional
groups of variables: the fuel burned in the boiler and
the efficiency of the boiler operation. "(I)t is very
important that everyone realize the impact fuel
preparation and handling have on boiler performance and
its effect on stack emissions" (McBurney 1976,p.166).
Boilers can be designed, built and, to a lesser extent

adjusted to accommodate a wide range of fuels. No matter

ho

5%
AB

Ac.

.5

30

how carefully the boiler is operated or how good the
emissions control systems are, unless the fuel supply is
consistent, the stack emissions will not consistently
meet standards. Boiler operators cannot continuously
adjust their equipment in response to changes in, for
example, fuel size or moisture content. When they don't,
violations often occur. The point is that given a
consistent fuel supply and existing emissions controls
systems, wood-fueled boilers can be operated cleanly.

At least one engineer believes that problems with
wood-fired boilers are often caused by operators- trying
to reduce emissions by installing emissions gontrol
systems rather than directly addressing the cause of the
problem which is usually either fuel or boiler related.
He suggests that "regulations might be more acceptable
and more practical if they addressed the cause of the
problem of emissions rather than the symptoms" (Johnson
1976,p.170). Johnson indicates a system of tax credits_.
for cleanly operated furnaces might produce better
results than the present, penalty-oriented monitoring of

stack effluent.

3. museum
Until this point in the discussion of wood-energy's

environmental impact, the focus has been on those impacts

31

which would be produced at the plant by burning the wood.
Wood-energy development will also produce environmental
impacts at the logging site. In general, these fall into
three classes: vegetation, water, and soil. Vegetation
damage is the most readily perceived of the three types.
As is discussed in Chapter V, recent research indicates
between one quarter and one half of the trees left in a
stand harvested using whole—tree methods will sustain
some kind of damage during a harvest. Two thirds of this
damage is caused by dragging entire trees from the stump
to the chipping site (Kelley 1983). What, if any, long
term impact this damage will have on the stand .is not
known. More research is needed in this area. Whole-tree
harvests for wood-energy will probably have a negligible
impact on water resources in the relatively flat ‘glacial
soils of Northern Michigan. In mountainous areas
clearcutting has produced serious soil erosion which in
turn decreases the quality of surface water below.
Because the study region's soils are, overall, flat or'.
contain only gentle slopes, erosion has not been a
problem.

The most serious and least understood harvest-site
impact of whole-tree chipping is soil nutrient depletion.
Traditional harvesting removes only the main tree stem,
leaving the branches and leaves. These decompose and
eventually become soil nutrients available to support new

vegetation. Little research has been done on this

32

subject (Kittleson 1979,p.24). The work that has been
done indicates that where soil nutrient supply is
marginal, whole-tree removal may seriously limit the type
of vegetation which can survive. As is detailed in
Appendix A, many of the study region's poorest timber
stands, the ones most desirable, from a wood-energy
perspective, are growing on soils which are very poor in
nutrients and very sandy. In fact, the reason the stands
are of such poor quality is that the soils are poor. It
is likely that some fertilization of these sites after
logging would be required.
1:. mm

A major independent engineering firm has recently
completed a multi-volume environmental and economic
comparison of energy production from coal and biomass
(Bechtel 1981). Although the study tends, predictably,
to treat emissions standards and applications of these
standards as though they were static, focusing instead on.-
the engineering issues, it does come to an interesting
conclusion regarding future environmental impacts of
coal and biomass as sources of energy.

'(S)mall biomass conversion plants do not
appear to have an overall environmental
advantage over coal counterparts on a relative
basis ... . This environmental standoff is
partly due to the extensive (air) pollution
controls used in the coal conversion plants
which reduce conventional pollutant emission to

about the same overall levels as those from
biomass conversion." (Bechtel 1981,p.vii).

33

A major Department of Energy study compared several
development scenarios in models through the year 2000
(Habegger,et a1 1981). Most of their conclusions are
similar to Bechtel's. Although both studies are based on
many assumptions, they both point toward wood-energy
developers having to work hard to meet attainment
standards and facing stiff competetion from coal.

In terms of economic viability, the Bechtel report
is even less enthusiastic about biomass. Following a
lengthy analysis of the cost of energy production with
comparable systems including the cost of emissions
controls, they conclude:

"From these comparisons, it is evident
that the coal conversion processes are more
likely to become major routes to clean fuels
than these biomass conversion processes,
primarily because of better economics." (Bechtel
1981,p.ix).

The two sources of energy become cost comparable»
when the average delivered price of coal is $25/ton and
the average delivered price of wood is nothing (Bechtel
1981. p.357). Sweeping. conclusions regarding the
environmental context for wood-energy are difficult to
draw, primarily because so much is uncertain. As
discussed above, questions exist concerning

interpretation and application of each of the major

pieces of legislation which determine environmental

3h

regulation. These environmental questions exist, in
turn, within a larger context of Constitutional issues.
By comparison to this legal and statutory confusion, the
technical aspects of wood-energy environmental impact are
reasonably well understood. Although most of this
knowledge is based on the use of wood residues in forest
products industries, experts believe they are valid for
wood-energy plants as well. The economic aspects of
environmental impact seem, in general, discouraging to
wood-energy development. However, there are few regions
of the U.S. where as abundant a supply of currently
unmarketable wood exists as Northern Lower Michigan. If
wood-energy development is to occur anywhere, the study
region is one of the best locations.

The most important factor in analyzing the legal
context of wood-energy development is that, as the above
discussion suggests, wood-energy can be combined with
other alternative sources of energy. For all practical
purposes relevant statutes, administrative procedures and
case law precedents treat wood-energy the same as they do
any other source of energy. Therefore, future
developments in energy law should be assumed to apply to
wood-energy, regardless of what raw material they refer
to specifically. Obviously the technical details will be
different and there will be exceptions. However, based
on developments so far, the legal context for wood-energy

development is not significantly different from any other

35

energy development.

Decisiumakens

Decisionmakers in all subsystems can be broadly
divided into three groups: (1) private and commercial
forest products users, (2) public resource managers and
resource policy-makers and (3) concerned citizens. In
general, private commercial decisionmakers will be
interested in the current research to the extent it can
improve their ability to achieve their goals. When the
model developed in this research is applied to actual
inventory data it will be a useful commercial planning
tool. Public sector decisionmakers will be able to use
the model to predict how wood-energy will affect publicly
owned resources and how they can optimize their
performance in the face of increasingly limited financial
and natural resources. Concerned citizens reactions to
this research will primarily be determined by their.
affiliation with a particular special interest group.

Raw material transportation costs are as important
in crop biomass conversion as they are in wood-energy. So
the model developed here will be valuable for crop-based
synfuel conversion, as well. Another bioenergy technology
that is becoming increasingly appealing to many urban
planners is solid waste conversion to either electricity

or gas. Waste transportation costs are a major factor in

36

determining the economic viability of this energy source.

Regardless of whether a bioenergy plant is based on
wood, crop biomass, or solid waste, the primary
instrument for obtaining long term private capital is the
corporate bond. Both bond ratings and bond coupon rates
will be heavily dependent on demonstrating the existence
of an adequate fuel supply within financially feasible
hauling distances. Therefore, estimates of fuel
transport costs should be valuable information for
financial analysts considering bioenergy investment.

All these decisionmakers face three major sets of
alternatives that will be affected by this model:

1. Should a bio-energy plant be built in a given
region at all? Who will benefit and who will
pay?

2. If so, where should it go? Public optimal
location probably differs from private optimal
location.

3. How large a plant will the regional resource

base support given present and future uses?

The model developed in this research addresses
these questions in several ways. First, it can be used
to predict how the introduction of a bioenergy plant will
affect transportation costs of the network of users
competing with it for raw material. Second, because

these costs are important to bioenergy and other forest

37

products, the model will address the issue of whether a
wood-energy plant should be built at all. Third, because
the model permits wood-energy plants of any size to be
simulated at any location, it can, in part, suggest where
it should go and how large it should be.

The model will not answer any of these questions
completely. It may, however, provide decisionmakers with
useful information on all three subjects. The model can
be applied with minor changes to industrial location
decisions in the area of crop biomass conversion to
alcohol. With more extensive modification, the model can
be applied to urban solid waste processing and disposal
location problems. Both of these applications are
discussed in Chapter IV. In general, the purpose of the
model is to provide decisionmakers with an analytical
tool for systematically comparing alternative impacts
which might come from development of wood-energy in a

given region.

CHAPTER III

M2421 An1.DBLi.QSEA

Whmmmmww
A.Timh.er.Ine

Each execution of Program I simulates the sale of
enough biomass stumpage to meet the combined demand of
all pulp and chip consumers in the regional network for
one year. The locations of simulated timber sales are
selected randomly from forested cells within the study
area. Individual consumers undoubtedly have a higher
preference for locations nearest their plants. However,
because: (1) the consumers are quite evenly disturbed
throughout the study area, and (2) at this time there is‘
no way to predict the propensity of particular owners to
sell stumpage, random sale locations are the most
realistic process. Acres of biomass available in each sale
are also generated randomly ranging from a minimum sale
size of 10 acres to the sale of the entire section (6A0
acres).

The type of timber in the sale is classified as

either hardwood, conifer, or mixed hardwood/conifer. The

38

 

39

probability of a particular sale being simulated as
hardwood, conifer or mixed is based on the proportion of
these types mapped in earlier satellite-based forest
biomass research (Appendix A). During this earlier
research, sample forests in the study area were mapped
from Landsat satellite imagery using a system developed
during the research. One result of the research was a
preliminary estimate of what proportion of the forests

were hardwood, conifer and mixed stands (Table 3.1).

Table 3.1. -- Area of forest types (Appendix A).

Ecnsst.szn Pennant Q£.E2££§1
Hardwood 42.82
Conifer 19.A0
Mixed 38.19

The current research extends this estimate to the-
entire study area. No quantitative basis exists for this
extension. But, by the same token, there is no reason to
suspect it is not accurate. The important point is that,
as is discussed in Chapter IV, an actual application of
the analysis system developed in this research must be
based on a regional biomass inventory, rather than the
small preliminary sample obtained for the system

development. The only cost effective, timely way to

HO

accomplish this at present is using Landsat satellite
imagery. The study reported in Appendix A developed and
tested such a mapping system. Regional Landsat biomass
mapping would produce actual proportions of hardwood,
conifer, and mixed types that could be substituted for
the preliminary estimates used here. In addition, no
variable probability allocation would be required because
the type of each forested cell in the system could be
computed using the data base. The same reasoning applies
to ten per acre estimates assigned to the three forest

types.

B. Biomass Tonnage.Pen Acne
Biomass tonnage per acre estimates used in this

research were also developed earlier (Table 3.2).

Table 3.2. -- Biomass of major forest types (tons/acre),
(Appendix A).

1112.1 Biomass
Hardwood 39.55
Conifer 33-32
Mixed 33-90

Table 3.2 estimates are based on a small sample obtained
during model development. Appendix A contains a detailed

description of the sampling procedure. They are

nerch

rate

A1

preliminary estimates and should not be interpreted as
accurate representations of the per acre biomass

available in the study area.

QWW

The most recent forest inventory information
available (Michigan Department Of Natural Resources 1981,
p.67) places the Michigan statewide annual forest growth
at 16.2 million cords of sub-merchantable timber and 4.7
million cords of merchantable timber. This growth occurs
on 19.A million forested acres. Assuming the ratio of
merchantable to sub-merchantable is the same for growth
rate as it is for area, this is an average growth rate of
.835 cords/acre/year for sub-merchantable timber and .2h2
cords/acre/year for commercial timber. An important
advantage of wood-energy supply is that it will be able to
utilize some of this sub-merchantable material. Therefore,
overall merchantability standards will be lower making
more low quality timber "commercial”. How much more‘
timber will become merchantable for wood-energy is not
known. In practice, minimum merchantability standards for
wood-energy harvests are so low that they are really
quantitative rather than qualititative. That is, wood-
energy buyers will not harvest a particular stand if it
contains less than a minimum amount of biomass. The only
reason some stands will not be harvestable for wood-energy

is that they do not contain sufficient biomass to make the

:Pans:
at°th1

c“DO:

QEErg.

hard'ic

A2

effort worthwhile. This is an important point because
many stands in the region are quite open because of either
site limitations, insect or disease problems or both as
discussed in Chapter I. Therefore, minimum stand density
will probably be a significant factor in determining
merchantability of many stands for wood-energy. How these
general principles will translate into quantitative
estimates is difficult to predict. During model
development it was assumed that an arbitrary value of 60
percent as an estimate of which stands would become
commercial for wood-energy. Forty percent of the stands
would then be non-commercial rather than the current 71
percent. Using these assumptions, preliminary “average
growth for the entire forest can be estimated in volume as
follows:

.60(.2A2)+.AO(.835) = .A79 cords/acre/year.
Translating this volume estimate into weight requires yet
another set of assumptions regarding the species
composition of wood-energy harvests.

Since one of the primary reasons for considering
the Northern Lower Peninsula as a location for wood-
energy is the availability of low quality northern
hardwood, particularly as discussed above for various oak
species, it is logical to conclude that these stands
would make up a large proportion of the harvested
stumpage. Again, ~translating this into quantitative

terms requires a subjective estimate. Logically, wood-

t

uh

b.

oak

are

oper

’v
o u

LU

set

only

will

4
a

one

‘
Alla

Va

1‘

done

-33

there

7%

(A

"3

energy buyers should focus on these poor quality stands
but two spatial constraints will prevent them from using
oak exclusively: (1) better stands of aspen and softwoods
are mixed with the northern hardwood, (2) wood-energy
operators will find cost incentives in harvesting these
better stands because they are growing near northern
hardwood stands where they already have their equipment
set up.

.Considering these factors, this model assumes that
only 70 percent of the stumpage harvested for wood-energy
will actually be mixed northern hardwoods, and the
remaining 30 percent will be composed of the more
valuable forest types in the study area: aspen, balsam
fir, pine (especially Jack pine) and spruce/fir which
happen to be growing near low quality stands.
Conversion of these assumptions into weight estimates was
done using wood industry standards (Table 3.3).

Based on Table 3.3:

.30(A525)+.70(5075) = #910 pounds/cord of wood-
energy fuel.
Therefore, in one year, a wood-energy stand should grow:
(A910 pounds/cord) x (.u79 cords grown/acre/year) =
= 2352 pounds/acre/year

= 1.18 tons/acre/year.

 

 

AA

Table 3.3. -- Weight of wood-energy stands.

Ins. WWI
(Source: Timber Mart 1981)

Aspen A650

Balsam fir A800

Pines A300

Spruce A350

(average a A525)

Mixed northern hardwoods 5350
Lighter mixed northern A800
hardwoods

(average = 5075)

D. Competition.zcn.fitumpasa In.The.Stud1.Ansa

Based on stumpage market price analysis, only
pulpwood harvesters should actually compete with wood-
energy harvesters for stumpage because both veneer and
sawlog consumers pay more for their stumpage (Table 3.A).
In addition to average stumpage price differences, both
pulp and chip harvesters commonly fell and skid trees
larger than 10 inches dbh (diameter breast high), leaving
them for either sawlog or veneer purchasers who negotiate
prices with landowner (personal field observation 1979).
Considering market prices and this informal practice, it

is apparent that sawlog consumers do not compete with

And On

an ant

all Of

 

A5

either pulp consumers or chip consumers for stumpage.
Since the model deals with only the network of stumpage
consumers who will compete with the wood-energy chip

consumers, sawlog and veneer consumers are not included.

Table 3.A. -- 1981 average stumpage prices in the study

region (Timber Mart 1981).

Insflfmsum Amnifllsflmnaae
eniee (dollars/green ton).
Veneer A0.A5 ($267.00/MBF)
Saw 15.60 ($103.00/MBF)
Pulp 5.27 ($12.38/Std.
Cord)
Whole-tree chips 12.50

9 Conversion factors: 1MBF weighs 6.6 tons green, 1
std. cord weighs 2.35 tons green (Timber Mart 1981).

Six major pulp and chip companies harvest stumpage
in the study area (Michigan Department Of Natural
Resources 1977). Five are located within the study area
and one is south of it (Table 3.5).

The existing pulp and chip using network will have
an annual demand of 910,675 tons of stumpage per year, if
all of the consumers are operating at capacity. Personal
communications with company officials and DNR personnel

failed to produce estimates of how close to capacity the

A6

companies were operating in 1981, although the consensus

Table 3.5. -- Pulp and chip consumers who harvest
stumpage from the study area (Michigan Department Of

Natural Resources 1977).

Consul Looaoion Plonioaoaoily.

(tons/year)

Abitibi, Inc. Alpena 156,950
Packaging Corp of Manistee 1A6,000
America. , .
S.D. Warren, Inc. Muskegon 87,600
Menasha, Inc. Otsego 82,125
Champion International, Gaylord 365,000
Inc.
Buskirk, Inc. Paris 73,000
Total = 910,675

was that that they were all well below capacity. It is
difficult to tell because the companies regard this as
proprietary information. The DNR faces the same type of
problem but because of their combined roles as a
management, regulatory and enforcement agency they
provide the companies with even less incentive to

cooperate. Forest products companies logically assume

seal:
node:
tree

metro

7,795

 

 

 

1:7

that information provided to one branch of the Department
for management could find its way into either enforcement
or regulation branches and be used against them. Because
of this problem, the model treated each consumer as
functioning at plant capacity during development and
testing, while recognizing that they do not do so
currently. However, in the event this information
becomes available, the model has been written to permit
different demands for each industry in each year. The
small- (25 Megawatt) wood-energy plant studied with the
model will need an estimated 285,000 tons/year of whole-
tree chips. When this plant enters the system, the
network demand for pulp and chips increases 31.35 to
1,195,675 tons/year.

Program I simulates stumpage sale: (1) location,
(2) timber type, (3) acreage, on the basis described
above until the entire network demand for one year is met
(flowcharts and source code listings for all programs are
in Appendix B). These sale locations, in two dimensional
(x,y) grid coordinates, and tons of biomass available at
each location simulated during one execution of Program
I, are the basis for one execution of Program II.
Program II computes transportation costs involved in
moving all stumpage sold in the simulation to all

possible combinations of network consumers.

A8

PnoanamILImoooziooionoosooomouooiion

Program II estimates the cost of transporting all
of the biomass harvested in the Program I simulation to
all consumers in the network. All possible combinations
of stumpage sales (origins) simulated in Program I and
biomass consumers (destinations) are identified and costs
of transporting biomass to each are computed. It is
basically an algebraic problem and the same two
dimensional spatial grid of the study area used in Program
I is used in Program II (Appendix B contains flowchart and
code). Program II's flow sequence is as follows:'

1. X and I coordinates for each origin and
destination are input,

2. straight line distances between all
possible origins and destinations are
calculated,

3. straight line distances are converted to
estimates of road travel distances »
required to cover the distance using a
regional transportation factor (RTF,
discussed below),

A. these distances are converted to round
trip cost estimates using a cost/mile
estimate.

The output from Program II, consisting of all

possible transportation cost combinations, are

coefi

by t!

IAEPQ

3.8

A9

coefficients of the objective functions which is minimized

by the Program III linear minimization.

A. WWW

Straight-line distance between all origins and all
destinations is computed using the Pythagorean Theorem
adapted to a two dimensional (x,y) axis grid. For

example, the distance between two points (x ,y ) and
1 1

(I 9y ) 18:
2 2
1. - X axis distance: the absolute value of the
difference between x and x
1 2
2. Y axis distance: the absolute value of the

difference between y and
1

Y
2
Therefore, the straight-line distance is:

3.8traight-line distance: the square root of the sum
of the x-axis distance
squared plus the y-axis

distance squared.

This simple process works for all destinations
(x ,y ) located within the study area. However, as
diicuised above, one of the major destinations for pulp
cut in the study area is located outside the study area.
The Menasha Corporation of Otsego, Michigan is 7A miles
south of the southern boundary of the study area. Menasha

is able to compete for stumpage in spite of this apparent

handicap because a limited access interstate highway (I-

50

131) extends from this plant into the heart of the study
area. making the raw material transport less expensive and
time consuming than it would be over two-lane roads. This
fact was compensated for in Program II by adding 7A miles
to all the straight-line distances computed for the Mensha
Corportation. This is done by simply adding 7A miles to
each straight-line distance estimated. The corrected set
of straight-line distances is converted to estimates of
actual mileage that would be required to cover each

distance using the regional transportation factor (RTF).

B. Rozionalrnanuomoionzaoioninm

The RTF used in Program II is 1.36 miles. It means
that an average of 1.36 miles of road travel is required
to move one straight-line mile in any direction within
the study area. The RTF substitutes for a computerized
transportation network map of the study area. The
Michigan Department of Transportation (MDOT) has
developed a computerized transportation map of Michigan.
This map was not used because: (1) the MDOT computer
system is not compatible with the MSU CIBER 750 at this
time, which would have severely limited the number of
model runs possible, (2) areas other than Michigan where
this model might be applied, in particular foreign
countries, probably would not have such a complete
computerized transportation network map. Considering

these problems, it was decided to develop and document

51

this model without using the MDOT data base. If such a
data base and the computer programs to exploit it for
trip distance computations are available, the RTF used in
Program II can be easily replaced with actual distances.
The RTF is based on a sample of the straight line
distances and actual road distances between A1 pairs of
random points in the study area. Straight line distances
were calculated using the Pythagorean Theorem method
described above. Road distances between the pairs of
points were estimated graphically using MDOT county road
maps. The graphic procedure assumed (1) the vehicle was
a large chip van so only the best roads were selected and
(2) alternative routes were compared and the one chosen
subjectively assesssed to be the best for this vehicle.
Since the study area is well roaded in an almost
exclusively east-west and north-south grid and contains
no toll roads or bridges, routing decisions were not
difficult. Each 'best' route was mapped, measured and--
the mileage recorded. The variance of the RTF estimate
is .378 miles so its standard deviation is .615 miles.
This means that if, for example, A0 trips are made
between any two points in the study area that are one
mile apart, thirty eight of them (95 percent) will
require between 1.17 miles and 1.55 miles of road travel,

and that all A0 trips will average 1.36 miles in length.

POUCC

52

c. WWW

A base year (1982) estimate of 9.3 cents per mile
was used in Program II (OTA 1980, Appendix I,p. 135).
The 9.3 cents per mile is measured one way for the round
trip and assumes a A0.000-50,000 pound payload. For
example, transporting chips from a harvesting site 10
miles to a processing facility would, using this
estimate, cost 9.3 cents/green ton/mile or 93 cents/green
ton for the trip of ten miles. In this example, the
round- trip is 20 miles but the cost of the empty run is
included in that of the loaded run. This estimate is for
a tandem-rear-axle tractor and a tandem-axle trailer
operated one shift per day.

Program II is designed to accommodate one or more
changes in either this cost estimate or the rate of
change of the cost estimate during a time series. For
developmental purposes, an arbitrary real annual
transportation cost growth rate of 5 percent was used in
all model runs. The potential impacts of changes in the
cost and the rate of changes is discussed in Chapter IV.

The output from Program II is the cost of
transporting one ton of chips from each simulated origin

to each destination in the system.

("V
H

..‘e
I!

lie
aeh:

t01

hatp

53

WIILWMW

Biomass stumpage sales simulated in Program I and
the transportation costs computed in Program II are the
input for this step in the Model. The minimization is
accomplished using the MPOS (Multi-Purpose Optimizing
System) software package available on the MSU CYBER 750
(Cohen and Stein 1978). IBM's MP8! and Control Data's
APEX will also work. The primary reason for using a
mainframe optimizing system rather than a microcomputer
algorithm is time savings. Minimization problems as
large as the ones developed in this research
(approximately 500 variables in the objective function
and and 100-150 constraint equations) are solved in under
15 seconds by MPOS. The same problem run on a
microcomputer would, according to the MSU Computer Lab's
senior statistical consultant, require at least 2A hours
to solve and possibly twice that long.

To facilitate translation of Programs I and II
outputs to a format MPOS can understand, an interactive
matrix generator program called MAGEN was developed on
the CYBER 750 (Appendix B). MAGEN permits the consumer
to take advantage of the MPOS matrix format input rather
than using the more time consuming equation format.

If (1) the supply of a raw material at any number
of origins, (2) the amount demanded at any number of

destinations and. (3) the cost of moving one unit of the

to th.
than

'icroi
Frogr
model

Dacka

scare

5A

raw material from each origin to each destination is
input, MAGEN can be used in conjunction with MPOS to
minimize the transportation costs for the entire network.

MAGEN should be particularly useful for teaching
purposes because it permits the use of MPOS without
requiring the student to understand the MPOS program
language. MPOS need only be called, told the name of the
data matrix, and a few general instructions and it can
solve the problem. It permits the focus of the learning
exercise to be on what is happening, rather than the
computational details.

MAGEN was not used for production modeling in this
research for two reasons: (1) it is expensive for large
problems, (2) since all the problems considered here are
similar, a system of creating MPOS equations using the
microcomputer screen editor was developed which permitted
input preparation on the microcomputer and later transfer
to the CYBER. The screen editor input is more efficient =
than MAGEN provided the programmer fully understands the
microcomputer editor, the CYBER editor, linear
programming theory and MPOS. Since most potential
modelers will not be familiar with these software
packages, MAGEN should be useful in the future. A sample
of MPOS input and output and the MAGEN flow chart and
source code are contained in Appendix B.

As microcomputers with larger memory capacity

bi

bi

fit

55

become available, MPOS-like optimizers will undoubtedly
be created for them and interactive mainframe costs will
no longer be a consideration.

The linear minimization of network transportation
costs conducted using MPOS is an example of what are
known as either transportation-type problems or classical
transportation problems. Properties that these problems
possess "enable one to solve them by methods that are
considerably more efficient “ than general linear
problems (Swanson 1980.p.1AA). These properties can be

most clearly stated symbolically. Symbols are defined as

follows:

X(i,j) = biomass shipped from origin (1)
to consumer (j) in tons,

A(i) = biomass harvested at origin (1)
in tons,

B(j) = biomass demanded by consumer (j)
in tons,

C(i,j) = the cost of shipping one ton of
biomass, from origin (1) to
consumer (j) in dollars,

m = number of origins,

n = number of consumers

(Dantzig 1963,p.299)

The objective function of the linear minimization

harvc

harvt

(I)
‘4
u
C 1'
n

the

Thi

bioma
iSUOr
biOna

coast

eQUal

56

is to minimize the cost of transporting a known quantity

of biomass to destinations which require known amounts:

m n
Minimize: z z C(i,J) X(i,j)
i=1 j=l
This minimization must be done subject to four

constraints.

First, all of the biomass available at a given
harvest location must be transported. In other words,
harvesting, but not transporting wood is not permitted.
Symbolically, this constraint can be stated as:

n
X x(1:J) 3 5(1)
i=1
Second, the amount of biomass needed by each of
the consumers in the network must be supplied to them.
This means the model cannot simulate transportation of
biomass only to consumers nearer harvest locations and
ignore those farther away. Every consumer's demand for
biomass must be met during each time period. This
constraint can be symbolically stated as:
m
2 1(1:3) 3 3(3)
i=1
Third, the amount of biomass harvested must be
equal to the total demand for the biomass in each time
period. Symbolically, this is:
m n
z A(i) = z 8(3)
i=1 jsl
Fourth, is a mathematical necessity which simply

reflects common sense. No empty shipments or returning

of biomass from consumers to origins is permitted.

Symbolically, this is:

57

X(i..1) 2 0.0

Constraint number three, that the amount harvested
must be equal to the amount demanded can be circumvented
by introducing artificial destination points in the

initial problem structure (Swanson 1980, p.155).

Timofiimulaoionmmoxonk

Program I, the biomass supply simulator; Program
II, the transportation cost computation; and Program III,
the MPOS minimization of those costs together constitute
one model run. Each model run produces an estimate of
network transportation costs for one year. These
estimates are not particularly meaningful by themselves.
However, as components in a time series, they become more
significant.

Estimates produced in this research cover the time
period from 1982 through 2012. The model produces an
estimate every five years beginning in 1982, for a total ~
of seven estimates per time-series simulation. No
separate computer program was required to produce the
time simulator because Program I and Program II are
written to permit changes in biomass growth and
transportation cost factors based on the passage of time.
As discussed above, running Programs I and II requires
specification of a calendar year. Both programs translate

this input into a definition of passage of time between

58

the specified year and the model base year (1982) and

recompute time sensitive variables as needed.

seri
Hich:

it I(

Chapter IV

Rooulio Q£.M£nolinz

Four series of model runs were made: (1) a series
to develop preliminary statistical parameters for cost
estimates produced in other model runs, (2) a benchmark
series based on only existing wood consumers in the
region with no wood energy plant in the system, (3) a
series simulating a wood-energy plant at Rose City,
Michigan and (A) a series simulating a wood-energy plant

at Idlewild, Michigan.

Pxooioion Qt Eiiinaloi

To obtain preliminary statistical parameters for
the transportation cost minimizations, nine model runs
were made holding all variables, including time,
constant. The only change made between these runs was
the generation of a completely new set of random timber
sales locations each time Program I was executed. These
runs were made using 1982 stand growth and transportation
cost estimates. Results of these runs are summarized in

Table A.1.

59

[\P
a

LA)
I

'5:
cost

60

Table A.1 -- Summary of nine estimate precision runs.

 

 

 

 

 

 

Consumer Mean Estimated 90$ Confidence
Transportation interval.
Cost($)

1. Abitibi 617,558 +/- .0725
2. PCA 619.750 +/- .0923
3. S.D.Warren 357.3A3 +/- .1051
A. Menasha J,299,01A +/- .0371
5. Champion 1,A78,992 +/- .0371
6. Buskirk 3A1,535 +/- .1261

Network Total: A,71A,192 +/- .01A1

' Expressed as percentage of mean estimated transportation
cost. Formula:
mean +/- (t(05.n-1))x(std. deviation/(n'i1/2))

 

 

As Table A.1 illustrates, model estimates are
precise. The high level of precision occurs because many -
Monte Carlo simulations of random timber sales were
completed in Program I. A total of 759 timber sales (an
average of 8A per run) were generated during the nine
full model runs. All possible transportation costs
between these origins and the six existing wood consumers
in the system were computed in Program II. These
possibilities were examined and total network cost

minimized in Program III. Cost estimates in Table A.1

61

are not minimized for each firm. They represent the
average annual cost to that firm when the transportation
cost for the network is minimized. Only the total,
$A,71A.192 has actually been minimized by the linear
programming algorithm.

This precision series demonstrates that estimates
made with this model will be grouped very closely
together if the assumptions are held constant. This
grouping occurs because:(1) Monte Carlo random generation
is used in Program I and, (2) a large number of these
random generations are required to simulate. enough
stumpage sales to meet network demand for one year. This
is an important result because it suggests that little
additional information will be obtained from making
multiple model runs under identical assumptions. In
other words, in this case the Monte Carlo simulation
reduces the advantage normally obtained by using
stochastic estimation. On the basis of this result, it
was decided to develop deterministic estimates for each =
set of assumptions in a given year during development and
testing of the model.

Results of the benchmark series (no wood-energy
plant present at any location) are summarized in Table
A.2. Tables A.3 and A.A present the model output for
simulated wood-energy development at Rose City and
Idlewild, respectively. The adjusted network totals in

Tables A.3 and A.A represent the totals for the entire

62

network, minus the transportation costs of the wood-
energy plant. That is, the adjusted total is the

transportation cost estimated for the existing network.

 

Net
Tot

63

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table A.2. Results of Benchmark Series (Series I).($)
Firm 1982 1987 1992 1997
1. Abitibi 377733? 830.327 W 1,A8A,28A
2. PCA 619.750 1,201,A89 792.A22 975.5A7
3. S.D.Warren 357.3A3 603,231 262,613 571,7AA
A. Menasha 1,299,01A 1.793.A16 1,836.780 2,A5A.500
5. Champion 1,A78,992 1,A1A,269 2,AA6.789 1.932.9A5
6. Buskirk 3A1.535 537.581 A31.A39 666.300
Network
Total A.71A.192 6.380.313 6,AO9.8A2 8.085.A2O
(Table A.2 continued)
Firm 2002 2007 2012 Firm
Total
1. Abitibi 1,273.95A 1.19A.755 1.166.928 7.207.605
2. PCA 1.133.399 1,038,0A3 1,632,677 7.393.A27
3. S.D.Warren 832.695 572.728 2,033.8A7 2,033.8A7
A. Menasha 1,996,03A 2,721,575 3.182,3A5 15,283,66A
5. Champion 3.999.351 2.768.955 3.870.913 17.912.21A
6. Buskirk 561,810 2A5.603 589.922 3.37A,190
Network
Total 9.797.2A5 8.5A1,659 12,A76.63A 56.605.301

 

 

6A

Table A.3. Results of Rose

City Series (Series II).($)

 

 

Firm

 

6.
7.

Abibiti

PCA

S.D.Warren

Menasha
Champion

Buskirk

Wood-energy

 

Network
Total (raw)

 

Network
Total (adj.)

 

1.
2.
3.
A.
5.
6.
7.

Abitibi

PCA

S.D.Warren

Menasha
Champion

Buskirk

Wood-energy

1982

A6A.A58
A29.983
197.527
1.115.101
1.632.671
269.507
8A0.925

A.9SO.172

A.109.2A7

 

 

Network
Total (raw)

 

Network
Total (adj.)

754.913
1.70A.639
260.A92
2.058.872
2,A69.277
558.331
1,A75.881

 

9.282.A05

7,806.52A

1

1987

 

655.901
511.187
109.037
1.365.095
2,003.99A
17A.AA8
1.661.031

6,A80.693

“,8199662

1992

1.166.727
832.195
525.631

1.776.806

2,275.666
AA6.665

1.298.A53

3.322.1A3

7.023.690

1997

 

955.1A5
81A.758
53A.70A
1.59A,878
2.582.555
236.567
2,09A.A31

8,813,038

6.718.607

Firm
Total

 

 

 

1,0A1,063
1,08A.573

3A3.A71
2.362,228
A.299.307

A10.088
2,153,690

1.7A3.5A0
82A.009
587.295
2.987.228
A.7A9.266
530.751
1.766.697

6,781.7A7
6,201,3AA
2.558.157

13,260,206

20.012.696

2.625.357

11,291,108

 

 

1.69A.A20

9.5A0.730

11.A22.089 51.nuo,507

 

 

Table A.A.

65

Results of Idlewild Series (Series III).($)

 

 

Firm

 

7.

Abitibi

PCA

S.D.Warren

. Menasha

Champion

Buskirk

Wood-energy

 

Network
Total (raw)

 

Network
Total (adj.)

 

5.
6.

Abitibi

PCA

S.D.Warren

Menasha
Champion

Buskirk

1982

396.306
767.285
27A.617
1.306.766
1,A8A.0A7
528,677
1.901.A6A

6.659.162

A.757.698

 

1.079.259
790.755
288.799

1.655.631

3.132.336
A59.913

7. Wood-energy 3.766.10A

 

Network
Total (raw)

 

Network

Total (adj.)

 

11.172.797

7.A06.693

 

 

 

 

 

 

 

 

1987 1992 1997
1,097,0u9 810.358 56A,712
1,072,188 961,285 1,18A,675

.368,206 375.A83 588,105
1,630,235 2,23A,817 1,726,A26
1.569.12A 2,261,787 2,825,379

628,205 916,651 75A,072
8.208.956 9.755.3A9 11,073,100

2007 2012 Firm

Total

925,1uo 2,110,253 6,983,077
2.070.791 2.372.835 9.219.81A

400,220 870,029 3.165.A59
2.53A.A38 3.703.170 1A.?91.A83
3.463.896 3.205.203 17.9A1.772
1.726.3A7 9A6.065 5.959.938
2,781,9A0 5,152,517 21,0A0,669

13.902.773 18.360.073 79.102.ZOA
11,120,833 13,207,556 58.061.535

 

 

of:
firn
teta

year

66

QomoaniooanSoniooIJIanoIII...

Many trends appear when Tables A.2, A.3 and A.A are
compared and it is possible to analyze them without
addressing the question of whether the trends are
statistically significant or not. More meaningful
interpretations can be made with probability theory. It
is not possible to compare individual cells within the
tables, for example a particular firm in a given year,
because the cells themselves are deterministic and have
no statisitical parameters. However, it is possible to
statistically compare the three series results in terms
of: (1) individual firm costs for all years, (2) all
firms' costs taken together for one year,(3) each firm's
totals over all years and (A) network costs for all
years. Meaningful combinations of these four types of
comparisons were tested for statistical significance
using the paired Student's-T method (Table A.6). For
example, the Series I (Benchmark) costs from Table A.3.
and the Series II (Rose City) costs from Table A.A which
accrue to firm number one (Abitibi) from 1982 through
2012. is shown in Table A.5.

The costs appear to be generally lower in Series II
than Series I and the total of the seven modeled years is
definitely lower for Series II. The question is, are
these differences statistically significant? While it is

not possible to test each annual cost estimate or the

00!

W00

ind

l l

67

grand total, it is possible to use paired T-testing to
compare Series I with Series II (and also Series III,
wood-energy plant at Idlewild, Michigan) as a whole for
individual firms.

Table A.5--Example of comparison between Benchmark and
Rose City for a single firm, Abitibi.

 

 

Bencimark RoseICity
1.91.1: com-4.11 92:11.1).
1982 617.558 A6A.A58
1987 830.327 655.901
1992 639.799 1.166.727
1997 1,A8A,28A 955.1A5
2002 1.273.95A 75A.913
2007 1.19A.755 1.0A1,063
2012 1.166.928 1.7A3.5A0
Total: 7.207.605 6.781.7A7

 

 

For example, comparing Series I for Firm 1 and
Series II for Firm 1 produces only a 27 percent
probability that they are in fact significantly different
data sets. That is, there is 73 percent probability of
stating they are different, when in fact they are not
(Type I error). A similar test for Series I, Firm 1
against Series III, Firm 1 produces similar results: an

89 percent probability of concluding they are

(a
y J».
an

sort
data
alte
test.
Cont.
fume.
Pith.
°°Apz

1eact

68

significantly different, when in fact they are not.

As discussed above, this testing procedure was used
to separate statistically significant differences
between the three Series from apparent, but statistically
insignificant differences. Obviously, the level of
confidence is very important in this kind of a test. In
evaluative research, values of 90 or 95 percent
(cumulative distribution function divided by 2) are
preferred for most T-tests (Sokol and Ralph 1970,p.133).
By contrast, the research discussed in this report is
essentially developmental; its purpose being to create and
test a bioenergy industrial location model. Within this
context, probability theory is most useful as a tool for
sorting meaningful results from the mass of meaningless
data, rather than as a definitive comparison of competing
alternatives. Therefore, developmental applications of T-
tests need not be as sensitive. Consequently, the minimum
confidence level (defined as the cumulative distribution
function divided by two) discussed in this report is .8,
rather than .9 or .95. Table A.6 presents all meaningful
comparisons between Series I, II and III which have at

least a .8 confidence level.

I~e
11.5

11.7

69

Table A.6-- Significant differences between results of
Series I, II and III.

