United States
                   Environmental Protection
Water Engineering
Research Laboratory
Cincinnati OH 45268
                   Research and Development
EPA/600/S2-87/024  June 1987
SEPA          Project Summary
                   Water  Main  Evaluation for

                   Robert M. Clark, Martin Allen, and D. Kelly O'Day
                     This report describes a series of
                   studies  that  identified factors
                   contributing to water main failures
                   and developed decision  strategies
                   on main  rehabilitation, replacement,
                   and preventive maintenance.  The
                   objective of this report is to provide
                   the water utility  industry with
                   practical  methods to assess long-
                   term changes in the  condition of
                   distribution systems and to assist in
                   rehabilitation/replacement planning
                     Written as a guidance manual, the
                   report discusses  the  following
                   1. A comparison of  utility  repair
                     recordkeeping practices with
                     recommended procedures for
                     analysis of leak and break trends.
                   2. Description  of  structural
                     properties of water mains and the
                     role  external  environmental
                     conditions and  corrosion
                     processes play  in water main
                   3. Description of  techniques to
                     evaluate water main conditions
                     with a recommended  physical
                     testing program that determines
                     internal  and external  corrosion
                   4. Presentation of economic models
                     related to leakage  control, main
                     rehabilitation,   and   main
                   5. Description  of suggested water
                     distribution system  inventory
                     procedures including geographic
                     referencing  systems,  computer
                     mapping, and hardware/software
                   6. Discussion of  approaches to
                     monitor distribution system
                     conditions, to  determine  repair
   rates  and  patterns,  and  to
   coordinate the cost of monitoring
   with long-term benefits.
 7. Presentation of  independent
   planning steps  that  can  be
   adopted by all utilities depending
   on  system needs, budget, and
   personnel. The steps  include
   standardized    assessment
   programs, economic models  for
   decision    making,    and
   computerized rehabilitation/re-
   This  Project  Summary was
 developed by EPA's Water Engineering
 Research Laboratory,  Cincinnati, OH,
 to announce key  findings of the
 research project  that  is  fully
 documented in a separate  report of
 the same title (see  Project Report
 ordering information at back).


   In recent years, the condition of water
 distribution  systems  nationwide has
 attracted  a great deal of media, political,
 and research attention. A number of
 divergent views have resulted regarding
 the magnitude  of water distribution
 problems and the  level of capital
 expenditures  needed to address  these
   These conflicting viewpoints arise
 from the  lack  of extensive  research
 undertaken  in the  past  into the
 effectiveness  of water distribution system
 planning  and maintenance activities.
 Standard  analytical procedures have  not
 been developed by the water utility
 industry to assess distribution system
 conditions.  For this  reason,  no
 benchmark or set of  procedures has
 been established to compare distribution
 system conditions.

   Recently, several  institutions  or
organizations  have surveyed  water
utilities nationwide to determine the type
and magnitude  of existing infrastructure
problems. The  surveys, conducted by
the American Water Works  Association
(AWWA) and  CH2M Hill Consultants,
indicate that a significant variation exists
within utility infrastructure improvement
planning systems.  Different sizes  of
systems tend  to emphasize  different
types of program. The larger systems
tend to  be more sophisticated systems.
Availability of useful  information  is
indicative also of  differing  levels  of
planning and programming.
   Research efforts in recent years  have
focused primarily on methods to assess
current  conditions of water  distribution
systems, including realistic evaluations of
the structural,  hydraulic,  and  water
quality  problems encountered.  These
important research needs have not been
addressed until recently by the industry.
   Results from the surveys as well  as
other sources of  information  have
indicated a number  of different types of
research needs.

Research Needs
   The  development  of  a  standard
condition assessment terminology that
can be  utilized  by all water utilities is a
major   research   priority.    The
establishment   of  standardized
nomenclature would  enable water utilities
to  compare  capital  programming
approaches  and results.  This  standard
should include an accepted set of  terms
and definitions  as  well as  a consistent
method  to assess  current  and  future
performance  levels  of  distribution
systems. Variations  in  distribution
system  problems  by  region of  the
country  should   not   hinder  the
development  of  such  a standard.
Although the type of distribution system
affected ranges from older,  established
systems in  the midwest and northern
regions  to younger, growing systems in
the southern and western regions, water
quality, hydraulic,  and  structural
deterioration problems are not limited to
any one region.
   Research must also address the  need
for an  increased  understanding  of the
structural deterioration process. There is
a  particularly critical  need for  more
research  into  the  causes  of internal
corrosion.  Internal  corrosion  of  water
mains can  lead  to  several  problems,
including the loss of hydraulic capacity,
water discoloration, and  a decrease in
the effective main  wall thickness. The
impact  of  internal  corrosion on  water
quality has taken a  back seat in the
research community compared  with
other  drinking  water  quality research
   Although most water quality research
has focused  on  the use  of  various
disinfection treatment  processes  to
control and   mitigate  system-wide
bacterial growth, the effect of a localized
deterioration of the distribution grid  on
water  quality must also be recognized. A
research  need is to  draw correlations
between internal corrosion  levels and
water  quality  deterioration.   Future
research must  seek new techniques and
materials for  the  replacement,
rehabilitation, and repair of water mams.
More  economical approaches should  be
found  to counter inflationary trends and
to assist utilities in many sections of the
country  to  increase their maintenance
   Research is needed to develop  an
objective framework  for  water  utility
managers  to  better  compare  and
understand various strategies available to
them  for  distribution   system
maintenance.  This  report  primarily
addresses this critical research need.
   In  addition  to defining  the above
mentioned research  needs,  the study
reported in this  project  summary was
conducted with five  major  objectives:
analyze leak and break patterns from a
selected  set  of utilities;  examine the
major  causes of water main deterioration;
survey analytical  techniques used  by
utilities  to  evaluate  water  mam
conditions;  discuss  the  economies  of
leakage control, main rehabilitation, and
replacement;   develop  a   series  of
recommendations focusing on the types
of information  necessary  in  building  a
water  distribution system  inventory;
provide a series of recommendations  for
monitoring   distribution  system
conditions; and,  present a series  of
planning steps  for rehabilitation that may
be adopted by  all size utilities depending
on their  system  needs, budget, and

