United States
Environmental Protection
Agency
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S2-83-121a,b,c Apr. 1984
Project Summary
Water Supply Simulation Model
Volumes I, II, and
Robert M. Clark, Richard Males, and Richard G. Stevie
This three-volume report describes
the development of a water supply
simulation model (WSSM), a system of
computer programs that allows for a
systematic evaluation of the physical
and economic characteristics of a water
distribution system in a spatial
framework.
The WSSM concept views a water
utility as a network overlaid upon a
spatial distribution of supply and
demand. The model explicitly deals
with the relationship of delivered water
costs to the service requirements of
spatially distributed demand. This spa-
tial representation is based on a charac-
terization of the water supply system as
a link-node network. Water is assumed
to enter and leave the system only at
nodes, which represent treatment
plants, junctions, demand locations and
storage tanks. Water is carried between
and among nodes through connecting
links. Costs are allocated to the various
facilities and system components based
on flow in the system.
Once an adequate hydraulic simula-
tion has been made, the model can be
used to determine costs, travel time,
and contaminant concentrations at
various points in the network. Results
from the model are particularly useful in
establishing the cost of service to
various spatially differentiated custom-
ers.
The model has been calibrated and
tested on several water supply systems,
including a small utility in New Vienna,
Ohio, and the Kenton County Water
District No. 1 in Kenton Couny, Ken-
tucky. Volume 1 of this study describes
the development of the model and its
underlying principals. Volume 2
discusses some of the engineering and
economic concepts used in developing
the model. Volume 3 is a users manual
for the model.
This Project Summary was developed
by EPA's Municipal Environmental
Research Laboratory. Cincinnati, OH,
to announce key findings of the
research project that is fully docu-
mented in three separate reports of the
same title (see Project Report ordering
information at back).
Introduction
Passage of the Safe Drinking Water Act
has intensified interest in problems
related to water supply and water utility
management. Economic analysis of the
regulations to be promulgated under the
Act indicates that some water utilities
may suffer adverse effects, which may be
most pronounced in small utilities. An
option often suggested for small systems
is to join with another larger system to
form a regional water supply utility. The
economies of scale associated with a
regional water system would supposedly
benefit small system consumers.
Many utilities find that a tradeoff exists
between the cost of building and opera-
ting facilities to meet demands for a
product and the cost of transportation.
High transportation costs and low facility
costs imply decentralization; the reverse
implies a few large, central facilities.
These factors must be considered in
planning, designing, constructing, and
operating water supply systems.
The water supply system can be
separated into two physical components:
(1) acquisition and treatment facilities
and (2) the delivery (transmission and
distribution) system. Each of these
components has a different cost
function. The unit costs associated with
treatment facilities are usually assumed
-------
to decrease as the quantity of service
provided increases. But the delivery
system is more directly affected by the
characteristics of the area being served.
The cost tradeoffs between the two
components will determine the cost of
delivering water to any portion of the
service area.
Few analytical instruments are availble
to study the economics of water supply
systems. The U.S. Environmental
Protection Agency (EPA) has therefore
initiated a program to develop techniques
and methods evaluating the regional
economics for water supply. A water
supply simulation model (WSSM) has
been designed to aid such an evaluation.
The model will also provide insights
into other water-related economic
issues such as spatial pricing and costing,
conservation policies, operating improve-
ments versus increased capital expendi-
ture, user class subsidization, and fire
protection capacity. In addition, the model
can also be used to analyze mixing
problems.
The WSSM incorporates a series of
submodels to describe the various
aspects of the economic, demographic,
and hydraulic systems that make up a
water utility. The logic used in developing
the model is discussed in the following
sections.
Model Structure
The WSSM is based on the concept of a
water utility as a network overlaid on a
spatial distribution of supply and demand.
The model explicitly deals with the
relationship of delivered water costs to
satisfy the service requirements of
spatially distributed demand. This spatial
representation is based on a characteri-
zation of the water supply system as a
link-node network. Water is assumed to
enter and leave the system only at nodes,
which represent treatment plants,
junctions, demand locations, and storage
tanks. Water is carried between and
among nodes through connecting links.
Costs are allocated to the various
facilities and system components based
on flow in the system.