 

 

Notxonk Gomoanison

One Minus

 

 

Network.Toials.£or.Sonioo. Probability of Type I
Error
I II! .85
II' III' .98
II - III 1.00 (.996)

' Adjusted network total=(Network total)-(wood-energy
plant costs)

Inoixioual Finn Qomoozioon

One Minus

 

 

 

Series-Firm vs. Series-Firm Probability of Type I
Error
II-2 III-2 .82
I-3 II-3 .89
I-3 III-3 .88
II-3 III-3 .83
I-A II-A .96
II-A III-A .87
I-6 III-6 .89
II-6 III-6 .97

II-7 III-7 .98

 

 

70

WAWHWMW
Network,xotals

Three significant results were produced by
comparing Series I,II and III network totals:

1. Existing consumer network will probably (.85 ')
have lower transportation costs if a wood-energy
plant is built at Rose City than if no wood-
energy plant is built at all.

2. Existing consumer network, not including a wood-
energy plant, will probably (.89) have lower
transportation-costs if the wood-energy plant is
built at Rose City. than if it is built at
Idlewild.

3. Consumer network, including the wood-energy
plant, will probably (.996) have lower
transportation costs if the wood-energy plant is
built at Rose City than if it is built at»

Idlewild.

The first result, that the existing wood consumers
will probably have lower raw material transportation
costs if a wood-energy plant is built in Rose City than

if no wood-energy plant is built, is based on a

' Significance level = 1 - (probability of Type I error.)

exi

Iii):
7982
1987
1992
1997
2202
2007

2012

//

71

statistical comparison of the Series I network totals for
all seven model years (Table u.2) with adjusted network
totals in Series II (Table b.3). This adjustment of
Series II and III network totals is needed to produce
existing network totals for these two series that do not
include the wood-energy transportation costs. The Series
II adjusted network cost is below the Series I cost for
five of the seven model years. The total savings is
$1,956,234, which translates to $6,2h1,002 over 30 years
and $808,033 per year.

This result contradicts intuition which suggests
that the net effect of adding an additional consumer to

the network should produce higher transportation costs

Table n.7-- Number of timber sale simulated in Series I

 

 

and II.

1132 32:12: I §££1£1.II Difference
1982 83 99 15
1987 69 84 20
1992 69 79 10
1997 53 73 20
2002 51 60 9
2007 33 59 16
2012 R6 52 6

Total: 96

 

 

rep

com]

con:

Idle
3tat
(adj!

with

72

for the network rather than lower. One hypothesis which
explains this contradiction involves the additional
number of timber sales required to meet network demand
including a wood-energy plant (Table u.7).

The hypothetical explanation suggests that the
additional 96 timber sales added to meet the wood-energy
plant's raw material requirements permits more efficient
allocation of sales to the other six consumers. This is
directly analogous to more suppliers coming into a market
in response to a perceived increase in demand. It
represents a shift in the supply curve to the right .

The second significant result obtained from
comparing network totals is that existing pulp and chip
consumers would probably experience lower raw material
transportation costs if the wood-energy plant were built
at Rose City (Series II) than if it were built at
Idlewild (Series III). This finding is based on the
statistical comparison of the Series II network costs-
(adjusted to remove the costs of the wood-energy plant)
with the Series III network costs (also adjusted).
Testing indicated a 981 probability that the lower Series
II costs represent a significant savings compared to
Series III (Table 4.8).

This prediction means that, in terms of the raw
material transportation costs of the existing network of

consumers, Rose City is a better location for the wood-

ener
aunt
expe
buil

F8811

Year

1982
1987
1992
1997
2002
2007

2012

Total

73

energy plant than Idlewild. In conjunction with result
number one, that the existing network as a whole should
experience lower transportation costs if the plant is
built at Rose City than if it isn't built at all, these
results suggest that existing consumers should encourage

Table 4.8 -- Adjusted. network transportation
costs for Series II and III.($)

 

 

 

 

 

 

 

 

Year Rose City Idlewild Savings at

Rose City
1982 4.109.247 4.757698 648.451
1987 4.819.662 6.365.007 1.545.345
1992 7.023.690 7.560.381 536,691
1997 6.718.607 7.673.373 954.766
2002 7.806.524 7,406,693 -399.831
2007 9.540.730 11,120,833 1.580.103
2012 11.422.089 13.207.555 1.785.467
Total: 51,440.549 58,091.541

6,650,992

5 wood-energy plant raw material transportation costs
removed for comparative purposes.

 

 

wood-energy development at Rose City. There may be
untested locations which are better than Rose City by
this criterion.

Consumer network raw material transportation costs,

including wood-energy plant costs, at Rose City will

74

(.996 probability) be lower than the same costs with the
plant at Idlewild (Table 4.9).

The $16.4 million total difference in Table 4.9 is
only for the seven years actually modeled. Because these
estimates are representative of the 30-year period as a
whole, it is reasonable to expand estimates to fill in
missing years. The mean difference in Table 4.9 is
$2,342,936 (standard error $1131). Therefore, the
Table- 4.9 -- Network raw material transportation costs

with ”a wood-energy plant at Rose City compared to
similar costs at Idlewild.($)

 

1'

Year Rose City Idlewild Difference

 

 

 

1982 4.950.172 6.659.162 1.708.990
1987 6.480.693 8.208.956 1,728,263
1992 8,322,143 9.755.349 1.9332205
1997 8.813.038 11,073,100 2,260,062
2002 9.282.405 11,172,797 1.890.392
2007 11,694,420 13,902,773 2,208,353-
2012 13,188,786 18,360,073 5,171,287

 

 

network should expect approximately $70.3 million savings
in raw material transportation costs if the plant is at
Rose City rather than Idlewild.

Two more major questions were answered using the
model: (1) what will the network costs be including the

wood-energy plant, and (2) which individual consumers

75

stand to gain or lose from wood-energy development? The
first question is essentially a public policy analysis
problem: where should public agencies encourage wood-
energy development in terms of minimized raw material
transportation costs? The second question, who among the
existing consumers will benefit and who will pay for wood
energy is obviously important to the operators
themselves. In addition, this question has a public
policy dimension because each of the consumers is a major
employer and therefore a locally important source of
social stability.

For the individual consumer, the important
question is, to whom would these savings accrue? while
the model developed here did not address all of the
subjects necessary to provide a comprehensive answer, it
did produce some insights within the context of the three

Series tested.

 

 

r:-

(I.

'16

1'!

(e

76

mmummmmnmmm
W

Testing all meaningful combinations of individual
consumer costs in Series I,II and III produced nine
statistically significant results (Table 4.6) and several
times as many statistically insignificant results. In a
general sense, these statistically insignificant results
can be thought of as results which this research might
have produced (because it dealt with these subjects), but
didn't. They are important because, in many cases there
is some evidence produced that tends to support them and
because of this, they may provide hints for future
research. However. regardless of how convincing the
evidence seems, since statistical testing indicates they
were not significant, they should not be interpreted as
results of this research. Each of the nine significant
results concerning individual consumers are discussed
below on a firm-by-firm basis.

mmuwwmm
WWW
The PCA would probably (.82) have lower raw

material transportation costs if a wood-energy plant were
built in Rose City than if it were built in Idlewild. 0n
the other hand, the difference between the PCA costs in

the Benchmark Series (Series I) and either the Rose City

77

Series (Series II) or Idlewild Series (Series III) was
not significant. However, if a plant is to be built at
one of the two tested locations, they should prefer Rose
City. The apparent reason is that, at the Idlewild
location, the wood-energy plant apparently competes too

effectively with the PCA for nearby low cost stumpage.

£12m numhsn.31.Sini.Earrsni.hnskssonx.hichisan

Three significant results were reached concerning
S.D. .Harren. They will : (1) probably (.89) have higher
raw material transportation costs if the wood-energy
plant is built in Rose City than if it is not built at
all, (2) probably (.88) have higher raw material
transportation costs if the wood-energy plant is built in
Idlewild than if no plant is built, and (3) probably
(.83) save more money on raw material transportation
costs if the plant is built in Rose City than if it is
built in Idlewild. Based on these results S.D. Warren
should: (1) discourage wood-energy development in either”.
location, (2) if forced to pick one, prefer the Rose City
location.

S.D. warren is located near the southern edge of the
heavily forested part of Michigan's Lower Peninsula and
the edge of the study area. They presently compete
closely with Buskirk,Inc. in Paris and Menasha, Inc. in
Otsego for raw material. It seems likely that the reason

the results of the modeling develop as they do is that

78

S.D. warren is too close to its competition to benefit
from the new stumpage supply drawn into the market by the
'wood-energy plant. whether the plant is built at Rose
City or at Idlewild. If a plant is to be built at one or
the other location, they should prefer Idlewild because it
is farther away from Muskegon.

Kim number. 4:. WW 95.8389... Michigan

Two significant results were reached for the
Menasha Corporation: (1) they would probably (.96) have
lower raw material transportation costs if the wood-
energy plant were built in Rose City than if it were not
built at all, and (2) the Rose City location is probably
(.87) better than the Idlewild location.

Menasha consistently, because of their location
outside the study area, has the highest cost per
delivered ton of raw material (Table 4.10).

Development of wood-energy in Rose City (Series II)
would, according to the model, stimulate enough new--
timber sales to allow Menasha a significant reduction in
raw material transportation costs without being in a
position to compete with Menasha for nearby stumpage.
However, when the location of the wood-energy plant is
Idlewild (Series III), the beneficial effect of the new
stumpage locations is offset by the increase in

competition between the two for the nearby stumpage.

79

Table 4.10 -- Mean annual projected transportation cost
per ton of biomass 1982-2012 (S/ton).

 

 

Consumer I II III
1. Abitibi 9.18 8.64 8.90
1* 1
2. PCA 10.14 8.94 12.63
2 2 2
3. S.D. Warren 11.91 5.84 7.22
3 3.4 4
4. Menasha 37.22 32.29 36.09
5. Champion 9.81 10.97 9.83
5 6 5.6
6. Buskirk 9.24 7.19 16.33
7 7
7. Wood-energy ---- 7.92 14.73

' Superscript digit indicates statistically significant
difference (above .8) with paired value.

 

 

mmnmmmmm
Paris, Michigan is located only 20 miles from

Idlewild, Michigan. Simulated development of a wood-.-
energy plant at Idlewild indicated Buskirk would probably
(.89) have higher raw material transportation costs than
if no plant were built (Series I). In addition, model
results also predict Buskirk would probably (.97) have
lower transportation costs if the plant were located in
Rose City (Series II) than if it were located in Idlewild
(Series III). It seems likely that this occurs because

Buskirk is in a position to take advantage of the new

annua
be 3
Farce
at I.
with

StUmp.
There

neces:

80

stumpage supply generated by the wood-energy plant
without having to compete so directly with it for nearby
stumpage. Buskirk's results, unlike those obtained for
the Menasha Corp., do not support the hypothesis that
Buskirk should favor wood-energy development at either
tested location. On the contrary, the results do suggest

that Buskirk should oppose development in Idlewild.

MWBWM

.A wood-energy plant in Rose City would probably
(.98) have lower biomass transportation costs than if it
were in Idlewild. Table 4.10 indicates the ‘average
annual cost per delivered ton between 1982 and 2012 would
be $7.92 at Rose City and $14.76 at Idlewild. an 86
percent increase. The reason for the difference is that,
at Idlewild, the plant would be competing more directly
with Buskirk, S.D. warren, the PCA and Menasha for
stumpage because they are relatively close to Idlewild.
There are no competitors near Rose City so it is not»
necessary to travel as far to get stumpage.

Optimal solutions for individual firms can be
obtained if the firm is treated as a network. The
objective function of the linear optimization then
becomes minimizing the sum of the transportation costs
for the firm, rather than the networks as was done in
this research. Since most firms are sufficiently large to

require rather complex wood procurement programs this

costs
indiv
likel;

Optin.

81

would probably not produce trivial solutions. An example
of the type of situation in which this optimization would
provide useful information is if a firm has 20 harvesting
locations, 6 long distance truck-loading decks and 5
different cost/ton/mile factors (functions of either
individual trucks or subcontactors). In this complex
environment. the linear minimization of transportation
costs would be helpful. In general, the more complex the
individual firm's wood procurement system is, the more
likely it is to benefit from transportation cost

optimization.

desi
ener
size
as
Alth
may
Yer:
finer
the

Gene

ind:

deVG

COHf

Prod
the

9f

82

Armament Inflnnil Results Kiln LBS: Than .8
Significance Lfilgl

Because the purpose of this research was to develop
a modeling system rather than evaluate alternative
models, large sample sizes were not particularly
desirable. Consequently, many results which might have
emerged as statistically significant with larger sample
sizes. are just below the minimum level of .8 (calculated
as cumulative distribution frequency divided by 2).
Although these are not statistically significant, they
may be important because they suggest trends which may be
verified if larger samples are taken. Four such trends
emerged in the results of this research: one concerning
the PCA, two for Champion International, and one
concerning Buskirk.

Comparing the PCA results in Series I and II
indicates they should perhaps (.63) encourage wood-energy
development at Rose City. Since they are a major factor
in the regional industry and .63 indicates a reasonably
strong trend, a larger sample could provide valuable
confirmation of this trend.

Champion International in Gaylord, the largest
production capacity facility in the region (and one of
the largest particle board plants in the world, capable

of producing 365.000 tons of board per year) had no

statist:
are a
such tho
trend:
from v<
Chaspio:
energy
higher
additioz
plant 1
Idlewild
differen
SUbJectq
Chaspior
t° 3UP;
discoure
Bx
materia:
City t1

II), 7

 

3amPle

Pesuits
high”

built

 

83

statistically significant results. Like the PCA, they
are a major regional influence on the industry and as
such their perspective on wood-energy is important. Two
trends emerged which indicate Champion might not benefit
from wood-energy development in either tested location.
Champion's estimated transportation costs with the wood-
energy plant in Rose City (Series II) might (.56) be
higher than if no plant is built (Series I). In
addition, they might (.67) expect lower costs if the
plant is built in Idlewild (Series III) and that the
Idlewild costs would be almost the same (probability of
difference is .014) as the no-plant costs.
Subjective comparison of the Series I ,II and III
Champion costs indicates they have no particular reason
to support wood-energy at Idlewild and should perhaps
discourage it at Rose City.

Buskirk, in Paris, might (.77) have lower raw
material costs if the wood-energy plant is built in Rose a
City than if no plant is built at all (Series I, Series
II). This is so close to the .8 threshold that a larger
sample would almost certainly produce significant
results.

The results of Series I,II and III indicate
Abitibi. Inc.. in Alpena, probably would not experience
higher or lower transportation costs if the plant is

built in either tested location. All Abitibi

s‘gniff

probabl
develop
due tc

region.

 

 

84

significance levels were below .3. On this basis, they
probably should have little concern over wood-energy
development at these locations. This result is probably

due to their relatively isolated location within the

region.

85

£211: 91.ncdslins

A model of the type developed here can be run using
a wide variety of computers. with the recent
introduction of microcomputers with over 300K
(approximately 300,000 bytes) of memory, it is possible
to run the entire model with no time-sharing costs at
all. However, a trade off may exist between time and
money. For example, it may be faster to use an existing
mainframe optimizing package to solve large linear
programs than to write and operate a similar
microcomputer package. However, time-sharing costs may
make the mainframe too expensive. Obviously. the expense
of modeling is, to a large extent, dependent on such
factors as hardware availability, the rate structure of
available time-sharing facilities and the skill and
backgrounds of programmers available. The large number
of possible combinations of administrative factors,
equipment and personnel precludes detailed prediction of
modeling costs under all circumstances. The development
costs discussed below are presented in the hope they may
help prospective modelers plan their research.

Programs I and II were run on a microcomputer,
therefore no time-sharing costs accrued. However, there
are other costs which occur in microcomputer programming

such as equipment purchase and maintenance which were not

86

included in this analysis. Program III, the minimization
of the transportation costs was conducted on the MSU
CYBER 750 mainframe using the MPOS optimizing package.
Final analysis and statistical testing was done using
small programs on both the microcomputer and the
mainframe which did not appreciably increase the cost.
These are modeling costs, not inventory costs. As
discussed in Chapter I, the biomass inventory on which
the modeling is based is a complex subject. Appendix A
contains a detailed discussion of the the requirements
for such a project.

Programs I and II occupy approximately 50 K of
memory when compiled. Program I requires a large random
number file and the inventory data base to execute.
These data files occupy 15K and 200K. respectively. on a
2-80 based microcomputer.

Program I, the timber sales simulator, took twenty
to thirty minutes to execute. Most of this time the.
program was reading through the inventory data base. A
typical linear program created by Programs I and II had
475 variables in the objective function and 85 constraint
equations. This size linear program required 62.080
(171.200 octal) words of CIBER memory, took six seconds to
solve, and cost roughly ten dollars when run at batch job
(low cost option) rates. The largest problem required

119,040 words (350.400 octal), took 15.5 seconds and cost

87

twenty dollars to execute. Nine such programs were run to
obtain statistical parameters for the network
minimization, and seven each for Series I,II and III, for
a total of 31 production runs. These runs cost roughly
four hundred dollars for memory time and required another
four hundred and fifty dollars of connect time and
administrative charges (e.g. mounting tapes and disk file
storage). Less well documented costs such as debugging
and various special program development costs (notably,
creation of the matrix generator program MAGEN) required
approximately five hundred dollars. The total mainframe
cost was approximately $1850. A substantial, but unknown
amount. was saved in both development and production runs
by creating most mainframe jobs on the microcomputer and
transferring the finished file to the CIBER, rather than
creating these files with the CYBBR editor. Technical
consultations were available at no charge through the MSU
Computer Laboratory. In terms of time required for model
operation, once models and hardware are configured and.
operating smoothly. three to four man hours are required

to complete one run.

ver
sic
mic
the

The

the

rea.
r001
arr:
But:
vitk
hece
out;
disk

avg?

88

WWHW

The two problems encountered during this research
were both caused by the limited memory capacity of
microcomputers used for Programs I and II. Because the
microcomputer memory was 64K it was not possible to have
the output of Program I read directly into Program II.
The relatively large number of computations required in
the simulation and the extensive input/output needed to
read through the inventory data in Program I. left no
room to dimension output arrays. In addition since the
arrays used in Program I are reused for simulating
subsequent stumpage sales, they could not be destroyed
with an 'in-place' output file. Therefore, it was
necessary to use intermediary disk files to transfer the
output from Program I to Program II. This increased the
disk read/write time by a factor of four and caused an -
average delay of 30 minutes in the execution of Programs
I and II.

The second problem was due to the number of bits
(on/off elements of memory) allocated to each byte (basic
unit of memory) by the microcomputer. The Z-80
microprocessor allocated eight bits per byte. Because of
the size of arrays required in the model, it was not

possible to assign more bits using FORTRAN's double

preci
occur
neede
consu
hecau
of P
the c
of

circu
by c
becau
readi
minin
also:
using
it tr
hits,
1000:
this
nicrc
main,
elimi

with

89

precision capability. Therefore, rounding errors
occurred in summation of the large amount of biomass
needed to meet the demand of the entire network of
consumers for one year. Correct summation is essential
because, unless the amount of biomass in a complete run
of Program I exactly equals the amount demanded by all
the consumers, no solution exists for linear minimization
of the transportation costs. It is possible to
circumvent this problem within the minimization program
by using a two step linear minimization. However,
because software for this two step algorithm is not as
readily available as software for the one step
minimization, it was decided to keep the one step
algorithm and adjust biomass tonnage totals for each run
using a small mainframe program. Both problems disappear
if the newer generation of microcomputer (those having 16
bits/byte and much larger memory capacity ranging up to
1000K) is used. On the basis of the experience gained in

this research, one should either use one of these newer I
microcomputers having at least 128K of memory or a
mainframe for this size model. Either option would
eliminate both the time and accuracy problems experienced

with the 2-80 microcomputer used in this research.

Chapter V
mmmnnnnmmm

Much of the value of this model as a
decisionmaking tool depends on how realistic it is. As
discussed in Chapter II, the primary constraint on this
realism is that some model components were simulated or
held constant during development and preliminary testing.
The purpose of this chapter is to explain how two of the
most -important of these may be altered to improve the
model's performance and to discuss decisions regarding
shock variable usage and stochastic estimation. The two
model components discussed are: (1) forest biomass
inventory and growth and (2) the regional transportation
network. In addition, this chapter provides an estimate
of the cost of applying this model to the study region
and discusses its applicability to crop biomass and urban

solid waste processing and disposal location decisions.

WWWAMMW

Estimating forest biomass represents a relatively
new challenge and therefore few studies have been
published (Kittleson 1979). In general, biomass
inventory is similar to traditional timber volume
inventory with the obvious exception of the special

physical problems encountered in actually weighing entire

90

91

trees. After individual-tree weight estimates are
developed, they can be expanded to stand weight estimates
using the same expansion factors developed for timber
volume estimation. Without the use of remote sensing,
further accurate expansion of these stand estimates to
entire forests or regions has been prohibitively
expensive. The preliminary research necessary to conduct
such a remote sensing-based expansion within the study
area has been completed during the first phase of this
research (Appendix A).

This research focused on expanding field estimates
of forest stand biomass to estimates“of the biomass of
stands as they appear on enhanced Landsat satellite
images. Based on the results of this research, it is
possible to produce a biomass map of the entire region
using Landsat imagery. As discussed below, the expense
and time required for such an inventory do not appear
unreasonable. Actual application of the model to the
region would require this type of complete biomass
inventory, as opposed to the preliminary samples used
during model development.

A more difficult obstacle to successful application
of the model. especially over longer periods of time is
estimating forest growth over large areas. While
identifying and measuring forests on a regional scale
has been made possible with remote sensing, no such

technical solution has emerged for predicting how fast

92

the forest will grow, nor does one seem imminent. For
developmental purposes, statewide growth estimates from
the MDNR were utilized. However, because of the low
quality of timber and poor growing site conditions which
typify the stands which are ideal for wood-energy,
statewide growth estimates are probably too high.
Several growth models have been developed for forests in
the Great Lakes area. With one exception, they can not
be applied to an area as large as the study area because
they require too much detailed information. The
exception is a generalized system developed recently
which requires only site index in addition to timber
parameters (Lundgren and Essex 1979). The combination of
the Lundgren and Essex system and the preliminary research
discussed in Appendix A provides an opportunity to include
site index estimates for each of the Landsat forest types
to produce growth estimates of wood-energy resources.
Better estimates would probably be produced if these-
growth projections were stratified based on major soil
groups in the region because it appears soil type is an
important factor in determining biomass, particularly as
the site index becomes critically low (OTA 1980

Vol.II,p.16).

93

WW

As discussed in Chapter II, transportation cost
estimation during model development was based on the use
of a general statistical factor (the RTF) which was used
to expand straight-line distances within the region to
estimated road-distance. This system will produce
reasonably accurate transportation costs if roads are
evenly distributed throughout a region. However, if roads
tend to be concentrated in some areas and scarce in
others as happens when there are major urban areas within
a region, model estimates for facilities operating in
heavily roaded areas will have unrealistically high
transportation costs while those in more rural locations
will have low transportation cost estimates. In an
application of the model, it would be desirable to
stratify the region and use several expansion factors
(RTFs) to account for this variability. In most cases,
this will probably be the only solution available.

In Michigan, however, a better alternative is
available. The Department Of Transportation maintains a
complete transportation network data base which includes
roads for which they have primary responsibility and
those for which state or federal funds have been spent.
It would be very desirable to incorporate this data base
into the model. The State data base was not included
during model development because the microcomputer

software used in model development is not compatible with

94

the data base software. However, according to the MSU
Computer Laboratory, this problem can probably be solved
by: (1) dumping the data base on a computer tape in a
generalized format. (2) loading the tape on the MSU
CIBER. (3) writing a program to convert the generalized
format to the CIBER format and (4) accessing the data
base from a microcomputer while it resides on the CYBER.
If this sequence works, it would be productive to
obtain improved estimates of the cost per mile per ton of
biomass from truckers in the region. If it does not
work. improved cost per mile estimates from truckers
mile would probably not’improve the model accuracy. To
produce accurate estimates of the actual costs, it would
be necessary to gain cooperation of the forest products
industry in the region. Their actual costs and, equally
importantly, components of these costs could then provide
the basis of the cost estimates. These components would

then become shock variables.

NW

A shock variable can be broadly defined as any
variable that is changed for the purpose of studying the
response of the model to change. Because the model's
optimization occurs within a repetitive (but not
recursive) time simulation, any variable in the model is
a potential shock variable for the next time period.

This would introduce an error compounding recursive

95

structure which would need to be monitored during
successive model years. In addition to this error
compounding problem, meaningful analysis of shock
variables often requires high data precision levels
because the internal impacts of the shocks must be
monitored as well as the output of the model itself.

In addition, new assumptions can be included in the
model in the form of new variables, which can produce an
entirely new range of values for a particular shock
variable. Two pivotal variables which should be used as
shocks in a production application of this model are the
transportation cost per ten per mile, discussed above,
and the forest growth factors discussed in the inventory
section of this chapter.

Assumptions made concerning individual components
of the transportation costs are particularly important
and complex. Therefore, they must be made explicit. As
an illustration. the price of diesel fuel for trucks is a»
major factor in determining transportation cost per ton
per mile. Within the last decade this price has been
extremely unstable because petroleum prices have
fluctuated a great deal. In general. all assumptions
should be made as explicit as possible in model
documentation. Special care should be taken to explain
assumptions made concerning shock variables. They must

be defined and in addition, because by definition values

96

of these variables change during the course of model
execution, functions defining these changes must be
defined. The direction. quantity and time of changes in
each shock variable must be specified clearly prior to
model execution to make possible meaningful

interpretation of results.

Sisshasiic.fisiimati2n

The decision to produce estimates with statistical
parameters for a particular variable in any given
application depends to a large extent on the
application's context. Two general observations may be
of value in this decision:(1) stochastic estimation is
important to the extent that results of modeling will be
a primary basis for decisionmaking and (2) random number-
based simulation tends to obviate stochastic estimation.

The latter situation occurs because estimates based
on large random samples are extremely precise but their
precision is a function of the data's randomness, rather-
than the precision of the estimator. As a rule, random
simulation is less expensive and is therefore tempting
when budgets are of concern. If simulation (using a
random number generator) rather than stochastic estimation
(using repeated simulations to obtain a distribution of
outputs) were used in a decisionmaking-directed
application of this model, it would be necessary to at

least statistically test several sets of randomly produced

97

data against sample field data to verify that they are
sufficiently representative. If, on the other hand,
stochastic estimation is chosen, it must be kept in mind
that. because the model is large and complex, each run
requires much time and effort. Consequently, a tradeoff
emerges between stochastic estimation. an accurate and
expensive approach and random simulation, a less
expensive but probably less accurate approach in most
cases.

.One way to approach this question is to present
decisionmakers with a series of three time/cost options:
(1) emphasizing stochastic estimation, (2) combined
stochastic estimation and random simulation, (3)
emphasizing random simulation. They could then select
the option which most closely met their needs.

Each of these decisions, from forest inventory
through transportation cost estimation and shock variable
use to decisions about estimator distributions, should be
made by the decisionmakers who will use model results
whenever possible. The more often decisionmakers are
included in these types of decisions, the more confidence
they will have in the model output.

In addition to involving decisionmakers in the
modeling as much as possible, it is desirable to try to
keep the model separate from the modeler in the eyes of
decisionmakers. It is quite obvious to the modeler that

the model's integrity, consistency and accuracy, as well

98

as its more obvious attributes such as speed of
computation, are distinctly different from his own. On
the other hand, the only knowledge decisionmakers would
normally have about the model must come from the modeler.
Therefore, the modeler and model are closely associated
from the decisionmakers' perspective. Obviously, to some
extent this is inevitable. The point is that when ever
possible, decisionmakers should be encouraged to focus
directly on the model itself. On the other hand, how
well a decisionmaker understands a model and its
assumptions, and how well he works with a modeler, is
probably as important as technical aspects of the model
application in determining the ultimate quality of the

decisions made using it, perhaps more so.

£2££.Q£.A22111n£ In: Madel.£on.£lannins

The cost of applying the model for decisionmaking
within the study area depends to some extent on which of-
the suggestions discussed above are included in the
application. In general, the best application would
include as many of the suggested additions as possible.
Cost of applying the model to the study area in the
Northern Lower Peninsula of Michigan can be divided into
three general catagories: biomass inventory, obtaining
forest products industry inputs and the cost of modeling.

A biomass inventory of the region would cost roughly

99

$11,000 in 1983:

- Landsat imagery $2,000
- Biomass mapping $5.000
- Site index factor experiments $2,000

- Computerized data base creation

and checking $2,000

$11,000

Obtaining inputs from the regional forest products
industry would cost approximately another $2,000. These
inputs would include: each plant's total requirements
subdivided by tree species (to facilitate comparison with
wood-energy). their recent past transportation costs and
projected transportation costs, and their harvest
schedules for as far into the future as possible.
Unfortunately. it is unlikely that operators would be
able to supply location specific harvest schedules more
than one year in advance. This is because in the past
operators made longer term contracts and were left with-1
too much wood when they reduced production. As a result,
now they are inclined to buy wood for the purpose of
keeping their plant yards full rather than to meet
projected production levels. This is a costly change
because it places operators at a disadvantage in
bargaining with wood growers. A plant's need for wood
can be easily determined simply by estimating the amount

in their yard. For purposes of modeling, their recent

100

stumpage purchases and harvesting schedules on their own
land are likely to be the best information available from
the industry. Even if this is true and no other data is
available, this input can be used to modify random
location simulations. A modified random simulation has
the important advantage of permitting construction of
spatial demand cones around individual operators.
Graphic presentation of these cones would make a major
improvement in the interpretability of the model's
assumptions and results.

The third major cost catagory, modeling, would cost
from $8,000 to $10,000 using the model developed in this
research. The breakdown of these costs would be very
similar to those experienced in model development and are
discussed extensively in Chapter III. On this basis, the
total cost of the application is estimated to be $21,000
- $23,000. This estimate, which includes both personnel
and supply costs, is based on five assumptions:

1. no subsidization of any kind occurs,

2. utilization of existing remote sensing,
cartographic and computing facilities (no
additional hardware or software needed).

3. no training of personnel beyond
familiarizing them with the existing
equipment and facilities,

4. no overhead costs are incurred,

101

5. technical specifications of the inventory
would be identical to those detailed in
the Appendix A discussion of developmental

research specifications.

In terms of time, this application, would probably
require four to eight months. . In addition, MSU is the
only facility in Michigan where the application could be
done this inexpensively within this time frame. This is
true .because the great diversity of resources available
here provides not just options that can be made to work,
but a range of alternatives from among which an efficient

system can be developed.

WMWWMM

Some resource planners regard potential competition
for land between food and energy as bioenergy's most
serious liability (Mitre Corp. 1977). The model
developed in this research can be adapted to study this=
conflict very easily. The only significant change that
would be needed is to include delivered price of the
harvested crop for each competing use. To the extent
crop biomass is more valuable per unit of weight than
wood. the price paid at the elevator is more important to
include than the price paid at the forest products plant
for wood. If, at some point in the future, wood-energy

industrial location analysis using models similar to the

102

one discussed here becomes more sophisticated, it will
probably be desirable to include delivered wood price.
Estimation of the impact of crop biomass-to-alcohol
conversion facilities on transportation costs is possible
using this model. The model can focus directly on the
question of when, under a given set of circumstances, an
economic incentive emerges for a particular farmer to
sell his crop for energy rather than food. It appears
likely that the trend away from food toward energy will
begin‘ to appear indirectly at first (OTA Vol.I,p.121).
Farmers now producing dairy silage (unless they are also
dairy products producers) will probably be the first
group tempted to switch markets. The argument for this
is that their crops are relatively high in bulk and low
in value compared to cash crops produced for human

consumption.

WUMMWMWW

Municipal and industrial trash disposal.
particularly in major urban areas, is becoming
increasingly difficult and expensive. In response to

rising costs, increasingly specialized equipment is being
employed. It is no longer simply a matter of refuse
being picked up and driven to the dump by one man with a
light truck. Larger trucks with teams of men pick up and
compress the rubbish. In our largest cities it is

transferred from these trucks. compressed again, loaded on

 

103

semi tractor/trailers and driven to a remote landfill.
It is a costly and complex process. Obviously,
locations of the landfill and the intermediate transfer
station, if one is needed, are very important factors in
the cost equation.

This application of transportation cost
minimization requires no product cost factors to account
for market value. In fact, it can be argued that a
benefit factor could be added to the transportation cost
minimization as an estimate of how much people are
willing to pay to have their trash removed.’ Trash
producers, in this application, are directly analogous to
biomass producers in wood-energy or crop-energy analysis.
Therefore, it would be necessary to aggregate individuals
into groups, for example city blocks or neighborhoods, so
they could be treated like timber sales locations or crop
fields were in the biomass analysis. Without this
aggregation, the linear program could become too large to--
be solved. As in the biomass analyses, this aggregation
would probably not seriously decrease the quality of the
model's estimates.

One advantage metropolitan areas have over rural
areas for purposes of this model is that almost all of
them have very well developed road networks. In
addition, most larger metropolitan areas have

computerized data bases which can be used to obtain

1oz:

actual point-to-point road distances rather than
estimates generalized for the entire area. With these
few changes, the model would be capable of comparing
either alternative trash disposal sites or transfer

sites.

Chapter VI

Summnmlumns

As discussed in the introduction, the overall
purpose of this research has been to develop a modeling
system to support wood-energy policy analysis. To this
end, two broad objectives were determined: (1) study the
impact a wood-energy plant at a particular location would
have on raw material transportation costs of the existing
regional network of forest products consumers and (2)
compare this impact with impacts produced with the plant
at alternative locations under different conditions. The
specific objectives of the research were as follows:

1. simulate a biomass supply based on actual

market prices and actual regional consumers,

2. develop a system for estimating both: (1) the

distance between the simulated biomass supply--
locations and the network of consumers and
(2) the cost of transporting biomass from its
location in the woods to consumers,

3. minimize transportation costs for the entire

network for one time period,

4. repeat steps one through three within a time

simulation which permits stochastic

minimization, forest resource growth and the

105

106

introduction of shock variables at specified
times to simulate anticipated future
structural changes in the economy and ecology

of the region.

These four objectives have been achieved. Program
I simulates biomass harvests until enough has been
harvested to meet the demand of all competing consumers
in the network for one year. Program II computes
distances between all possible combinations of simulated
biomass harvest locations and consumers. It then
converts these to estimated road-distances that would be
required to move the biomass from each harvest ~location
to each consumer and estimates of the transportation
cost. Program III minimizes transportation costs for the
entire network. Appendix B contains flowcharts and
source code listings for Programs I and II and a sample
of linear programming input and output as well as the~
small batch program needed to execute the Program III
minimization. No independent computer program was
required to simulate the passage of time because Program
I and II were written to accomplish this during their
other computations. Since Program III's only input is

from Program I and II, its time simulation is automatic.

107

mammal:

Four sets of experiments were conducted using this
model: (1) a series of 1982 simulations to provide
preliminary estimates of the statistical parameters of
other estimates, (2) a Benchmark Series to model the
existing network of consumers, (3) a series placing a
hypothetical wood-energy’ plant at Rose City and (4) a
similar series with the plant at Idlewild. The precision
series demonstrated that the model system will produce
extremely precise (less than one percent apart at the 90
percent confidence level) estimates.

Comparison of existing (benchmark) network costs
with cost estimates which included the wood-energy plant
at Rose City and Idlewild produced three statistically
significant results. First, the existing consumer network
will probably have lower transportation costs if a wood-
energy plant is built at Rose City than if no plant is
built at all. Second, existing consumers, not including
the wood-energy plant, will probably have lower
transportation costs if the wood-energy plant is built at
Rose City than if it is built at Idlewild. Third, the
consumer network, including the wood-energy plant will
probably have lower transportation costs if the wood-
energy plant is built at Rose City than if it is built at
Idlewild. Overall, the model's demonstrated ability to

distinguish between alternative potential locations for a

108

wood-energy plant at a statistically significant level is
the most important result. This indicates that the model
probably has the potential to provide policymakers with
useful analytical support in more complex, real-world
situations. In addition to these results, statistically
significant differences between transportation costs to
individual firms were produced. Although transportation
cost minimization actually occurs only for the entire
network's cost, individual firms can use the model output
with .some limitations. Most importantly, individual
firms must realize that the goal of the minimization is
still to minimize the network transportation costs as a
whole, not their own costs.

The only component of the model that was not
computerized was a master program which would repeatedly
execute Programs I,II and III. The master program was
not possible because Programs I and II were run on a
microcomputer while Program III ran on a mainframe
computer. Under these circumstances, the most a master
program could have done was to repeat Programs I and II
several times. Their output would then have been
transferred to the mainframe where another master program
would have run repeated minimizations. With these
limitations, it was determined that the master programs
required more effort than they were worth. Single runs
of Programs I, II and III do not not represent a

significant burden.

109

The model should offer decisionmakers a useful
tool for studying bio-energy industrial location when it
is combined with a biomass inventory. Its utilization of
the optimizing linear program within a simulation
framework requires explicit objective functions and
constraint specification while still permitting realistic
simulation.

The primary objective of this research was to
design and develop this model. Therefore, the
preliminary estimates discussed extensively in Chapter
III should not be interpreted as accurate estimates of
network transportation costs. They are only examples of
what the model can produce. although at this time they
are the best estimates available. Model results will
only be useful for decisionmaking if the assumptions upon

which they are made are realistic (Chapter V).

Wmummmm

Five substantive catagories of further research
have emerged during this study: (1) development of a
-forest resource data base for modeling from Landsat
imagery of the study region, rather than from existing
forest cover type maps, (2) creation of a general forest
growth projection system which is compatible with forest
type mapping done from satellite imagery, (3) changes in

the model to permit differentiation among individual

firms in terms of their efficiency in minimizing raw
material transportation costs, (4) expansion in the model
to permit direct interactive input of each factor
determining cost/mile, (5) integration of transportation

cost impact assessment, more general economic impact

assessment (i.e. changes in stumpage price) and
environmental impact assessment into one model. These
needs are discussed sequentially below. They are viewed

as being additions to further tests on the technical
details of the model developed in this research. The two
most important tests needed are: (1) making each
estimate stochastic rather than deterministic, and (2)
general sensitivity testing.

Appendix A reports the results of research which
provides basic information necessary to conduct a regional
forest biomass inventory. An inventory of this kind would
produce a data base which should be substituted for the
cartographically-derived data base used in the development.
of this modeling system. .Linking satellite-based forest
inventory with the transportation cost analysis model
developed in this research would provide a useful overall
system for analysis of bio-energy location decisions. As
detailed in Appendix A, no technological barrier exists to
this link. The major obstacle is money. Conducting a
computerized inventory of a large area is expensive. As

the availability of computer systems capable of coping

Pro.

for«

are

Whit
com]
bio:
data
has:

PP01

con:
the
They
3De]
is
decj
Prev
him
is
°bJe
tram
eati

theP

111

with enormous amounts of data increases, the seriousness
of this obstacle will decrease.

Research concerning regional forest growth
projection systems also a high priority. Growth
forecasting has traditionally been confined to small
areas because the primary factors which determine growth
are not easily generalized. One approach to this problem
which seems to have promise is integrating the
computerized data base produced during a satellite-based
biomass inventory with computerized soil and hydrologic
data bases. This three layered system could provide the
basis for expanding site-specific growth tables to
provide estimates for the entire region.