Analysis of  Utility Leak and
Break Patterns
   The study team analyzed  records on
broken  water  mains  for six  utilities:
Denver Water Department, East Bay
Municipal  District (California), Kenosha
Water Utility  (Wisconsin), Louisville
Water Company (Kentucky), New York
City  Water  Supply  Bureau,  and
Philadelphia Water  Department. The
utilities varied  in  size of service  area,
geological conditions, and  geographic
location. Comparative information  on
mam  break  patterns by  age,  diamete
construction  period, and  material  in th
six  utilities studied  is presented in thi
section.  Each  utility's  recordkeepm
system was  examined. A recommende
procedure is included for the analysis  c
leak and break repair data. Eight exhibit:
including a  sample repair  form and
series of data plots, are presented.
  It was concluded from this study th<
water  main leak and break patterns var
widely among utilities and within a  give
utility. Variations in pipe  materials, joir
materials,  design  practices, constructio
practices,  soil  conditions,  corrosio
conditions,  external loadings, an
weather  conditions all  contribute   t
variations in  leak and break experience;
Even with these variations across utilities
there  are similar patterns  which must b

Water Main Deterioration
  The   causes  of  water   mai
deterioration are the focal  point of thi
section  Structural  properties  of  wate
mains are   discussed,  including
technical  subsection describing thi
interplay  of  external  loads,  beddmi
conditions,  and   corrosion  on  thi
deterioration  process   The  genera
principles of corrosion are described witl
respect to both internal and  external pipi
conditions.  A series of  photographs  c
pipe  samples that  underwent variou
types  of  corrosion  is presented.  Thesi
samples  were  obtained   under thi
Philadelphia  Water Department's materie
testing program.  A discussion of thi
relationship  between the age of  wate
mams and  their  rate of deteriorate
encompasses changes  in   castmi
techniques,  construction  practices, am
materials over time.
  It was concluded  that a water main wi
fail  when the forces exerted on the man
exceed  the  main's effective  strength
This strength decreases from its origine
design strength as a result of internal an<
external corrosion,  reducing the mam'
ability to  withstand various appliei
forces.  Many  environmental factor
should be investigated for their roles  n
this process.

Assessing  Water Main Conditions
  A survey of analytical techniques use<
by  utilities  to  evaluate  water  mail
conditions is provided in this section
Descriptive analysis techniques evaluati
the frequency and patterns of  mail
breaks for several utilities, including Eas
Bay Municipal Utility District (California]
Philadelphia Water  Department,  am
several  utilities in  Great  Britain am

Canada. Three  predictive  model  case
studies from  Des Moines, the USEPA,
and  the Massachusetts  Institute  of
Technology  are reviewed.  Physical
analysis to  estimate  water  main
conditions is  outlined. Analyses  from
Vancouver,  Philadelphia, and the Army
Corps of  Engineers  Construction
Engineering  Research Laboratory are
   The predictive model approach offers
utilities  an  evaluation  method for
assessing the interplay of environmental
and structural  factors which lead to water
main  deterioration and failure. Utilities
are cautioned that  this approach will
require continual enhancement to refine
predictive capabilities as each  utility
collects information  on the interplay of
loads, environmental factors, and failures
of its main.

Economics of Distribution System
   The economics of leakage control,
main  rehabilitation, and replacement are
examined in  this section.  Economic
trade-offs of  alternative maintenance
and  capital  programs  are  shown in
models and discussed through sensitivity
analyses. Simple  work  sheets  are
presented so that managers can examine
and develop unit costs for their individual

Water Distribution System
   This section  presents recommen-
dations  focusing  on  information
necessary in building a water distribution
system inventory. Geographic reference
system  requirements,  inventory
procedures,  computer  mapping
capabilities,  and hardware/software
requirements  are  discussed.  The
recommended inventory system includes
a  combination of coordinate, link node,
and street address reference capabilities.