The WSSM requires that the system be
described as a network of pipes, storage
tanks, treatment plants, demands, and
other hydraulic elements. Information
concerning the network is stored in a
network data base, which also stores
additional descriptive or calculated
information about each element (such as
size of pipe, geographic location of each
demand, population associated with a
demand, connectivity of pipes, etc.). Cert-
ain basic information must be stored for
the system to operate, but other informa-
tion is elective and is a function of the
particular uses and analysis to which the
WSSM is to be put. Other program
modules communicate with the data base
through standardized data base access
methods, which consist of routines to
extract or insert information into the
network data base.
Figure 1 illustrates the way in which
the WSSM operates. Data are entered
into the data base (link and node files)
through an establishment module. Once
the data base is established, various
program modules manipulate it. The
general elements of the WSSM are the
network data base, data base establish-
ment and editing modules, data base
access methods, hydraulic network
analysis models, other physical and
economic models, and display and
reporting modules.
The original concepts and approaches
of the model were tested in a pilot study of
the Cincinnati, Ohio, Water Works
system. A contour map of zonal costs for
delivery of water to various locations
within the service area (Figure 2) was
developed from a pilot version of the
WSSM. Further WSSM development was
encouraged by the ability of such displays
to synthesize easily the results of
complex physical, policy, and economic
situations. A revised, more general-
purpose version uses the New Vienna,
Ohio; Kenton County, Kentucky; and
Tampa, Florida; water systems as
testbeds. The New Vienna example is
described in detail in this report.
^
Link-Node
of Water Supply
System
File Establishment
Module
File Access Routines
\ T 4 1
1
1
]
1
f
Link
File
?
Node
File
•
1
1
1
J
I I
Display
Module
Lister
Module
Figure 1. Basic structure of the water supply simulation model.
-------
Zone Cost ($/1000 gal)"
1
2
3
4
5
6
7
*1973 Costs
.15-.20
.20 - .29
.29 - .38
.38-.47
.47 -.57
.57-.66
.66-.76
Y/7/A Not included in
service area
•- Treatment Plant
Figure 2. Cost contours for the Cincinnati Water Works area based on the WSSM. Source:
Goddard, H. C.. Stevie, R. G., and Trygg. G. D., "Planning Water Supply: Cost-Rate
Differentials and Plumbing Permits," EPA-600/5-78-008, U.S. Environmental
Protection Agency, Cincinnati. Ohio, 1978.
All software modules of the WSSM are
written in Fortran IV and adhere as
closely as possible to ANSI Standard
Fortran IV, thus increasing the ease with
which the model can be transported from
one computer to another. Hardware envi-
ronments may exist, however, in which
the system will require modifications.
This problem has been minimized and
localized by separating the file access
routines into a separate module and
preparing careful internal documentation
of those parts of the code that are more
susceptible to specific computer
dependency.
The seven software modules used in
the model include the establishment
module, the editing module, the display
module, the listing module, the hydraulic
analysis module, the system solver
module, and the (Input-Output) I/O
module. The establishment of a data base
for the WSSM requires the use of four of
these modules. The establishment
module prepares a list of possible errors
that the user may make while creating
the files; the input data are not corrected
while the files are being created. The
display and listing modules are then used
to verify the contents of the files, and
necessary corrections are made with the
editing module. Because the integrity of
the information is so important, all
modules except those for key establish-
ment and editing are designed to
preclude destruction or distortion of
these data.
Case Study
To display some of the features of the
WSSM and to serve as a small,
manageable system for testing various
WSSM elements, the water supply
system in the village of New Vienna,
Ohio, was selected as an example. New
Vienna is a village of approximately 1000
population (1980) located in Clinton
County in Southwestern Ohio. The water
supply system serves approximately 900
residents, with some 340 residential
meters. In addition, light industry,
laundromats, and schools are served
from the system. Average metered use in
the village is approximately 1.7 million
gal per month.
Water is supplied to the system from
two sources: a well field and treatment
plant operated by the Village, and
purchased water from Highland County.
Because the Village is required to
purchase a set amount of water each
month from Highland County, it operates
so as to purchase that amount and then
switches over to its local sources.
Typically, two-thirds of the water is
purchased each month. The two sources
do not operate simultaneously.
Development of the New Vienna Data
Base consisted of the following steps:
a. Delineation of the system in link-
node form,
b. Determination of physical system
characteristics (pipe size, diameter,
etc.) and spatial coordinates,
c. Development of demand data, and
d. Development of cost data.
Some interplay must occur among these
efforts to ensure that the link-node repre-
sentation does portray the important
changes in physical system character.
Delineation of System in
Link-Node Form
A link-node representation of the New
Vienna water system was laid out on a
gridded overlay to the 1 in. = 200 ft map.