From the perspective of either an existing chip-
consumer or a hypothetical wood-energy plant, results of
the current study should appear tantalizingly useful.
They demonstrate how the technique might work but don't
spell out the details. Another important area research
is to increase the model's applicability too.
decisionmaking in individual firms. The major constraint
preventing this is that the transportation cost
minimization accomplished with linear programming, if it
is used within the model structure tested here, has the
objective of minimizing of the entire network's
transportation costs rather than one consumer. If the
entire network is defined as the subject of the research,

there is simply no way to minimize an individual firm's

rf

t4
tr
tc
wo

Th

Dr:
trt
prc
cos
res.
cOn<
tern

sDec

indu;
otheI

thGSe

112

costs. However, individual firms could make good use of
the model structure, and probably much of the data as
well. by treating each of their harvesting locations as
an origin in the system and their plant or wood yard as
the destinations. This concept could also be useful for
harvest scheduling.

It would also be helpful to change the model to
permit individual consumers to be treated differently in
terms- of their efficiency in minimizing their
transportation costs. For example, if a consumer is able
to have consistently lower transportation/mile costs, he
would have a comparative advantage over his competition.
This factor could be included in the model.

An application of this model not discussed in the
present research is the increasing role of independent
truckers in transporting raw material to Michigan forest
products industries. An operator could minimize his
costs using the model structure as it is defined in this»
research. Applied research translating the general
concepts of linear transportation cost minimization into
terms that are meaningful to the transportation
specialist is needed.

Long-term research priorities for bio-energy
industrial location research are similar to those in many
other regional resource areas. Decisionmaking concerning

these subjects is difficult for many reasons. A

a:

QC

Pe

Ch:
of

9P0

113

synthesis of economic and ecological impact forecasting
is needed to provide a comprehensive policy model.
Industrial bio-energy is a new use of a resource and it
will probably occur on a large enough scale to be
economically feasible. Therefore, it will probably
develop relatively quickly and be important to both the
economic and ecologic health of a region. So the need to
integrate economic and ecological models is particularly
pressing.

.As discussed extensively in Appendix A, the issue of
soil nutrient depletion from whole-tree harvesting is
important in wood-energy planning. This is primarily an
ecological impact that would occur away from the wood-
energy plant. A» similar problem is damage to timber
reproduction due to dragging (known as skidding) of whole
trees from their growing locations to chipping locations.
Recent research in forests similar to those of this study
region suggests that between 27 and 47 percent of
unharvested trees are wounded during whole-tree harvesting.
and that skidding causes at least two-thirds of the damage
(Kelley 1983). This kind of information should be
coordinated with growth forecasts, especially if
reharvesting is anticipated.

Air and water pollution caused by burning the
chipped wood at the power plant are the other major class
of environmental impacts which need to be studied. To

provide policymakers with comprehensive models for

Ce
1:
in
th
en
en'
de‘
W1]

the

Cha
ahaj
301s
Dart
only
°°rrq
diff:

who

114

planning, models must include these factors as well as

the purely economic considerations.

mailman].

At this time it is not possible to predict when
wood-energy will begin to become a profitable enterprise,
if ever. Obviously, the price of oil is an important
consideration. However, there are many other factors,
including such imponderables as what the prevailing
interest rates might be at any time in future. High on
the list of uncertainties faced by prospective wood-
energy producers is the question of how the social and
environmental impacts of industrial scale wood-energy
development will be evaluated by regulatory agencies.
Will developers meet with assistance or resistance to
their plans?

No reason has emerged why this model should not be
a useful tool for wood-energy (and, as discussed in..
Chapter IV, bio-energy in general) industrial location
analyses. Like most other tools produced by modern
science, this model cannot itself prevent misuse or, in
particular, biased application. In any application, the
only real guarantee available that the model is being
correctly applied is the intent of the modeler, which is
difficult if not impossible to ascertain. Decisionmakers

who will use the model output are, by default,

115

responsible for providing a context which encourages

unbiased applications.

REGION

 

 

APPENDIX A

REGIONAL FOREST BIOHASS INVENTORY AND ANALYSIS SYSTEM

A Regional Forest Biomass Inventory
and Analysis System

(Final Report on Title V funded
system development)

Kyle Ki ttl eson
Resource Development Department

Michigan State University
East Lansing, MI 48824

this
5: re s t
:caassi
:egixal

‘-
‘r'ecas .

res-ed as

:iividua
{Size and
3;37285 as
The I
1'1.- tOIY]
:Uhlixugx
Costs. Th
ll; Users

S:311141’101)

1‘s

.. a biwné
We Con

Abstract

This report (1) describes the development and testing of a regional scale
forest biomass inventory system. (2) presents preliminary
biomass-per-unit-area estimates and (3) discusses ongoing developnent of a
regional economic nodeling system which will use the biomass inventory to
forecast effects of bio-energy resource develognent. The inventory system is
based on Landsat satellite imagery. Maps produced using the Landsat imagery
are compared to Michigan Department of Natural Resairces (DNR) forest cover
type naps of the study areas in the northern lower peninsula of Michigan.
Detailed stand composition of the Landsat derived naps is obtained frcm the
111R maps. Bianass—per-acre estimates of DNR types are developed by expanding
individual tree biomass equations to stand equations using DNR stocking data
(size and density) supplemented with field observations. In turn. these stand
biomass estinates are expanded to the more general Landsat forest types.

The regional economic model necessary to effectively utilize the
inventory has been developed and is being doomented. It is a hybrid oarputer
model based cm a linear progranming minimization of biomass transportation
costs. The model allocates bianass resources anong the network of carpeting
pulp users in the region. Application of the linear programing model within a
simulation framework will permit sane of the eooncmic inpacts of developnent

of a bio-energy industry to be forecast under a variety of possible future

econanic conditions.

The E
selysis o
is iment
Estes: bio
portion v
efficientlj
trod-burn:
mastery 1
:a-rbinatn
Elli fiEld (
:eiium Sc
helm”; e
Passible US

Alter

“his“

INI‘ROIIETION

 

The purpose of this study was to develop a systan for inventory and
analysis of forestry bianass in northern Michigan. The specific objective of
the inventory portion of the study was to create a systen for napping Michigan
forest biomass on a regional scale. The specific objective of the analysis
portion was to create a regional economic modeling system which could
efficiently use the inventory output to predict some potential impacts of a
wood-burning electric power plant on the region. Oarparison of alternative
inventory methods resulted in the decision to develop a system based on a
combination of enhanced Landsat satellite imagery. existing forest type naps
and field checks. Several alternatives were explored including the use of
medium scale color infrared aerial photography. The major reason for
including esdsting type maps in the inventory system was to make the best
possible use of existing sources.

Alternative economic modeling systems were evaluated to determine their
potential for effeciently utilizing the inventory data and providing useful
inpact predictions. It was determined that no single existing nodeling systen
could meet the requirements of the project as well as a modeling system that
could be developed within the Resource Develognent Department. This system
has been created.

It is a hybrid. computer-based model containing a linear programming
minimization of regional forest products network transportation costs within a
stochastic sinulation framework . It takes advantage of the potential of linear
progranming for efficient optimization by placing it in emnomic environments

that simulate the future based on a variety of assumptions.

3y cc
ire intent;
:‘erelopmei
irrelopnen‘

nheocnpz

rsteis sh:
a method 1
ievelogner
sathenatic

... 1
wage.

3““lh.
. VIL‘L’

Initi
c.
win the 1
“E'lelopme!
hintire/S.
arm-m the 1

Peso“93s.. r

By combining this model with accurate. timely information gathered using
the inventory system it will be possible to predict impacts of bio-energy
development on this region. It will also be possible to predict the optimum
development location. An important additional capability of the combined
system is that it permits the impacts of develqment at alternative locations
to be ompared against each other quickly enough to be meaningful within
realistic time constraints. Applied together. the inventory and analysis
system shcmld assist public and private decisiomakers by providing them with
a method for. making systematic comparisons of alternatives for bio-energy
development by combining recent scientific advances in remote sensing.

mathematical modeling. regional economics and system simulation in a useable
package.

FUNDING

Initial development of the inventory and analysis systems and
quantification of the inveitory for modeling was supported by a Title V grant
from the Cooperative Extension Service at Michigan State University.
Development of the regional economic model is being supported by
McIntire/Stennis Act project number 1075. This research is being conducted
within the Resource Development Department. College of Agriculture and Natural

Resources. Michigan State University.

I. INVENIUH SYSTEM DEVELOPMENT AND TESTING

 

The inventory system was developed and tested in four phases:
a) sample area selection and map base selection arri reproduction.
b) Landsat image enhancement and mapping,
c) field surveying and

d) analysis.

A. Sample Area and lap Base Selection and Reproduction

The primary objective of this phase was to select a study area that vas
both representative of the region and. if possible. of special significance
for wood—energy development. Early proposals focused on forests around
Phrsey. Michigan because Consumers Power Company plannned to construct a
wood-fired electric plant there. waever. before research began. Consumers
abandon the Hersey site. Subsequent discussions with Consumer's planners.
analysts in the Michigan Public Service Commission's Scientific and Technical
Research Section and foresters from the Michigan Department of (Natural
Resources revealed roughly a dozen locations in the northern lower peninsula
were being considered as potential locations for such a plant.

Analysis of wood supply area radii for these locations resulted in cm
significant findings. Approximately half of the potential locations of fuel
supply radii overlapped forested portions of central Osceola county and fuel
supply radii from all the potential locations included some of Michigan's
poorest soils. Low productivity. sandy soils are of special concern in
wood-energy because the most cost effective harvesting method. whole-tree
chipping, removes the entire tree, leaves and all. Numerous stuiies have

sham that whole-tree chipping can adversely effect nutrient recycling on

sites where nutrient supply is marginal (Walkonen 1973 and Kittleson 1979).
In addition. a preliminary survey of over sixty (60) biomass prediction
equations and tables confirmed the common sense idea that forest bicnass is
often significantly lower on poor soils than on better soils (Hudson and
Kittleson 1978) .

Based on these considerations. two sample areas were selected: one in
central Osceola Commty and one of lower productivity. sandy soil in Kalkaska
County. In Osceola Oomty. the tomships were:

l..Tl9N.R8N:Tl9N.WW

2. ' namasnmav. new.
In KaDtaska County. the townships were :

l. 'I'26N.R5W:T26N.R6W

2. T25N.RSN:T25N.R6W.
In both of these areas a majority of the forested land is owed by the State
of Michigan. Therefore, access for field checks was not a problem and forest
cover type maps were readily available whidn were typical of the quality and
vintage of maps in the region. Osceola County type maps were updated in 1968.
Kalkaska County type maps were updated in 1969. Both counties will be
rerapped in the near future based on U.S. Forest Service forest inventory
data. Each study area included all state land within the 144 square mile
(four township) areas.

Because the region forest oovertype has been mapped extensively on medium
scale color-infrared aerial photography. the reliability. expense and time
required are relatively well known. By contrast. how well Landsat imagery and
existing forest cover type maps compare and might be effectively used together

has not been extensively studied. To be as efficient as possible. any

 

inventory
as made i
scale CIR

"he 1'

~6

of a nose
:irite ma;
type map
1:24.000 5

inset in.

E'lar over

inventory system must make use of existing data. Consequently. the decision
was made to develop a system using the existing forest type maps. Medium
scale CIR type mapping can be substituted directly for these maps if needed.
The map base for each sample area was a 1 : 24. GOD scale mylar reproduction
of a mosaic made from U.S. Geological Survey topographic maps (7.5 ard 15
minute maps were used) Michigan Department of Natural Resource (DNR) forest
type maps of the areas were reduced slightly in scale from 1:15.840 to
1:24.000 and registered to the topographic base using a cartographic pin bar.
Laxdsat imagery was interpreted and manually mapped directly on pin registered
mylar overlaying these maps. The product of this process was three layers of
maps: (1) topographic map mosaic. (2) DNR forest type map and (3) Lardsat map.

ProceduremusediniandsatmappingarediscussedinSectionB.

B. Iardsat Mapping of Forest Types

 

One midsummer 1978 Landsat image covered both the Kalkaska Cotmty and
Osceola County study areas. The black ard white transparencies of this image
were contrast-stretched to enhance the difference bet-.weei major forest types.
Enhancement was based on densitametric sampling of the major forest types in
each of the four spectral bands of the image. Diazo color copying of the
black and white images produced a color transparency composed of a yellow bard
4 image, a magenta band 5 image, ard a cyan bard 7 image. Placed together,
these images form a color composite similar in appearance to a typical Kodak
2443 color-infrared (CIR) image exposed through a Wratten 12 filter. This
similarity is an important advantage in interpretation because most
photointerpreters have experience with CIR and are. therefore. able to

correctly interpret the similar Landsat image. This greatly increases the

speed and accuracy of mapping. For this reason. the "pseudo CIR" Landsat
image was selected for mapping in spite of the fact tlat other cambirations of
bands produced better edge definition between forest types. For example. a
contrast-stretched ratio of positive 5 and negative 7 (+5/-7) bands produces

significantly better contrast between jack pines (Pinus bauksiana Lamb.) and

 

mixed swamp conifers. Distinguishing between these two types is sometimes a
problem is this region using the "pseudo CIR" Landsat image. Howaver. the
+5/-7 ratio produces color combinations that are so unlike any other
film/filter combination that it is impossible to interpret them visually
without additional training. Imagery of unusual ratios of various bands has
beei successfully used to supplement "pseudo-CIR" imagery in northern Michigan
(Sicuranza and Carpenter 1980) . Use of this imagery requires extensive
experience in regional natural vegetation mapping and dual image projection
systems. Most problems interpreting natural vegetation on a pseudo-CIR
Landsat image can be solved by an interpreter with intimate knowledge of the
regional vegetation patterns if he uses topographic maps during
interpretation.

The diazo color carposite was copied at 1:1 on Kodachrome 25 film using a
Hasselblad copy camera system because diazo film fades quickly under the
intense liq-1t of cartographic projection equipment. napping from the Landsat
imagewasdcnecnthebGURercrteSeising Center's cartographic equipmentusing
16X magnification to bring the Landsat scale firem l:l.000.000 to 1:62.500.
This 1:62.500 scale map was later enlarged to 1:24.000 for analysis.

The sample areas were mapped into the following catagories: hardwood
forest. conifer forest. mixed hardwood and conifer forest and nonforested.

these being the Level II forest categories in the Michigan Land Cover/Use

Classification System (Michigan Land Use Classification and Referencing
Committee 1975). The only supplemental reference used during mapping was the
topographic mosaic to which the projected Lardsat image was registered on the
mapping surface. Tb duplicate production inventory mapping as closely as
possible no reference was made to other aerial photography or the DNR forest
type maps of the region to resolve questions. Questionable stands were
identified and overall 120 locations required field checks to determine the

correct type.

C. Field Cheddng Procedure

 

Field checking Landsat interpretations required approximately 18
man-days. An average of 13 stands per day were visited per day by a two-man
team using a fair-wheel drive vehicle. These Landsat field checks cerprcrmised
approximately one half of the actual field time for the project. The other
half of the field was devoted to checking the accuracy of the DNR type maps
ard gathering data on timber stand size and density to supplement mR stocking
estimates. In general. the DNR type maps were accurate in both counties.
Differences between the maps ard the stands today are due mostly to growth
since the maps were published. Ninety-four percent of all stards checked were
typed correctly. However. over the years size ard density has changed of the
classes changed in approximately one-eighth of the stands . Due to time
constraints. no comprehensive statistical sample of the maps was possible.
Therefore. these figures should be interpreted as preliminary orders of
magnitude. not statistically reliable estimates .

In addition to the Lardsat field checks. sixty stands in both sttdy areas

were cruised for type. diameter (dbh). height and basal area. One plot was

recorded for each stand. The location of the plot was systetatic and
subjective ard was based on wfnat was jtdged the representative or typical part
of each stard. Identification of sample stards was done randomly from the DNR
maps. Fifty unique types were identified which were representative of the
types in the study area. These plus ten samples allocated to aspen and
harduaod reproduction made Lp the stard size sample (Figure 1). The sample
drawn from mixed types was very erall ard diversity encountered in than large.
In light of these problem ard time constraints. it was decided to use the
average of conifer and hardwood size/ density figures for mixed types.

Figure 1. represents the results of a sample of only 60 plots.
Considering the large number of other variables that must be assured to be
constant to provide meaningful statistical analysis and the developmental
purposes of this research. it was determined to use this data for
developmental purposes without distributional parameters. Stochastic

estimates are necessary before this type of data can be used in a production

inventory.
D. Analysis

Based on average stand size and density as determined in the field ard
from the DR data (Figure 1). individual tree weights (Figure 2) were expanded
to stand weights per acre (Figure 3). Over two hundred forest types composed

of carbinations of these types were used as estimates of

Figure 1 . Average Hardwood and Conifer Stand Sizes

 

 

 

 

Size/ Density .
C1ass* dbh (inches) height (feet) state / acre

w
1 3 25 250
2 II II 550
3 " " 1000
4 7 50 94
5 II II 2%
6 II II 375
7 12 65 31
8 II II ‘70
9 " " 127

Conifer
1 3 15 250
2 II II 550
3 II II 1%
4 7 ‘ 39 ' 94
5 II II 2%
6 II II 375
7 12 7O 23
8 II II 51
9 II II 94

*Michigan Department of Natural Resources numbering system (Michigan
Department of Conservation. 1968) .

 

(predominantly Abies. Picea.
Thuja spp.)

 

 

 

. Figure 2. Individual Tree Weight (green pounds)
Pole Saw
Forest Type Restocking Timber Timber

1. Aspen 38(3) 401(3) 1515(3)
(Populus tremoides. Mich*)

2. Maple 70(1)* 437(2) 1594(2)
(Acer spp.)

3. (bk 42(4) 429(4) 1884(4)
(mercus 5139-)

4. Mixed lowland l'lar'dwood“Hr 70(1) 437(2) 1594(2)
(predominantly Fraxinus.
Ulmus. Acer spp.)

5. E1831! fir 41(10) 421(6) 1990(8)
(Abies balsamea L.)

6. Ehmlock 21(7) 396(8) 2118(8)
(T‘suga canadensis (L.) carr)

7. Jack pine 84(12) 629(12) 2475(11)
(Pinus banksiana Lamb.) ‘

8. lbd pine 30(11) 363(6) 2358(6)
(Pinus resinosa Ait.)

9. Spruce 109(5) 449(6) 2111(6)
(Picea app.)

10. Vhite cedar 25(7) 261(9) 817(9)
(Thuja occidentalis L.) I

11. Finite pine 30(11) 410(8) 2344(8)
(Pinus strobus L.)

12. Mixed lowland conifer"** 67 355 1464

* Sources for bierass equations appear by number in Appendix A.
** Maple data used because no mixed lowlard hardwood biarass data available .

*** Spruce ard cedar averages used because no mixed lowland conifer bimass
data available.

Figure 3. Stand Weight Per Acre (pure types only. in greem pounds)

Coniferous Tygs

 

 

 

 

 

 

SPMEE HEMLGX CEDAR (WHITE)

(Picea spp.) (Tsuga canadensis (Thuja
(L.) Carr.) occidentalis L.)

S1 27.250 [-11 5.250 C1 62.500
82 59.950 HZ 11.550 (:2 13.750
S3 109.000 HB 21.000 C3 25.000
S4 42.206 H4 37.224 C4 24.534
SS 92.494 [-15 81.576 C5 53.766
86 168.375 H6 148.500 06 97.875
S7 48.553 H7 48.714 C7 18.791
88 107.661 [-8 108.018 C8 41.667
S9 198.434 HQ 199.092 C9 76.798

BALSAM FIR “BITE PINE RED PINE
(Abies balsamea L.) (Pinus strobus L.) (Pinus resincsa Ait.)
F1 10.250 W1 7.500 R1 7.500
F2 22.550 W2 16.500 R2 16. 500
F3 41.000 W3 30.000 R3 30.000
F4 93.574 W4 38.540 R4 34.122
F5 86.726 W5 84.460 R5 74.778
F6 157.875 wa 153.750 R6 136.125
F7 45.770 W7 53.912 R7 54.234
F8 101.490 VB 119.544 R8 120.258
F9 187.060 W9 220.336 R9 221.652

JACK.PINE bflRII)I£MHIflHD<IXIEflfl§3

 

(Pinus banksiana Lamb.) (predcminantly Abies. Picea. Thuja spp.)
J1 21.000 01 16.750
J2 46.200 02 36.850
J3 84.000 03 67.000
J4 59.126 04 33.370
J5 129.574 05 73.130
J6 235.875 C5 133.125
J7 55.775 07 33.672
J8 123.675 (B 74.664
J9 227.950 09 128.666

Hardwochypes

 

ASPEN OAK MAPLE

(Pepulus treruloides Michx . ) (Qaercus spp. ) (Acer spp. )
A1 9.500 01 10. 500 M]. 17.500
A2 20. 900 02 23. 100 MZ 38. 500
A3 38.00) 03 42.000 M3 70.000
A4 37.694 04 40. 326 M4 41.078
A5 82.606 05 88. 374 [‘5 90. 022
A6 150. 375 06 160. 875 N5 163. 875
A7 46 . 655 07 58. 435 M7 49. 414
A8 105.350 08 131.950 PB 111.580
A9 191. 135 09 239. 395 DB 202. 438

.MIXED LCWLAND HARDWOODS

 

(predominantly Fraxinus. Ulmus. Acer spp.)

17.500
38.500
70.000
41.078
90.022
163.875
49.414
111.580
202.438

53533548535389

Over two hundred forest types composed of combinations of these pure types
occured in the study. Double type tonnage (for example A504. an aspen-oak
pole sized stard) was calculated on a 60%/40% basis:

.6(A5 tons/acre) + .4(O4 tons/acre) = A504 tons/acre.
Triple types were calculated on a 50%/ 30%/ 20% basis:

_ M5MJ1 tons/acre = .5(M5 tons/acre) + .3(A4 tons/acre) + .2(J1 tons/acre).
Individual tree weights (Figure 2) were obtained by applying equations from
the sources indicated. These weights were multiplied by the stems/acre in the
stand (Figure l) to obtain stand weights per acre (Figure 3). Figure 4

illustrates the sequence.

Figure 4. Summary of Stand Biomass Estimation Process.

Step Activity
1. Average stand dbh. height and trees/acre data gathered from
existing sources and supplemented with field surveys (Figure l) .
2. Weight per-tree obtained from best available source (Figure 2) .
3. (Trees per acre) x (weight per tree) = stand weight per-acre
(Figure 3).
4. (habitation types weights calculated:

Double: 60% of dominant type weight per acre plus 40% of
subordinate type weight per acre.

Triple: 50% of dominant type weight per acre plus 30% of
intermediate type weight per acre plus 20% of minor type
weight per acre.

 

II. I“???
fcrest c

1.

 

II. MAPPIBB PRmESS
Comparison of the Landsat forest cover type maps (Level II) with the DNR
forest cover type maps was conducted as follows :

l. Overlay pin-registered Landsat and DNR maps with a dot grid
(5 acre/dot).

2. For each DNR stard record:
a. INR type
b. Level II type
c. number of dots
The major goal of this analysis was to detenmine which mR types were
mapped as hardwood. conifer. mixed. or nonforested on the Level II Landsat
map. Another important objective was to develqa a bicrass—per-acre estimate
for hardwood..oonifer and mixed types tlat could provide a starting point for
future Landsat biarass mapping projects.
Three FORTRAN programs were written for a micro computer which analysed
and summarized the data that was extracted from the map carparison. All data

analysis was conducted on a 64K RAM Vector Graphics. Inc. micro computer

equipped with dual 635K disk drives.

Results and Conclusions of Inventory System

 

M thousand one hundred and nine (2.109) DNR stand were mapped on
seventy thousard three hundred fifteen (70.315) acres. Sixty-one thousand tea

(61.010) acres or 87% were forested. Paramod types dominated (Figure 6).

Figure 6. Level II Acreage Sumary (both study areas. based on DIR data).

 

 

31.85. Acres % of Tbtal % of Forested
Hardwood 29.820 42 49
Conifer 13.640 ' 19 23
Mixed 17.550 26 29

Non-forested 9. 305 13 —

The small non-forested area (13%) is a consequence of both study areas
being contained within state forests. It should not be taken as
represeitative of the region as a whole. Figure 7 summarizes the Level II ard

DNR acreage results by township. *

 

*wanships are numbered as follms:

 

Number Name ' Location
1 Cedar. Osceola Oo. T 1&1. R9W
2 Osceola. " T 18N. RSW
3 Ease Lake. " T 19N. R941
4 Wdc. " T 19N. RSW
5 Oliver. Kalkaska Oo. T 2m. RSW
6 Bear Lake (Middle). Kalkaska Co. T 231. R34
7 " (South) " T 23x1. RSW
8 Garfield. Kalkaska 00. T 231. RSW

iFigure 7. Landsat Mapping Accuracy and System.Error Level. by'wanShip.

 

 

 

'Downship % Correctly Mapped on.Landsat*
Bummer
Cartographic and
NOn- analysis-induced
Eamdwocd Conifer Mixed forested error**
1 85 95 80 99 0
2 -98 98 9O 96 O
3 93 100 86 95 1
4 88 99 88 98 1
5 89 98 67 85 5
6 80 9O 70 99 0
7 98 95 85 84 2
8 100 88 80 93 1

 

* Assumes DNR.type:maps are correct

‘** Total Landsat acres minus total actual acres (measured as Total Level II
acres munus total DNR.acres)

Figure 8. Problem areas. by type and township.

 

Problem %*
Township MShj-P
Number I-hrdwxnd Conifer gig Nonforested Average
1 15 5 20 l 10
2 3 3 10 4 5
3 7 0 l4 6 7
4 11 1 12 2 6
5 10 2 32 15 15
6 20 10 31 1 15
7 2 5 14 16 9
8 0 12 20 7 10

 

* (Landsat acreage - actual acreage) /Tota1 acreage mapped in township.

Figure 8 provides a clearer picture of where Landsat mapping errors
occured. Mixed forest mapping errors accounted for roughly 49% of all errors .
Of this 49%. 40% came from townships 5 ard 6. Based on acreage estimates.
townships 5 and 6 should have accounted for only 29% of the error. The
difference between Landsat and DNR maps of mixed types in these townships is
due to the shrubby nature of the vegetation occupying poor sites on outwash
plains. These areas were mapped as upland shrub (U) m DNR maps but mapped as
mixed forest on Landsat. Typically. they contain shrubs (primarily various
cherry species) having no main stem or leader. From an aerial perspective.
they appear to be tree croms. However. from the ground they are obviously
open-crowned shrubs. Stereo interpretation of medium scale CIR. imagery. as
opposed to Iardsat. would have prevented this discrepancy. Regardless of
whether they are incorrectly mapped as mixed forest or correctly as shrub.
these sites contain little biorass.

Figure 9 summarizes the results of the study in broad terms. Overall.
89% of the Landsat mapping was "correct" (accepting the INR maps as completely
accurate). This means that 89% of the area mapped on Landsat as either
hardwood. conifer. mixed or nonforested was mapped the same way on the DNR
maps. As irdicated earlier. 94% of the sampled types were correct on DNR maps
when field checked. However. the 89% accuracy figure for Landsat is
aggregated for an entire township ard therefore are not directly corparable to

the location-by-location accuracy of the [NR maps .

Figure 9. Whighted Average Landsat Accuracy.

 

 

1 2 3 4 5
% Correct on weighted
'DownShip Acreage % of Tbtal Landsat overall value
EEEEEEE 1555333 (rounded) in twpu (rounded)*
1 6455 9.18 89.49 8.21
2 3170 4.51 95.19 4.29
3 1495 2.13 93.27 1.98
4 2550 3.63 93.42 3.39
5 8660 12.32 84.67 10.43
6 11970 17.02 84.66 14.41
7 19365 27.54 90.38 24.89
8 16650 23.67 90.11 21.34
TOTAL 70315 100.00 88.94

‘Weighted average % correct landsat mapping (on townShip‘basis) = 88.94% (89%)

 

* weighted walue = % of total acres mapped x % correctly mapped over all Level
II types in townShip (column 3 x column 4 = column 5).

The 89% accuracy figure for the landsat shoild be interpreted as meaning
that. for an entire township. one could expect an 11% difference between the
acreage actually growing there and the acreage a Landsat survey (using similar
methods) predicts is growing there at Level II. The important components of
this definition are:

l. aggregation of acreages to township levels

2. similar mapping methods

3. Level II mapping.

The results of this study do E5 mean that for any point in these
townships the probability that it is correctly mapped at Level II is 89%. The
probability of the Landsat map being accurate on a location-by-lccation basis .
although unknown. is certainly lower than 89%.

Figure 10 indicates that just over 80% of the area mapped was in Kalkaska
county. This occured because the size of the state forest holdings in the
study areas were four times as large in the Kalkaska county area as they were
in the Osceola county area. The imbalance has the effect of weighing the
biorass estimates from the poor Kalkaska comty sites four times as heavily as
those from the Osceola county sites . This produces what can be interpreted as
a conservative biomass estimate since the poor Kalkaska sites produce less

biorass/acre than the Osceola county sites.

Figure 10. Chrparison of acreage mapped in Osceola and Kalkaska study
areas.
Osceola Kalkaska

l . Acres 13. 670 56. 645

2. % of 'Ibtal 19.5 80.5

III. BICMASS CI" FOREST TYPES

 

Biomass was estimated for each Level II forest type. Estimates were
aggregated to township level as follows:

1. For each stand on the DNR map. the biomass per acre was multiplied

by the number of acres to obtain the stand biotass.

2. This biomass was allocated to the Level II type given it during the

Landsat mapping.

3. These stard biotass estimates were summed for the township by Level
II type and divided by the acreage the type occupied in the
township. The result was the Level II bicmass per acre estimate for
that type in that township. ‘

Biomass was allocated on the basis of Level II Landsat mapping. not DNR
maps. This means. if a stand was mapped as hardwood based on its Landsat
image. it was included in the hardwood estimate regardless of its actual (DNR)
type. Consequently. several coniferous. mixed and nonforested stands are
included in the level II hardwood data. A similar situation exists for the
Level II coniferous and mixed types.

Estimates were produced this way because the purpose of the study was to
develop‘a regioral biotass inventory system. An application of this system
would probably result in similar mistyping. Within this context. it is not
significant whether the Level II types are pure. as long as the the mix of
species composing each type is constant. Figure 11 contains Level II biotass

per acre estimates for each township in the study.

Figure 11 .

 

 

* No conifer types were mapped in this township.

45.3
58.0
61.3

55.3

28.9
58.4

29.4

28.1 '

Conifer

39.9

35.6

29.3

40.3
47.1
24.2

28.1

Level II Biorass Estimates (greet tons per acre).

Mixed

46.5
56.7
60.0
53.4

30.9
39.1
29.7
26.7

Average weighted biotass for each Level II type was computed as follows:
1. Level II type acreage in each township multiplied by its biarass per acre

equals its weighted value.

2. Weighted values sumed for each type over all townships and divide by the
total of all acrse mapped in the type in all townships. The result is

the weighted biorass per acre estimate for each type (Figure 11) .

For example. in one township 4.380 acres were mapped as hardwoods.
Hardwood biomass for the township was estimated to be 45.3 tons/acre.
Therefore. the weighted biomass/acre for hardwood in this township is :
4580 x 45.3 = 207.474. County and total weighted biomass estimates are
presented in Figure 12. Data was not aggregated on a county level to reach

the overall biorass estimates.

Figure 12. Comty and Overall Biorass Estimates (greei tons per acre).

Hardwood Cbnifer Mixed
Osceola 52. 3 36. 2 50.3
County
Kalkaska 32.7 33.3* 31.6
County
Overall 39.5 33.3* 33.9

 

* Actual data: Kalkaska Co. = 33.291. Overall = 33.324. The numbers are
nearly identical because Kalkaska County contained 99% of the conifers mapped.

As Figure 12 indicates. biomass estimates in the Osceola (bunty strdy
area were higher for all three types than were the same estimates from
Kalkaska County. Because the same set of [NR type biotass per acre tables
were used in both comties. the difference between the counties suggests there
may be a wide range of biorass values ocoiring on different quality sites.

The causes of these differences. although beyond tre scope of this study.
may be related to differences in site quality or the way in which type
differences are translated into spectral signature of a forest type on
Landsat. How these signature differences are affected by the contrast

enhancement process etployed here probably also influences the estimates .

IV. REEIONAL WC mm STATUS

 

Development of an analysis system to utilize data gathered from a
regional inveitory based cm the above process has been recently completed and
will be documented in my dissertation. Basically. the system consists of a
linear programming minimization of network biomass transportation costs within
a simulation fratework.

For develqmental purposes it is corposed of four models:

I. A Simulation of regional biotass supply using biorass estimates and
variable probabilities for sampling developed in this study.

II. Calculation of a transportation costs matrix from each supply location
to each pulp user in the region. I

III. Linear programming minimization of annual regional transportation

costs.

IV. Simulation of time and repeated runs of models 1. 2 ard 3 to obtain
annual estimates of network transportatim costs .

The simulation of biomass supply (Model I) may be elminated when a
regional biotass inventory is available. However. during the developmental
phases. Model I made repeated random supply locations possible in the linear
program. This has the effect of outlining the approximate supply area
(spatial detand cone) of each pulp user.

Using this four-model system. hypothetical new bio-energy using plants
were introduced to the region. The impact of these plants on the existing
wood users was predicted for alternative plant locations. The time simulation

tracked the cumulative effect of modeling errors.

SAVER

Seventy thousand three hundred fifteen (70.315) acres were mapped at
Level II in Kalkaska ard Osceola Counties. Michigan. TWenty-one hundred and
nine (2.109) forest stands covering sixty-one thousard ten (61.010) acres were
corpared to Michigan DNR forest cover type maps to estimate (1) the accuracy
of the Level II Landsat maps and ( 2) forest biomass for Level II types in
Michigan. Results irdicated Lardsat mapping was 89% correct on a township by
tomship basis. Irdividual tree bicnass prediction equations were exparded to
produce stardbiorass predictim equations using a corbination of [NR size and
density data and field surveys. These stand biomass estimates formed the
basis of preliminary Level II biotass estimates presented. -

No major obstacles were encountered during the development ard testing of
this system: therefore. a regional biorass inventory based or a similar system
should be successful . Altl'ocgh requiring coisiderable corputational support.
the analysis system of linear programming minimization of biomass
transportation costs within a simulation model is an effective metlod of
converting the regional biomass inventory data into information that

decisionmakers should fird useful .

AppendixA

Sources for individual tree biomass (Figure 2). Numbers indicate

footnotes in Figure 2.

Average data used for restocking biomass. Average of biomass estimates
from sources 4. 7. 8 and Young. H.E.. 1972. Biorass sampling methods for
puckerbrush stands. Forest Biotass Studies. pp. 179-192. Oroo. Naine:
University of Whine.

 

Steinhilb. H. M ard Winsaver. S. A.. 1973. Sugar Maple: Tree and Pole
Weilts. Volumes. Centers of Gravity ard Logging Residues. NEE-132. St.
Paul. Minnesota: U.S. Forest Service.

 

Schlaegel. B.F.. . 1975. Estimat' As Volume and Weight for Individual
Trees. Diameter Classes. or Entire Stands. LIE-20. St. Paul. Minnesota:
U.S. Forest Service.

 

Wiant. H. V.. 1977. Sampling for Weight in Appalachian Hardwoods.
Agricultural Experiment Statioi Paper 1516. Dbrgantown. West Virginia:
West Virginia University.

Whittaker. R.H.: Bormann. F.H.: Liken. G.E.; and Siccera. T.G.. 1974.
The Hubbard Brook ecosystem study: forest biorass and production.
Ecologcal Nbrographs. 44: pp. 233-252. Dirhem. N.C.: Dike University
Press.

Steinhilb. H.M. ard Erickson. J. R.. 1972. Weights ard Centers of Gravity
for Red Pine. White Spruce and Balsam Fir. [It-75. St. Paul. Minnesota:
U. S. Forest Service.

 

 

Yang. H. E. ard Carpenter P. N.. 1967 . Weight. Nutrient Element and
Productivity Studies of Seedling. and Saplings of Eight Tree Species in
Natural Ecosystem. Tech. Bull. 8. Oroo. Maine: UniverSity of Paine.
Ag. Experiment Station.

 

Young. H.E.; Strand. L.. and Altenberger. R.. 1964. Preliminary Fresh
and Dry Weight Tables for Seven Tree Species in thine. _T€ch. Bull. 12.
Oroo. Paine: UniverSity of Paine, Ag. fbcpeflment Station.

 

Dryer. R.F. . 1967. Fresh and Dry Weight. Nutrient Elerents ard Pul
Characteristics of Northern white Cear (Thuja accidentalis). Tech. Bill.
27. Oroo. mine: University of Whine. Ag. Experiment Statim.

10. Baskerville. G.L.. 1965. Dry matter production in immature balsam fir

stands. Forest Science anograph. No. 9. washington. D.C.: Society of
American Foresters.

11. Schlaegel. B.E.. 1975. Yields of four 40—year-old coiifers ard aspen in
adjacent stards. Can. J. For. Res.. 5. 278. 280.

12. Green. D.C. ard Grigal. D.F.. 1978. Generalized Biomass Esti_m_ation
Equations For Jack Pine (Pinus banksiana Lamb.). Minnesota Forest
Research Note No. 268. St. Paul. Minnesota: University of Minnesota.
College of Forestry.

Literature Cited

Hudson. D.H. and Kittleson. K. . 1978. Identification o f wo o d e n e r gy
resources in central Michigan. —Remote Sensing Project Publications.
Fast Lansing. Michigan: Michigan State University.

Kittleson. K.. 1979. A Discussigon of Recent Forest Biomass Literature

Concerning Survey Designs. gSampling and Nbdeling Procedures. Variable
Selection ard Nutrient Reserves. East Lansing. Midnigan: (Mimecgraph).

Malkonen. E.. 1973. Effect of complete tree utilizations on the nutrient

reserves of forest soils. IUFRO Biotass StLdies. pp. 375-386. Oroo.
Paine: lhiversity of whine.

Michigan Department of Conservation. 1968. Forest Type Mapping Instructions
and Type Symbols. Department Letter #176. Lansing. Michigan: Michigan
Department of Natural Resources . -

Michigan Lard Use Classification ard Referencing Committee. 1975. Michigan
land cover[use classification system. Lansing. Michigan. Michigan
Department of Natural Resources (mimeograph) .

 

Sionranza. L.P. ard Carpenter. G.F.. 1980. Future Pipeline Development in
the Northeastern Michigan Coastal Zone and Niagaran fiend: A Regional
Analysis of Environmental Resources ard Land Use. Lansing. mchigan:
Michigan Public Service Commission. Scientific Research ard Evaluation
Office.

APPENDIX B
1. PROGRAM FLONDIAGRAMS AND FORTRAN SOURCE CODE

LISTINGS FOR PROGRAMS I AND II AND MAGEN

2. EXAMPLE OF MPOS TRANSPORTATION COST MINIMIZATION
INCLUDING: BATCH PROGRAM, MPOS PROBLEM PROGRAM

AND MPOS OUTPUT

\

PRQGRAM.I

! START '
{II}...
I
l DIMENSION ACRES,TYPE,FMT,LASTT
I REAL ACRES,TYPE TONS,INDEM,LAST,LAST1
I INTEGER YESNO,X.Y,RANUM,FMT
l

 

I
(DATA INDEM = 1195575yYEAR = 1982,
GRONTH = 1.16, G? = YEAR-1982
FMT(1)=ZH1
FMT(2)=ZHX,
PMT(6)=ZH(/
FMT‘7)=ZH),
FMT(8)=ZH1X
FMT(9)=ZH,2
FMT(10)=ZH(I
FMT(11)=233,
FMT(12)=ZH1X
FMT(13)=2H))

 

I=1.150

 

 

ACRES(I)=0.
LASTT(I)=0.
TYPE(I)=0.
TONS(I)=0.