Monitoring Distribution System
   By monitoring its distribution system,
a  water utility can obtain the important
information needed to establish  a  cost-
effective  rehabilitation/replacement
planning  system.  A  series  of
recommendations  that  water utility
managers can use to monitor distribution
system conditions is  presented. Trend
analyses of both leak and break rates as
well as hydrant, valve,  and service
repairs are  suggested as  means to
assess system conditions.  The need for
annual  assessment  of distribution
systems  is  examined. These annual
assessments include analysis of current
and  anticipated  future trends  and
identification of those  portions of the
system  experiencing  higher or lower
failure rates.

Developing a Rehabilitation
Planning System

   Several water  distribution  planning
system approaches developed by water
utilities across the  country are presented
in this section.  The series  of planning
steps presented may be  adopted by all
utilities depending on  their  system
needs, budget,  and personnel.  These
planning  steps include a- standardized
main repair recordkeeping system for the
annual  assessment  of distribution
network conditions, the analysis of repair
patterns  by  geographic  area,  the
establishment of repair  pin maps, a main
testing program,  the  development of
economic  models  for  maintenance
decision  making,  the  preparation of a
computerized  inventory  of  the water
distribution system, and the development
of both water main condition predictive
models   and   computer-based
rehabilitation/-replacement  screening
and  selection systems.  Reactive  and
predictive  planning system  approaches
by four utilities are illustrated in  this
section  as  case studies.  Various
combinations of the   planning system
components  described above  are used
by each utility.

Summary and Conclusions
   Designing and  implementing a water
main  decision planning  system is an
on-going  evolutionary  process. It begins
with  an assessment of the  distribution
system condition.  From this  assessment
a  reactive planning  system  may  be
designed and implemented. This, in turn,
can  lead to  the development  of a
predictive planning system.
   Regardless  of  the  level  of
sophistication  and  computerization
involved,  each utility should collect
information  concerning  system
conditions,  implement  preventive
maintenance  measures,  identify
chronically failing  mains, and prioritize
these mains for repair or replacement.
The utility policy and replacement criteria
may change from  year to year, but the
basic process  remains the same.  Case
studies of live utility  planning systems
were  presented  to illustrate  various
approaches  to these  primary planning
   If  a utility  decides  that a predictive
planning method is  desirable, the  first
step  is  the  development  of  a
comprehensive water main inventory and
maintenance history  database.  This
database should include  the  location of
inventory,  maintenance history,  and
environmental  description  for each
record.  A  variety   of  options  for
inventorying water mains are available
and span a wide cost range.
   The   second   step   in   the
implementation of a predictive  planning
approach  is the development  of  a
physical condition  assessment  model.
This  model  enables  the  utility to
approximate the deterioration  process
based  on engineering principles,  and
estimate the relative  break and  leak
frequency  for the main.  This ability to
better estimate the future  condition of the
main makes preventive maintenance and
replacement  activities more  cost-
effective  due  to  more  precise
prioritization and scheduling capabilities.
   Additional criteria assessment models
such as hazard assessment or economic
evaluation models  may  also be
developed.  These criteria assessment
models may be coupled  with a physical
condition assessment  model to  select
and prioritize mains  for  replacement or
other  maintenance activities.  Field  data
collection and testing should  take place
to verify critical model assumptions.
   Increased   planning   system
sophistication    will    increase
implementation cost, particularly the cost
of building  and updating a water main
information  database. These costs must
be weighed against the benefits of more
effective  decision-making  regarding
repair, replacement,  abandonment,  and
preventive maintenance  activities.
Improved budgetary  planning can  be a
means of justifying additional  funding for
system improvements.  Improved system
reliability and water quality conditions are
other  important benefits which should be
measured.  These costs and  benefits
should be investigated  thoroughly before
a  commitment is made to implement a
predictive planning system.
   The full report was  submitted in
fulfillment of Cooperative  Agreement No.
CR810988  by  American  Water Works
Association Research  Foundation under
the   sponsorship  of  the  U.S.
Environmental Protection Agency.

  The EPA author. R. M. Clark (also the EPA Project Officer, see below), is with
   Water Engineering Research Laboratory, Cincinnati. OH 45268; M. Allen is
   with American Water Works Research Foundation, Denver, CO 80235; and
   D. Kelly O'Day is with Peer Systems, Inc.. Philadelphia, PA.
  The complete report,  entitled "Water Main Evaluation  for Rehabilitation/
 | Replacement,  (OrderNo. PB87-175378/AS; Cost: $18.95. subjecttochange) \
 '; will be available only from:                                              '
         National Technical Information Service
         5285 Port Royal Road
         Springfield, VA 22161
         Telephone: 703-487-4650
  The EPA Project Officer can be contacted at:
         Water Engineering Research Laboratory
         U.S. Environmental Protection Agency
         Cincinnati, OH 45268
United States
Environmental Protection
Center for Environmental Research
Cincinnati OH 45268
Official Business
Penalty for Private Use $300