Vous êtes sur la page 1sur 239

GWRC Compendium.

qxd 4/5/10 3:23 PM Page 1 (2,1)

Strategic Asset Management

Compendium of Best Practices in

Water Infrastructure Asset Management


Compiled and edited
JN Bhagwan

Global Water Research Coalition

November 2009


This study was jointly funded by GWRC members. GWRC and its members assume no responsibility
for the content of the research study reported in this publication or for the opinion or statements of fact
expressed in the report. The mention of trade names for commercial products does not represent or
imply the approval or endorsement of GWRC and its members. This report is presented solely for
informational purposes.

Copyright 2009
Global Water Research Coalition
ISBN 978 90 77622 22 3


The Global Water Research Coalition (GWRC) is a non-profit organisation that serves as a
collaborative mechanism for water research. The benefits that the GWRC offers its members are
water research information and knowledge. The Coalition focuses on water supply and wastewater
issues and renewable water resources: the urban water cycle. GWRC was officially formed in April
2002 with the signing of an agreement of collaboration and a partnership agreement was signed with
the U.S. Environmental Protection Agency in July 2003. GWRC is affiliated with the International
Water Association (IWA).
The members of the GWRC are:
Anjou Recherche Water Operations Research Center of Veolia Water (France); EAWAG Swiss
Federal Institute for Aquatic Science and Technology; KWR Watercycle Research Institute
(Netherlands); PUB National Water Agency of Singapore; SUEZ Environmental CIRSEE
International Research Center on Water and Environment (France); Stowa Foundation for Applied
Water Management Research (Netherlands); TZW - Water Technology Center of the German
Waterworks Association; UKWIR - UK Water Industry Research; Water Environment Research
Foundation (USA); WQRA - Water Quality Research Australia; WRC - Water Research Commission
(South Africa); Water Research Foundation (USA); WSAA - Water Services Association of Australia.
These organisations have national research programs addressing different parts of the water cycle.
They provide the impetus, credibility, and funding for the GWRC. Each member brings a unique set of
skills and knowledge to the Coalition. Through its member organisations GWRC represents the
interests and needs of 500 million consumers.
Water supply and wastewater utilities are organisations where physical infrastructure assets are
important and critical factors in achieving its business objectives and effective service delivery. These
water and wastewater infrastructures are increasingly challenged by a growing demand due to
urbanisation, problems related to aging and sometimes disintegration of the existing infrastructure,
and the impact of climate change. Not surprisingly, asset management is currently receiving great
attention in the water and wastewater industry and recognised as essential for the sustainability of the
water supply and wastewater services.
Therefore, the members of the Global Water Research Coalition selected Asset Management (AM) as
one of the priority areas of the GWRCs research agenda. In choosing asset management as a
priority area, the GWRC was interested in identifying the potential for collaborative research.
Compiling a Compendium of AM Case Studies was one of the joint activities of the GWRC to address
these challenges and needs in the field of asset management. Other activities include the
development of tools for Risk Management and Benefit Cost analyses, and the development of a
format for key asset date and performance indicators.
GWRC expresses the wish that this Compendium will be useful to all who are active in the field of
asset management of water and wastewater infrastructure.


The Global Water Research Coalition wishes to express its appreciation to the Water Research
Commission (WRC) of South Africa for acting as the GWRCs lead organisation for this joint effort.
The efforts by the compiler and editor of the compendium Jayant N Bhagwan (Director Water Use and
Waste Management at WRC) is gratefully acknowledged.
The compiler and editor is especially grateful to the GWRC partners for their support in the collection
and Ms Ingrid Buchan of the WRC for the editing of case studies.
I wish to thank the following people and institutions who have contributed to this compendium:
Steve Allbee (EPA, USA), David Cox (Water Services Association of Australia), Scott Haskins (Seattle
Public Utilities, USA), Maureen Hodgins (Water Research Foundation, USA), David M. Hughes
(American Water, USA), Max Maurer (Swiss Federal Institute for Environmental Science and
Technology), Roy Ramani (Water Environment Research Foundation, USA), Loet Rosenthal (PWN,
NL), Chris Royce (Anglian Water, UK), Andrew Smith (Yorkshire Water Services Limited, UK), Steve
Whipp (United Utilities, UK), Yunita Tan (Public Utilities Board -PUB, Singapore), Frans Schulting
(GWRC) .
I am also grateful to many of the above organisations who have provided permission to use their
materials in the compilation of the compendium. Finally, acknowledgements to Dr Kevin Wall of the
CSIR (South Africa) for preparing the cases from South Africa, as well as those named and unnamed
authors of the contributions in this compendium. Thank you to you all.