 

 

SUMTNS=0.
1:0.
1:0.

 

\ WRITE: YEAR, "HAVE YOU SET THE SKIP LOOP /
\ CORRECTLY? (Y=YES.N=NO)' /

 

 

 

 

 

\ READ: YESNO /
\ /
\ /
\ /
I
I
I I
I I
I I
I I
I I
I YESNO=NO I ---YES--
I I g
I 7 I I I I
I I I I
I I GO TO SOOOI
0 C C C
I I I I
I
NO
I
/
\ SKIP USED RANDOM NUMBERS IN DATA FILES /
\ NUMBER 8.9 AND 10. /
\ /
\ /
I
I
I=1.150 I
I
I
/
\ LASTT(I),ACRES(I).TYPE(I) /
\ /
\ /
\ /

 

 

 

GFA=GFIGROWTH
THDNDzGFA+39.55
TCON=GFA+33.32
TMIX=GFA+33.9O

 

I=1.150

 

LAST=0

LAST1=0

LAST2=0

LAST2A=0

LASTZB=0

LASTZC=0

REWIND 5

RENIND 7
LAST=LASTT(I)
LAST1=LASTT(I)-1.

 

 

 

 

 

 

 

 

 

/
WRITE LASTI /
READ LASTZA,LASTZB,LASTZC /
(CONVERTS INTEGER T0 ALPHANUMERIC) /
/
I
WRITE: LASTZA,LASTZB,LASTZC /
/
I
READ X,Y /
I
I l
I
I I
I I
I I
I I
I I
LASTZ 1570 AND I- YES-
I s 5390 I I
I I
I 7 I I
I I I X=X+50
I I g .
I
I
I
I I
I I
I I
I I
- 1.1312 5393 AND .
.S 10751 I -133-
I I I
I 7 I I
I I I
e I | X=X+100 I
I I g I

 

 

 

 

 

 

 

 

 

E E
E E
E E
E E
I LASTz 10752 I --YES-
E E I
I 7 I I
I I I
I I I X=X+150
E E I
E
I
E
E E
E E
E E
E E
E E
I TYPE(I)$ 42.41 I I --YES--
I I I
E '2 g I
E E EEE
E E E E
I I I GO TO I
I I 200 I
I EEE
EEE
E E
I GO TO I
- 21o -
EEE
I
EEE
E E
I 200 I
E E
EEE
I
I TONS(I)=THDNDIACRES(I) I
I I I
\ WRITE: I,LAST,X,Y,ACRES(I),TONS(I) /
\ /
I
\ NEITE:"STUMPAGE TYPE IS HARDWOOD.a /
\ /
I
EEE
E E

I GO TO I

I 202 I
III

I I I
I
I I I
I I
I 230 I
I I
I I I

TONS( I ) =TCONIACRES( I)

I I
I
I GO TO
I 230
I I I
I--IES--
I
I
I

 

\

\

\

\

I
WRITE: I,LAST,X,Y,ACRES(I),TONS(I)

 

I
WRITEz'STUMPAGE TYPE IS CONIFER"

 

I I
' 220 I
I I
III

I
I TONS(I)=TMIX'ACRES(I) I
I

I
wnITE:I,LAsr,x,Y,ACRES(I),rons<x)

 

 

I
WRITE: "STUMPAGE TYPE IS MIXED."

 

I I
' 202 i
I I
III

I
I sourus=suurN3+rons(I) I
I ' I
I
WRITE: sunrus «Tons SOLD so FAR."

 

I I
I I
I I
I I
I I
* SUMTN82,INDEM *--IES--
I I
I I
'I I
I I I
I I I
I I
l
I I I
I I
9 222 i
I I

 

 

 

\ WRITE: "TOTAL INDUSTRIAL NETWORK /
\ DEMAND HAS BEEN MET. SIMULATION /
\ COMPLETE." SUMTNS “TONS SOLD." /

\ /

l

\

\ WRITE: I,"LOCATIONS, ACREAGES, AND TYPES /
\ VALUES USED IN THIS RUN." /
\ /

\ /

I
I I I
I I
' GO TO 5
' 1005 '

I I I
I
I I I
I I
' 5000 '

I I
I I I

I
\ NRITE: ' THEN DO IT now.-

 

\ /
I
I I I
I I
5 1005 '
I I

' END '
I I I I I I I

23:
24:
25:
26:
27:
28:
29:
30:
31:
32:
33:

352
36:
37:

39:

41:
42:
43:
44:
45:

47:
48:
49:
50:
51:
52:
53:
54:
55:
56:
57:
58:
59:

MAM].

C (HEATION AND SJPPLY SMJLATOR. . .)
C

DIMENSION ACRES( 150) ,TYPE( 150) ,TONS( 150) , FMI'( 13) ,LASTN 150)

CIIIIIIIIIIIIIIII***II*III***II***III***I***IIIIIIIIIIIIIIIIIIIIII
C ***REMEMBER TO:

C REDDdENSICNAILARRAYSTUlwWHENADDINGWDODENEIGYPLANT”
c*****************************************************************

C
C*** ‘IHINGS'IUIIDBEFOREEACHRUN:

C LSE'I‘SGPREAD
C 2.SE1‘YEAR
C

REAL ACRES,TYPE,TONS, INDEM,L.ASTI‘,LAS'I‘,I.AST1
INTEISER YBSMD.X.Y.RANLM.FMI'

(SEI' INIIJSTRIAL DEVIAND. . .)
910675 IS IND. DEMAND FOR BASIC SET,NO WOOD mm PLANT
1195675 WITH I‘DOD ENEEGY PLANT
[ETA EDEN/1195675. /
[ETA YEAR/2012./
ETA MIR/1.16/

@YEAR-1982 .

0000

0

EMI‘Il)=2H(1
M(2)= mo
FMI‘(6)= 21W
M(7)= 21):
EMT(8)= 2HlX
FMI‘(9)= 2H,2
M(10)= 2H(I
FMI‘(11)=2H3.
EMT(12)=2H1X
FMI‘(13)=2H))

C (ZEIUOUTARRPXS ANDVARIAEIS...)
II) 13 I=1.150

make.
ms'rr(1)=o.
m(1)=o.
m(1)=o.
13 CONTINUE
ammo.
x=o
Y‘=0

WRITE(1,234)YEAR
VRITE(2,234)YEAR
234 EDMT(///.IX.'** YEAR 18 '.Fs.o.' **'./)
WRITE(1,1001)
1001 mm<1x.s(/).70('*').////.1x.' HAVE You
+SETTTIES<IPLOOPANDGRCNTHFACIORCDRRE€PLY???
+ (Y=YES,N=NO) ',/)
C
READ(1,1002)YESI\U
1002 FORMATUAI)
IF(YESNO.ED.1HN) G0 '10 5000
C

Ck'k*********************‘k***‘k‘k******************************************

61:
62:
63:
64:
65:
66:
67:
69:
70:
71:
72:
73:
74:
75:
76:
77:
78:
79:
81:
82:
83:
84:
85:
87:

90:
91:
92:
93:

95:

97;

OOOOOOOOOOOOOOOOQOOOOOO00000000

1006

ck*******************************************************************
(READ RAW ACRES AND mus. . .)

C

110

CI*************************************

C

(USE SUP ILDP ONLY AFTER 18T KIN...)

'IO 9GP) '

(NEWACRESANDTYPES

64 W12
64 PNl3
38 PN14
51 PN15
43 H116
16 W14CIDRREXZI‘ION
46 PN17
99 1182
84 1187
74 1192
73 1197
60 1102
59 1107
52 ' 1112
109 11182
% 11187
76 11192
77 11197
59 11102
56 11107

READ(9. 1006)

READ(8. 1006)

READ(10. 1006)

FOMAN IX, 1298( / ) )

MAINS, 110)LAS'IT( I)
READ(9, 110)ACREB(I)
FORM(1X.F10.0)

READ(10. 110)TYPE( I)
(DNI‘INUE
G‘WWI‘H
MAI-39.55
WMBBJZ
MX=GE'P:I-33.90

ID 222 1=1.150
EST-=0
LAS'I'1='-O.
1A8T2=0
IASTZAFO
LASTZB=0
IASTZGO

X=0

Y=0

—--= WWWLBEDSOFAR(LINFS

SUM

64
128
166
217
260
276
322
421
505
579
652
712
771
823
932
1030
1106
1183
1242
1298

DATE

(5000 OF EACH) CREATED 12JAI>BZ...)

7FEB82
7FE882
7FE882
7FE882
7FE882
10FEB82
?
13FEB82
14FE882
15FEB82
15FE882
15FE882
16FEB82
?
m2
m2
GMAIBZ
W2
HVIAR82
m2

MAINlmP

151:
152:
153:

HEW-IND 7
IASIELASTIT I)
IAST1=IASTI‘( I ) -l .
C
Cue-III***I***-I*I***IIIIIIIIIIIIIIIIIIIII*I mULIZE
REWIND 6
WRITE(6 , 164)I.AS'I'1
RBVIND 6
READ(6, 165) IASTZA. MBIIAS‘IQC
164 FOHIAT( 1X, F10.0)
165 FDRMAT( 5X,A1 , 2A2)
CIIIIIIIIIIII**IIIIIIIIIIIIIIIII***IIIIIII CRT VERIFICATION PRINT
WRITE‘ 1 , 166) IASTZA. LASTZBgLASTZC
166 FUWAT(1X,' SKIP=',1X,A1,2A2)
C
M ( 3 )= IASTZA
M(4)= IASTZB
FMI‘( 5)= IASTZC

)FO
Y=0
READ(7,FMI‘)X,Y

IE‘(LAST.GE.1670. .AND.LAST.LE.6392. )x=x+50
IF(LAST.GE.6393. ....ANDLASTLE 10751. )x=x+100
IF(I.AST.GE. 10752. )x=x+150

c

C *****************************************
IF(TYPE(I).LE.42.41) GO ID 200
Go '10 210

200 'IDNS(I)=4'HDND*ACRES(I)

c ORIG HDWD mamss
mm(1,111)I.msr,x,Y.ACREs(I),TONsu)
WRITE(2.111)I,IAST,X,Y,ACRES(I),TONSu)

111 mmATI/,1x,' Doc. W '13,' RANLM= ',F6.0,/
+,1x,' S'I'IMPAGE SALE LOCATION COORDINATES = ',
+2(1x.15),/,1x,' ACRES IN SALE =- ',1X,F6.0,/,1X.' TONS 015‘ BICMASS
+ SOLD -- '.1X.F10.0)

mITE(1.112)
VRITE(2,112)

112 mmux,‘ STLMPAGE TYPE IS HARDNOOD ')

00 TO 202

210 IF(TYPE(1) .LE.61.81) G0 '10 230
(I) '10 220

230 IWS(I)#I'(DN*ACRES(1)

C ORIG CDNIFER 'IDNS'=33.32
WITE(1,111)I,IAST,X,Y,ACRES(I),TONS(1)
W(2,111)1,1AST.X.Y,ACRES(I),TONS(1)
WRITE(1,113)

WRITE(2,113)
113 FOHGATUX,‘ STIMPAGE TYPE IS CDNIFER ')
GD '10 202

C

C

220 TUNS(I)=‘IDE)(*ACRES(1)

C ORIG MIXED TONS-133.90
WRITE“. 111)1,LAST,X.Y,ACRES(1) ,TONS(I)
WRITE(2,111)1,LAST,X,Y,ACRES(I) ,TONS(I)

 

 

 

181: WRITE(1.114)
182: warm 2, 114)

183: 114 mMTuX,‘ STLMPAGE TYPE IS MIXED ')
184: C
185: 202 SMTNS=SMNSVPONS(I)
186: C
mmuaeflsmms

187:

188: WRITE(2.365)SINI‘NS

189: 365 FORMAT(40X,F15.0, ' TONS SOLD SO FAR ')
190: IF(SIMINS.GE.1NDEM)GO '10 203

191: C

192: 222 mm
193: Cki*************************************************************

WRITE“, 115) SIMI'NS

194: 203
19s: WRITE(2,115) SIMI‘NS
196: 115 SOMAT(//,1x,' TOTAL INDUSTRIAL NETWORK DEMAND HAS BEEN m.
197: + SIMULATION ompIEI‘ED. ',/,1x,F15.0.' TONS SOLD ',/)
198: C

199: WRITE(1.1004)I

200: WRITE(2,1004)I

201: 1004 EOMT(/.1X,15,' LOCATIONS,ACREAGE AND TYPE VALUES
202: +IBED1N'IHIS RUN.')

203: C .

204: GD '10 1005

205: 5000 CDNTINUE

206: WRITE(1,1003)

207: 1003 mm(1x,10(/),1x,' THEN Do IT NCW YOU 01mm ')

208: C
209: 1005 CDNI‘INUE

210: END

 

 

 

 

BBQQEAM Z

I I I I I I I
' START '

IIIIIII
I
DIMENSION ORIGIN,DESTIN,XDIST,YDIST,
DIST,RDIST,NITSED,COST

I
I
I
INTEGER 0,D,1ESNO I
I
REAL NITSED I

I

 

I
DATA RTF/1.36/ I
NITSED/134.,153.79..12.,9..13., I
25.,55.,131.,92.,h6.,52.,55.,39./I
I
I
l

D/7/

 

1:1,10

 

J=1,2

 

—~_*-—~u-

 

DESTIN(I,J)=0.

 

 

1:1,1100

 

XDIST(I)=0.
YDIST(I)=0.
DIST(I)=0.

RDIST(I)=0.

 

 

 

\

\

WRITE:"WHAT YEAR 18 I

\

READ:

\

YEAR

COUNTzYEAR-IQBZ
C=COUNT’.05
COSTPM=.093

I
I
I
I
I
I
T

7

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\ "CORRECT? (Y:YES,N=N0)

\

/I l
I I
/ I I=1,D I
/ I I
I
/I I
I I
/ I '17-‘92 I
/ I I
I DESTIN(I,J)=NITSED(J,I) l
I I
I\ I
I \ I
I 5 I
I \ I
\ WRITE:'HOH MANY ORIGINS D0 I00 HAVE IN THIS
\ SYSTEM? (MAX IS 150 NITHOOT REDIMENSIONING /
\ ORIGIN.)"
\ /
I
\ READ:0
\ /
I I
/ I I
/ 6 I I
' I
\ NRITE:'ENTER (x,y) COORD 0F ORIGIN IN
\ (88.0,1x,ru.0) FORMAT" /
\
I
\ READ: (ORIGIN(I,J),J=1,2)
\
I
\ WRITE: (ORIGIN(I,J),J=I,2)

 

 

 

 

 

\ READ: YESNO /
/
I
I
I I
I I
I I
I I
I I
' IESNO=IES 9 --YES--
I I I
I 7 I I I I
I I I I
0 I I 00 T0 35 I
I I I I
I I I I
I
N0
- I
\ NRITE: "ENTER CORRECT (x,I) COORDS FOR ORIGIN"/
/
I
I I I
9 00 T0 9
' 2979 '
I I
I I I
I
I
I
I
I
I I I
I I
I 35 I
I I
I I I
I
I xc=1 I
I K3=O'D I

 

 

 

 

I
I XDIST(KC):ABS(0RIGIN(I,1)-DESTIN(J,1) I
I YDIST(KC)=ABS(ORIGIN(I,2)-DESTIN(J,2) I
I DIST(KC)=SQRT(XDIST(KC)992+YDIST(KC)992I
I I

 

 

 

 

 

I I
I I
I I
I I
I I
9 J = u 9 --YES--
I 9 I
I ‘z 9 I
9 9 I DIST(KC)=
I I I DIST(KC)+7h
I I I
9 I
I I
I
I I
I KC:KC+1 I

1:1,!3

 

 

I RDIST(I)=DIST(I)9RTF
I COST(I)=RDIST(I)9COSTPM
I

 

 

H—..-...-

\ WRITE: "ORIGINS 1N TH S SET ARE...9 /

 

 

 

 

\ NRITE:(0RIGIN(I,J),J=1,2) /
\
I
\ I
0\ I
\ I
I
\ WRITE: "ORIGIN DESTINATION TRANSPORT COSTS /
\ (3/GREEN TON)"
I
I KK=1
I
I I
/I I
11 I 131.0 I
I I
I
I I
/I I
12 I J=1,D I
I I
I
\ WRITE: KK,I,J,COST /
\
I
I KK=KK+1
I
I
\ I
2\ ‘I
\ I
I
\ WRITE: LINE FEED /
\
I
\ I
1\ I
\ I
I
\ WRITE: PAGE FEED /
\

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

END
IIIIIII

 

1: PROGRAM 2

2:(2
: DIMENSION ORIGIN(1SO, 2) ,DESI'IN(150, 2) ,XDIST( 1100) ,YDIST(1100) .
4: + DIST(1100),RDIST(1100) ,NITSED(2,7) ,COST(1100)
5:(2
6: INTEGER o,D,YESNO
7: REAL NITSED
9: c EUR 11 SERIES, mOD ENERGY PLANT AT ROSE CITY(90.120;
10: C FOR 111 SERIES. mOD ENERGY PLANT AT IDLEWILD $55.39
11: c LJ) NOT (x,Y)1u
12: DATA RTF/1.36/
13: C
14: DATA NITSED/134.,153., 79.,12., 9.,14., 25.,55., 131.,92.,
15: + 46.,52..55.,39./
16:(2
17: [IRA D/7/
18: C ***************************************I***************************
19: C (ZERO OUT ARRAYS. . .)
20: DO 200 I=1,10
21: DO 200 J=102
22: ORIGIN(I,J)=O.
23: DESTIN(1,J)=O.

24:
25:
26:
27:
28:
29:
30:
31:
32:
33:
34:
35:
36:
37:
38:
39:
40:
41:
42:
43:

45:

47:

49:
50:
51:
52:
53:
54:
55:
56:
57:
58:
59:
60:

200 CONTINUE
DO 201 1=1,1100
XDISI‘(I)=0.
YDIST(I)=0.
DEST(I)=0.
RDIST(1)=0.

(3

201 CONTINUE
D0 222 1=1.10
DO 222 J=1.10

222 CONTINUE

C ***************************************************************

C SET YEAR. . .
WRITE(1,9999)

9999 1501214142010

C
mm(1.545)

545 mmAT(/,1x.' WHAT YEAR IS IT? ',/,1x,'(Es.0)'./)
READ(1,546)YEAR

546 FOMTIFS.O)
WYEAR-1982.
o-OOUNT'OS
COSTEM=.O93+C*(.093)

c
00 211 I=1,D
DO 211 J=1.2
IES'I‘INII.J)=N1'ISED(J,I)

211 CONTINUE

C
WRITEu,37)

37 mmAT(//,' HCW MANY ORIGINS Do YOU HAVE IN THIS SYSTEMTJ

+ 1x, ' (MAX IS 150 WITHOUT REDIMENSIONING ORIGIN.) ',/)

READI1.38)O

38 mmum)

C

c (READ OOORDS OF ORIGINS FRCM CRT. . .)
DO 35 I=1.O

CI:
62: 34
63:

64:

65: 2979

66: 33
67:

68: 333
69:(3
70:

71 : 2976

72: C
73:
74:
75:
76:

77: 2978

78:
79:<2
80:1fi5
81:(2
82:
83:
84: C
85:
86:
87:
88:
89:
90: C
91:
92:
93: 20

94: C

95:

96:

97:¢C

98:

99: 31
100: C
101:
102:
103:
104:
105:
106:

234

2977

L07:
.08:
09:
10:
11:
12:

236
235
C

237

VRITEII.34)1
TORIAT(//,1x,'ENTER (x,Y) COORD OF ORIGIN',13.' IN
4» (T4.0,1X,E4.0)

+ './)
READI1,33) (ORIGIN(I,J) ,J=1,2)

EOMAT(F4.0.1X,F4.0)
WITE(1,333) (ORIGDI(1.J) ,J=1, 2)

mmAT(1x,2(E4.O, 1x))

VRI'I‘EI1,2976)
FORMAT(/,1X,' CORRECT ?

READ( 1, 2977)YESNO

FORMAT(1A1) I
IF'IYESNO.E0.1HY) GD '10 35

IRITE(1,2978)I
FDR‘IATUJX.’ ENTER CORRECT (x,Y) COORDS EOR ORIGIN',I3,/)

(13102979

(Y=YES,N=NC) ',/)

(DNTINUE

IC=1
K3=O*D

II) 20 I=1,0

m 20 J=1.D
XDIST(KC)=ABS(ORIGIN(I, l)-DESTIN(J. 1))

YDIST(I<C)=ABS(ORIGIN(I. 2)-DEST1N(J, 2))
DIST(IC)=SQRT( (XDIS'T(KC)**2)+(YDIST(KC)**2))
+74 MILES EOR DESTINATION 4

IF(J.ED.4)D15T(KC)=D1$T(KC)+74.
IC-‘KC-I-l
(IDNI‘INUE

EDIS'I‘(I)=DIST(1)*RTF
CDS'I‘(I,J) CALCULATED ON S/GREEN TON/MILE

(DSTI I)=RDISI‘( 1)*COSI‘PM
CDNTINUE

WITE(1 . 234)

mITE(2.234)
FORMAT(///,1X.' ORIGINS IN THIS SET ARE...',/)

II) 235 I==1,0
WRITEKI, 236) 1. (ORIGIN(I.J) ,J=1. 2)

WRITEXZ, 236) I, (ORIGIN(I,J) ,J=1. 2)
mmAT(1X, 13. 1X,2(F4.0, 1X))

CDNTINUE

WTEK 1 , 237)
WRITE( 2, 237)
FORMAT( // , 10X, ' ORIGIN DESTINATION TRANSPORT (0813

+ (s/GREEN TON) ')

KK=1
II) 30 I==1,0

mlTE( l, 117)KK,I,J.(IJST(KK)

WRITE( 2, 117)KK.1.J,(I)SI‘(KK)
KK=KK+1
CIDNTINUE

 

 

1.4L 3
122:
123:
124:
125:
126:
127:
128:
129:

186

117

METER .L , 18b)

WRITE(2, 186)

EOFMATU)

CONTINUE
EOMT(1X,3(13.7X) .F10.2)

WRITE(2,9999)
WRITE(2.9999)
EDH)

PROGRAM MAGEN

9 START 9
I I I I I I I

I
DIMENSION COST,BOTUM,SO,DD,COSTT

I I
I I
I INTEGER OUTPUT I
I INPUT=1 I
I Npos=2 l
I BOUNCE=3 I
I OUTPUT=4 I
I I

 

I
\ NR1TE:'NELCONE TO THE RESOURCE DEVELOPMENT /

-\ DEPARTMENT TRANSPORTATION COST LINEAR PROGRAM/

 

 

 

 

 

 

\ HOW MANY ORIGINS DO YOU HAVE?(MAXINUM IS /
\150)? DON'T USE A DECIMAL POINT!" /
\ /
I
\ READ:NO /
\ /
l
\ HRITE'ROU HAN! DESTINATIONS DO YOU HAVE /
\ (MAXIMUM IS 50)? DON'T USE DECIMAL POINT /
\ HERE. EITHER.) /
\ /
I
\ READ:ND /
\ /

 

' l
\ NR1TE:91F YOU RANT TO USE A SCALING FACTOR, /

\ ENTER IT. (USE DECIMAL POINT.) IF NOT, TYPE /
\ 1.0 AND TYPE RETURN." /
\ /

 

\ READ:SCALE /

SCALE=O.

 

SCALE=1.

 

 

 

I=1,N0

 

SO(I)=0.

 

 

J=1,ND

 

COST(I,J)=O.

 

 

 

1:1,ND

 

 

DD(I)=0.

 

 

YESNO=O.

YESNO1=O.
YESN02=0.
YESN03=O.
YESNOASO.
YESN05=O.
YESNOGSO.

 

I
\ HRITE:"1NPUT COST MATRIX STARTING 1N UPPER LEFT /
\ HAND CORNER. ENTER ONE ROW AT A TIME, FROM /
\ LEFT TO RIGHT.(TYPE RETURN AFTER ENTERING /
\ EACH ELEMTNT). (DON'T FORGET TO USE A. /

 

 

 

 

 

 

 

 

 

 

\ DECIMAL POINTI)' /
\ /

I

I xxx=1 I

I I
I I
I I
l I=1,NO I
I I
I
I I
I I
I I
I I
I

\ NRITE:xxx /
\ . I
I
\ READ:COST(I,J) . /
\ /

I

I xxx=xxx+1 l

I
\ NR1TE:9INPUT SUPPLY AT ORIGINS 1 THROUGH 9 NO /
\ 9(IN ORDER)(ENTER ONE SUPPLY VALUE (WITH A I
\ POINT) AND TYPE RETURN.)' /
\ /
\ /

I xxx=xxx+1 I

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

/I I
/ I I
/ 5 I I=1,NO I
/ I i
\ NRIIszxx /
\ /
I
\ READ:SO(I) /
\ /
I
I xxx=xxx+1 I
I I
I\ I
I \ I
I 6 \ I
I_______\ I
\ HRITE:"INPUT DEMAND AT DESTINATIONS 1 THROUGH" /
\ ND '(IN ORDER) ENTER ONE DEMAND VALUE (NIIE /
\ DECIMAL POINT) AND TYPE RETURN.)" . /
\ /
I
I KKK=I I
I l
/I I
/ I I
/ I I
\ NRIIszxx /
\ /
I
\ READ:DD(I) /
\ l
I
I xxx=xxx+1 I
I I
I\ I
I \ I
I 7 \ I
I \ I
I

 

\ NRITE:'PLEASE CHECK ALL YOUR INPUT CAREFULLY. /
\ THE FOLLOWING ARE YOUR COST MATRIX ELEMENTS. I
\ VERIFY OR CORRECT EACH ELEMENT AS IT /

\ APPEARS.” /

\ /

 

 

 

 

 

 

 

 

 

I an I
I
I I I
I I
I I
I I=1,NO I
I I
I
I I
I I
I J=1,ND I
' I
\ NHITE'FROM ORIGINn I "TO DESTINATION" J /
\ ”UNIT TRANSPORT COST IS' COST(I,J) /
\ I
- I
\ WRITE: "IS THIS THE CORRECT VALUE?(Y=YES,N=NO) "/
\ /
I
\ READ:YESNO1 /
\ /
I
I
I I
I I
I
I I
I I
0 YESN01=NO I --YES--
I I
I 7 I
O O
C O O
I I
I I
I O
I I I
C
I GO TO 37 0
I
I I I
I
...
I
I 33 I
I
...
I
\ WRITE:'HHAT IS THE CORRECT VALUE? (USE A /
\ DECIMAL POINT.)' /
\ /

 

' GO TO 38 '
I

—D\

\ READ:'COST(I,J)

 

 

 

 

 

 

 

 

 

 

 

\ /
I
I
I
I
I
I
I
I
I
I
...
37
...
I
O
O O
O
O
C
' YESNOU=NO ' --YES--
. I
I 7 III
O O
. GO TO ”5
E C O
. ...
I
\ WRITE3THREE LINE FEEDS /
\ /
I
...
n7
...
I
I
I=1,NO I
I
I
\ HRITE3'THE SUPPLY AT ORIGIN. I 'IS' SO(I) /
\ /
I
\ WRITE: 'IS THIS THE CORRECT VALUE?(Y=YES,N=NO) '/
\ /
I
\ READ3YESNOZ /
\ /

 

O/

 

 

 

g G
O D
l I
I I
- YESN02=NO 0 --TEs--
I O I
I 7 I I
l I .5.
II C I C
' ' ' GO TO no '
I O I
I III
5"
I I
I GO To 39 *
I G
.3.
I
5'.
G G
I no I
O I
CC.
I
\ HRITE:"HHAT IS THE CORRECT VALUE? (USE A /
\ DECIMAL POINT.)' /
\ /
I
\ READ:SO(I) /
\ /

 

 

11

 

 

I I
I I
I I
I I
E YESN05=NO E --YES--
I ' I
I ? ' I
I I I I I
I I I I
E E E GO TO 53 E
I I I
I I I I
I
\ WRITE3THREE LINE FEEDS /
\ /
I
I I I
I I
I 50 I
I I

I
I
I
I=1,ND I
I

 

I
\ HRITE:"THE DEMAND AT DESTINATION," I 'IS' DD(I) /

 

 

 

 

 

\ /

I
\ WHITE: '18 THIS THE CORRECT VALUE?(Y=YES,N=NO) "/

\ /

I
\ READ:YESN03 /

\ /
I
I

I I
I I
I I
I I
I I
' YESN03=NO ' --YES--
I ‘ I
I 7 ' I
I I I I I
I I I I
U P ' GO TO 42 F
I I

 

 

 

 

 

 

I I
EOOTOME
I I
III
I
III
I I
I 42 I
I I
III
I
\ HRITEz'HHAT IS THE CORRECT VALUE? (USE A /
\ DECIMAL POINT.)' /
\ l
I
\ READ:DD(I) /
\ /
l
I
I
I
I
III
I I
I III I
I I
III
I
I
I I
I I
I I
I I
I I
' YESN06=NO ' --YES--
I I I
I 7 I I
I I III
I I I I
' ' ' GO TO 52 5
I I I
I III
I
III
I I
I ‘I5 I
I I
III
I
\ NRITE:"FINAL INPUT CHECK SEQUENCE. COST MATRIX/
\ ELEMENTS” /

\ /

 

 

 

 

 

 

 

 

 

 

 

I=1,NO
NRITE:I,SO(I) /
/
I
\ I
II I
\ I
I
HRITE:'ARE ALL THESE VALUES CORRECT? /
(Y=YES,N=NO).' /
/
I
READ3YESNOS /
/
I
I
I I
-I 'I
I I
I I
I I
E IESN05=NO E --YES--
I ' I
I ? ' I
I I I I I
I I I
E E E GO TO A?
I I
I I I I
I
HRITE:THREE LINE FEEDS /
I
I
I I I
I I
I 52 I
I I
I I I

NRITE:'DEMAND AT DESTINATIONS 1 THROUGH" ND /
/

I=1,ND

 

 

 

NRITE:I,DD(I) /

11

U'I/

 

\

 

 

 

 

 

 

\ WRITE:"ARE ALL THESE VALUES CORRECT? /
\ (Y:YES.N=NO).' /
\ /
I
\ READ:YESN06 /
\ /
I
I
I I
I I
I I
I I
I I
' IESNOG=NO ' --YES--
I ' I
I 7 I I
I I III
I I I I
E E -E601'050E
I I I
I III
I
IK=1 I
I I
I=1,NO

 

 

 

I
I
I
I
I
I
I
l
I
I
)

I COSTT(K)=COST(I,J
I K=K+1 I
I I

 

 

12

 

 

I=1,NO

 

l1

BOTUM(I)=SO(I)

 

N:NO+ND

H=NO+1
J=1

 

 

\8

 

 

 

)
J
I.\
D
D
.-
\l
I
(
“1
0+
TJ
O.-
" BJ
’
H
: III
I
I
9
’1
/

 

 

 

K2=NO'ND

 

 

 

)

I

(

T

T

S

O

nu

2 :

K E

a T

1. I

= R

I N

\.

_/
_0
1,2
/.
_/

 

ONE=1.0
ZERO=0.0
K=NO+1
L=ND-1

 

 

 

I=1,K

 

’2

13

II=1,ND

HRITE:ONE

 

 

 

/
,2

\

 

III=1,NO

 

I
,2

IIII=1,ND

'ZERO

I

NRITE

 

 

\

 

I
I2

 

 

 

1:1,ND

HRITE:ONE

 

 

\

 

I
’2

 

 

L

1

.-

J
\ I!!!

l
\ 6
l2
\5
2 l
\

Ill... /

1H

 

 

FL

 

 

I=1,NO

’2

 

 

NRITE:ONE

\

II=1,L

 

 

 

I
,2

NRITE:ZERO

\

 

 

 

 

\8 \7
\ \

NRITE:ZERO

\

 

 

\6
\

1:1,!

,2

 

 

NRITE:ONE

\

 

 

 

1"...

l3

HRITE:ZERO

\

 

 

 

\0

15

NRITE:ONE

 

 

1,N

NRITE:BOTUM(I)

 

 

\

 

 

STOP '

16

PRCXSRAM WGEN(TAPE1,TAPE2.TAPE3,TAPE4)
DIMENSION CDSI'(150, SO) ,KYI‘LMUSOO) ,
+ 80(150) ,DD(SO) ,(DS'I'I‘WSOO)

INTEGER GH'PUI‘

LABEL LUN

INPUT

CIJTPUI'
MPOS

m

wN-bI-E

m==NmBER OF ORIGINS
ND=NLMBER OF DESTINATIONS
WIDE FACIDR
OOST(I,J)=OOSTOI'mVImONEUNITERMITDJ
SO(I)-SUPPLY AVAILABLE AT I
DD(J)=DIMAND AT J
mn(J)=- Rm OE CONSTRAINT RIGHIHANIBIDES
m1
MPOS=2
BOUNCE=3
W

00000000000000000 0

333 OONTINUE
‘ WRITE(OU'IPUI‘,1000)
1000 IomAT(1x,IO(/),7O('*')/)

YMTE(OIH'PUI‘,1001)
1001 FORM(1HL5X,' WELCCME TO THE RESOURCE ',

+ ' DEVELOPMENT IEPARIMEN'I‘ '/5x,' TRANSPORTATION (DST ',
+ ' LINEAR PROGRAM ',///70('*'),5(/),10x.' ch MANY ORIGINS ',
+ ' DOYDU HAVE (MNCIMUM Is 150)? '/10x,' (DON"T (BE A '.
+ ' DEcmAL POINT1)'/)
c
READ(INPUI,1002)NO
1002 mmuz)
c
WRITE(OUTPU1‘,1003)

1003 FORMATU/lOX.’ Hm MANY DESTIM‘I‘IONS Do YOU HAVE (MAmeM '.
+ ' IS 50)? '/10x.' (DON"T [BE DECIMAL POINT HERE, EITHER)

+ '/) ‘
C
READ( INPUT, 1002)ND
C
WRITE(GJTPUT, 1004)

1004 FOMT('O',1OX,' IF YOU WANT To LEE A SCALING FACTOR. ',

+ ' ENTER IT. (USE DECIMAL POINI‘.)'/)

MTEwUI'PUT, 1026)
1026 IDmATux,‘ IF NUI‘, TYPE 1.0 AND TYPE RETURN. W
C

READ(INPUI,1005)SCNE

1005 I03MAT(F60.4)
c

C
C (SETSCALEhL IFNOSCALEENI'ERED...)

61:
62:
63:
64:
65:

C

00038

32

33

C

IF(SCALE.m.O.) GO 'ID 30
GO TO 31
m1.
CDNTINUE

(ZERO ARRAYS AND VARIABLES. . .)
DO 32 I=I,NO
so(I)=o.
DO 32 J=1.ND
ODSTII.J)=O.
OONTINUE
DO 33 I=1,ND
OD(I)=O.
OONTINUE
W.
YESNO1=0.
YESNO2=0.
YESNO3=O.
YESNO4=0.
YESNOS=0.
YESNO$=O.

WRITE(OUTPUI‘, 1006)

1006 IDmAT('O',sx,' INPUT COST mm STARTING IN UPPER '/5x.

+'LEFTHAND00RNER. ENTERONERavAT'/sx.'ATmE'.
: +‘FRIMIEET‘IORIG-l'1‘. (TYPEREI'URNAFTERENI‘ERDBEAQ—I'.

89: +'ELEMENI‘.) '/5X.' (DQI"TEOK§EI"IOUSEADECD4AL',
: + ' POINT!) '/)
91: C
92: KKK-=1
93: II) 34 I=1,N0
: 11) 34 J=1,ND
9S: WRITE(0UI'PUI‘,396)KKK
: 396 EOIMATUXJS,‘ = V)
97: READ(INPU1‘,1005) COST(I.J)
: WKKlG-l
: 34 mm
100: C
101: WRITE(OUTPUT. 1007)N0
102: 1007 FOIMATGX,’ INPUT SIPPLY AT ORIGINS 1 'IHROUG-I ',12,' (IN ',
103: + ' ORDER) '/5x.' (ENTERONE SUPPLYVALUE (WI'IHA DECIMAL ',
104: + ' POINT)AND TYPE RETURN.) '/)
105: KKK-=1
1%: m 35 I=1,NO
107: WRITE(OUTPUI'. 396)KKK
108: IEAD(INPUT,1005)SO(I)
109: KKK=KKK+1
110: 35 mm
111: C
112: WRITE(0UTPUI', 1008)ND
113: 1008 FORMAT(SX,' INPUT DEMAND AT DESTINATIONS 1 Tim-I ',12.
114: +' (IN ORDER) '/SX,' ENTER ONE DEMAND VALUE (WITH DECIMAL '.
115: + ' POINI‘) AND TYPE RETURN.) '/)
116: C
117: KKK-“=1
118: II) 36 I=1,ND
119: WRITE(0UI'PUI‘, 396) KKK

finial Gm"..- 1| AAP\ ARI -\

121:
122:
123:
124:
125:
126:
127:
128:
129:
130:
131:
132:
133:
134:
135:
136:
137:
138:
139:

141:
142:
143:
144:
145:
146:
147:
148:
149:

151:
152:
153:
154:
155:
156:
157:
158:
159:
160:
161:
162:
163:
164:
165:
166:
167:
168:
169:
170:
171:
172:
173:
174:
175:
176:
177:
178:
179:

8%

00000

KKK?KKK&1
CONTINUE
(END OF INPUT SECTION...)
(BEGIN INPUT CHECKING SECTION...)

WRITE(4,1010)

**********************'k*************************************

1010 FORWATUX,‘ PLEASE cm ALL YOUR INPUT CAREFULLY. THE 'o
+' FOILOWING»AREIYOUR.OOST MATRIX.ELEMENTS. '/1XQ' VERIFY ':

£0 00

1009
(2

1011
(3

1012
38
1013

C3

37

800033

1016

+ ' ORCDRREIII‘EACHELEENI‘ASITAPPEARS. '/)
(mammooRRECrION PROCEDURE...)