Nodes were located in continuous
lengths of pipe, based on their serving as
centers of aggregation of demand. Nodes
were also located at pipe junctions and at
changes in pipe diameter or type. The
system was also laid out so that nodes
were located at major changes of pipe
direction, thus providing an accurate
geometric and topologic map. Each link
and node was numbered sequentially.
The New Vienna link-node representa-
tion consists of approximately 50 links
and 50 nodes (Figure 3).
Determination of Physical
System and Spatial
Characteristics
Pipe lengths, types, and diameters are
available from the base map. This
information was transferred to the link-
node system and associated with each
individual link number. Topological data
in the form of the upstream and down-
stream node numbers for each link were
also taken from the base map and
recorded. This information sets the
convention for flow throughout the
WSSM and provides topologic connectiv-
ity. By convention, flow in the pipe from
upstream to downstream node is positive,
and flow into a node is positive.
The coordinate locations of each node
were digitized through hand take-off from
the gridded overlay and recorded.
Elevations were obtained by placing the
overlay on top of the base map and
reading the elevation at each node from
-------
the contours on the base map. Contours
on the base map are at 3-ft intervals.
Data for both the links and nodes were
encoded and prepared as input to the file
establishment module of the WSSM.
Development of Demand Data
Metered information is availableforthe
majority of the Village, but demand data
were developed for other than industrial
users by performing house counts within
demand zones. These zones were drawn
on the link-node overlay for an arbitrary
association of demands with nodes. Each
node has a demand zone, and it is
assumed that any demands falling within
that area are aggregated to the node.
Development of Cost Data
At its current state of development, the
economic allocation procedures require
a single annual cost representing the
combination of amortized capital and
operating and maintenance (O&M) costs
for each node and link. For the case of
New Vienna, actual construction costs
were not available for most elements.
Capital costs in general were estimated
as current replacement costs and then
revised to the actual year of installation
through use of the three-digit
Engineering News Record Construction
Cost Index (CCI). The existing well and
water treatment system costs were
estimated with cost curves derived from
EPA reference material. Bid data for
replacement of the elevated storage tank
were used to estimate its replacement
cost. Pipe costs were developed based on
unit prices in a construction bid for the
area. All estimated costs were first
calculated on a 1981 basis before being
revised to year of installation. The base
year of the CCI is 1913, thus all system
elements known to be installed before
that date were treated as if they had been
installed in 1913.
Annual O&M costs are taken from
Village records. Total 1981 O&M cost,
including debt service and administrative
costs, was projected to be $46,721. Costs
were allocated as indicated above. Node
and link cost summaries were calculated
showing the 1981 construction cost, year
of construction, CCI factor, original
construction cost (computed), and 1981
O&M cost (Figure 4).
After the basic description of the link-
node network and its physical character-
istics were obtained, the data were
encoded. These data were then used as
input to the data base establishment
programs, ESTBLINK and ESTBNODE.
Figure 3. New Vienna link-node representation.
«
Figure 4. Cost contours (cents/'1000 gals.) for New Vienna (Scenario 1).
Program ERCHECK is run to test for
topologic errors. At this point, it is
frequently desirable to be able to visually
inspect the data base and to be able to
refer directly to the base map.
The full report was submitted in fulfill-
ment of Contract No. 68-03-2941 by W.
E. Gates and Associates, Inc., under the
sponsorship of the U.S. Environmental
Protection Agency.
-------
The EPA authors Robert M. Clark (also the EPA Project Officer, see below) and
Richard G. Stevie are with the Municipal Environmental Research Laboratory,
Cincinnati, OH 45268; Richard M. Males is with W. E. Gates and Associates.
Inc., Batavia, OH 45103.
The complete report consists of three volumes, entitled "Water Supply Simulation
Model," (Set Order No. PB 84-143 908; Cost: $32.00)
"Volume I. Model Development," (Order No. PB 84-143 916; Cost: $11.50
"Volume II. Literature Review and Background Research," (Order No. PB
84-143 924; Cost: $13.00)
"Volume III. Documentation,"(Order No. PB 84-143 932; Cost: $13.00)
The above reports will be available only from: (cost subject to change)
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officers can be contacted at:
Municipal Environmental Research Laboratory
U.S. Environmental Protection Agency
Cincinnati. OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES PAID
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
U.S. GOVERNMENT PRINTING OFFICE: 1984-759-102/910
------- |