There are many terminologies and acronyms which have recently developed and evolved in the
enhanced management of water and wastewater infrastructure. Some of these are terms such as
Infrastructure Asset Management, Total Asset Management, Comprehensive Asset Management,
Strategic Asset Management, Strategic Infrastructure Asset Management and so on. The list is long
and growing, though differing slightly in meaning, the intention or objectives are all common, that
being improved management of water infrastructure. Gone are the days where the management of
this expensive investment was managed on the back of a matchbox or by thumb rules. The
increased scientific knowledge of the management of water infrastructure, is forcing the application of
radical and innovative techniques, that goes beyond seen pipes, tanks and pumps as infrastructure in
the ground, but as assets which have an operational life and effective and efficient use of these
component impacts on our ability as water providers and utilities to continue to provide high quality,
high assurance water and sanitation services at reliably and affordably. They are now being seen as
the lifeblood of the community, since they protect public health and they insure that local economies
and national economies continue to run.
For the purposes of this compendium I have kept the definition of Asset Management to encompass
all the terminologies used. Simply described, Asset Management is an integrated process of
decision-making, planning and control over the acquisition, use, safeguarding and disposal of assets
to maximise their service delivery potential and benefits, and to minimise their related risks and costs
over their entire life.
According to Steve Albee, who has for many years relentlessly lobbied for improved Asset
Management, People have not had the experience in essence of having to make massive and
significant reinvestment in systems from a generation ago. Its going to change the way people think
about the finance of these systems. Its become such a part of who we are and how we live that we
take all these systems for granted, yet theyre amazingly important in terms of our quality of life and if
you think about it, it still is not that long since that was not common.
Mounting evidence internationally suggests that the integrity of drinking water and wastewater
infrastructure is at risk without a concerted effort to improve the management of key assets
pipelines, treatment plants, and other facilities - and a significant investment in maintaining,
rehabilitating, and replacing these assets. For example in the USA, the Environmental Protection
Agency estimates that the projected investment needs range from $485 billion to nearly $1.200 billion,
and in South Africa the Government estimates $100 billion to be spent on all infrastructure over the
next few years. The brevity of this situation is now being seriously taken by many national
Recognising asset management as a fruitful area for research is relatively new. Asset management
covers a broad area and has roots in many disciplines. The application of asset management
principles in the global water industry is of particular interest at the present time due to pressures from
governments, regulators, shareholders and consumer groups in many countries to provide costeffective and sustainable water services at least cost to customers and the environment. Recognising
that the application of good science and innovation was on a rapid incline, it was deemed necessary
to capture these novel and creative applications, techniques and process which were contributing to
this new area of importance.
It has been a privilege and a huge leap in learning in compiling this publication on best practice, since
it demonstrate the advancements and progress being made in applying innovative and novel
techniques and processes in the management of water and sanitation infrastructure. The cases


highlight examples in these different countries on strategic initiatives at the highest level, through the
development of the policy and legislation towards ensuring that asset management becomes a legal
requirement for all water services providers; innovation techniques for infrastructure risk assessment;
decision- making techniques for capital investments; studies on the implementation of Asset
management in utility practice, the use of GIS and IT technology, and the various little techniques of
meter replacement, pressure management, continuous leak detection and rehabilitation of
This tapestry of best practice from around the globe demonstrates the giant strides the sector in
taking in ensuring that the good principles of Strategic Asset Management are implemented, applied
and built on. This wealth of this information is just a small taste of many initiatives which are taking
place and is encouraging in realizing that the professionals around the world are taking asset
management seriously.
In the words of Steve Albee, bringing about a paradigm shift in water services cannot be solely a
bottom-up phenomenon. Broad-based improvements in applied asset management practices provide
for a huge step toward bringing about sustainable systems. A collaborative understanding among
federal, state and local governments of how water systems should be managed would improve the
publics acceptance of whats really required. The pathway to sustainability is a quest that will last a
generation or more. Progress starts with taking small steps in the right general direction.
Strategic Asset Management is growing and will shape how we provide water services into the future.
For now, we must learn, absorb knowledge and put good practice to use, because things will not get
better if we do nothing.