DO 37 J=1,ND
WRITE(4,1009)I.J,COSI‘(I.J)
mRIAT(1x,/' E1204 ORIGIN '.Is.'
+ /.1x.' UNIT TRANSPORT OOST Is ',F15.4)

WRITE(4, 1011)

TO DESTINATION I.I5:

FORMAT(1X,' IS THIS THE CORRECT VALUE? (Y=YES,N=NO) '/)

READ(1.1012)YESNOI
FORMAT(1A1)
IF(YESNOI.EQ.1HN) GO TO 38
GO TO 37

VRITE(4, 1013)

IKfiOmflleg"WHAT IS THE‘OORRECT‘VALUE? '/1X;' (USE.A ',

+" DECIMAL POINT.) '/)

READ(1,1005)OOST(I.J)
(IINPDNUE
IF(YESNO4.EO.1HN) GOITO 45

(CHECK SUPPLY INPUTS. . .)
WRI'rE(4,3675)
FDMT(1X.3(/))

DO 39 I=1,NO
WITE(4. 1015) I.SO( I)

FOREIA'NIX,‘ THE SJPPLY AT ORIGIN ',IS,' IS ',F15.4)

WRITE(4.1011)
READ(1,1012)YESNOQ
IF(YESNOQ.EQ.1HN) GOITO 40
GO TO 39

WRITE(4. 1013)
READ(1,1005)SO(I)

CONTINUE

IF(YESNOS.EQ.1HN) GOITO 53

(CHECK DEMAND INPUTS. . .)
WRITE(4. 3675)

DO 41 I=1,ND

WRITE(4.1016)I.DD(I)

FORMATHX, ' THE DWAND AT DESTINATION ',IS,'
WRITE(4.1011)

READ(1,1012)YESNO3

mlm‘n‘: an Inn“ an m 4")

IS ',F15.4)

CI) '10 41
42 WRITE(4,1013)
READ(1.1005)DD(I)
41 CDNTINUE

IF(YESM£.ED.1HN) GO TO 52
C
C (FINAL INPUT (THEIR...)
C
45 WRITE(4.1017)

1017 FOWATPOHIX,‘ FINAL INPUI‘ CHECK SEQUENCE. ‘/)
54 WRITE(4,1019)

1019 EOHVIATUX,‘ COST MATRIX ELElVlENl'S '/)
C

II) 43 I=1.NO

II) 43 J=1,ND

WITE(4,1009)I,J,CI)SI‘(I,J)

43 CDNTINUE

C
WRITE(4,1018)

1018 FORMAT(1X,' ARE ALL THESE VALUES CORRECT? '/1x.
+ ' (y=YES,N=NO) V)

C
READ(1,1012)YESNO4
IF(YESNO4.ED.1HN) G0 '10 44
C
FRITE(4,3675)
53 WITE(4, 1020)NO

1020 FOREIAT(1X,' SUPPLIES AT ORIGINS 1 THERE"! '.IS)
II) 46 I=1,NO

WRITE(4.1015)I.SO(I)

CINTINUE

“A

WRITE(4,1018)
READ(1.1012)YESNC5
11"(YESDDS.ED.ll-IN) GO TO 47

(CHER DEMANDS...)

000

VRITE(4.3675)

52 WRITE(4.1021) ND

1021 FUWAT(1X,' DEMAND AT DESTINATIONS 1 THKXJGH ',IS)
11) 49 I=1.ND
WITE(4.1016)I.DD(I)

49 mm
WRITE(4,1018)
READ(1.1012)YESM£
IF(YESNI£.ED.1HN) G0 '10 50

u

C

Ck*************************************************************
Ck*************************************************************

C
C (WRITE SJPPLIES,DEMANDS. AND (DST [ETA “IO DATFIL
C FDR MPOS PROCESSING. . .)
C (CONVERTGJSP'IOAVECIOR...)
K31
II) 55 I=1,NO
II) 55 J=1,ND
CDS'I'I‘(K)=COST(I,J)
K=K+l

241:
242:
243:
244:
245:
246:
247:
248:
249:
250:
251:
252:
253:
254:
255:
256:
257:
258:
259:
260:
261:
262:
263:
264:
265:
266:
267:
268:
269:
270:
271:
272:
273:
274:
275:
276:
277:
278:
279:

:5

Eé’é?

N
m
3‘“ n u

287

(3(38}

213

730

750
740
20
C!

60
C

(WRITE amps! AND DEMAND VECI'OR...)
DO 62 I=1,NO

W(I)=SO(I)
CONTINUE

NhN0+1
EHNOWND

J=l

OO 63 I=M,N
W(I)=DD(J)
J==J+1

CONTINUE

(WRITE OBJECTIVE FUNCTION. . .)
K2=NO*ND

DO 213 I=1,K2

WRITE(MPOS,1200) comm
CONTINUE

CNE=1.0
ZERO=0.0
K=No-1
L=NO-1 ‘
ORIGIN CONSTRAINTS
DO 20 I=1,K
DO 730 II=1,NO
WRI'TE(2,1200) ONE
CONTINUE
DO 740 III=1,NC
DO 750 IIII=1,NO
WRITE(2,1200) ZERO
(DNI'INUE
(IJN'I‘INUE
CDNTINUE

DO 60 I=1,ND
WRITE(2,1200)ONE

.(ITHHNUE

C DESI‘.CDNST.

. C

”8

101
100

DC 90 J=1,L

DO 70 I=1,No
WRITE(2,1200) ONE
DO 80 II=1,L
WRITE(2.1200) ZERO
CDNTINUE

CONTINUE
‘WRITE(2,1200) ZERO
CONTINUE

DO 100 I=1,K
WRITE(2,1200) ONE
DO 101 II=1,L
WRITE(2,1200) ZERO
CONTINUE

CONTINUE

meIfi 1fim\ A‘m

301: C

302: 1200 FOM‘I‘(1X,F15.4)

303: II) 770 I=1.N

304: WRITE(MPOS,1200) BOTLMU)
305: 770 (IJNTINUE

306: END

301: C

302: 1200 FORMAT(1X,F15.4)

303: II) 770 I=1,N

304: WRITE(MPOS,1200) BO'I'LMU)
305: 770 (IDNTINUE

306: END

MPOS Batch Job Example

The following is a sample of the batch Job run on the MSU
CYBER 750 from an interactive terminal. It attaches the
permanent file KMKIIBZ (which is the MPOS input file) and
runs MPOS from the HAL unsupported library. It then catalogs
the output (which is listed beginning on the following page

of this appendix) and catalogs the dayfile to trace crashes.

KYLE,CM370000,JCZSOO,PN554872,RGB.
ATTACH,INP,KMKII02.
HAL,L'UNSUP,HPOS,I=INP,0=OUTPUT.
‘CATALOG,OUTPUT,KMKIIOZOUTPUT,RP=999.
EXIT,C,S.

DAYFILE.
CATALOG,DAYF,KMKIIOZDAYFILE,RP=999.

NOWHWN—a
II II II II II II II

T NPOS VERSION 4.0 NORTHIESIERN UNIVERSITY
§§§iifii§§§{§{{§§i§§ii§§{{i§§{§§§§§§§
o
N P O S i
O
I

VERSION 4.0

*****

i
I HOLTI-PURPOSE OPTIMIZATION SYSTEH !
E O
iliiiiiiii§i{fliiiiiiii§l{{§§ii§§i§ii

(Ihe following section is the N905 probln statnent.)

an: PROBLEM NUMBER I am

TITLE

I182.1982.9? ORIOINS,693 ELE. N.E. PLANT AT ROSE CITY.

REGULAR

VARINBLES

XI TO X693

NININIIE

f OBJECTIVE FUNCTION:
O.17Xl+13.21I2+16.70X3+21.12I4+6.12X5+9.70I6+l.20l7+
I9.9618+!.6619+7.0SI10+17.36I11+12.9II12+5.38I13+13.86XIQ+
II.25II§+8.86116+14.beI7+20.26XIO+4.SOXIO+7.92!20+6.15X21+
7.70122+12.63123+l7.28!24*22.06I25+4.05X26+IO.25X27+2.74I28*
12.43129+9.33I30+12.40131+17.14X32+7.82X33+5.36X34+5.53X35+
23.64136+8.I4X37+l.31138+14.83139+17.94X40+5.85!41+16.58142+
15.68X43+4.7BX44+10.45145+17.I9X46+8.95X47+4.55I40+9.651490
4.56X50+19.59X51+22.62X52+26.75153+9.bOI54+15.00X55+5.22X56+
4.28X57+18.45!58+21.77!59+25.96!60+O.49!61+I§.87I62*4.17X63+
20.52!64+0.81X65+O.51X65+IO.48167tl3.23IOO*6.52!69+14.65X70+
b.94X71+14.25172+20.77x73+26.15x74+l.60X75+l3.9bX76+6.98I77+
2.95X70+17.69179+21.95X80+26.36181+6.00XOZIIQ.93!OS+4.36!84*
7.70X85+12.79186+17.I9IOT+21.87IOO+4.52X89+10.14X90+2.28X9I+
9.41X92+13.43193+20.91X?4*26.07I95+2.74196+I4.45X97+9.29I98+
I4.90X99+6.9bIIOO+I0.08X101+15.51X102+9.33X103i3.15X104+O.16X105+
ll.90X106+13.79I107+22.66XIOO+29.90X109+8.27XIIO+I7.27XIII+14.0IXIIZ+
6.06I113+17.19I114+20.00X115+24.IQIIIO+O.4OXIII+13.I9XIIO+2.79XII9*
II.IIX120+15.331121+23.43X122+29.741123+5.84X124+17.24X125*12.291126+
9.58X127+II.32X128+IO.25X129+24.061130+l.48X131+I1.68I13297.32X133+
20.13II34+3.39X135+6.36113b+15.74I137+I3.53I138+4.07XIS9+13.651140+
14.55X141+5.55I142+12.31X143+I9.011144+7.33X145+6.38X146+9.18X147+
5.16!148+16.IOXI40+I9.58X150+24.20I151+6.451152+12.86I153+2.161154+
12.861155+8.66I156+12.02X157+lb.961158+7.03X159*4.99X160+6.09!lbl+
21.52X162+I.73I163+7.93IIO4+IO.59I165+14.27X166+6.09XIO7*IS.SSXIOO+
12.74X169+12.15XI70+12.09XI7I+Ib.94X172*10.28Xl73+6.£61174+5.47X175+
IS.54II76+4.57XI77+II.93X178+19.13XI79+8.15X180+6.49XIOI+IO.24X182+
14.14X183ib.IOIIOQ+II.78X185+IT.941186+7.I9XIO7+5.371188+O.25XIO9+
5.36XI90+IO.33XI9I+24.841192+30.02XI93+5.73XI94+17.951195+9.44X196+
19.21XI97+5.10I190+6.061199+13.89X200+I3.19X201+2.161202+12.4I!203+
9.031204+11.59X205+15.37x206+20.621207+Q.9OXZOO+O.831209+2.97X210*
3.36XZII+17.28X212+22.86X2!3+27.69IZI4+4.97X215+15.85X216+6.451217+
7.40X218+I‘.78X219+17.00I220+22.05X221+7.15X222+10.89I223+0.SOIZZII

02/24/82

.19.07.38.

FREE

1

17.351225+2.801226+10.96X227+I9.081228+9.87x229+b.53l230§11.95X231+
5.001232+15.121233+20.40X234i25.23x235+3.49X236+I3.3GX237+4.40I238+
23.I9X239+7.63X240+I.GSXZIltlI.63X242+I7.421243+5.421244+16.17x245+
16.10X246+5.I9!247+9.41X24G+15.84I249+9.GIXZSO+3.20X251*9.67X252+
6.48X253+IG.38I254+25.23X255+30.58!256+6.01X257+IO.42X258+10.361259+
22.IIX260+6.67X261+3.031262*14.I31253+16.24X264+4.40X265+15.151266+
I4.15X267+6.50X268+I4.32X269+21.28I270+6.30I271+8.b4!272+9.83x273+
I4.97X274+13.54I275+12.24I276+16.29I277+13.10I278fb.99X279t7.901280+
6.86!2GI+I4.0IIZGZ+I7.981283+22.§6I284+5.42X285+10.95I286+I.I9X287+
G.3#XZGG+II.77XZG9+I7.19X290+22.30129I+2.821292+10.24I293t4.25X294+
5.55X295+I4.55X296+I9.74X297+24.57I29G+3.33X299+12.72X300+3.9SI301+
2.67X302+19.04X303+22.92X304+27.21I305*G.15X306+IS.95I307+$.21X308+
23.44X309+O.17I310+1.441311’14.571312+I7.79X313+5.651314+16.36I315*
2.79X316+IG.30X317+23.G?X318+ZG.691319+5.GOI320+IO.G7ISZI+7.28X322+
4.12I323+Ib.33l324+21.94X325+26.GZX326+4.00X327+14.93X328+5.851329+
5.59X330+15.24X331+21.331332+26.461333+2.05X334+I4.403335+6.36X336+
23.‘0X337+b.9|!33G+2.391339+15.41I340*17.34X341+5.76X342+16.491343+
6.67X344+I4.201345+18.I7!346+22.651347+5.391348+II.IQX349+I.26X350+
IG.25!351+9.201352+7.011353+II.05X354+13.87X355+2.39X356+10.99X357+
16.131358+4.051359+II.86!360+I9.391361+G.62I362*6.771363+IO.91X364+
4.441365+16.23X366+25.451367i30.47!368+6.561369+IG.50X370+9.43137I+
I3.01I372+7.15I373+12.9GX374+IO.94X375+6.271376+6.44X377+7.45X378+
7.9GX379+14.G7X380+22.01!3Gl+27.bGI382+2.97I383+15.36X384+G.95X385+
16.84X386+3.49X387+II.OOX388+19.741389+9.221390+7.15X391+II.71X392+
I4.53X393+I3.221394+22.04X395+29.231396+7.70!397+Ib.601398+14.22I399+
12.01I400+12.53I401+13.61X402+I7.671403+9.821404+7.09I405+4.75X406i
T NPOS VERSION 4.0 NORTHNEGTERN UNIVERSITY

”cannon": “if"

I PROOLEN NUNOER I i
i:i§§§£i§:§§§iii§§§il

USING REGULAR
1182.1982.99 ORIGINS,693 ELE. N.E. PLANT AT ROSE CITY.

II.GOIQOT+G.32XQOG*IQ.69IIO9+20.61X410+I.OIIGI1+8.15X412+7.06X413+.
9.IOI4II+12.36XOIS+19.I9I416+25.3214I7+1.4II418*12.94X419+7.97XIZO+
7.631421+14.76X422+17.65X423+21.871424+7.37I425+10.761426+0.77X4270
8.74X4ZO+II.881429+16.151430+20.GOI431+4.901432+9.111433+2.74X434+
12.30I435+14.0GIISG+I4.35X437+IG.3IIISG+II.151439+G.25!440+5.37144l*
6.84I442+17.801443+20.101444+24.16I445+9.601446+13.£21447+3.SGI44G+
IQ.63X449+5.48X!50+IZ.38X451+19.151452t7.35I453+6.5II454+9.32I455+
6.66I456+16.611457+23.47X458+28.G9I459+4.24I460+16.69XIbI+9.15X462+
IO.9GX463*7.35X464+5.GSI465+II.bTIIbb+I3.TGX467+I.3IX468+II.871469+
21.34X470+6.ITIQ7I+3.69X472+13.57X473+15.541474+3.62I475+14.37I476+
0.661477+20.57X470+25.48X479+29.96X400+8.55X4GI+IG.44I482+7.97X483+
6.001484+Il.44IQOS+20.SSX486+25.45I487+2.13X488+13.411489+5.55I490+
II.02XQ9I+I3.b9!492+12.9OX493+16.9GIQ94+12.38I495+7.36!496+6.99x497+
21.98149G+2.50I499+7.16X500+IO.20X501+14.85X502+6.721503+15.85X504+
2.351505+IG.531506+24.071507+28.821508+6.06X509+17.041510+7.SSXSII+
4.52I512+Ib.931513+20.641514+24.951515+6.971516+I3.67X517+2.92XSIG+
5.60X319+16.741520+19.931521+24.15X522+7.651523+13.03I524+2.35X525+
5.29!526+17.28X527+20.441528+24.641529+G.00X530+13.56I531+2.901532+
22.261533+7.09X534+2.73X535+13.95X536+16.50X537+4.49X538+15.25X539+
5.99!540+25.65X54I+29.32X542+33.47XSQ3+14.081544+22.44X545+II.70X346+

02/24l82

.19.07.38.

PAGE

2

4.061547+IV.GOXS4Gt23.02X549+27.17x550+9.591551+16.161552+5.501553+
15.891554t4.251555+ll.05X556+lO.21X557+G.75!558+5.571559t10.231560+
5.80I561+I4.33X562+19.78X563+24.70I564+2.791565+l2.78X566+4.4IX567+
10.65X568+12.55X569+14.59X570+18.77!57I+G.441572+7.79X573+3.35X574+
9.96I575+12.771576+15.18x577+19.4II57G+7.861579+8.31!580+2.66X581*

I1.24X582+9.04X583+IS.57X584+21.7l1585+3.671586+9.25X587+7.19X588+

IO.4IX589+8.261590+6.56X59l+lI.I6X592+l3.621593+1.55!594+ll.21l595+
15.761596+5.191597+13.63X598+21.241599+7.G71600+G.60X601+II.31!602+
2.791503+19.391604+23.IbI605+27.42X606+G.571607#16.2!!608+5.451609+
9.30!610+12.01X611+IV.OGX612+24.90I613+I.31X614+12.52X615+7.74Iblb+

T.52Ibl7+lG.04I618+25.191619+30.72X620+6.00I62I+IG.4OX622+10.961623+
G.47X624+Ib.53!625+23.94X626+29.70X627+5.001628+17.36I629*10.731630+

4.52X631+16.93!632+20.64X633+24.95X634+6.971635+13.67X636+2.921637+
13.07X638+7.19X639+12.71X640+IG.59X64Itb.52X642+6.091643t7.301644+

5.4GI645+15.95!646+22.231647+27.4IIb4G+3.28Xb49+15.33I650+7.23Xb§l+
20.36X652t3.281653+6.31X654+I§.94Ib§5013.72X656+4.33I657+I$.90X658+
II.31X659+10.09X660+13.531661+IO.23I662+6.93I663+6.49X664+4.491665+
Ib.21!666+5.421567t9.I4X668+IS.43!669+10.09I670+2.79!67I+9.66I672+

20.IIX673+7.52167404.731675+ll.73!676+14.GSXOTT+2.39I678+12.991679+

I2.49I680+l4.28X681+I4.38!682+IO.36I683+Il.44IGG4+G.371685+5.62X686+

10.36X687+13.51X688+15.24X689*19.34X690+G.95X691+G.571692t3.19X693
S

(The following are the constraint equations.)

T

I.
2.
3.
4.
5.
b.
7.
G.
9.
10.
II.
12.
13.
I4.
15.
lb.
17.
IO.
19.
20.
21.
22.
23.
24.
25.

S

S

CONSTRAINTS

i

II+IZ+13+I4+X5+I6+173Il075
XG+IV+X10+XII+XIZtXI3+I14314517
IIStIlb+I17+IIO+XIV+X20+X21316481
I22tl23+124+l25+l26+127+X28318132
129+I30+131+I32+133+I34+135=I440!
X36+X37+ISG+I39+I40+X4I+I42=9279
I43+I44+I45+I46+I47+I4G+X49=21014
I50+ISI+152+I53+X54+I55+156=54I4
157+I58+I§9+160+I61+X62+X63=3136
I64tI65+166+167+IGG+I69+X70=I574!
X71+172+I73+X74+175+X76+I77810983
ITO+X79+XGO+XOI+IGZ+I83+XG4=21883
I85+XOG+IGT+IGG+XGV+190+IVI=10947
192+193+194+I95+196+X97+X90818408
X99+I1009X101+I102+1103+I104+1105'6099
I106+1107+XIOG+1109+IIIO+XIII+XIIZ'20921
XII3TXII4+II15+XI16+X117+IIIG+IIl9il4OVI
X120+IIZI+1122+I123+XI24*1125+1126=2606
1127+XIZG+II29+X130+XI3I+I132+II3387128
IIS4+XIS§+X136+I137+XISO+I139+II40=IV475
II41+Il42+II43tII44+Xl45+XI46+II47'7436
II4G+I149+!150+!151+1152t1153t1154=21307
1155+I156+I157+X158+1159+1160+I161310136
X162+I163+!164+I165+I166+X167+I168=20468
XleOXIIOtI171+X172+X173+X174+Il75815372

NPOS VERSION 4.0 NORTNNESTERN UNIVERSITY

§§{!§§{§§§§§§§{§§§§§G

02124/82

.19.07.38. PAGE

3

e PROOLEN NUHOER l e
eeeeeeeeemmeeeeee

USING REGULAR
“82.198239 ORIGINS,693 ELE. N.E. PLANT AT ROSE CITY.

26. 1176+!177+II78tII79+XIGO+XIGI+IIGZ=I4621
27. XIO3+XIG4+X185+!l86+XIG7+IIOO+XlG9=l3OI
28. Il90+Il9ItXI92+XI93+XI94+II95+XI96315921
29. 1197+!198+!199+1200+X201+I202+I203=19729
30. I204+X205+1206+I207+I208+X209+X2I0=3403
31. X21I+X212+1213+X214+1215+X216+1217=5409
32. X218+I219+X220+IZZI+1222+1223+1224=4707
33. 1225+1226+1227+1228+X22941230+1231'13013
34. X232+X233+X234+X233+XZ36+I237+X23G=I1849
35. 1239+1240+X24l+1242+1243+1244+1245=9502
36. X246+1247+I24G+I249+X250+I25I+1252=20677
37. 1253+1254+I255+X256+1257+1258+I259=16326
3G. 1260+X26l+X262+I263+IZ64+I265+X266=18987
39. X267+X268+X269+X270+I27I+X272+1273=16185
40. X274+I275+X276+I277+I278+I279+X280=435
41. X28ItX282+X283+XZG4+X285+1286+1287822248
42. X2GG+I289+X290+129I+X292+X293+I294=9797
43. 1295+X296+X297+X298+1299+X300+1301=13352
44. X302+X303+X304+X305+X306+X307+X300=GZO3
45. X309+I310+I31I+I312+X3I3+X3I4+I315=16440
46. 1316+I3I7+X3IG+I3I9+X320+I321+X322=21427
47. X323+X324+1325+I326+X327+X328+I329=4731
4G. I330+I33I+I332+I333+I334+I335+I336=II749
49. I337+I338+X339+X340+l34l+l342+l343=21102
50. X344+I345+X346+X347+X348+X349+I350=16008
31. X351+X352+I353+I354+1355+I356+I357=8807
52. I3SG+I339+X360+I36I+1362+X363+X364=16443
53. I365+I366+I367+X36O+X369+I370+I371318848
54. x372+x373+xs74+x375+x376+x377+x37a=2221
55. X379+X3GO+I3GI+X382tX383+I3G4+I385=723G
56. I3G6+I3G7+X3GG+I389+I390+I391+X392=GGI6
57. X393+X394+I395+X396+X397*139G+I399=959G
- 38. X400+I40l+X4OZ+I403+X404+I405+X406=13503
59. I407+X408+I409+X410+X4lI+X412+I4I3=932
60. X4I4+X415+X4I6+I417+X4IG+I4I9+I420=IZG70
61. X421+X422+I423+I424+X425+X426+I427=I7890
62. I428+X429+X430+I43I+I432tl433+X434=5760
63. 1435+!436+X437+I438+X439+I440+I441=15012
64. I442+I443+X444+I445+I446+X447+X44G=21029
65. X449tX450+X45I+I452+X453+X454+X455=15967
66. X456+I457+X458+I459+I460+X46I+I46231307
67. I463+X464+X465+X466+X467+I468+X469=10946
68. I470+I47I+X472+X473+I474+X475+X476=16995
69. I477+X4784X479+X480+X4GI*X402+X4G3=2126G
70. X4G4tX485+X486+X487+X4Gth489+X490=14334
71. I491+1492+I493tI494+X495+X496+I497=19193
72. I498+I499+X500+I§01+X502+X503+X504=3663
73. 1505+I506+X507+X§08+X§09+XS10+ISII=I909I
74. X512+X513+X5I4+I515+I516+X517+XSIG=12139

 

 

75L
76a
77}
7G.
79.
80H
81.
82.
83.
84.
83.
86.
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
T

15l9+X520+X521+1522+1523+1524+X525=l8590
X526+X527+1528+1529+X530+!531+l532=3541
1533+X534+1535+X53b+1537+X538+X539=9201
1540+XS41+1542+XS43+X544+X545+X546=2187
X547+XS40+X549+1550+155l+X552+X553=2757
1554+1555+1556+X557+X558+1559+1560=8369
156l+X562+l563+X564+1565+X566+x567=l1971
X568+X569+1570+157l+X§72+1573+XS74=17547
X575+1576+X577+x578+1579+1580+158l=20038
1582+XSBS+1584+X585+1586+x587+l588=14221
1589+XS90+X591+1592+1593+X594+1595=10957
X596+X597+X598+x599+l600+l601+1602=10674
1603*X604+X605+X606+1607+X608+1609=18364
XblO+Xbl1+X612+X613+X6l4+l615+lblb=la469
1617+1618+X619+lbzo+1621+X622+1623=10060
1624+X625+X626+X627+1628+1629+1630=12269
1631+X632+Xb33+l634+X635+X636+Xb37=16819
Xb38+l639+lb40+164l+X642+Xb43+lb44812055
lb45+l646+lb47+1648+X649+1650+X65l=1720
l652+lb§3+l654+1655+X656+X657+1658I5080
1659+X660+X66l+l662+1663+1664+X665=2197
X666+X667+lbbB+lbb9+XbTO+XbTl+l672=13256

NPOS VERSION 4.0 NORTHNESTERN UNIVERSITY

“WNW

If PROOLEN NUNOER l e
museueeumeen

INSING REGULAR
IIGZ.1902.99 ORIGINS,693 ELE. N.E. PLANT AT ROSE CITY.

97.
9G.
99.

100.

101.

X673+I674+1675+I676+I677+I678+I679=G93
X680+X6GI+1682+X6G3+I684+I685+168686326
I687+I6GGtI6G9+I690+I69I+I692+X693836G

{

e DESTINATION I
II+XG+I15+IZ2+I29+I36+I43+I50+I§7+X64+I7I+I7G+I85+I92+X99+I106+!I13+
X120+1127+XI34+XI4I+II4G+IISS+Xl62+II69+XI76+IIO3+XI90+II97+1204+I211+
IZIG+IZ25+X232+1239+I246+X253+1260+1267+X274t1281+I288+X295+I302+X309+
l3I6+I323+X330+I337+I344+133l+X358+I365+X372+I379+X386+I393+X400+X407+
I4I4+X421+X428+I435+I442+X449+X456+X463+X470+I477+I484+I491+1498+I$05+
ISI2+I5I9+X526+IS33+I540+1547+1554+1561+X568+I575+1582+1589+1596+I603+
1610+X617+!624tI63I+X638+1645+I652+1659+X666+X673+1680+X687'

156950

5

e DESTINATION 2
IZ+X9+I16+I23+I30+I37+I44+ISI+ISG+I65+I72+I79+IG6+I93+I100+!IO7+II14+
IIZIOIIZOtX135+!142+!I49+X156+I163+II70+1177+XIG4OXI9I+XI9G+1205+X212+
12I9+X226+1233+1240+X247+I254+1261+126Gt1275+x282+X289+1296+X303+I310+
I3I7+I324tx33I+I33G+I345+1352+X359tX366+X373+X380+X387+X394+X40I+I408+
I415+X422+I429+X436+X443+X450+X457+X464+X47l+X478+I4G§+I492+X499+1506+
1513+X520+XSZ7+I534+154l4I540+1355+1562+1569+X576+X583+IS90+X597+1604+
I61I+I61G+I625+X632+Ib39+l646+1653*X660+I667+Xb74+l681+I688=

146000
e

02/24/82

. .l9.07.38.

PAGE

4

i DESTINATION 3

POZ. 13+!10+117+YZ4+X31+I38+X45+152+159+166+X73+180+187+X94+I101*!108+II15+
1122+1129+1136+XI43+I150+!157+1164+1171+1178+1185+I192+1199+1206+1213+
I220+1227+I234+1241+I248+1255+X262+X269+1276+X283+1290+1297+I304+I311+
X318+X325+X332+I339+I346+X353+1360+1367+X374+I381+X388+1395+X402+X409+
I416+X423+I430+X437+X444+I451+1458+X465+1472+1479+X486+I493+1500+1507+
I514+I521+1528+1535+1542+XS49+1556+1563+1570+1577+1584+1591+1598+X605+
1612+1619+X626+1633+1640+Xb47+1654+1661*!668+Xb75+1682+16898
87600
e
e DESTINATION 4

103. I4t111+X18+125+X32tX39+X46+153+160+167+I74+181+188+X95+I102+1109+X116+
X123+1130+XI37+I144+I151+I158+XI65+1172*I179+1186+X193+1200+1207+1214+
1221+X228+X235+1242+1249+X256+1263+1270+I277+X284+1291+1298+1305+X312+
I319+I326+X333tI340tl347+1354+l361+1368+I375+I382+X389+1396+!403+X410+
I417+X424+X431+1438+I445+X452+I459+X466+X473+X480+X487+X494+I501+1508+
I515+1522+X529+1536tX543+1550t1557+1564+1571+I578+1585+1592+1599+X606+
1613+1620+I627+Xb34+1641+1648+16550166291669+1676+1683+1690=
82125
A -

i DESTINATION 5

104. 15+!12+!19+126+133+I40+X47+I54+161+168+X75+X82+189+X96+I103+I110+I117+
1124+II31+1138+1145+I152+XI59+I166+II73+X180+!187+1194+1201+1208+1215+
1222*1229+1236+XZ43+X250+1257t1264+1271+1278+1285+1292+1299+1306+X313+
I320tX327+I334+I341+1348+1355+X362+X369+X376+I383+I390+1397+X404+I411+
I418+X425+I432+I439+1446+X453+I460+I467+X474+X481+1488+I495+1502+X509+
X516+!523tx530+1537+1544+I551+155841565+1572+1579+!586+1593+1600+1607+
I614+1621+1628+X635+I642+X649+1656+1663+1670+1677+X684+Ib91=
365000
f
e DESTINATION 6

105. 16+!13+I20+127+I34+X41+X48+X55+X62+X69+176+183+I90+X97+I104+!111+1118+
1125+!132+XI39+I146+X153+I160+X167tI174+1181+I188+1195+I202*1209+1216+
122341230+1237+1244+1251+1258+1265+X272+!279+1286+1293+I300+I307+1314+
I321t1328+X335+I342+1349+I356+X363+I370+I377+I384+X391+1398+X405+I412+
I419+X426+1433tx440+X447+X454+X461+X468+X475+X482+X489+I496+1503+I510+
1517+X524+X531*1538+I545+1552+X559t1566tI573+1580+1587+1594+X601+1608+
I615+1622+1629+1636+X643+X650+I657+1664+1671+1678+I685+I692=

73000
e

5
. * DESTINATION 7

106. 17+!14+XZ1+128+135+X42t149+156+163+I70+IG4+X91+X98+I105+!112+!119+
1126+II33+X140+1147+!154+II61+!168+II75+1182+I189+1196+1203+X210+
1217+1224+1231+1238+X245+1252+I259+1266+1273+X280+1287+1294+1301+
I308+X315+X322+I329t1336+I343+X350+X357+X364+I371+1378+I385+1392+
I399+I406+X413+X420+I427+I434+I441+1448+X455+X462+1469+X476+I483t

T NPOS VERSION 4.0 NORTNNESTERN UNIVERSITY

{SIARSSQSNESRRSSERSRC

e PROOLEH NUNOER l !
emeueeeeueseeeeee

USING REGULAR
1182.1982.99 ORIGINS,693 ELE. N.E. PLANT AT ROSE CITY.

02l24/82

.19.07.38.

PAGE

5

1490+14W+1504+1511+1518+1525+1532+1539+1546+1553+1560+1567+1574+
1581+1588+1595+1602+1609+1616+1623+1630+1637+1644+1651+1658+1665+

 

1672+1679+1686+1693=
285000
e
e
e
SOUNDS
11 TO 1693.8E.0
11111011 500000
OPTINIZE
T NPOS VERSION 4.0 NORTHNESTERN UNIVERSITY 02/24182
eumeuneeumen
e PROBLEN NUTIBER 1 e
sueeeeeeuueneme
(Fro. here to the end of this listing is an exuple of the NPOS
output.1 -
USING REGULAR
1182.1982.99 ORIGINS,693 ELE. N.E. PLANT AT ROSE CITY.
PROBLEN INPUT SUNNNIY
CONSTRAINTS VARIABLES Nm-ZEROS PARMETERS SOUNDS
E858 106 INT3 0 NUNBER= 1385 TOL= .100E-08
LES! 0 TOT AL: 693 PERCENTI 1.89 EPSsnET-‘AULT
SSS: 0 NOIIB8 0 LINIT=OEFAULT
TOTAL8 106 NOLBI 693 RSCALE= .100801
VARIABLE TABLE
1-11 2-12 3-13 4-14 5-15 6-16
7-17 8-18 9-19 10-110 11-111 12-112
13 - 113 14 - 114 15 - 115 16 - 116 17 - 117 18 - 118
19 - 119 20 - 120 21 - 121 22 - 122 23 - 123 24 - 124
25 - 125 26 - 126 27 - 127 28 - 128 29 - 129 30 - 130
31 - 131 32 - 132 33 - 133 34 - 134 35 - 135 36 - 136
37 - 137 38 - 138 39 - 139 40 - 140 41 - 141 42 - 142
43 - 143 44 - 144 45 - 145 46 - 146 47 - 147 48 - 148
49 - 149 50 - 150 51 - 151 52 - 152 53 - 153 54 - 154
55 - 155 56 - 156 57 - 157 58 - 158 59 - 159 60 - 160
61-161 62-162 63-163 64-164 65-165 66-166
67 - 167 68 - 168 69 - 169 70 - 170 71 - 171 72 - 172
73 - 173 74 - 174 75 - 175 76 - 176 77 - 177 78 - 178
79 - 179 80 - 180 81 - 181 82 - 182 83 - 183 84 - 184
85-185 86-186 87-187 88‘188 89-189 90-190
91 - 191 92 - 192 93 - 193 94 - 194 95 - 195 96 - 196
97 - 197 98 - 198 99 - 199 100 - 1100 101 - 1101 102 - 1102
103 - 1103 104 - 1104 105 - 1105 106 - 1106 107 - 1107 108 - 1108

. 19.07.38. PAGE

6

355 - 1355

109 - 1109 110 - 1110
115 - 1115 116 - 1116
121 - 1121 122 - 1122
127 - 1127 128 - 1128
133 - 1133 134 - 1134
139 - 1139 140 - 1140
145 - 1145 146 - 1146
151 - 1151 152 - 1152
157 - 1157 158 - 1158
163 - 1163 164 - 1164
169 - 1169 170 - 1170
175 - 1175 176 - 1176
181 - 1181 182 - 1182
187 - 1187 188 - 1188
193 - 1193 194 - 1194
199 - 1199 200 - 1200
205 - 1205 206 - 1206
211 s 1211 212 - 1212
217 r 1217 218 ~ 1218
223 - 1223 224 - 1224
229 - 1229 230 - 1230
235 - 1235 236 - 1236
241 - 1241 242 - 1242
247 - 1247 248 - 1248
253 - 1253 254 - 1254
259 s 1259 260 s 1260
265 - 1265 266 - 1266
271 - 1271 272 - 1272
277 - 1277 278 - 1278
283 - 1283 284 - 1284
289 - 1289 290 - 1290
295 - 1295 296 - 1296
301 - 1301 302 - 1302
307 - 1307 308 - 1308
313 - 1313 314 - 1314
319 - 1319 320 - 1320
325 - 1325 326 - 1326
331 - 1331 332 - 1332
337 - 1337 338 - 1338
343 - 1343 344 - 1344
349 s 1349 350 - 1350
.NPOS VERSION 4.0
MAWEM‘IM‘H“
! PROOLEH NUMBER 1 I
555555555555555555555
USING REGULAR

356 - 1356

111 - 1111
117 - 1117
123 - 1123
129 - 1129
135 - 1135
141 - 1141
147 - 1147
153 - 1153
159 - 1159
165 - 1165
171 - 1171
177 - 1177
183 - 1183
189 - 1189
195 - 1195
201 - 1201
207 - 1207
213 - 1213
219 - 1219
225 - 1225
231 - 1231
237 - 1237
243 - 1243
249 - 1249
255 - 1255
261 - 1261
267 ' 1267
273 - 1273
279 - 1279
285 - 1285
291 - 1291
297 - 1297
303 5 1303
309 - 1309
315 - 1315
321 - 1321
327 - 1327
333 - 1333
339 - 1339
345 - 1345
351 - 1351

NORTNNESTERN UNIVERSITY

1182.1982.99 ORIGINS,693 ELE. N.E. PLANT AT ROSE CITY.

112 - 1112
118 - 1118
124 - 1124
130 - 1130
136 - 1136
142 - 1142
148 - 1148
154 ' 1154
160 - 1160
166 s 1166
172 - 1172
178 - 1178
184 - 1184
190 - 1190
196 - 1196
202 - 1202
208 - 1208
214 - 1214
220 - 1220
226 - 1226
232 - 1232
238 ' 1238
244 - 1244
250 - 1250
256 - 1256
262 - 1262
268 ' 1268
274 - 1274
280 s 1280
286 - 1286
292 s 1292
298 - 1298
304 - 1304
310 - 1310
316 - 1316
322 - 1322
328 - 1328
334 - 1334
340 - 1340
346 - 1346
352 - 1352

VARIABLE TABLE

357 - 1357

358 - 1358

113 - 1113
119 - 1119
125 ' 1125
131 - 1131
137 - 1137
143 - 1143
149 - 1149
155 - 1155
161 ' 1161
167 - 1167
173 - 1173

I 179 - 1179

185 - 1185
191 - 1191
197 - 1197
203 ' 1203
209 - 1209
215 - 1215
221 - 1221
227 - 1227
233 ° 1233
239 - 1239
245 - 1245
251 - 1251
257 - 1257
263 - 1263
269 - 1269
275 - 1275
281 - 1281
287 - 1287
293 - 1293
299 - 1299
305 - 1305
311 - 1311
317 - 1317
323 - 1323
329 - 1329
335 - 1335
341 - 1341
347 - 1347
353 - 1353

359 - 1359

114 - 1114
1N-1uo
123 - 1126
132 - 1132
138 - 113B
144 - 1144
150 - 1150
151 - 1153
132 - 1102
1m-1u8
174 - 1174
1m-1wo
1u-1w6
192 - 1192
1m-1w8
204 - 1204
210 - 1210
Ab-nm
222 - 1222
228 - 1228
234 - 1234
240 - 1240
243 - 1243
252 - 1252
258 - 1258
234 - 1204
270 - 1270
273 - 1273
282 - 1282
288 - 1288
294 - 1294
300 - 1300
306 - 1306
nz-uu
ma-uw
324 - 1324
330 - 1330
330 - 1336
342 - 1342
34B - 1348
354 - 1354

02I24l82

360 - 1360

.19.07.38.

PAGE

7

361 5 1361
367 5 1367
373 5 1373
379 5 1379
385 5 1385
391 5 1391
397 5 1397
403 5 1403
409 5 1409
415 5 1415
421 5 1421
427 5 1427
433 5 1433
439 5 1439
445 5 1445
451 5 1451
457 5 1457
463 5 1463
469 5 1469
475 5 1475
481 5 1481
487 5 1487
493 5 1493
499 5 1499
505 5 1505
511 5 1511
517 5 1517
523 5 1523
529 5 1529
535 5 1535
541 5 1541
547 5 1547
553 5 1553
559 5 1559
565 5 1565
571 5 1571
577 5 1577
583 5 1583
589 5 1589
595 5 1595
601 5 1601
607 5 1607
613 5 1613
619 5 1619
625 5 1625
631 5 1631
637 5 1637
643 5 1643
649 5 1649
655 5 1655
661 5 1661
667 5 1667
673 5 1673
679 5 1679
685 5 1685

362 5 1362
368 5 1368
374 5 1374
380 5 1380
386 5 1386
392 5 1392
398 5 1398
404 5 1404
410 5 1410
416 5 1416
422 5 1422
428 5 1428
434 5 1434
440 5 1440
446 5 1446
452 5 1452
458 5 1458
464 5 1464
470 5 1470
476 5 1476
482 5 1482
488 5 1488
494 5 1494
500 5 1500
506 5 1506
512 5 1512
518 5 1518
524 5 1524
530 5 1530
536 5 1536
542 5 1542
548 5 1548
554 5 1554
560 5 1560
566 5 1566
572 5 1572
578 5 1578
584 5 1584
590 5 1590
596 5 1596
602 5 1602
608 5 1608
614 5 1614
620 5 1620
626 5 1626
632 5 1632
638 5 1638
644 5 1644
650 5 1650
656 5 1656
662 5 1662
668 5 1668
674 5 1674
680 5 1680
686 5 1686

363 5 1363
369 5 1369
375 5 1375
381 5 1381
387 5 1387
393 5 1393
399 5 1399
405 5 1405
411 5 1411
417 5 1417
423 5 1423
429 5 1429
435 5 1435
441 5 1441
447 5 1447
453 5 1453
459 5 1459
465 5 1465
471 5 1471
477 5 1477
483 5 1483
489 5 1489
495 5 1495
501 5 1501
507 5 1507
513 5 1513
519 5 1519
525 5 1525
531 5 1531
537 5 1537
543 5 1543
549 5 1549
555 5 1555
561 5 1561
567 5 1567
573 5 1573
579 5 1579
585 5 1585
591 5 1591
597 5 1597
603 5 1603
609 5 1609
615 5 1615
621 5 1621
627 5 1627
633 5 1633
639 5 1639
645 5 1645
651 5 1651
657 5 1657
663 5 1663
669 5 1669
675 5 1675
681 5 1681
687 5 1687

364 5 1364
370 5 1370
376 5 1376
382 5 1382
388 5 1388
394 5 1394
400 5 1400
406 5 1406
412 5 1412
418 5 1418
424 5 1424
430 5 1430
436 5 1436
442 5 1442
448 5 1448
454 5 1454
460 5 1460
466 5 1466
472 5 1472
478 5 1478
484 5 1484
490 5 1490
496 5 1496
502 5 1502
508 5 1508
514 5 1514
520 5 1520
526 5 1526
532 5 1532
538 5 1538
544 5 1544
550 5 1550

556 5 1556‘

562 5 1562
568 5 1568
574 5 1574
580 5 1580
586 5 1586
592 5 1592
598 5 1598
604 5 1604
610 5 1610
616 5 1616
622 5 1622
628 5 1628
634 5 1634
640 5 1640
646 5 1646
652 5 1652
658 5 1658
664 5 1664
670 5 1670
676 5 1676
682 5 1682
688 5 1688

365 5 1365
371 5 1371
377 - 1377
383 5 1383
389 5 1389
395 5 1395
401 5 1401
407 5 1407
413 5 1413
419 5 1419
425 5 1425
431 - 1431
437 5 1437
443 5 1443
449 5 1449
455 5 1455
461 5 1461
467 5 1467
473 5 1473
479 5 1479
485 5 1485
491 5 1491
497 5 1497
503 5 1503
509 5 1509
515 5 1515
521 5 1521
527 5 1527
533 5 1533
539 5 1539
545 5 1545
551 5 1551
557 5 1557
563 5 1563
569 5 1569
575 5 1575
581 5 1581
587 5 1587
593 5 1593
599 5 1599
605 5 1605
611 5 1611
617 5 1617
623 5 1623
629 5 1629
635 5 1635
641 5 1641
647 5 1647
653 5 1653
659 5 1659
665 5 1665
671 5 1671
677 5 1677
683 5 1683
689 5 1689

366 5 1366
372 5 1372
378 5 1378
384 5 1384
390 5 1390
396 5 1396
402 5 1402
408 5 1408
414 5 1414
420 - 1420
426 5 1426
432 5 1432
438 5 1438
444 5 1444
450 5 1450
456 5 1456
462 5 1462
468 5 1468
474 5 1474
480 5 1480
486 5 1486
492 5 1492
498 5 1498
504 5 1504
510 5 1510
516 5 1516
522 5 1522
528 5 1528
534 5 1534
540 5 1540
546 5 1546
552 5 1552
558 5 1558
564 5 1564
570 5 1570
576 5 1576
582 5 1582
588 5 1588
594 5 1594
600 5 1600
606 5 1606
612 5 1612
618 5 1618
624 5 1624
630 5 1630
636 5 1636
642 5 1642
648 5 1648
654 5 1654
660 5 1660
666 5 1666
672 5 1672
678 5 1678
684 5 1684
690 5 1690

691 5 1691 692 5 1692 693 5 1693
INPUT TRANSLATION TINE =

T NPOS VERSION 4.0

1.8140 SECONDS

NORTHVESTERN UNIVERSITY 02l24/82 .19.07.38. PAGE 8

{iii‘INEN‘IVAHSHHEV

e PROBLEN NUNBER 1 e
eeeeeneeuuuuesu

USING REGULAR
1182.1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.

iITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITND.
ITNO.
ITNO.
ITNO.

1 IN VAR5
2 IN VAR5
3 IN VAR5
4 IN VAR5
5 IN VAR5
6 IN VAR5
7 IN VAR5
8 IN VAR5
9 IN VAR5
10 IN VAR5
11 IN VAR5
12 IN VAR5
13 IN VAR5
14 IN VAR5
15 IN VAR5
16 IN VAR5
17 IN VAR5
18 IN VAR5
19 IN VAR5
20 IN VAR5
21 IN VAR5
22 IN VAR5
23 IN VAR5
24 IN VAR5
25 IN VAR5
26 IN VAR5
27 IN VAR5
28 IN VAR5
29 IN VAR5
30 IN VAR5
31 IN VAR5
32 IN VAR5
33 IN VAR5
34 IN VAR5
35 IN VAR5
36 IN VAR5
37 IN VAR5
38 IN VAR5
39 IN VAR5
40 IN VAR5
41 IN VAR5
42 IN VAR5
43 IN VAR5

1
8
15
22
29
33
43
50
57
34
71
78
2
85
9
92
13
99
106
23
30
113
37
120
127
44
134
141
51
58
148
35
72
155
79
112
3
10
86
119
17
93
171

VAR5 800 TO LOVER ROUND
VAR5 801 TO LOVER ROUND
VAR5 802 TO LOVER ROUND
VAR5 803 TO LOVER BOUND
VAR5 804 TO LOVER ROUND
VAR5 805 TO LOVER ROUND
VAR5 806 TO LOVER ROUND
VAR5 807 TO LOVER OOUND
VAR5 808 TO LOVER ROUND
VAR5 809 TO LOVER ROUND
VAR5 810 TO LOVER ROUND
VAR5 899 TO LOVER ROUND
VAR5 811 TO LOVER ROUND
VAR5 1 TO LOVER ROUND
VAR5 812 TO LOVER ROUND
VAR5 8 TO LOVER ROUND
VAR5 813 TO LOVER ROUND
VAR5 814 TO LOVER BOUND
VAR5 15 TO LOVER ROUND
VAR5 22 TO LOVER ROUND
VAR5 815 TO LOVER BOUND
VAR5 29 TO LOVER ROUND
VAR5 816 TO LOVER BOUND
VAR5 817 TO LOVER BOUND
VAR5 36 TO LOVER ROUND
VAR5 818 TO LOVER ROUND
VAR5 819 TO LOVER BOUND
VAR5 43 TO LOVER ROUND
VAR5 50 TO LOVER ROUND
VAR5 820 TO LOVER ROUND
VAR5 57 TO LOVER ROUND
VAR5 64 TO LOVER ROUND
VAR5 821 TO LOVER ROUND
VAR5 71 TO LOVER BOUND
VAR5 822 TO LOVER ROUND
VAR5 900 TO LOVER ROUND
VAR5 2 TO LOVER BOUND
VAR5 78 TO LOVER BOUND
VAR5 823 TO LOVER ROUND
VAR5 9 TO LOVER BOUND
VAR5 85 TO LOVER ROUND
VAR5 824 TO LOVER BOUND
VAR5 16 IO LOVER ROUND

10

5VNIN=
5VNIN=
5VNIN8
5VNIN=
5VNIN8
5VNIN=
5VNIN8
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN8
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN8
5VNIN=
5VNIN8
5VNIN8
5VNIN=
5VNIN3
5VNIN=
5VNIN8
5NNIN8
5VNIN=
5VNIN=
5VNIN8
5VNIN=
5VNIN=
5VNIN=
5VNIN8
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=

-2319200.000000
-234o110.000000
-23o7204.000000
-227094o.000000
-2242138.000000
52223580.000000
~2181552.000000
-2170724.000000
-2134452.000000
52132966.000000
-2111000.000000
-2077450.000000
-2037234.000000
-2055300.000000
-204534o.000000
52026266.000000
-2008524.000000
51996326.000000
-1993304.000000
-1957040.000000

51954484.000000'

51928238.000000
51926302.000000
51921090.000000
51909680.000000
51906834.000000
51867884.000000
51867652.000000
51856824.000000
51853012.000000
51850552.000000
51819066.000000
51810398.000000
51797100.000000
51790126.000000
51785450.000000
51763300.000000
51753334.000000
51749190.000000
51734266.000000
51731440.000000
51718446.000000
51701304.000000

VAR5 92 TO LOVER BOUND
VAR5 825 TO LOVER ROUND
VAR5 826 TO LOVER BOUND
VAR5 99 TO LOVER BOUND
VAR5 23 TO LOVER BOUND
VAR5 827 TO LOVER ROUND
VAR5 106 TO LOVER BOUND
VAR5 30 TO LOVER BOUND
VAR5 37 TO LOVER ROUND
VAR5 828 TO LOVER ROUND
VAR5 113 TO LOVER ROUND
VAR5 901 TO LOVER ROUND
VAR5 829 TO LOVER BOUND
VAR5 120 TO LOVER BOUND
VAR5 830 TO LOVER ROUND
VAR5 127 TO LOVER ROUND
VAR5 831 TO LOVER BOUND
VAR5 3 TO LOVER BOUND
VAR5 44 TO LOVER ROUND
VAR5 51 TO LOVER BOUND
VAR5 832 TO LOVER ROUND
VAR5 10 TO LOVER BOUND
VAR5 58 TO LOVER ROUND
VAR5 134 TO LOVER ROUND
VAR5 141 TO LOVER BOUND
VAR5 833 TO LOVER ROUND
VAR5 65 TO LOVER ROUND
VAR5 17 TO LOVER BOUND

5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN8
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN8
"VIIIN'I
5VNIN=
5VNIN8
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN8
5VNIN=
5VNIN8
5VNIN=

NORTHVESTERN UNIVERSITY

1182. 1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.

ITNO. 44 144 444414- 24
17448. 45 144 44444- 100
17448. 44 144 444444- 183
17448. 47 144 474444— 190
17448. 48 144 VAR- 107
17448. 49 144 444444— 31
17448. 50 144 44444- 197
17448. 51 144 44444- 114
17448. 52 144 474444- 38
17448. 53 144 444444- 45
17448. 54 144 444444- 204
17448. 55 144 47444- 121
17448. 54 144 VAR- 4
17448. 57 144 44444- 211
17448. 58 144 4444- 128
17448. 59 144 44144- 218
17448. 40 144 VAR- 135
17448. 41 144 44444- 225
17448. 42 144 9444- 11
17448. 43144 471444- 52
17448. 44144 44444- 59
17448. 45 144 474444- 232
17448. 44 144 474444- 18
17418. 47 144 47044— 44
17448. 68 144 VAR- 142
17448. 49 144 VAR- 149
17448. 70 144 44444- 239
17448. 71 144 444444- 73

7 NPOSVERSION 4.0
*“ifH‘IENEEENENflH
e PROBLEN NUI'IBER 1 4
Ifiiiifliiflfiifliiiii
4151448 REGULAR
17448. 72144 44444- 25
17448. 73 144 444444- 244
77448. 74 144 444444- 80
17448. 75 144 444444- 154
17448. 74144 VAR- 32
17448. 77 144 94444- 253
17148. 78 144 VAR- 143
17448. 79144 9444- 87
17448. 80 144 474444- 39
17448. 81 144 474444- 5
17148. 82 144 94444- 240
17448. 831N 444444- 44
17448. 84 144 vm— 94
77448. 85 144 44444- 170
17448. 84144 4748- 12
17448. 87 144 m- 247
17448. 88 144 444444- 177
17448. 89 144 VAR- 101

VAR5 834 TO LOVER ROUND
VAR5 72 TO LOVER BOUND
VAR5 148 TO LOVER ROUND
VAR5 24 TO LOVER BOUND
VAR5 835 TO LOVER BOUND
VAR5 155 TO LOVER BOUND
VAR5 79 TO LOVER ROUND
VAR5 31 TO LOVER ROUND
VAR5 902 TO LOVER BOUND
VAR5 836 TO LOVER BOUND
VAR5 38 TO LOVER ROUND
VAR5 86 TO LOVER BOUND
VAR5 162 TO LOVER BOUND
VAR5 4 TO LOVER BOUND
VAR5 837 TO LOVER BOUND
VAR5 169 TO LOVER BOUND
VAR5 93 TO LOVER BOUND
VAR5 45 TO LOVER BOUND

11

5VNIN8
5VNIN8
5VNIN=
5VNIN3
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VNIN=
5VN1N=
5VNIN=
5VNIN=
5VNIN=

51694624.000000
51689204.000000
51686602.000000
51682426.000000
51665040.000000
51654760.000000
51640584.000000
51636238.000000
51617680.000000
51615302.000000
51612402.000000
51610250.000000
51608492.000000
51607190.000000
51597674.000000
51592934.000000
51588260.000000
51588100.000000
51575652.000000
51564824.000000
51562234.000000
51559066.000000
51558552.000000
51553984.000000
51539112.000000

"51538536.000000

51527066.000000
51526104.000000

51519532.000000
51505100.000000
51496498.000000
51489840.000000
51478178.000000
51476226.000000
51461334.000000
51461038.000000
51446000.000000
51445526.000000
51442480.000000
51439440.000000
51435290.000000
51423850.000000
51407552.000000
51404546.000000
51402624.000000
51400452.000000

02I24/82

.19.07.38.

PAGE

9

T

ITV0.
ITV8.
ITV0.
ITV0.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITVO.
ITV0.
ITV8.
ITV0.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITV8.
ITV0.
ITV0.
ITNO.
ITV0.
ITV0.
ITV0.
ITNO.
ITV8.
ITV0.
ITVO.
ITVO.
ITVO.
ITV0.
ITV0.
ITNO.
ITNO.
ITV0.
ITV0.
ITNO.

90 IV VRR- 53
91 IN VAR- 19
92 IV VAR- 108
93 IV VAR- 60
94 IV VAR- 67
95 IN VAR- 184
96 IV VAR- 274
97 IV VAR- 281
98 IV VAR- 191
99 IV VAR- 26
100 IV VRR- 74
101 IV VAR- 115
102 IV VAR- 198
103 IV VAR- 81
104 IN VAR- 288
105 IV VRR- 33
106 IV VRR- 122
107 IV UAR- 129
108 IV VAR- 295
109 IV VAR- 205
110 IV VAR- 136
111 IV VAR- 40
112 IV VAR- 212
113 IV VAR- 88
114 IV VAR- 219
115 IV VAR- 302
116 IV VAR- 47
117 IV VRR- 226
118 IV VRR- 309
119 IV VAR- 95
120 IV 988- 143
121 IV VAR- 233
122 IV VRR- 150
123 IV VAR- 54
124 IV VAR- 316
125 IN VRR- 102
126 IV VRR- 61
127 IV VAR- 240
128 IV VRR- 68
129 IV VAR- 109
130 IV VAR- 247
131 IV VAR- 157
132 IV VAR- 323
133 IV VRR- 75
134 IV VRR- 164
135 IV 988- 330
136 IV VRR- 116
137 IV VAR- 82
138 IV VRR- 254
139 IV VRR- 337
140 IV VRR- 123
141 IV VAR- 171
142 IV VRR‘ 130

VPOS VERSION 4.0

VAR- 11 TO LOVER BOUVO
VAR- 100 TO LOVER BOUND
988- 52 TO LOVER BOUVO
VAR- 59 TO LOVER ROUND
VAR- 176 TO LOVER BOUND
VAR- 838 TO LOVER BOUND
VAR- 839 TO LOVER BOUND
VAR- 183 TO LOVER ROUND
VAR- 18 TO LOVER ROUVD
VRR- 66 TO LOVER ROUND
VVR- 107 TO LOVER ROUND
VAR- 190 TO LOVER BOUND
VAR- 73 TO LOVER BOUND
VRR- 840 TO LOVER BOUND
VAR- 25 TO LOVER BOUVO
VAR- 114 TO LOVER BOUVO
VAR- 121 TO LOVER ROUVO
VAR- 841 TO LOVER BOUND
VAR- 197 TO LOVER ROUVO
VRR- 128 TO LOVER BOUVO
VAR- 32 TO LOVER BOUVO
VAR- 204 TO LOVER ROUND
VAR- 80 TO LOVER ROUVO
UAR- 211 TO LOVER ROUND
VAR- 842 TO LOVER BOUVO
VAR- 39 TO LOVER ROUVO
988- 218 TO LOVER ROUVO
VAR- 843 TO LOVER BOUVO
VAR- 87 TO LOVER ROUVO
VAR- 135 TO LOVER BOUND
VAR- 225 TO LOVER BOUVD
VRR- 142 TO LOVER ROUVO
VRR- 46 TO LOVER ROUVO
VAR- 844 TO LOVER ROUND
VRR- 94 TO LOVER ROUND
VRR- 53 TO LOVER ROUND
VRR- 232 TO LOVER ROUND
VRR- 60 TO LOVER ROUND
VAR- 101 TO LOVER ROUVO
VRR- 239 TO LOVER ROUND
VRR- 149 TO LOVER ROUVD
VRR- 845 TO LOVER ROUVO
VAR- 67 TO LOVER ROUVO
VAR- 156 TO LOVER ROUVO
VAR- 846 TO LOVER ROUVO
VAR’ 108 TO LOVER BOUVO
VAR- 74 TO LOVER BOUVO
VAR- 246 TO LOVER BOUND
988- 847 TO LOVER BOUND
VAR- 115 TO LOVER BOUND
VRR- 163 TO LOVER BOUND
908- 122 TO LOVER OOUVO
VRR- 253 TO LOVER BOUND

-VVIV=
-VNIV=
-VVIN=
-VVIV=
-VVIV=
~VVIV=
-VHIN=
-VNIV=
~VNIV=
-VVIN=
‘VNIV!
~VNIV=
-VNIV=
-VVIN=
-VVIV=
-VNIV8
-VVIV=
-VVIN=
-VNIV=
'VVIVS
‘VVIN8
-VNIV=
-VVIV=
-VNIV=
-VVIV=
'VNIVS
-VVIV8
-VNIV=
-VNIV=
-VVIV8
'VVIN8
'VNIVS
-VNIV=
-VNIV=
-VVIV=
-VNIV=
-VNIV'
-VVIV=
-VVIN=
-VVIV=
-VNIV=
~VN1V8
-VVIV=
~VNIV=
-VVIV8
~VNIV'
-VVIV=
~VVIV8
-VVIN=
-VVIV=
~VVIN=
-VNIV=
'VVIV=

VORTRVESTERV UNIVERSITY

12

-1394816.000000
-1390426.000000
'1389624.000000
-1383352.000000
-1375304.000000
-1375182.000000
'1374312.000000
-1372702.000000
-1361854.000000
'1351866.000000
'1348584.000000
-1340860.000000
~1329900.000000
-1329816.000000
'1325590.000000
'1320402.000000
-1315190.000000
-1310222.000000
'1301402.000000
-1300934.000000
'1296788.000000
*1294592.000000
'1286134.000000
'1283774.000000
'1283118.000000
'1278230.000000
'1274360.000000
-1266552.000000
71264240.000000
-1261984.000000
'1248334.000000
‘1247112.000000
°1236202.000000
-1233672.000000
'1227424.000000
71225374.000000
-1224636.000000
'1219102.000000
71215226.000000
-1205632.000000
‘1204498.000000
-1190818.000000
'1187616.000000
'1184226.000000
'1181316.000000
'1173384.000000
'1165650.000000
-1164278.000000
'1157818.000000
'1145202.000000
'1143290.000000
'1139990.000000
-1131626.000000

02I24/82

.19.07.38.

PAGE

10

VVVVWHVVVVVVHV‘V"

1 PROBLEM RUBBER 1 4'
"4444444444444444444

USIVS REGULAR
1182.1982.99 ORIEINS,693 ELE. V.E. PLAVT AT ROSE CITY.

ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITNO.
ITV0.
ITV0.
ITV8.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITV0.
ITV8.
ITNO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.

143 IN VAR- 261
144 IN VAR- 137
145 IN VAR- 89
146 IV 988* 344
147 IV VAR- 178
148 IV VAR- 96
149 IV VAR- 268
150 IV VAR- 144
151 IN VAR- 351
152 IV VAR- 185
153 IV VAR- 192
154 IV VAR- 151
155 IN VAR- 358
156 IV VAR- 103
157 IV VOR' 275
158 IV VRR- 282
159 IN VRR- 110
160 IV VAR- 199
161 IV VAR‘ 365
162 IV 908- 158
163 IV 908- 289
164 IN VAR- 206
165 IV VAR- 117
166 IN VRR- 165
167 IV 988- 213
168 IV VAR- 296
169 IV VAR- 372
170 IV 988- 220
171 IV VRR- 379
172 IV 988- 227
173 IV VRR- 124
174 IV VRR- 386
175 IV VAR- 131
176 IN VOR- 303
177 IV VRR- 172
178 IV VRR- 138
179 IV VVR- 393
180 IV 988- 234
181 IV 988- 310
182 IV VRR- 400
183 IV 988- 179
184 IN 988- 241
185 IN VRR- 145
186 IV 988- 317
187 IV VAR- 248
188 IN VVR- 407
189 IV VAR- 414
190 IV 988- 186

VAR- 129 TO LOVER ROUND
VAR- 81 TO LOVER BOUND
VAR- 848 TO LOVER BOUVO
VAR- 170 TO LOVER ROUVO
988- 88 TO LOVER ROUND
948- 260 TO LOVER ROUND
VAR- 136 TO LOVER BOUVO
VAR- 849 TO LOVER ROUND
VAR- 177 TO LOVER BOUVO
VRR- 184 TO LOVER ROUVO
988- 143 TO LOVER ROUVO
VRR- 850 TO LOVER ROUVO
VRR- 95 TO LOVER BOUND
VAR- 267 TO LOVER ROUND
908- 274 TO LOVER ROUVO
VRR- 102 TO LOVER BOUVO
VAR- 191 TO LOVER ROUND
VRR- 851 TO LOVER BOUVO
VAR- 150_TO LOVER ROUND
988- 281 TO LOVER BOUND
VAR- 198 TO LOVER ROUVU
VAR- 109 TO LOVER ROUVO
988- 157 TO LOVER ROUND
VAR7 205 TO LOVER BOUVO
VRR- 288 TO LOVER OOUVO
VRR- 852 TO LOVER ROUVO
988- 212 TO LOVER ROUVO
988- 853 TO LOVER OOUVO
VAR- 219 TO LOVER ROUNO
VAR- 116 TO LOVER BOUVO
VRR- 854 TO LOVER ROUND
988- 123 TO LOVER ROUND
988- 295 TO LOVER ROUND
VAR- 164 TO LOVER ROUND
988- 130 TO LOVER ROUND
UAR- 855 TO LOVER ROUND
VRR- 226 TO LOVER ROUVO
VVR- 302 TO LOVER ROUND
VAR- 856 TO LOVER ROUND
988- 171 TO LOVER ROUVO
VRR- 233 TO LOVER BOUVO
988- 137 TO LOVER BOUVO
VAR- 309 TO LOVER ROUND
988- 240 TO LOVER BOUVO
VAR- 857 TO LOVER ROUND
VAR- 858 TO LOVER BOUND
VRR- 178 TO LOVER ROUND
VAR- 144 TO LOVER BOUND

13

-VNIV=
-VVIV=
-VNIN=
-VVIN=
-VVIV=
-VVIN=
-VVIV=
-VNIN*
-VVIV=
-VNIN=
-VNIV8
-VVIN=
'VNIV3
-VNIV=
'VNIV=
-VNIV=
-VNIN=
'VVIV=
-VVIV=
-VVIV=
-VVIV=
-VVIV=
-VNIV8
-VNIV=
'VVIV=
“VVINB
-VNIV8
-VVIV8
-VNIV=
-VNIN'
-VNIV=
-VNIV8
-VVIV=
-VVIV8
-VVIV8
-VVIV8
~VVIN=
-VVIN8
-VVIN'
-VN1V8
-VVIV8
-VVIN3
-VNIN=
-VNIV=
-VN1V=
-VNIN=
-VVIN=
-VVIN=

-1125734.000000
-1121884.000000
'1115614.000000
'1112546.000000
-1099990.000000
-1093652.000000
-1086784.000000
-1083598.000000
-1083304.000000
71080702.000000
-1071912.000000
-1065984.000000
‘1063174.000000
-1061282.000000
'1060412.000000
-1050976.000000
'1048860.000000
-1033098.000000
-1029298.000000
'1015916.000000
-1009402.000000
'1009134.000000
'1009026.000000
'1002592.000000
*996322.000000
-995402.000000
-991774.000000
-990960.000000
'982360.000000
'980952.000000
-976484.000000
'975740.000000
'969218.000000
-968090.000000
'961484.000000
~958852.000000
'956334.000000
“952652.000000
~939656.000000
-937346.000000
-932636.000000
-922534.000000
-919772.000000
-913632.000000
-912650.000000
-910786.000000
-908104.000000
-907662.000000

ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
T

191 IN VAR- 152
192 IN VAR- 193
193 IV VAR- 421
194 IN VAR- 324
195 IV VAR- 200
196 IV VAR- 255
197 IV VAR- 331
198 IN VAR- 159
199 IN VAR- 428
200 IV VAR- 166
201 IV VAR- 338
202 IV VAR- 262
203 IN VAR- 435
204 IN VAR- 207
205 IV VAR- 214
206 IN VAR- 221
207 IV VAR- 442
208 IV VAR- 228
209 IN VAR- 173
210 IV VAR- 345
211 IV VAR- 269
212 IN VAR- 235
213 IV VAR- 180

VPOS VERSIOV 4.0

{Viiiflflfliiiiiiflfl

I PROBLEV VUNRER 1 4
441444414444444444444

USING REGULAR
IIG2.1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.

ITVO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.

214 IN VAR- 352
215 IV VAR- 276
216 IN VAR- 283
217 IV VAR- 449
218 IV VAR- 242
219 IV VAR- 359
220 IV VAR- 187
221 IV VAR- 194
222 IV VAR- 249
223 IV VAR- 456
224 IV VAR- 463
225 IV VAR- 290
226 IV VAR' 366
227 IV VAR- 6
228 IV VAR- 201
229 IV VAR- 470
230 IV VAR- 297
231 IN VAR- 256
232 IN VAR- 13
233 IV VAR- 373
234 IV VAR- 304
235 IV VAR- 380
236 IV VAR- 477

VAR“ 185 TO LOVER ROUVO
VAR- 859 TO LOVER BOUND
VAR- 316 TO LOVER BOUVO
VAR- 192 TO LOVER ROUVO
VAR- 247 TO LOVER BOUVO
VAR- 323 TO LOVER ROUND
VAR- 151 TO LOVER BOUVO
VAR- 860 TO LOVER BOUVO
VAR- 158 TO LOVER BOUND
VAR- 330 TO LOVER ROUVO
VAR- 254 TO LOVER ROUVO
VAR- 861 TO LOVER BOUND
VAR- 199 TO LOVER BOUND
VAR- 206 TO LOVER BOUND
VAR- 213 TO LOVER ROUVO
VAR- 862 TO LOVER ROUVO
VAR- 220 TO LOVER ROUND
VAR- 165 TO LOVER ROUND
VAR- 337 TO LOVER ROUVO
VAR- 261 TO LOVER BOUND
VAR- 227 TO LOVER ROUVO
VAR- 172 TO LOVER ROUND
VAR- 344 TO LOVER BOUND

-VVIN8
-VVIN=
-VVIV=
-VVIN=
-VVIV=
~VVIV=
-VNIN=
-VNIN=
'VVIN=
-VNIN=
-VVIV=
-VNIV=
-VNIV=
'VNIN=
-VNIV=
-VVIV=
-VNIV=
-VVIN=
-VVIN=
'VVIN3
-VNIN=
-VNIV8
-VNIV=

VORTHVESTERV UVIVERSITY

VAR- 268 TO LOVER ROUND
VAR- 275 TO LOVER ROUVO
VAR- 863 TO LOVER ROUVO
VAR- 234 TO LOVER BOUND
VAR- 351 TO LOVER ROUND
VAR- 179 TO LOVER BOUVO
VAR- 186 TO LOVER ROUND
VAR- 241 TO LOVER ROUVO
VAR- 864 TO LOVER ROUVO
VAR- 865 TO LOVER BOUND
VAR- 282 TO LOVER BOUND
VAR- 358 TO LOVER ROUVO
VAR- 903 TO LOVER ROUVO
VAR- 193 TO LOVER ROUVO
VAR- 866 TO LOVER BOUND
VAR- 289 TO LOVER OOUVO
VAR- 248 TO LOVER BOUND
VAR- 5 TO LOVER BOUVO
VAR- 365 TO LOVER ROUND
VAR- 296 TO LOVER BOUND
VAR- 372 TO LOVER BOUND
VAR- 867 TO LOVER ROUVO
VAR- 200 TO LOVER BOUND

14

'VVIV=
~VNIV=
-VVIN=
-VVIN'
-VVIV=
'VVIV8
-VNIV=
-VVIV=
-VNIN3
-VVIV8
-VVIV=
°VVIV¢
-VVIV=
'VNIV3
-VVIV=
-VNIV=
-VVIV=
-VNIN=
-VNIV=
-VNIV=
-VVIN=
-VNIN=
-VNIV=

'905502.000000
'885046.000000
-876918.000000
'873660.000000
~872278.000000
-867416.000000
-865048.000000
-849266.000000
-844776.000000
-843918.000000
'839626.000000
-837746.000000
'834202.000000
-827392.000000
'816574.000000
'807722.000000
-807160.000000
'803840.000000
'801714.000000
'801652.000000
-781134.000000
~773096.000000
'769698.000000

'769282.000000
-768412.000000
-765664.000000
‘757436.000000
-752084.000000
-743854.000000
-741252.000000
~738432.000000
'733730.000000
'731116.000000
'723916.000000
-719198.000000
-716000.000000
“709410.000000
~709224.000000
-704322.000000
-697078.000000
-693850.000000
-681502.000000
-677218.000000
-677060.000000
-675234.000000
-669952.000000

02/24782

1.19.07.38.

PAGE

11

ITNO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
T

237 IV VAR- 208
238 IN VAR- 20
239 IV VAR- 263
240 IV VAR- 215
241 IN VAR- 387
242 IV VAR- 311
243 IV VAR- 222
244 IN VAR- 394
245 IN VAR- 229
246 IN VAR- 484
247 IV VAR- 27
248 IV VAR- 318
249 IN VAR- 270
250 IV VAR- 401
251 IV VAR- 236
252 IV VAR- 491
253 IN VAR- 408
254 IV VAR- 415
255 IV VAR- 34
256 IV VAR- 277
257 IN VAR- 284
258 IV VAR- 243
259 IV VAR- 325
260 IV VAR- 332
261 IN VAR- 250
262 IV VAR- 422
263 IV VAR- 7
264 IV VAR- 41
265 IN VAR- 498
266 IV VAR- 505
267 IV VAR- 339
268 IV VAR- 291
269 IV VAR- 48
270 IV VAR- 14
271 IN VAR- 429
272 IV VAR- 257
273 IN VAR- 298
274 IV VAR- 436
275 IN VAR- 512
276 IV VAR- 21
277 IV VAR- 346
278 IN VAR- 55
279 IV VAR- 305
280 IV VAR- 264
281 IN VAR- 519
282 IV VAR- 62
283 IV VAR- 443
284 IV VAR- 69

VPOS VERSION 4.0

{VMVHVVVVVHVVH’VV

4 PROBLEV ROVOER 1 9
444444444444444444444

USING REGULAR

VAR- 12 TO LOVER BOUND
VAR- 255 TO LOVER BOUND
VAR- 207 TO LOVER ROUVO
VAR‘ 379 TO LOVER ROUND
VAR- 303 TO LOVER BOUVO
VAR- 214 TO LOVER ROUND
VAR- 386 TO LOVER BOUVD
VAR- 221 TO LOVER ROUVO
VAR- 868 TO LOVER BOUND
VAR- 19 TO LOVER ROUND
VAR- 310 TO LOVER ROUVO
VAR- 262 TO LOVER ROUVO
VAR- 393 TO LOVER ROUND
VAR- 228 TO LOVER ROUND
VAR- 869 TO LOVER BOUVO
VAR- 400 TO LOVER BOUVO
VAR- 407 TO LOVER ROUVO
VAR- 26 TO LOVER BOUND
VAR- 269 TO LOVER BOUND
VAR- 276 TO LOVER ROUVO
VAR- 235 TO LOVER ROUND
VAR- 317 TO LOVER BOUVO
VAR- 324 TO LOVER BOUND
VAR- 242 TO LOVER ROUVO
VAR- 414 TO LOVER GOUVO
VAR- 904 TO LOVER BOUVO
VAR- 33 TO LOVER ROUVO
VAR- 870 TO LOVER ROUND
VAR- 871 TO LOVER ROUVO
VAR- 331 TO LOVER ROUND
VAR- 283 TO LOVER GOUVO
VAR- 40 TO LOVER ROUVO
VAR- 6 TO LOVER ROUND
VAR- 421 TO LOVER ROUVD
VAR- 249 TO LOVER ROUVO
VAR- 290 TO LOVER BOUVO
VAR- 428 TO LOVER ROUVO
VAR- 872 TO LOVER ROUND
VAR- 13 TO LOVER ROUND
VAR- 338 TO LOVER ROUVO
VAR- 47 TO LOVER ROUVO
VAR- 297 TO LOVER BOUVO
VAR- 256 TO LOVER BOUND
VAR- 873 TO LOVER ROUND
VAR- 54 TO LOVER GOUVO
VAR- 435 TO LOVER OOUVO
VAR- 61 TO LOVER ROUVO
VAR- 20 TO LOVER ROUVO

-VVIV=
‘VVIN8
-VVIV=
-VNIV=
-VNIN=
-VVIN=
-VNIV=
-VNIV=
'VVIV'
-VNIV=
-VNIN=
-VNIV‘
-VNIN=
-VVIN=
-VVIN=
-VVIV=
-VVIV'
-VNIN'
-VNIN8
~VNIV=
-V11IN=I
-VNIV8
-VNIN=
-VVIN8
-VNIV=
-VNIN=
-VVIV8
*VNIV’
-VNIN8
°VNIV8
-VNIV=
'VVIV!
-VVIV8
’VVIN‘
-VNIVB
°VVIV8
“VNIV8
-VNIN'
-VVIV=
‘VVIN'
-VVIV8
-VNIV8
-VVINI
-VNIV8
-VVIN3
'VNIV8
-VNIV8
-VIIIN'I

NORTHVESTERV UVIVERSITY

15

'664816.000000
-664426.000000
-663142.000000
-662584.000000
'660652.000000
7652324.000000
'644952.000000
‘642910.000000
'632698.000000
-631854.000000
-627772.000000
-626452.000000
7625756.000000
-616884.000000
'604030.000000
'598750.000000
'596886.000000
-595590.000000
-594082.000000
'593212.000000
'593186.000000
-584918.000000
-575416.000000
‘574182.000000
-571146.000000
7570000.000000
-566788.000000
-565644.000000
'558318.000000
'551918.000000
'548716.000000
'548230.000000
-547850.000000
'535366.000000
-532828.000000
'529122.000000
'523846.000000
'520136.000000
-518816.000000
’509714.000000
-506202.000000
'502018.000000
'500176.000000
'495858.000000
-495374.000000
-493822.000000
-489102.000000
'485854.000000

02/24I82

.19.07.38.

PAGE

12

1182.1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.

ITVO.
ITVO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITNO.
ITNO.