Jayant N Bhagwan
Compiler and Editor



Foreword ............................................................................................................................. 5
Acknowledgements ............................................................................................................ 6
Introduction to the Compendium by the Editor ................................................................ 7
Table of contents ................................................................................................................ 9
Overview of water and wastewater infrastructure Asset Management ......................... 11
Historical Background ................................................................................................................. 11
Asset Management Definition ..................................................................................................... 12
Concepts used in Infrastructure Asset Management .................................................................. 13
Some guiding principles for IAM (sourced from Department of Water affairs south Africa,
integrated asset management strategy) ..................................................................................... 16
Asset Management around the Water World ............................................................................. 17
Challenges and future needs ...................................................................................................... 18
Global cooperation on Strategic Asset Management ................................................................. 19

Contributions from KWR Netherlands .......................................................................... 20

Meter-replacement optimisation at Duinwaterbedrijf Zuid-Holland, The Netherlands ................ 20
Distribution network design at PWN Water Company North-Holland ......................................... 30

Contributions from Water Environment Research Foundation (WERF) USA ............ 35

Programme sustainability A case study City of Hamilton, Canada ......................................... 35
Application of practice of understanding the condition of buried assets City .......................... 42
of Atlanta, Georgia (Atlanta Department of Watershed Management) ....................................... 42
Application of practice of setting the required level of service - City of Columbus Water Works
(CWW), GA. ................................................................................................................................ 50
Application of reliability centred maintenance at their Deer Island advanced wastewater
treatment plant as a best practice O&M strategy - Massachusetts Water Resources Authority
(MWRA), MASS. ......................................................................................................................... 56
Seattle Public Utilities (SPU) takes aim at critical assets using risk-based decision-making and
strategic asset management plans - City of Seattle (Washington State) ................................... 67

Contributions from Water Research Foundation ............................................................ 75

Decision-making for capital investments - Seattle public utilities case study: Introduction ........ 75
Continuous leak detection to monitor condition of water distribution pipes - American Water
case study ................................................................................................................................... 83
Use of electronic mobile and field solutions by Las Vegas Valley Water District - Las Vegas
Valley Water District.................................................................................................................... 89
The use of GIS to support EPCORs business processes - EPCOR Water Services ................ 95
Main replacement and rehabilitation program Louisville Water Company ............................ 102

Contributions from UKWIR............................................................................................. 107

Effective risk modelling of water infrastructure assets in the United Kingdom ......................... 107
Anglian Waters strategic investment approach - A case study on its initial application to
wastewater infrastructure .......................................................................................................... 115


System planning and the asset management process adopted within Northumbrian Water
Limited. ..................................................................................................................................... 122
A case study on water mains rehabilitation planning and implementation, United Utilities, UK
.................................................................................................................................................. 133
Change and asset management in Thames Water: building the (im)perfect beast ................. 139
Capital maintenance Good practice guide Leading Edge Asset Decisions Assessment (LEADA)
- Yorkshire Water ...................................................................................................................... 149

Contributions from PUB, Singapore .............................................................................. 166

Active, Beautiful, Clean (ABC) Waters Programme for Singapores sustainable assets
management ............................................................................................................................. 166

Contributions from South Africa.................................................................................... 172

European Union/DPLG Pilot Project Asset Planning for Four Municipalities ........................... 172
Free State Province and South African national monitoring of water and wastewater quality and
improvement of asset management ......................................................................................... 180
Johannesburg Water and its "Operation Gcin'amanzi" service upgrading, asset management
and demand management programme .................................................................................... 187
DWAF National Water Services Infrastructure Asset Management Strategy ........................... 194
Sebokeng/Evaton water saving through pressure management .............................................. 200
Rand Water major pipeline asset management ....................................................................... 207
Rustenburg public-private partnership for upgrading, operation and maintenance of water and
wastewater treatment infrastructure ......................................................................................... 216
Western Cape Provincial Government collaborative programme. - Asset management planning
support for local municipalities .................................................................................................. 221