285 IV VAR- 28
286 IN VAR- 312
287 IV VAR- 353
288 IV VAR- 271
289 IV VAR- 360
290 IV VAR- 526
291 IN VAR- 76
292 IN VAR- 319
293 IV VAR- 450
294 IV VAR- 533
295 IV VAR- 35
296 IV VAR- 83
297 IV VAR- 540
298 IV VAR- 278
299 IV VAR- 285
300 IV VAR- 547
301 IV VAR- 367
302 IN VAR- 554
303 IV VAR- 42
304 IV VAR- 457
305 IV VAR- 464
306 IN VAR- 326
307 IV VAR- 561
308 IN VAR- 49
309 IV VAR- 333
310 IV VAR- 471
311 IN VAR- 90
312 IV VAR- 374
313 IV VAR- 381
314 IV VAR- 292
315 IN VAR‘ 568
316 IV VAR- 340
317 IV VAR‘ 388
318 IV VAR- 97
319 IV VAR- 299
320 IV VAR- 478
321 IV VAR- 56
322 IV VAR- 395
323 IV VAR- 63
324 IV VAR- 575
325 IV VAR- 70
326 IN VAR- 306
327 IV VAR- 347
328 IN VAR- 402
329 IV VAR- 104
330 IN VAR- 313
331 IN VAR- 111
332 IV VAR- 485
333 IV VAR- 84
334 IV VAR- 582
335 IN VAR- 409
336 IN VAR- 416
337 IN VAR- 354

VAR- 304 TO LOVER BOUND
VAR- 345 TO LOVER ROUND
VAR- 263 TO LOVER BOUVO
VAR- 352 TO LOVER BOUND
VAR- 874 TO LOVER ROUVO
VAR- 68 TO LOVER ROUND
VAR- 311 TO LOVER ROUVO
VAR- 442 TO LOVER BOUVO
VAR- 875 TO LOVER ROUND
VAR- 27 TO LOVER ROUND
VAR- 75 TO LOVER ROUVO
VAR- 876 TO LOVER ROUND
VAR- 270 TO LOVER BOUVO
VAR- 277 TO LOVER ROUND
VAR- 877 TO LOVER ROUND
VAR- 359 TO LOVER BOUND
VAR- 878 TO LOVER BOUVO
VAR- 34 TO LOVER ROUND
VAR- 449 TO LOVER ROUVO
VAR- 456 TO LOVER ROUVO
VAR- 318 TO LOVER ROUND
VAR- 879 TO LOVER ROUND
VAR- 41 TO LOVER BOUVO
VAR- 325 TO LOVER ROUVO
VAR- 463 TO LOVER ROUVO
VAR- 82 TO LOVER ROUND
VAR- 366 TO LOVER ROUVO
VAR- 373 TO LOVER ROUND
VAR- 284 TO LOVER BOUVO
VAR- 880 TO LOVER ROUVO
VAR- 332 TO LOVER ROUVO
VAR- 380 TO LOVER OOUVO
VAR- 89 TO LOVER ROUVO
VAR- 291 TO LOVER BOUND
VAR- 470 TO LOVER BOUND
VAR- 48 TO LOVER ROUND
VAR- 387 TO LOVER BOUND
VAR- 55 TO LOVER ROUND
VAR- 881 TO LOVER BOUVO
VAR- 62 TO LOVER ROUVO
VAR- 298 TO LOVER BOUVO
VAR- 339 TO LOVER ROUVO
VAR- 394 TO LOVER ROUND
VAR- 96 TO LOVER ROUND
VAR- 305 TO LOVER ROUVO
VAR- 103 TO LOVER ROUND
VAR- 477 TO LOVER ROUVO
VAR- 69 TO LOVER ROUND
VAR- 882 TO LOVER BOUVO
VAR- 401 TO LOVER ROUVU
VAR- 408 TO LOVER ROUND
VAR- 346 TO LOVER ROUND
VAR- 484 TO LOVER BOUND

16

-VVIV=
-VVIN=
'VVIV8
-VNIN=
-VVIN=
-VVIN=
-VNIN=
-VVIN=
-VNIN=
-VNIV=
-VVIN=
-VVIV=
'VNIV=
-VRIV=
-VNIV=
-VVIN=
-VNIV=
-VNIN=
-VVIN=
-VNIN=
'VVIN=
'VNIN=
-VVIV=
~VVIN=
-VNIV=
’VVIN8
*VVIN‘
-VVIV8
-VVIN=
'VNIVS
-VVIN8
-VVIVI
-VVIV=
-VVIN=
-VHIV=
~VNIV3
‘VVIV'
-VNIV=
-VVIN=
-VVIV=
-VVIN=
-VVIV=
-VVIV=
-VNIN=
-VNIV=
-VVIN=
-VVIN=
-VVIV=
-VVIV=
-VVIV=
~VVIV=
-VVIN=
-VNIN=

~485452.000000
-477698.000000
-462202.000000
-460084.000000
-458678.000000
~457616.000000
-452572.000000
-451764.000000
-451596.000000
'449590.000000
-435650.000000
-433194.000000
-429832.000000
-428962.000000
-428820.000000
-427198.000000
-423306.000000
‘420788.000000
-419830.000000
-417216.000000
-409718.000000
-406568.000000
-402230.000000
-400216.000000
-395324.000000
'391884.000000
~389502.000000
~385060.000000
-384466.000000
'382626.000000
-376718.000000
-370584.000000
'369990.000000
'364872.000000
“361334.000000
-360202.000000
'352952.000000
'349374.000000
'347532.000000
-343102.000000
'337768.000000
-334514.000000
-333756.000000
-333174.000000
-321202.000000
'320976.000000
-318798.000000
-311616.000000
-307456.000000
“306750.000000
-304886.000000
-302498.000000
-290130.000000

ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
T

338 IV VAR- 492
339 IV VAR- 320
340 IN VAR- 361
341 IV VAR- 423
342 IV VAR- 118
343 IV VAR- 589
344 IV VAR- 91
345 IV VAR- 596
346 IN VAR- 368
347 IN VAR- 499
348 IN VAR- 125
349 IV VAR- 98
350 IV VAR- 132
351 IV VAR- 327
352 IV VAR- 506
353 IV VAR- 430
354 IN VAR- 334
355 IN VAR" 603

VPOS VERSIOV 4.0

WVMHHHV‘H'V‘IVH

5 PROBLEN VUVBER 1 4
umununnuaa

USING REBULAR
IIG2.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITNO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.

356 IV VAR- 437
357 IN VAR" I39
358 IV VAR- 375
359 IV VAR- 341
360 IV VAR- 382
361 IV VAR- 105
362 IV VAR- 513
363 IV VAR- 444
364 IN VAR- 610
365 IN VAR- 112
366 IV VAR- 389
367 IV VAR- 146
368 IV VAR- 520
369 IV VAR- 396
370 IV VAR- 153
371 IV VAR- 348
372 IV VAR° 617
373 IV VAR- 451
374 IN VAR- 403
375 IV VAR- 119
376 IV VAR- 527
377 IN VAR- 624
378 IV VAR- 355
379 IN VAR- 534
380 IV VAR- 160
381 IV VAR- 410
382 IV VAR- 417
383 IV VAR‘ 458

VAR- 312 TO LOVER ROUVO
VAR- 353 TO LOVER BOUVO
VAR- 415 TO LOVER ROUVD
VAR- 110 TO LOVER BOUVO
VAR- 883 TO LOVER BOUVO
VAR- 83 TO LOVER BOUND
VAR- 884 TO LOVER BOUND
VAR- 360 TO LOVER ROUND
VAR- 491 TO LOVER BOUND
VAR- 117 TO LOVER BOUND
VAR- 90 TO LOVER BOUVO
VAR- 124 TO LOVER BOUNO
VAR- 319 TO LOVER ROUVO
VAR- 498 TO LOVER ROUND
VAR- 422 TO LOVER ROUVO
VAR- 326 TO LOVER ROUVO
VAR- 885 TO LOVER ROUVO
VAR- 429 TO LOVER ROUND

-VNIV=
-VNIV=
-VVIV=
-VNIV=
-VNIV=
-VNIV=
-VVIV=
-VNIV=
-VVIN=
-VVIN*
-VVIV=
°VNIN=
-VVIN=
-VVIN=
-VVIN8
-VNIN=
-VNIV=
-VVIN8

NORTHVESTERV UVIVERSITY

VAR‘ 131 TO LOVER ROUVO
VAR- 367 TO LOVER ROUVO
VAR- 333 TO LOVER ROUND
VAR- 374 TO LOVER ROUVO
VAR- 97 TO LOVER ROUND
VAR- 505 TO LOVER ROUND
VAR- 436 TO LOVER ROUVO
VAR- 886 TO LOVER ROUVO
VAR- 104 TO LOVER ROUVO
VAR- 381 TO LOVER ROUND
VAR- 138 TO LOVER ROUVO
VAR- 512 TO LOVER ROUVO
VAR- 388 TO LOVER ROUVO
VAR- 145 TO LOVER ROUND
VAR- 340 TO LOVER BOUND
VAR- 887 TO LOVER BOUVO
VAR- 443 TO LOVER ROUVO
VAR- 395 TO LOVER OOUVO
VAR- 111 TO LOVER ROUND
VAR- 519 TO LOVER ROUVO
VAR- 888 TO LOVER ROUND
VAR- 347 TO LOVER BOUND
VAR- 526 TO LOVER BOUND
VAR- 152 TO LOVER ROUVO
VAR- 402 TO LOVER ROUVO
VAR- 409 TO LOVER BOUVO
VAR- 450 TO LOVER BOUND
VAR- 118 TO LOVER BOUND

17

-VNIN'
-VVIN=
~VVIN=
-VVIV=
-VNIN=
-VVIV=
-VVIV=
~VNIV=
-VNIV=
~VNIN=
-VNIN=
-VNIV=
-VNIV8
-VVIV=
~VVIV=
-VVIV8
-VNIV=
-VVIN=
-VNIN=
-VNIN=
-VNIN=
-VVIN=
'VVIV8
-VNIV*
-VNIV=
-VVIV=
-VNIN=
-VVIN=

-288322.000000
-284884.000000
'279146.000000
~279134.000000
'279014.000000
-267850.000000
-257100.000000
~251998.000000
-251744.000000
'250952.000000
-245956.000000
'245740.000000
‘245468.000000
'244418.000000
-243366.000000
‘235966.000000
'235752.000000
7231846.000000

-231484.000000
-214302.000000
-212468.000000
'209860.000000
'209140.000000
'206236.000000
'201822.000000
'199024.000000
'196942.000000
'195384.000000
-192534.000000
-181958.000000
'177752.000000
'177662.000000
~170264.000000
7162086.000000
’159764.000000
'158556.000000
-155100.000000
-144778.000000
'141966.000000
'138248.000000
-137696.000000
-135048.000000
'131550.000000
-129686.000000
-I27830.000000
~126918.000000

02/24/82

.19.07.38.

PAGE

13

ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.

384 IV VAR- 126
385 IV VAR- 465
386 IN VAR- 133
387 IV VAR- 362
388 IV VAR- 541
389 IV VAR- 631
390 IV VAR- 548
391 IV VAR- 167
392 IN VAR- 555
393 IV VAR- 140
394 IV VAR- 424
395 IV VAR- 472
396 IV VAR- 562
397 IN VAR- 369
398 IN VAR° 638
399 IN VAR- 174
400 IV VAR- 479
401 IV VAR- 569
402 IV VAR- 147
403 IV VAR- 431
404 IV VAR- 645
405 IV VAR- 438
406 IV VAR- 652
407 IN VAR- 154
408 IV VAR- 376
409 IV VAR- 659
410 IV VAR- 383
411 IN VAR- 181
412 IN VAR- 666
413 IV VAR- 576
414 IV VAR- 390
415 IV VAR- 486
416 IV VAR- 445
417 IV VAR- 673
418 IV VAR- 188
419 IN VAR- 397
420 IV VAR- 680
421 IN VAR- 195
422 IV VAR- 161
423 IV VAR- 687

ENTERIVG PHASE II

T VPOS VERSIOV 4.0
444444444444444444444

4 PROBLEV VUVBER 1 1
444444444444444444444

USING REGULAR

VAR- 457 TO LOVER ROUND
VAR- 125 TO LOVER BOUVO
VAR- 354 TO LOVER BOUVO
VAR- 533 TO LOVER ROUND
VAR- 889 TO LOVER BOUVO
VAR- 540 TO LOVER ROUVO
VAR- 159 TO LOVER ROUVO
VAR‘ 547 TO LOVER ROUVO
VAR- 132 TO LOVER BOUND
VAR- 416 TO LOVER BOUND
VAR- 464 TO LOVER BOUVO
VAR- 554 TO LOVER ROUVO
VAR- 361 TO LOVER ROUND
VAR- 890 TO LOVER BOUND
VAR- 166 TO LOVER BOUND
VAR- 471 TO LOVER BOUND
VAR- 561 TO LOVER ROUND
VAR- 139 TU LOVER BOUVO
VAR- 423 TO LOVER BOUVO
VAR- 891 TO LOVER ROUVO
VAR- 430 TO LOVER ROUND
VAR- 892 TO LOVER ROUND
VAR- 146 TO LOVER ROUND
VAR- 368 TO LOVER ROUND
VAR- 893 TO LOVER BOUVO
VAR- 375 TO LOVER ROUND
VAR- 173 TO LOVER ROUND
VAR- 894 TO LOVER ROUVO
VAR- 568 TO LOVER ROUND
VAR- 382 TO LOVER BOUND
VAR- 478 TO LOVER ROUND
VAR- 437 TO LOVER ROUND
VAR- 895 TO LOVER ROUVO
VAR- 180 TO LOVER ROUVO
VAR- 389 TO LOVER ROUND
VAR- 896 TO LOVER ROUND
VAR- 187 TO LOVER ROUND
VAR- 153 TO LOVER BOUVO
VAR- 897 TO LOVER BOUND
VAR- 898 TO LOVER ROUND

-VVIN=
-VVIV=
-VVIV=
~VVIV=
-VVIV=
-VNIV=
-VNIN=
~VVIN=
-VVIN=
°VVIN=
-VVIV8
-VVIV=
-VVIV=
-VVIN=
-VVIV8
'VNIN'
-VNIN8
-VVIV'
-VNIV=
-VVIV=
’VVINa
-VVIN8
-VVIV=
-VVIV=
-VVIV=
-VVIV8
-VVIN=
-VVIV=
-VVIV=
'VVIV’
-VNIN=
-VVIV=
-VVIN=
'VVIV8
-VVIV8
'VVIV'
-VV1N=
-VVIN'
-VVIN8
-VVIV8

NORTVVESTERV UVIVERSITY

II82.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

ITVO.
ITVO.

ITVO. 426 IV VAR- 594

VAR- 195 TO LOVER ROUND

18

-125216.000000
'121706.000000
-120634.000000
‘119294.000000
'117428.000000
'114920.000000
'114776.000000
“109406.000000
'107450.000000
'103946.000000
-103324.000000
-92668.000000
-87748.000000
'83790.000000
-73840.000000
-69334.000000
'68726.000000
'68500.000000
-68166.000000
-59680.000000
756646.000000
'56240.000000
'53628.000000
-50052.000000
-46080.000000
-45610.000000
-43096.000000
'41686.000000
-33632.000000
731134.000000
'26798.000000
-26622.000000
-15174.000000
'13854.000000
'13502.000000
-13388.000000
-11252.000000
'11014.000000
-736.000000
0.000000

424 IV VAR- 311 VAR- 396 TO LOVER BOUND -ZVIV= '12757064.819991
425 IN VAR- 678 VAR- 673 TO LOVER BOUND -ZVIV= -12707940.889991

-ZVIV= -12451648.939991

02/24/82

.19.07.38.

PAGE

14

ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.

427 IV VAR- 75
428 IV VAR- 657
429 IV VAR- 477
430 IV VAR- 241
431 IN VAR- 96
432 IN VAR- 671
433 IV VAR- 175
434 IN VAR- 339
435 IV VAR- 525
436 IV VAR- 456
437 IV VAR- 110
438 IN VAR- 442
439 IV VAR- 466
440 IV VAR- 262
441 IN VAR- 597
442 IV VAR- 532

VAR- 589 TO LOVER BOUND
VAR- 486 TO LOVER BOUVO
VAR- 313 TO LOVER ROUVO
VAR- 76 TO LOVER ROUND
VAR- 652 TO LOVER ROUVO
VAR- 160 TO LOVER ROUVO
VAR- 243 TO LOVER ROUVO
VAR- 666 TO LOVER BOUVO
VAR- 479 TO LOVER BOUNO
VAR- 98 TO LOVER ROUVO
VAR- 458 TO LOVER ROUND
VAR- 444 TO LOVER GOUVO
VAR- 341 TO LOVER GOUVO
VAR- 575 TO LOVER ROUND
VAR- 520 TO LOVER ROUND
VAR- 465 TO LOVER ROUND

19

-ZVIV8 '12113747.979991
-ZNIN= ~12030938.439991
-ZVIN= -12017443.799990
-ZVIN= -11810003.319990
-ZVIN= '11790143.049990
-ZVIV= ~11534850.969990
-ZVIN' -11420924.019990
-ZVIV= -10990287.589990
-ZVIV= -10730837.689990
-ZNIV= -10713376.189990
72NIV8 710665641.029990
-ZVIV= -10439967.599990
’IVIN= -10290568.959990
'ZNIV' ~10178782.659990
-ZVIN= -10068105.169990
-ZVIN= '9898662.169990

T

ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITNO.
ITVO.
ITNO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.

443 IV VAR- 452
444 IV VAR- 553
445 IV VAR- 518
446 IV VAR- 124
447 IV VAR- 414
448 IV VAR- 639
449 IV VAR- 131
450 IV VAR- 354
451 IV VAR- 38
452 IV VAR- 546
453 IV VAR- 511
454 IV VAR- 583
455 IV VAR- 421
456 IN VAR- 474
457 IN VAR- 145
458 IV VAR- 473
459 IV VAR- 200
460 IV VAR- 19
461 IN VAR- 660
462 IN VAR- 450
463 IV VAR- 488
464 IV VAR- 65
465 IV VAR- 581
466 IV VAR- 9
467 IV VAR- 565
468 IV VAR- 251
469 IV VAR- 137
470 IV VAR- 681
471 IN VAR- 224
472 IV VAR- 135
473 IV VAR- 614
474 IN VAR- 568
475 IV VAR- 280
476 IV VAR- 287
477 IV VAR- 163
478 IV VAR- 418
479 IN VAR- 428
480 IV VAR- 491
481 IV VAR- 48
482 IV VAR- 586
483 IV VAR- 676
484 IN VAR- 592
485 IV VAR- 350
486 IV VAR- 408
487 IN VAR- 411
488 IV VAR- 177
489 IV VAR- 432
490 IV VAR- 104
491 IV VAR- 663
492 IN VAR- 628
493 IN VAR- 574
494 IV VAR- 540

NPOS VERSIOV 4.0

"44444444444451!!!"

VAR- 527 TO LOVER GOUVD
VAR- 548 TO LOVER GOUVO
VAR- 112 TO LOVER BOUND
VAR- 445 TO LOVER BOUVD
VAR- 596 TO LOVER BOUVD
VAR- 126 TO LOVER BOUND
VAR- 264 TO LOVER BOUVD
VAR~ 133 TO LOVER DOUVD
VAR- 513 TO LOVER BOUND
VAR- 541 TO LOVER ROUND
VAR- 638 TO LOVER ROUVD
VAR- 417 TO LOVER ROUVD
VAR- 451 TO LOVER BOUVD
VAR- 42 TO LOVER ROUND
VAR~ 355 TO LOVER BOUVD
VAR- 474 TO LOVER ROUND
VAR- 147 TO LOVER BOUND
VAR- 582 TO LOVER ROUVD
VAR- 659 TO LOVER ROUVO
VAR- 506 TO LOVER ROUVD
VAR- 201 TO LOVER OOUVO
VAR- 485 TO LOVER DOUVD
VAR- 70 TO LOVER ROUND
VAR- 576 TO LOVER BOUVD
VAR- 19 TO LOVER ROUND
VAR- 14 TO LOVER DOUVD
VAR- 452 TO LOVER DOUVD
VAR- 174 TO LOVER DOUVD
VAR- 424 TO LOVER BOUVD
VAR- 562 TO LOVER ROUVD
VAR- 681 TO LOVER OOUVD
VAR- 222 TO LOVER ROUVD
VAR- 278 TO LOVER DOUVD
VAR- 140 TO LOVER ROUND
VAR- 610 TO LOVER ROUVD
VAR- 569 TO LOVER ROUND
VAR- 431 TO LOVER ROUND
VAR- 250 TO LOVER ROUND
VAR- 414 TO LOVER BOUVD
VAR- 167 TO LOVER ROUND
VAR- 678 TO LOVER ROUND
VAR- 285 TO LOVER BOUND
VAR- 137 TO LOVER ROUND
VAR- 583 TO LOVER ROUVD
VAR- 410 TO LOVER GOUVO
VAR- 408 TO LOVER DOUVD
VAR- 49 TO LOVER BOUVD
VAR- 428 TO LOVER OOUVD
VAR- 660 TO LOVER ROUVD
VAR- 348 TO LOVER ROUVD
VAR- 105 TO LOVER BOUND
VAR- 546 TO LOVER ROUND

°ZVIN=
’ZVIV=
-ZVIV=
~ZVIN=
'ZVIV=
-ZVIN=
-ZVIV=
-ZVIN=
-ZVIV=
-lVIV=
-ZVIV=
-ZVIN=
-ZVIV=
'ZNIN8
'ZVIN8
-ZVIV=
'2VINt
-ZVIV=
-ZVIV=
-ZVIV=
-ZVIN=
-ZVIV=
'7VIN8
-ZVIV=
'ZVIN=
-ZVIV8
‘ZNIVS
-ZVIV'
-ZNIV=
-ZVIV'
-ZVIV=
'ZVIV8
‘ZVIVS
-IVIV=
-ZVIV8
'ZIIIV'I
-ZNIV=
-ZVIVa
-ZVIN8
-ZVIN=
-ZVIN8
-ZVIV8
'INIVa
-ZVIV=
'IVIVS
-ZNIV8
-ZVIV=
-ZVIV=
-2VIN=
'ZVIN!
-ZVIN=
-INIV=

NORTVVESTERV UVIVERSITY

20

-9869600.669990
~9717496.979990
-9670080.659990
'9611729.309990
-9586103.729990
79534101.769990
-9309943.369990
-9198754.649990
-9130133.919990
-9032878.029990
-9027206.829990
'9014394.089990
-9000921.329990
‘8841610.029990
'8789493.969990
-8641568.769990
'8609432.679990
'8570401.379990
-8509961.909990
78444710.269990
-8128508.029990
-8083624.949990
'7750312.799990
'7613753.289990
-7518910.029989
'7389728.369989
-7361376.869989
-7320507.349989
-7308399.099989
~7284437.939989
-7236264.269989
-7192704.809989
77181677.559989
‘6888459.299989
“6832737.689989
'6813693.799989
-6702180.199989
'6605268.319989
-6579259.339989
-6479383.399989
~6463568.369989
-6409277.649989
'6331507.749989
'6319394.469989
-6315786.249989
-6305369.389989
-6236833.889989
'6221025.439989
'6189784.099989
-6162335.499989
-6150047.899989
'6121595.029989

02/24/82

.19.07.38.

PAGE

15

! PRODLEV VUVRER 1 *
184444444414444444441

USING REGULAR
II82.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITNO.
ITNO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
ITVO.
T

495 IN VAR- 505
496 IV VAR- 693
497 IV VAR- 497
498 IV VAR- 226
499 IV VAR‘ 302
500 IN VAR- 19
501 IN VAR- 316
502 IV VAR- 686
503 IV VAR- 26
504 IN VAR- 427
505 IV VAR- 547
506 IV VAR- 78
507 IV VAR- 642
508 IN VAR- 387
509 IV VAR- 535
510 IN VAR- 441
511 IV VAR- 359
512 IV VAR- 139
513 IV VAR- 460
514 IV VAR“ 655
515 IV VAR- 649
516 IV VAR- 621
517 IN VAR- 406
518 IN VAR- 137
519 IV VAR- 50
520 IV VAR- 665
521 IV VAR- 277
522 IV VAR- 448
523 IN VAR- 57
524 IV VAR- 365
525 IN VAR- 159
526 IN VAR- 89
527 IN VAR- 494
528 IV VAR° 453
529 IV VAR- 637
530 IV VAR- 653
531 IN VAR- 211
532 IV VAR- 202
533 IV VAR- 665
534 IV VAR- 160
535 IN VAR- 600
536 IV VAR- 187
537 IV VAR- 89

VPOS VERSION 4.0

if!!!“ VVIVVVHHVVVV

4 PROBLEV VUVRER 1 I
444444444144441444444

VAR- 568 TO LOVER ROUVD
VAR- 687 TO LOVER ROUVD
VAR- 511 TO LOVER ROUVD
VAR- 492 TO LOVER ROUVD
VAR- 624 TO LOVER ROUVD
VAR- 306 TO LOVER ROUVD
VAR- 491 TO LOVER ROUND
VAR- 21 TO LOVER ROUVD
VAR- 680 TO LOVER ROUND
VAR- 320 TO LOVER ROUVD
VAR- 553 TO LOVER ROUND
VAR- 421 TO LOVER ROUVD
VAR- 229 TO LOVER ROUVD
VAR- 639 TO LOVER ROUVD
VAR- 181 TO LOVER ROUVD
VAR- 390 TO LOVER ROUVD
VAR- 534 TO LOVER ROUND
VAR- 28 TO LOVER ROUVD
VAR- 594 TO LOVER ROUVD
VAR- 456 TO LOVER ROUND
VAR- 645 TO LOVER ROUVD
VAR- 438 TO LOVER ROUND
VAR- 657 TO LOVER ROUVD
VAR- 84 TO LOVER ROUND
VAR- 362 TO LOVER ROUVD
VAR- 617 TO LOVER ROUVD
VAR- 280 TO LOVER ROUVD
VAR- 56 TO LOVER ROUVD
VAR- 63 TO LOVER ROUVD
VAR- 665 TO LOVER ROUVD
VAR- 161 TO LOVER ROUVD
VAR- 442 TO LOVER ROUVD
VAR- 403 TO LOVER ROUVD
VAR- 89 TO LOVER ROUVD
VAR- 137 TO LOVER ROUVD
VAR- 369 TO LOVER ROUVD
VAR- 655 TO LOVER ROUND
VAR- 215 TO LOVER ROUVD
VAR- 497 TO LOVER ROUVD
VAR- 450 TO LOVER ROUVD
VAR- 139 TO LOVER ROUND
VAR- 188 TO LOVER ROUVD
VAR- 663 TO LOVER ROUVD

-ZVIN=
-ZVIV8
'ZVIV=
'ZVINI
-ZVIV=
°ZVIV=
-ZVIV=
~2VIV8
-2VIV8
-ZVIV=
‘ZNIV‘
-ZVIV=
-ZVIV'
-ZVIV8
-ZNIV8
-ZVIV=
-ZVIN=
’lVIV’
-ZVIV=
-ZNIV8
-ZVIV'
-ZVIV=
-ZVIV=
-ZVIV=
-ZNIV8
-ZVIV8
-IVIV=
-ZVIV=
-ZVIV8
-ZNIV8
-ZNIV=
'ZVIN8
-ZNIV¢
-ZVIV=
'ZNIVB
'IVIV3
-ZVIV=
-ZVIV8
-ZVIV8
-ZVIV'
-ZVIV8
°ZVIV=
-ZVIV=

VORTVVESTERN UVIVERSITY

21

-5946435.429989
'5942030.469989
-5891788.4599R9
-5846935.929989
-5793746.079989
'5760798.8799R9
75614443.219989
'5596867.4699R9
°5555808.599989
'5534451.399989
'5511568.299989
-5407052.0399R9
'5340333.239989
-5318349.239989
'5299441.779989
-5235569.309989
'5228350.769989
'5199892.229989
'5191227.669989
'5186460.669989
-5171221.469989
'5161904.449989
'5148115.969989
-5120815.159989
'5108129.599989
'5098386.019989
'5096415.469989
-5090253.229989
'5080374.829989
-5074174.369989
-5037279.329989
-5024480.6899R9
-5006150.609989
-4995023.129989
~4990644.729989
-4983923.4R9989
'4969874.859989
'4963228.299989
-4960462.549989
-4955463.049989
-4951362.169989
-4950425.449989
-4950171.049989

02I24182

.19.07.38.

PAGE

16

USIVG REGULAR

1182.1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.

VAR VAR
VO VAVE

1 X1
2 I2
3 13
4 14
5 I5
6 I6
7 I7
8 18
9 I9
10 110
11 111
12 112
13 113
14 114
15 115
16 116
17 I17
18 118
19 119
20 I20
21 121
22 122
23 I23
24 124
25 I25
26 126
27 127
28 128
29 129
30 I30
31 131
32 I32
33 I33
34 I34
35 I35
36 I36
37 I37
38 I38
39 139
40 I40
41 I41
42 142
43 143
44 I44
45 I45
46 146
47 I47

ROV STATUS

L8
L8
L8
L8
L8

SUVVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

11075.0000000

0.0000000

14517.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

16481.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

18132.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

14401.0000000

0.0000000
0.0000000

9279.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

22

OPPORTUVITY
COST

-5.3700000
712.4500000
'14.2500000
-8.7900000
-2.6800000
-9.1000000
0.0000000
-16.2600000
0.0000000
-4.5000000
-4.1300000
-R.5700000
-3.8800000
-11.7600000
-7.3100000
“6.9600000
'11.0700000
-6.7900000
0.0000000
’6.1800000
-3.8100000
-4.2900000
711.2600000
“14.2200000
-9.1200000
0.0000000
-9.0400000
-.9300000
-5.3000000
74.2400000
'5.6200000
-.4800000
“.0500000
-.4300000
0.0000000
-21.9800000
'8.5200000
0.0000000
-3.6400000
-15.6400000
-6.3900000
~16.5200000
-R.9300000
-.0700000
-4.0500000
-.9100000
-I.5600000

LOVER

ROUVD
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

UPPER
ROUND
IVF
IVF
INF
INF
IVF
IVF
IVF
IVF
INF
IVF
INF
IVF
IVF
IVF
INF
IVF
IVF
INF
IVF
IVF
INF
IVF
INF
IVF
INF
IVF
IVF
INF
INF
INF
INF
IVF
INF
IVF
IVF
IVF
INF
IVF
INF
INF
IVF
INF
INF
INF
INF
IVF
IVF

48 148 -- R
49 149 -- L8
50 150 -- R
51 151 - L8
52 152 -- L8
53 153 - L8
54 154 -- LB
55 155 - L8
56 156 -- L8
57 157 - 8
58 158 - L8
59 159 - L8
60 160 -- LB
61 161 - L8
62 162 - 18
63 163 - L8
64 164 - L8
65 165 - R
T VPOS VERSIOV 4.0
444444444444444444444
i PRORLEV VUVRER 1 *
runnunnuuuu

USIVG REGULAR

21014.0000000

0.0000000

5414.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

3136.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

15743.0000000

0.0000000
'4.5000000
0.0000000
-17.0700000
718.4100000
-12.6600000
-4.4800000
-13.4400000
-2.2600000
0.0000000
716.2100000
'17.R400000
'12.1500000
'3.5700000
-12.7900000
'1.4900000
'17.6700000
0.0000000

VORTVVESTERN UVIVERSITY

1182.1982.99 ORIGIVS,693 ELE. V.E. PLANT AT ROSE CITY.

VAR VAR ROV STATUS

V0 VANE V8

66 166 '- L8
67 167 - L8
68 168 '7 L8
69 169 -' L8
70 170 - LR
71 171 - L8
72 172 -' L8
73 173 - L8
74 174 '7 L8
75 175 - 8
76 176 -- L8
77 177 '- L8
78 178 ’7 R
79 179 - L8
80 180 - L8
81 181 '7 L8
82 182 - L8
83 183 -' L8
84 184 -- L8
85 185 - L8
86 186 - LD
87 187 '- L8
88 188 -- LR

SUVVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

10983.0000000

0.0000000
0.0000000

21883.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

23

OPPORTUVITY

COST
'6.0100000
'6.1000000
-9.7400000
'5.8700000

-13.4000000
'5.9800000
'15.3300000
-20.1600000
'15.6600000
0.0000000
'15.2000000
-7.6200000
0.0000000
-16.7800000
-19.3500000
'13.8800000
-3.0900000
-14.1800000
-3.0100000
'3.8200000
-10.9500000
-13.6600000
-8.4600000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

LOVER

ROUND
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF
INF
IVF
INF
IVF
IVF
IVF
INF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
INF

UPPER
ROUVD
IVF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF

02l24/R2

.19.07.38. PAGE

17

89 189 - R 1272.0000000 0.0000000 0.0000 INF'
90 190 -‘ LR 0.0000000 -8.4600000 0.0000 IVF
91 191 -- R 9675.0000000 0.0000000 0.0000 IVF
92 192 -- LR 0.0000000 -7.3100000 0.0000 IVF
93 193 -- LR 0.0000000 -13.3700000 0.0000 INF
94 194 - L8 0.0000000 -19.1600000 0.0000 INF
95 195 -- L8 0.0000000 '15.2400000 0.0000 IVF
96 196 - R 18408.0000000 0.0000000 0.0000 IVF
97 197 -- LR 0.0000000 -14.5500000 0.0000 IVF
98 198 - L8 0.0000000 -8.7900000 0.0000 INF
99 199 - L8 0.0000000 -9.5500000 0.0000 IVF
100 1100 - LR 0.0000000 -3.6500000 0.0000 INF
101 1101 -‘ LR 0.0000000 -5.0800000 0.0000 INF
102 1102 -- LR 0.0000000 -.6300000 0.0000 INF
103 1103 -- LR 0.0000000 -3.3400000 0.0000 INF
104 1104 - R 6099.0000000 0.0000000 0.0000 INF
105 1105 -- LR 0.0000000 '4.4100000 0.0000 IVF
106 1106 - LR 0.0000000 -7.2700000 0.0000 IVF
107 1107 - LR 0.0000000 -8.2000000 0.0000 IVF
108 1108 - LR 0.0000000 '15.3800000 0.0000 IVF
109 1109 -- L8 0.0000000 -12.7400000 0.0000 INF
110 1110 - R 20921.0000000 0.0000000 0.0000 IVF
111 1111 7' LR 0.0000000 '11.8400000 0.0000 INF
112 1112 '7 LR 0.0000000 -8.7800000 0.0000 IVF
113 1113 - LR 0.0000000 '1.6700000 0.0000 IVF
114 1114 - LR 0.0000000 ~14.8400000 0.0000 IVF
115 1115 -' LR 0.0000000 '15.9600000 0.0000 INF
116 1116 - LR 0.0000000 '10.2200000 0.0000 INF
117 1117 - L8 0.0000000 -3.4500000 0.0000 INF
118 1118 - L8 0.0000000 '11.0000000 0.0000 IVF
119 1119 -- 8 14091.0000000 0.0000000 0.0000 IVF
120 1120 - LR 0.0000000 -5.9100000 0.0000 IVF
121 1121 - L8 0.0000000 '12.1700000 0.0000 IVF
122 1122 -' L8 0.0000000 '18.5800000 0.0000 INF
123 1123 "- LR 0.0000000 '15.0100000 0.0000 IVF
124 1124 - R 2606.0000000 0.0000000 0.0000 INF
125 1125 - LR 0.0000000 -14.2400000 0.0000 IVF
126 1126 - L8 0.0000000 -8.6900000 0.0000 INF
127 1127 - L8 0.0000000 -R.7400000 0.0000 IVF
128 1128 -' LR 0.0000000 '12.5200000 0.0000 INF
129 1129 '7 L8 0.0000000 -17.7600000 0.0000 INF
130 1130 - LR 0.0000000 -13.6900000 0.0000 INF
T NPOS VERSION 4.0 VORTVVESTERN UVIVERSITY 02l24/82 .19.07.38. PAGE 18
{44445414444444444444
i PRORLEV VUVRER 1 4
411444144444514454444
USING REGULAR
1182.1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.
SUNVARY OF RESULTS
VAR VAR ROV STATUS ACTIVITY OPPORTUNITY LOVER UPPER

24

VO VAVE
131 1131
132 1132
133 1133
134 1134
135 1135
136 1136
137 1137
138 1138
139 1139
140 1140
141 1141
142 1142
143 1143
144 1144
145 1145
146 1146
147 1147
148 1148
149 1149
150 1150
151 1151
152 1152
153 1153
154 1154
155 1155
156 1156
157 1157
158 1158
159 1159
160 1160
161 1161
162 1162
163 1163
164 1164
165 1165
166 1166
167 1167
168 1168
169 1169
170 1170
171 1171
172 1172
173 1173
174 1174
175 1175
176 1176
177 1177
178 1178
179 1179
180 1180
181 1181
182 1182
183 1183
184 1184

LR

LR

LEVEL

7128.0000000

0.0000000
0.0000000
0.0000000

19475.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

7436.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

21307.0000000

0.0000000
0.0000000
0.0000000
0.0000000

923.0000000
9213.0000000

0.0000000
0.0000000

20468.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

15372.0000000

0.0000000

14621.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

25

COST

0.0000000
“13.0400000
“8.0800000
“14.7000000
0.0000000
“1.2800000
“.7800000
“7.4600000
“.8400000
“9.8200000
“7.8600000
“.9000000
“5.9700000
“2.7900000
0.0000000
“1.8900000
“4.0900000
“1.4000000
“14.3800000
“16.1700000
“10.9100000
“2.0500000
“11.3000000
0.0000000
“5.6700000
“3.5100000
“5.1800000
“.2400000
0.0000000
0.0000000
“.5000000
“17.7500000
0.0000000
“4.5100000
“5.2900000
“9.8600000
“5.3200000
“13.3800000
“5.6700000
“7.1200000
“6.1700000
“.3400000
“2.5700000
“1.5900000
0.0000000
“8.9300000
0.0000000
“5.6700000
“2.9900000
“.9000000
“2.0800000
“5.2300000
“7.2900000
“1.3700000

ROUVD
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

4 0.0000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

ROUVD
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
INF
IVF
INF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
INF
IVF
IVF
IVF
INF
IVF
INF

185 1185 ““ L8
186 1186 ““ L8
187 1187 ““ 8
188 1188 ““ L8
189 1189 ““ L8
190 1190 ““ L8
191 1191 ““ L8
192 1192 ““ L8
193 1193 ““ L8
194 1194 ““ R
195 1195 ““ LR
VPOS VERSION 4.0
VMMVV‘IVVWVRVA
! PRORLEN VUVRER 1 !
{44444044444444444444

USING REGULAR

0.0000000
0.0000000

1301.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

15921.0000000

0.0000000

“5.2800000
“1.5600000
0.0000000
“.7200000
“3.0000000
“.2700000
“15.2800000
“20.1000000
“15.4000000
0.0000000
“15.0600000

VORTVVESTERN UNIVERSITY

IIG2.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

VAR VAR

NO NAVE
196 1196
197 1197
198 1198
199 1199
200 1200
201 1201
202 1202
203 1203
204 1204
205 1205
206 1206
207 1207
208 1208
209 1209
210 1210
211 1211
212 1212
213 1213
214 1214
215 1215
216 1216
217 1217
218 1218
219 1219
220 1220
221 1221
222 1222
223 1223
224 1224
225 1225

ROV STATUS

L8
L8
L8
L8

8
L8

8
L8
L8
L8
L8
L8

8
L8
L8

8
LR

- L8

L8
L8
L8
L8
L8
L8
L8
L8
L8
L8

8
L8

SUVVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000
0.0000000
0.0000000
0.0000000

16988.0000000

0.0000000

2741.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

3405.0000000

0.0000000
0.0000000

5409.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

4707.0000000

0.0000000

OPPORTUNITY

COST

“5.9500000
“14.8500000
“2.7800000
“2.0500000
0.0000000
“8.1900000
0.0000000
“9.6500000
“4.6900000
“9.2900000
“11.8800000
“6.7500000
0.0000000
“6.6900000
“.2300000
0.0000000
“15.9600000
“19.8500000
“14.8000000
“.8700000
“14.6900000
“4.7000000
“5.2400000
“14.6600000
“15.9900000
“10.3600000
“4.3500000
“10.9300000
0.0000000
“12.5100000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

LOVER

ROUND
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
INF
INF
IVF

UPPER
ROUVD
IVF
INF
INF
INF
IVF
INF
IVF
INF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
INF
INF
IVF
INF
IVF
INF

02l24/82

.19.07.38.

PAGE

19

226 1226 - B
227 1227 - L8
228 1228 - LB
229 1229 -- L3
230 1230 -“ L8
231 1231 - L8
232 1232 -- LA
233 1233 - L8
234 1234 -- LB
235 1235 - L8
236 1236 -- B
237 1237 -- LB
238 1238 -- LR
239 1239 -- L8
240 1240 -- LB
241 1241 - B
242 1242 -- L3
243 1243 -“ L8
244 1244 - LR
245 1245 - L8
246 1243 “- LB
247 1247 - L8
248 1248 -“ 18
249 1249 - L8
250 1250 - LB
251 1251 “- 8
252 1252 - LS
253 1253 - L8
254 1254 - L8
255 1255 -“ L8
256 1256 ““ LB
257 1257 -“ B
258 1258 ““ L8
259 1259 - LB
260 1260 - LA
I VPOS VERSIOV 4.0
4944444444444944444
0 PROBLEV VUVRER 1 1
444444444444444444444
USING REGULAR

13013.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

11849.0000000

0.0000000
0.0000000
0.0000000
0.0000000

9502.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

20677.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

16326.0000000

0.0000000
0.0000000
0.0000000

0.0000000
“6.4700000
“4.7100000
“4.3900000
“3.8900000
“8.7100000
“2.1500000

“14.3100000
“17.9000000
“12.8500000
0.0000000
“12.7300000
“3.1500000
“20.9900000
“7.4700000

0.0000000

“2.9000000

“14.5800000
“5.4200000
“15.5700000
“10.7000000
“1.8300000
“4.3600000

“.9100000
“3.8000000

0.0000000
“5.8700000
“1.1100000

“15.0500000
“20.2100000
“15.6800000
0.0000000
“15.2500000
“6.5900000
“18.7300000

VORTVVESTERN UVIVERSITY

IIR2.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

VAR VAR
NO VAVE
261 1261
262 1262
263 1263
264 1264
265 1265
266 1266

ROV STATUS

L8

8
L8
L8
L8
L8

SUVVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000

18987.0000000

0.0000000
0.0000000
0.0000000
0.0000000

27

OPPORTUNITY

COST
“5.3300000
0.0000000
“1.2200000
“12.2200000
“3.2200000
“13.3700000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

LOVER

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
INF
INF
INF
IVF
INF
INF
IVF
IVF
IVF
INF
IVF
IVF
INF
INF
INF
INF
INF
INF
IVF
INF
INF
INF
INF
IVF
IVF
IVF

UPPER
ROUND
INF
INF
IVF
INF
IVF
INF

02124l82

.19.07.38.

PAGE 20

267 1267
268 1268
269 1269
270 1270
271 1271
272 1272
273 1273
274 1274
275 1275
276 1276
277 1277
278 1278
279 1279
280 1280
281 1281
282 1282
283 1283
284 1284
285 1285
286 1286
287 1287
288 1288
289 1289
290 1290
291 1291
292 1292
293 1293
294 1294
295 1295
296 1296
297 1297
298 1298
299 1299
300 1300
301 1301
.302 1302
303 1303
304 1304
305 1305
306 1306
307 1307
308 1308
309 1309
310 1310
311 1311
312 1312
313 1313
314 1314
315 1315
316 1316
317 1317
318 1318
319 1319
320 1320
321 1321

L8
L8
L8
L8

L8
L8
L8
L8
L8

0.0000000
0.0000000
0.0000000
0.0000000

16185.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
435.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

22248.0000000

0.0000000
0.0000000
0.0000000
0.0000000

9797.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

13552.0000000

0.0000000
0.0000000

8283.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

16440.0000000

0.0000000
0.0000000
0.0000000
0.0000000

21427.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

28

“8.4900000
“2.8800000
“9.0100000
“6.0900000
0.0000000
“5.1800000
“5.7700000
“8.2100000
“8.8200000
“5.8300000
0.0000000
“5.7000000
“2.4300000
“2.7400000
“4.0700000
“13.2900000
“15.5400000
“10.1400000
“1.9900000
“10.3600000
0.0000000
“6.1600000
“11.6300000
“15.3600000
“10.5900000
0.0000000
“10.2600000
“3.6700000
“2.8600000
“13.9000000
“17.4000000
“12.3500000
0.0000000
“12.2300000
“2.8400000
0.0000000
“18.4100000
“20.6000000
“15.0100000
“4.8400000
“15.4800000
“4.1400000
“21.6500000
“8.4200000
0.0000000
“3.2500000
“15.3600000
“6.0600000
“16.1700000
0.0000000
“17.5500000
“21.4500000
“16.3700000
“2.3700000
“16.2800000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF
INF
INF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
INF
INF
INF
INF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
INF
INF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
INF
INF
INF
IVF
INF
IVF
IVF
IVF
INF
IVF
INF
INF
IVF

322 1322
323 1323
324 1324
325 1325

L8
L8
L8
L8

T VPOS VERSIOV 4.0

MVVVVVVAVVVVVVVVVV

i PROBLEV VUVRER 1 4
444414414444444441444

USING REGULAR

0.0000000
0.0000000
0.0000000
0.0000000

“6.0900000
“.7600000
“15.0100000
“18.9300000

VORTVVESTERN UVIVERSITY

II82.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

VAR VAR

NO NANE
326 1326
327 1327
328 1328
329 1329
330 1330
331 1331
332 1332
333 1333
334 1334
335 1335
336 1336
337 1337
338 1338
339 1339
340 1340
341 1341
342 1342
343 1343
344 1344
345 1345
346 1346
347 1347
348 1348
349 1349
350 1350
351 1351
352 1352
353 1353
354 1354
355 1355
356 1356
357 1357
358 1358
359 1359
360 1360
361 1361
362 1362

ROV STATUS

L8

8
L8
L8
L8
L8
L8
L8

8
L8
L8
L8
L8

8
L8
L8
L8
L8
L8
L8
L8

ru—
U

E; E; 5; an E; G; S; 8; an E; 53 S; an E; S;

SUNVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000

4751.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

11749.0000000

0.0000000
0. 0000000
0.0000000
0.0000000

21102.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

16008.0000000

0.0000000
0.0000000
0.0000000

8807.0000000

0.0000000
0.0000000
0.0000000
0.0000000

16443.0000000

0.0000000
0.0000000
0.0000000

29

OPPORTUNITY

COST

“13.9300000
0.0000000
“13.7700000
“4.0900000
“3.5800000
“15.2700000
“19.6700000
“14.9200000
0.0000000
“14.5900000
“5.9500000
“20.6600000
“6.2100000
0.0000000
“3.1400000
“13.9600000
“5.2200000
“15.3500000
“3.8100000
“13.3800000
“15.6600000
“10.2600000
“1.8900000
“10.4800000
0.0000000
“16.7300000
“9.7200000
“5.8400000
0.0000000
“11.7100000
“3.0700000
“11.0700000
“10.0400000
0.0000000
“6.1200000
“3.7700000
“1.8900000

0.0000
0.0000
0.0000
0.0000

LOVER

ROUVD
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF
IVF
INF
IVF

UPPER
ROUND
INF
IVF
INF
INF
IVF
IVF
INF
IVF
IVF
INF
INF
IVF
INF
IVF
INF
IVF
IVF
INF
INF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
INF
IVF
IVF
INF
INF
INF
IVF

02124782

.19.07.38. PAGE

21

363 1363 -“ L8
364 1364 ““ L8
365 1365 ““ 8
366 1366 ““ L8
367 1367 ““ L8
368 1368 ““ L8
369 1369 ““ L8
370 1370 ““ L8
371 1371 ““ L8
372 1372 ““ L8
373 1373 ““ L8
374 1374 ““ L8
375 1375 ““ L8
376 1376 ““ R
377 1377 “- L8
378 1378 ““ L8
379 1379 ““ L8
380 1380 ““ L8
381 1381 ““ L8
382 1382 ““ L8
383 1383 ““ R
384 1384 ““ L8
385 1385 ““ L8
386 1386 ““ L8
387 1387 ““ 8
388 1388 ““ L8
389 1389 ““ L8
390 1390 ““ LR
T NPOS VERSION 4.0

444444544144444444444

i PRORLEV VUVRER 1 4

444440444445141454444

USING REGULAR

0.0000000
0.0000000

18848.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

2221.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

7238.0000000

0.0000000
0.0000000
0.0000000

8816.0000000

0.0000000
0.0000000
0.0000000

“2.8800000
“6.4200000
0.0000000
“13.8300000
“21.3600000
“16.5000000
“1.4800000
“16.2600000
“6.5900000
“7.3800000
“3.5600000
“7.7000000
“3.7800000
0.0000000
“3.0100000
“3.4200000
“5.6500000
“14.5800000
“20.0300000
“15.8200000
0.0000000
“15.2300000
“8.2200000
“11.3100000
0.0000000
“6.6200000
“4.6800000
“3.0500000

VORTVVESTERN UVIVERSITY

IIR2.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

VAR VAR
NO VAVE
391 1391
392 1392
393 1393
394 1394
395 1395
396 1396
397 1397
398 1398
399 1399
400 1400
401 1401
402 1402
403 1403

ROV STATUS

E; E; £3 E; 53 83 an EELS; E; 83 £3 53

SUVVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

9598.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

30

OPPORTUNITY
COST

“3.8200000
“7.7800000
“7.4700000
“8.2000000
“15.3300000
“12.6400000
0.0000000
“11.7400000
“8.7600000
“5.6600000
“8.2200000
“7.6100000
“1.7900000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

LOVER

ROUND
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

INF
IVF
INF
INF
IVF
IVF
IVF
IVF
INF
INF
INF
INF
IVF
INF
INF
INF
INF
INF
INF
IVF
INF
IVF
IVF
IVF
IVF
INF
INF
IVF

UPPER

IVF
IVF
INF
IVF
IVF
IVF
IVF
INF
INF
IVF
INF
INF
IVF

02l24l82

.19.07.38. PAGE 22

404 1404
405 1405
406 1406
407 1407
408 1408
409 1409
410 1410
411 1411
412 1412
413 1413
414 1414
415 1415
416 1416
417 1417
418 1418
419 1419
420 1420
421 1421
422 1422
423 1423
424 1424
425 1425
426 1426
427 1427
428 1428
429 1429
430 1430
431 1431
432 1432
433 1433
434 1434
435 1435
436 1436
437 1437
438 1438
439 1439
440 1440
441 1441
442 1442
443 1443
444 1444
445 1445
446 1446
447 1447
448 1448
449 1449
450 1450
451 1451
452 1452
453 1453
454 1454
455 1455

L8
L8

L8
L8
L8
L8

$55554.

aSGSGSQGGaGSSEuSGSQEGQSuS

55555

8855';ng

L8
L8

1 14405 VERSION 4.0

“VVHVHVVHVVVVVVVV

0.0000000
0.0000000

13503.0000000

0.0000000
0.0000000
0.0000000
0.0000000
932.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

12870.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

17890.0000000

0.0000000
0.0000000
0.0000000
0.0000000
5760.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

15012.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

21029.0000000

0.0000000
0.0000000
0.0000000
0.0000000

15967.0000000

0.0000000
0.0000000

“2.8300000
“2.9400000
0.0000000
“7.8300000
“6.3900000
“11.0700000
“7.1100000
0.0000000
“6.3800000
“4.6900000
“8.4100000
“13.6300000
“19.0700000
“15.0200000
0.0000000
“14.3700000
“8.8000000
“5.2600000
“14.4300000
“15.6300000
“9.9700000
“4.3600000
“10.5900000
0.0000000
“4.4800000
“9.6600000
“12.2400000
“7.0900000
0.0000000
“7.0500000
“.0800000
“5.3300000
“9.1500000
“7.7300000
“1.8400000
“3.5400000
“3.4800000
0.0000000
“1.6600000
“14.6600000
“15.2700000
“9.4500000
“3.7800000
“10.4400000
0.0000000
“7.9200000
“.8100000
“6.0200000
“2.9100000
0.0000000
“2.0000000
“4.2100000

VORTVVESTERN UNIVERSITY

31

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
IVF
INF
INF
IVF
IVF
INF
IVF
INF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
INF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
INF
INF
INF
INF
INF
INF
INF
INF

02/24/82

.19.07.38.

PAGE

23

! PROBLEH VUHRER 1 i
444444444444411444444

USING REGULAR

1182.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

VAR VAR

NO VANE
456 1456
457 1457
458 1458
459 1459
460 1460
461 1461
462 1462
463 1463
464 1464
465 1465
466 1466
467 1467
468 1468
469 1469
470 1470
471 1471
472 1472
473 1473
474 1474
475 1475
476 1476
477 1477
478 1478
479 1479
480 1480
481 1481
482 1482
483 1483
484 1484
485 1485
486 1486
487 1487
488 1488
489 1489
490 1490
491 1491
492 1492
493 1493
494 1494
495 1495
496 1496
497 1497
498 1498
499 1499

ROV STATUS

L8
L8
L8
L8

8
L8
L8
L8
L8
L8

8
L8
L8
L8
L8

uSSSSuSSSSSwSSSQES555588555qu

SUVVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000
0.0000000
0.0000000
0.0000000

1307.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

10946.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
3089.0000000
13906.0000000
0.0000000
0.0000000
0.0000000
21268.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
14334.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
19193.0000000
0.0000000
0.0000000
0.0000000
0.0000000
3663.0000000

32

OPPORTUNITY
COST

“3.0600000
“15.0500000
“20.2200000
“15.7600000
0.0000000
“15.2900000
“7.1500000
“16.8400000
“7.2500000
“4.0600000
0.0000000
“11.0000000
“1.3700000
“11.3300000
“17.3000000
“4.4700000
0.0000000
0.0000000
“10.8600000
“1.7800000
“11.9300000
0.0000000
“22.0500000
“25.1700000
“19.7700000
“7.2500000
“19.9800000
“8.9100000
“4.5100000
“14.9900000
“19.2100000
“14.4300000
0.0000000
“14.1200000
“5.6600000
“6.5700000
“8.2800000
“5.8800000
0.0000000
“4.2900000
“2.1100000
“1.1400000
“17.4400000
0.0000000

LOVER

ROUND
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

UPPER
ROUVD
IVF
INF
IVF
IVF
IVF
INF
IVF
IVF
IVF
IVF
IVF
IVF
INF
INF
IVF
INF
IVF
IVF
IVF
INF
INF
INF
IVF
IVF
IVF
IVF
IVF
INF
INF
INF
IVF
INF
INF
INF
INF
INF
IVF
IVF
INF
IVF
IVF
INF
INF
IVF

500 1500
501 1501
502 1502
503 1503
504 1504
505 1505
506 1506
507 1507
508 1508
509 1509
510 1510
511 1511
512 1512
513 1513
514 1514
515 1515
516 1516
517 1517
518 1518
519 1519
520 1520

T VPOS VERSION 4.0

{HAVEHRHVVAVVV‘EVV

! PROBLEM NUMBER 1 i
141444414444144414444

USING REGULAR

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

19091.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

12139.0000000

0.0000000
0.0000000

“2.9700000
“4.1300000
“9.6700000
“4.3800000
“12.9100000
0.0000000
“18.0200000
“21.8700000
“16.7400000
“2.8700000
“16.6900000
“6.4000000
0.0000000
“14.4500000
“16.4700000
“10.9000000
“1.8100000
“11.3500000
0.0000000
“1.6500000
“14.8300000

VORTVVESTERN UVIVERSITY

IIR2.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

VAR VAR

NO NAVE
521 1521
522 1522
523 1523
524 1524
525 1525
526 1526
527 1527
528 1528
529 1529
530 1530
531 1531
532 1532
533 1533
534 1534
535 1535
536 1536
537 1537
538 1538
539 1539
540 1540

ROV STATUS

L8
L8
L8
L8

8
L8
L8
L8
L8
L8
L8

8
L8
L8

8
L8
L8
L8
L8

8

SUVVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000
0.0000000
0.0000000
0.0000000

18590.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
.0.0000000
3541.0000000
0.0000000
0.0000000
9201.0000000
0.0000000
0.0000000
0.0000000
0.0000000
2187.0000000

33

OPPORTUNITY
COST

“16.3300000
“10.6700000
“3.0600000
“11.2800000
0.0000000
“.7900000
“14.8200000
“16.2900000
“10.6100000
“2.8600000
“11.2600000
0.0000000
“19.1800000
“6.0500000
0.0000000
“1.3400000
“12.7800000
“3.6100000
“13.7700000
0.0000000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

LOVER

ROUVD
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF
INF
INF
INF
IVF
INF
INF
INF
INF
IVF
INF
INF
IVF
IVF
INF
IVF
IVF
IVF
INF
INF
IVF

UPPER
ROUVD
IVF
IVF
INF
INF
IVF
INF
INF
IVF
INF
INF
INF
INF
IVF
IVF
IVF
IVF
IVF
INF
IVF
IVF

02I24I82

.19.07.38. PAGE 24

541 1541 ““ L8
542 1542 ““ L8
543 1543 ““ L8
544 1544 ““ L8
545 1545 ““ L8
546 1546 ““ L8
547 1547 ““ 8
548 1548 ““ L8
549 1549 ““ L8
550 1550 ““ L8
551 1551 ““ LR
552 1552 ““ L8
553 1553 ““ L8
554 1554 ““ L8
555 1555 ““ R
556 1556 ““ L8
557 1557 ““ LR
558 1558 ““ L8
559 1559 ““ L8
560 1560 ““ L8
561 1561 ““ L8
562 1562 ““ L8
563 1563 ““ L8
564 1564 ““ L8
565 1565 ““ R
566 1566 ““ L8
567 1567 ““ L8
568 1568 ““ L8
569 1569 ““ L8
570 1570 ““ L8
571 1571 ““ L8
572 1572 ““ L8
573 1573 ““ L8
574 1574 ““ R
575 1575 ““ L8
576 1576 ““ L8
577 1577 ““ L8
578 1578 ““ L8
579 1579 ““ L8
580 1580 ““ L8
581 1581 ““ R
582 1582 ““ L8
583 1583 ““ L8
584 1584 ““ L8
585 1585 ““ L8
7 VPOS VERSION 4.0
454440445114541444441
* PRORLEV NUNRER 1 *
141444544414444441444
USING REGULAR

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

2757.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

8369.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

11971.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

17547.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

20038.0000000

0.0000000
0.0000000
0.0000000
0.0000000

“21.7000000
“23.6800000
“17.9500000
“7.4500000
“18.6500000
“7.3100000
0.0000000
“17.7800000
“19.3100000
“13.5800000
“4.8900000
“14.3000000
“3.0400000
“9.6000000
0.0000000
“5.1100000
“2.3900000
“1.8200000
“1.4800000
“5.5400000
“3.6500000
“14.2200000
“17.9800000
“13.0200000
0.0000000
“12.8300000
“3.8600000
“5.7000000
“9.6400000
“9.9900000
“4.2900000
“2.8500000
“5.0400000
0.0000000
“5.7000000
“10.5500000
“11.2700000
“5.6200000
“2.9600000
“6.2500000
0.0000000
“8.2100000
“8.0500000
“12.8900000
“9.1500000

VORTVVESTERN UVIVERSITY

I182.1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.

34

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

INF
INF
INF
INF
IVF
INF
IVF
INF
INF
INF
IVF
INF
IVF
IVF
INF
INF
IVF
IVF
INF
INF
INF
IVF
INF
IVF
INF
INF
IVF
IVF
INF
IVF
IVF
IVF
IVF
INF
INF
IVF
IVF
INF
IVF
INF
INF
IVF
IVF
IVF
IVF

02/24782

.19.07.38.

PAGE

25

 

 

VAR VAR

NO NAVE
586 1586
587 1587
588 1588
589 1589
590 1590
591 1591
592 1592
593 1593
594 1594
595 1595
596 1596
597 1597
598 1598
599 1599
600 1600
601 1601
602 1602
603 1603
604 1604
605 1605
606 1606
607 1607
608 1608
609 1609

610 1610 .

611 1611
612 1612
613 1613
614 1614
615 1615
616 1616
617 1617
618 1618
619 1619
620 1620
621 1621
622 1622
623 1623
624 1624
625 1625
626 1626
627 1627
628 1628
629 1629
630 1630
631 1631
632 1632
633 1633
634 1634
635 1635
636 1636

ROV STATUS

NO

8
L8
L8
L8
L8
L8

8
L8
L8
L8
L8

5; an 53 5; an 83 £3 53 8; £3 $3 an £3 £3 £3 £3 £3 5; an 53 E; £3 £3 £3 £3 £3 E; 83 83 In E; E; on E; C; an

E; E; E; E;

SUVNARY OF RESULTS

ACTIVITY
LEVEL
14221.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
10957.0000000
0.0000000
0.0000000
0.0000000
0.0000000
5792.0000000
0.0000000
0.0000000
4882.0000000
0.0000000
0.0000000
18364.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
18469.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
10060.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
12269.0000000
0.0000000
0.0000000
8883.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

35

OPPORTUNITY

COST

0.0000000
“8.4200000
“5.7600000
“16.7800000
“8.6700000
“5.2800000
0.0000000
“11.3500000
“2.1200000
“11.1800000
“8.5300000
0.0000000
“6.7500000
“4.4800000
0.0000000
“3.5700000
“5.6800000
0.0000000
“18.6400000
“20.7200000
“15.1000000
“5.1400000
“15.6200000
“4.2600000
“8.6300000
“13.3800000
“18.7600000
“14.7000000
0.0000000
“14.0500000
“8.6700000
“2.1600000
“14.7200000
“20.1800000
“15.8300000
0.0000000
“15.3200000
“7.2000000
“4.1100000
“14.2100000
“19.9300000
“15.8100000
0.0000000
“15.2000000
“7.9700000
0.0000000
“14.4500000
“16.4700000
“10.9000000
“1.8100000
“11.3500000

—.

LOVER

ROUVD
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

UPPER
ROUND
INF
INF
INF
INF
IVF
INF
INF
INF
IVF
IVF
IVF
IVF
INF
IVF
INF
INF
IVF
INF
INF
IVF
IVF
INF
INF
IVF
INF
IVF
IVF
INF
INF
INF
IVF
INF
INF
INF
IVF
IVF
INF
INF
INF
INF
IVF
INF
IVF
INF
IVF
IVF
IVF
INF
INF
INF
INF

 

T

637 1637 ““ R
638 1638 ““ L8
639 1639 ““ L8
640 1640 ““ L8
641 1641 ““ L8
642 1642 ““ R
643 1643 ““ L8
644 1644 ““ L8
645 1645 ““ L8
646 1646 ““ L8
647 1647 ““ L8
648 1648 ““ L8
649 1649 ““ R
650 1650 ““ LR

VPOS VERSION 4.0
444444444444444444444
! PROGLEV VUVRER 1 4
444444444444444444444
USING REGULAR

7936.0000000

0.0000000
0.0000000
0.0000000
0.0000000

12055.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

1720.0000000

0.0000000

0.0000000
“7.1900000
“3.3500000
“7.1800000
“3.1800000

0.0000000
“2.4100000
“3.0200000
“2.8400000

“15.3500000
“19.9400000
“15.2400000

0.0000000
“14.8900000

VORTVVESTERN UNIVERSITY

II82.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

VAR VAR
NO NAVE
651 1651
652 1652
653 1653
I 654 1654
655 1655
656 1656
657 1657
658 1658
659 1659
660 1660
661 1661
662 1662
663 1663
664 1664
665 1665
666 1666
667 1667
668 1668
669 1669
670 1670
671 1671
672 1672
673 1673
674 1674
675 1675
676 1676
677 1677

ROV STATUS

£3 C; an S; E; E; E; C; an £3 £3 £3 E; S; S; C; G; E; G; S; an E; 8;

L8
L8

L8

SUNVARY OF RESULTS

ACTIVITY

LEVEL
0.0000000
0.0000000

5080.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

2197.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

13256.0000000

0.0000000
0.0000000
0.0000000
0.0000000
893.0000000
0.0000000

36

OPPORTUNITY

COST

“6.1900000
“15.0400000
0.0000000
“1.3400000
“1.0900000
“7.7600000
“1.2100000
“10.1800000
“5.2200000
“6.0400000
“7.7900000
“2.6100000
“.2000000
“2.6000000
0.0000000
“11.2200000
“2.4700000
“4.5000000
“.9100000
“4.4600000
0.0000000
“6.2700000
“17.9100000
“7.3600000
“2.8800000
0.0000000
“12.0100000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

LOVER

ROUVD
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

IVF-
INF
IVF
INF
INF
IVF
INF
INF
INF
INF
INF
INF
INF
IVF

UPPER
ROUND
INF
IVF
INF
IVF

’ INF

INF
IVF
IVF
INF
IVF
INF
IVF
INF
INF
INF
INF
INF
INF
IVF
INF
INF
INF
IVF
INF
INF
INF
INF

02/24/82

.19.07.38.

PAGE

26

 

678 1678 ““ L8
679 1679 ““ L8
680 1680 ““ L8
681 1681 ““ L8
682 1682 ““ L8
683 1683 ““ L8
684 1684 ““ L8
685 1685 ““ L8
686 1686 ““ R
687 1687 ““ L8
688 1688 ““ L8
689 1689 ““ L8
690 1690 ““ L8
691 1691 ““ L8
692 1692 ““ L8
693 1693 ““ R
694 ARTIF““ D“ 1 L8
695 ARTIF““ D“ 2 L8
696 ARTIF““ D“ 3 L8
697 ARTIF““ D“ 4 L8
698 ARTIF““ D“ 5 L8
699 ARTIF““ D“ 6 L8
700 ARTIF““ D“ 7 L8
701 ARTIF““ D“ 8 L8
702 ARTIF““ D“ 9 L8
‘703 ARTIF““ D“ 10 L8
704 ARTIF““ D“ 11 L8
705 ARTIF““ D“ 12 L8
706 ARTIF““ D“ 13 L8
707 ARTIF““ D“ 14 L8
708 ARTIF““ D“ 15 L8
709 ARTIF““ D“ 16 L8
710 ARTIF““ D“ 17 L8
711 ARTIF““ D“ 18 L8
712 ARTIF““ D“ 19 L8
713 ARTIF““ D“ 20 L8
714 ARTIF““ D“ 21 L8
715 ARTIF““ D“ 22 LR

T VPOS VERSION 4.0
444444444444444444444
4 PRORLEV NUVRER 1 4
{VVVVNVIVI‘I‘DMVNVI
USING REGULAR

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

6326.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

368.0000000

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

“2.3900000
“12.3900000
“5.2700000
“9.1000000
“7.5100000
“1.6100000
“3.5800000
“3.3500000
0.0000000
“5.5700000
“10.7600000
“10.8000000
“5.0200000
“3.5200000
“5.9800000
0.0000000
1.2000000
2.1000000
2.3400000
1.8100000
5.5300000
.0600000
5.1500000
2.9600000
2.6800000
1.2500000
“.6400000
1.3500000
2.2800000
.5000000
3.7500000
6.0300000
2.7900000
3.6000000
“.7600000
3.8300000
5.0900000
2.1600000

VORTVVESTERN UNIVERSITY

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

1182.1982.99 ORIGINS,693 ELE. V.E. PLANT AT ROSE CITY.

SUVNARY OF RESULTS

VAR VAR ROV STATUS ACTIVITY OPPORTUNITY LOVER
NO VANE NO LEVEL COST ROUND
716 ARTIF““ D“ 23 LR 0.0000000 5.5900000 0.0000
717 ARTIF““ D“ 24 LR 0.0000000 2.1700000 0.0000
718 ARTIF““ D“ 25 LR 0.0000000 5.4700000 0.0000

37

INF
INF
INF
IVF
IVF
INF
INF
INF
INF
INF
INF
INF
IVF
INF
INF
INF
IVF
INF
INF
INF
INF
IVF
INF
INF
INF
INF
INF
INF
INF
INF
IVF
INF
INF
INF
IVF
INF
INF
INF

UPPER
ROUND
INF
IVF
IVF

02I24IR2 .19.07.38. PAGE 27

719 ARIIF““ D“
720 ARTIF““ D“
721 ARTIF““ D“
722 ARTIF““ D“
723 ARTIF““ D“
724 ARTIF““ D“
725 ARTIF““ D“
726 ARTIF““ D“
727 ARTIF““ D“
728 ARTIF““ D“
729 ARTIF““ D“
730 ARTIF““ D“
731 ARTIF““ D“
732 ARTIF““ D“
733 ARTIF““ D“
734 ARTIF““ D“
735 ARTIF““ D“
736 ARTIF““ D“
737 ARTIF““ D“
738 ARTIF““ D“
739 ARTIF““ D“
740 ARTIF““ D“
741 ARTIF““ D“
742 ARTIF““ D“
743 ARTIF““ D“
744 ARTIF““ D“
745 ARTIF““ D“
746 ARTIF““ D“
747 ARTIF““ D“
748 ARTIF““ D“
749 ARTIF““ D“
750 ARTIF““ D“
751 ARTIF““ D“
752 ARTIF““ D“
753 ARTIF““ D“
754 ARTIF““ D“
755 ARTIF““ D“
756 ARTIF““ D“
757 ARTIF““ D“
758 ARTIF““ D“
759 ARTIF““ D“
760 ARTIF““ D“
761 ARTIF““ D“
762 ARTIF““ D“
763 ARTIF““ D“
764 ARTIF““ D“
765 ARTIF““ D“
766 ARTIF““ D“
767 ARTIF““ D“
768 ARTIF““ D“
769 ARTIF““ D“
770 ARTIF““ D“
771 ARTIF““ D“
772 ARTIF““ D“
773 ARTIF““ D“

26
27
28

30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56

58
59
60
61
62
63
64
65
66
67

68 -

69
70
71
72

74
75
76
77
78
79
80

0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000
0.0000000

38

5.0100000
5.2500000
3.4900000
2.7600000
2.7400000
1.7600000

.5600000
3.2400000
1.2500000

.6000000
3.8000000
3.7700000
1.7800000
4.0600000
5.1600000
1.1900000

.5800000
1.0900000
1.0700000

.1900000
1.1900000
1.7600000

.4100000
1.1400000
1.2600000
“.0800000
4.4900000
2.8400000
4.0300000

.7300000
3.9300000
5.4600000
4.7500000
2.3700000
“.8300000

.7700000
2.6600000
5.3700000
3.5800000
5.1100000
2.0000000

.5400000
2.4400000
“.9400000
“.1100000
5.8500000
2.9400000

.9500000
2.9200000
2.3500000
2.9000000
1.4800000
4.3900000
2.4600000
4.6900000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

INF
INF
INF
INF
IVF
INF
INF
IVF
IVF
INF
INF
INF
INF
IVF
IVF
IVF
INF
INF
IVF
IVF
IVF
INF
INF
IVF
INF
INF
IVF
IVF
IVF
INF
INF
INF
INF
IVF
INF
INF
INF
IVF
IVF
INF
IVF
IVF
IVF
INF
INF
INF
INF
INF
IVF
INF
INF
INF
IVF
INF
INF

 

774 ARTIF““ D“ 81 LR 0.0000000 .5500000 0.0000 INF
775 ARTIF““ D“ 82 LR 0.0000000 3.3500000 0.0000 INF
776 ARTIF““ D“ 83 LR 0.0000000 2.6600000 0.0000 INF
777 ARTIF““ D“ 84 LR 0.0000000 1.4300000 0.0000 INF
778 ARTIF““ D“ 85 LR 0.0000000 .0300000 0.0000 INF
779 ARTIF““ D“ 86 LR 0.0000000 5.6300000 0.0000 IVF
780 ARTIF““ D“ 87 LR 0.0000000 1.1900000 0.0000 IVF
T VPOS VERSION 4.0 VORTVVESTERN UNIVERSITY 02l24/82 .19.07.38. PAGE 28
444444444444444444444

4 PRORLEV NUVRER 1 4

444444444444444444444

USING REGULAR

IIR2.1982.99 ORIGINS,693 ELE. V.E. PLAVT AT ROSE CITY.

SUVVARY OF RESULTS

VAR VAR ROV STATUS ACTIVITY OPPORTUNITY LOVER " UPPER

NO NANE NO LEVEL COST ROUVD ROUVD
781 ARTIF““ D“ 88 LR 0.0000000 “.9300000 0.0000 INF
782 ARTIF““ D“ 89 LR 0.0000000 3.7600000 0.0000 INF
783 ARTIF““ D“ 90 LR 0.0000000 2.7600000 0.0000 INF
784 ARTIF““ D“ 91 LR 0.0000000 2.9200000 0.0000 INF
785 ARTIF““ D“ 92 LR 0.0000000 4.2800000 0.0000 INF
786 ARTIF““ D“ 93 LR 0.0000000 1.0400000 0.0000 IVF
7R7 ARTIF““ D“ 94 LR 0.0000000 3.7200000 0.0000 IVF
788 ARTIF““ D“ 95 LR 0.0000000 4.4900000 0.0000 INF
7R9 ARTIF““ D“ 96 LR 0.0000000 3.3900000 0.0000 INF
790 ARTIF““ D“ 97 LR 0.0000000 .6000000 0.0000 IVF
791 ARTIF““ D“ 98 L8 0.0000000 5.6200000 0.0000 INF
792 ARTIF““ D“ 99 LR 0.0000000 3.1900000 0.0000 IVF
793 ARTIF““ D“ 100 L8 0.0000000 1.6000000 0.0000 INF
794 ARTIF““ D“ 101 L8 0.0000000 “.4400000 0.0000 INF
795 ARTIF““ D“ 102 L8 0.0000000 1.2500000 0.0000 IVF
796 ARTIF““ D“ 103 L8 0.0000000 11.1300000 0.0000 IVF
797 ARTIF““ D“ 104 L8 0.0000000 2.2400000 0.0000 INF
798 ARTIF““ D“ 105 L8 0.0000000 “.6000000 0.0000 INF
905 ARTIF““ D“ 106 8 0.0000000 0.0000000 0.0000 INF

VINIVUN VALUE OF THE OBJECTIVE FUVCTION = 4950171.049989

CALCULATION TIME VAS 9.3640 SECONDS FOR 538 ITERATIONS.

DATA STORAGE NEVORY =254054(OCTAL1 TOTAL VENORY 8 306IOOIOCTALI
2TOTAL TIVE FOR THIS PRORLEV VAS 12.261 SECONDS

 

39

REFERENCES

 

Bechtel National, Inc. 1981. Enyizgnmgntal an1.££8n2n1£
88411112212811 8228.18.28.81me 8f.

8881. nnnhinmm mammal. Bechtel. Inc-:
San Francisco, Calif.

Boelter, A.H. 1977. 111.811.1281: mummmd
using companies. Lansing, Michigan: Michigan

Department Of Natural Resources.

Cohen, C. and Stein, J. 1978. H3111 nunnn:n.22£1m115£121

81.41.13.811 2 1.18.828 11111.18 1.82.11.8n £1.20. Evanston.Ill-=
Northwestern University Computer Center.

Dantzig, 6.3. 1963. Linna2.n2nznnmnina and nxtnnninns.

Princeton, N.J.: Princeton University Press.

Darrell, J.A. 1982. mummmfimm.

Presented at American Bar Association Annual
Meeting, 1982. San Francisco, Calif.: American Bar
Association.

Daverman Associates, Inc. 1979. A,feasihili&1 £1381 £22 a
188.22.118.81 81.88.8218 meaning nlann. Grand Rapids.

Michigan.: Daverman Associates,Inc.

DenUyl. 11.3.. 1982. 5288.1. 1.81282 shinning. W
and mm 282 minhinanis 2.82.881. 223.411.21.11. 3
nuliminnu ennnnmin 1min. MICHU-SG-BZ-SOO- East
Lansing, Michgan. Michigan State University
Cooperative Extension Service,

Fisher, J.M. 1980. Allocating decisionmaking in the field
of energy resource development: some questions and

suggestions. Anizgna.Lax,ngxigw. 22:785-870. Tucson
Arizona: College Of Law, University Of Arizona. -

Flick, A.A. 1976. Pulping industry experience with control
of flue gas emissions from bark and wood fired
boilers. Win28... 1:82.881. 228.811.81.11 11.88.882.811
3981811,V212.lfllfi. Forest Products Research Society:
Atlanta, Ga.

GAO (Government Accounting Office). 1982. A, market

8231:. (PAD-BZ-ISA). General Accounting Office:
Gaithersburg,Md. '

Hinshaw. P.O. 1982. 8.881.881.28811818882881888818888888
Wazammmmmm
WWWHWMflm.
Masters Thesis, Michigan State University: East

Lansing, Michigan.

Hisner, M.L.,ed.1977. Elnsfisfllnznwnl £21££n22.811l18n 88

W W 88 1188.8 9.11.91.81- Lansing.
Michigan: MIchigan State Public Service Commission.

Hudson, D.H. and Kittleson, K.M. 1978. ldtntititgtitn 8i

8288:888281 288282888 in finntznl. Mi8nisan» East
Lansing, Michigan.: Michigan State University Center

For Remote Sensing.

James, L.M, Heinen, S.E., Olson, D.D. and Chappelle, D.E.
.1982. 1.1.8882 828.11.118.88 8.8888101 1411 14118111388. Research
Report 446, Michigan State University Agricultural
Experiment Station, East Lansing. Michigan State
University: East Lansing, Michigan.

Johnson. 8. 1983. £88828 .1888 1481.1. in 88888.18 14488.88. Wed.
March 23,1983. p. B1. Detroit Free Press:Detroit,
Michigan.

Johnson, R.C. 1976. Reduction of emissions from bark fired
furnaces by good fuel management. 828888818288
28288.8?2888888118888881158818811188. 1.918... 1182.888
WWWWB...

MBAmaimmmnWW

18.182m8nt,h82818888, Feb.1983, Journal Of Forestry.
Soceity Of American Foresters: Washington, D.C.

Kittleson. K.H- 1979. A. 8188888188 82 288888 £82888.
mm lit8288028 88888281888881.81 8.88.1888...
WMWWWW

888 nntnim 288.82.11.88. East Lansing.
Michigan.:(Mimeograph).

Lubin, B.L. and Pickholz, M.G. 1980. Introduction to the
powerplant and industrial fuel use act of 1978:
securing exemptions for utilities and major

industrial users. Anni-£3.11 W 1.8.11
£81i81,29:485-513 Spr.80. Washington, D.C.: American

University.

Lundgren, A.L. and Essex, B.L. 1979. A gtntnniiztn,£8288t.

1128111211 wmwmmmm
- . St. Paul, Minn.: U.S. Forest Service

North Central Forest Experiment Station.

McBurney, V.E. 1976. Controlling emissions on wood fired

industrial boilers. 228888818881 282888 22888888
B8888288.S888181 N811.1318. Forest Products Research
Soceity: Atlanta, Ga.

Michigan Department Of Natural Resources. 1981. Mithiggnit

888881188 {22281.288232881.21838.52218.1211.L128£Llo
Lansing, Michigan: Michigan Department Of Natural
Resources.

Michigan Legislature. 8888828181 E82888. R888218. A88.
Public Act No. on (1925).

Michigan Legislature.8888828181 £82888 3888218. A88
8888888881, Public Act No. 393 (1980).

Mitre Corporation. 1977. 3111188188281 8188888.£8288.1811
1. McLean, Virginia: Mitre Corporation.

Norris. F.N. 1981. 118882 8828.== 882881.818818881.181111.
N882. Highland, North Carolina: F.N. Norris.

Office Of Technology Assessment (OTA). 1980. Engtgz,£ngm

8181881881 828888888. Washington. D-C- 0.8.
Government Printing Office.

Roddis, L.H. 1981. 8188888.8882811 288828.8£.888.888281

28888288 88118821. 88828. 888881 88 8188888.
Nashington, D.C.: Department Of Energy.

Sokol, T. and Ralph, A. 1970. Bigmgtzx, New York: Wiley.

Stobaush. R. and Yersin. D. 1979. 888281.888828 -- 288828
88. 888 888281 8281888, 88, 888 8821828 88818888
338291,. New York: Ballantine. -.

Supreme Court Of 0.8... 2888281. £88281. 3888188821.

88881881881 88.811.11 818818818811.88 811. 102 S.
Ct. 2126, 1982.

Swanson. L.W. 1980. 118882 Rnflfllflnn1n£.::,h8£1£.£h£2£1 888
888118881888. New York: McGraw Hill, Inc.

U.S. Congress, £1888 Lin 888 (as ammended in 1977). #2
U.S. Code, Sections 1857 et seq.

0-8- Congress. 2818281888 888.1888882181 £881.A88. 88
1318 (PIFUA), hz U.S. Code, Sections 8301-8h83.

0.8- Congress. 22182181 888281. 2281888. A88 81 1888
(PEPA), s. 1308, 96th Cong., 1st Sess. (1979).

U.

S.

Congress. £88118 8811181888888188821888 £81181
A88 81‘. 1181(PURPA),Pub. L. No. 95-617. 92 Stat.

3117.

Walker, R.W. 1980. New frontiers of energy development and

conservation: the search continues and potential
liability increases. £8288 15:610-17 Spr. 80.
Section or Insurance Negligence and Compensation
Law, American Bar Association: Chicago, Illinois.