Historical Background
In the water sector we have seen an evolution of practice and this has a strong linkage to the
terminology we utilize. In the past we have used terms such as equipment management, pipeline
management, material management etc. These related to very isolated or fragmented activities
related to the management of different water services equipment. Then we saw the emergence of the
use of term infrastructure is and this gave realization to the fact of the interdependency of different
water supply components and their impact on performance of these components. This phase brought
some realization of the need for a more integrated approach; however this was still heavily technically
biased and is still commonly used in the management of water and wastewater components. The
meaning given in a good dictionary of this is basic services, and equipment needed for country or
organization to function properly. The sudden realization in many parts of the world, of many of these
infrastructure components reaching their useful operational life, and requiring more than maintenance
but rehabilitation and replacement has raised many concerns especially due to the magnitude of costs
involved and its impact on the consumer. There has been this sudden change in paradigm from
depreciation to appreciation, which has led to the whole concept of Asset Management.
The word asset derives from the Latin ad satis, meaning sufficiency. It was adopted long ago by the
accounting profession to denote real property of sufficient value to offset the debt, or liability held by
an individual or organization. The term asset management first appeared in the banking industry to
describe an investment practice that built wealth through investments in different types of financial
vehicles. These vehicles bonds, stocks, real estate, etc. could be valued and posted on the ledger
as assets. Moving money between these vehicles or managing money became known as
managing assets.
The principles of Asset Management are well established in the petro-chemical, car manufacturing
and mining industries. In the water industry, the first comprehensive introduction was in the UK in
1989. At the time of privatization of the water industry, there was a need to provide investors with
information about the condition and investment requirements associated with the takeover process.
Further, it was needed in order to establish equitable pricing and required the private companies to
develop detailed asset management plans identifying how they would ensure the maximum return on
the public investment already made in the infrastructure of the utilities they were to acquire. Thus in
the UK Asset Management was seen as essential tool for water companies, which amongst other
functions enables them to meet obligations for the delivery of financial statements. Company
accounts are a statutory requirement in terms of legislation.
Implementation of Asset Management in New Zealand and Australia started primarily from the reform
of Local Government in the first half of 1990s. The process stimulated the reassessment of a typical
Councils engineering infrastructure assets and advanced the integration of their management with
accounting standards and state regulations. In 1993, the Australian Accounting Standards Board
issued the Australian Accounting Standard 27 (AAS27). AAS27 required municipalities to capitalize
and depreciate infrastructure assets rather than expense them against earnings. Thus, infrastructure
roads, sewers, fire hydrants and the like the domain of the engineer, became a complex set of
problems for accountants to manage as well.
Asset Management continues to evolve and provides a foundation for strategic financial planning and
annual reporting of stated targets and performance. In many parts of the world Asset Management
continues to get legislated because of its importance to national priorities. A good example to



highlight is that of South Africa, where Asset Management is regulated through five different line
ministries, from Treasury to Local Government.

Asset Management Definition

There are many terminologies and definitions on the subject of asset management. In the water and
wastewater environment one will come across many terms such as Strategic Asset Management,
Infrastructure Asset Management, Total Asset Management etc. The reason for this is that the
science and innovation on this subject is in an ongoing development and though the approaches and
definitions may vary, they all have a central objective being an integrated approach to the
management and improvements in the management of water infrastructure.
The International Infrastructure Management Manual (IIMM) defines strategic asset management in
terms of its goal, namely to meet present and future required levels of service in a cost effective way
through the creation, acquisition, maintenance, operation, rehabilitation and disposal of assets. The
principles of asset management guide the decision making process in order to attain this goal.
The Federal Highway Administration (USA) definition states that Asset management is a systematic
process of maintaining, upgrading, and operating physical assets cost effectively. It combines
engineering principles with sound business practices and economic theory, and it provides tools to
facilitate a more organized, logical approach to decision making. Thus, asset management provides a
framework for handling both short and long-range planning.
The United States Government Office of Accounting describes Comprehensive Asset Management
and the asset management approach which minimizes the total cost of buying, operating, maintaining,
replacing, and disposing of capital assets during their life cycles, while achieving service goals.

The US EPA Definitions of Asset Management are as follows:

For wastewater management utilities, asset management can be defined as managing infrastructure
capital assets to minimize the total cost of owning and operating them, while delivering the service
levels customer's desire. It is continuous process that guides the acquisition, use, and disposal of
infrastructure assets to optimize service delivery and minimize costs over the assets entire life.
The Department of Water and Environment Affairs of South Africas definition is that Infrastructure
asset management (IAM) is an integrated process of decision-making, planning and control over the
acquisition, use, safeguarding and disposal of assets to maximise their service delivery potential and
benefits, and to minimise their related risks and costs over their entire life. Thus IAM includes
operation of infrastructure assets, and also planned maintenance and repair, refurbishment and
renewal, and provision for replacement of the infrastructure. This definition indicates that IAM:
(a) Takes an organisation-wide perspective and draws upon applicable principles and techniques
in the management, engineering, accounting and social sciences (including human
(b) Has an outcomes focus (i.e. a focus on outcomes such as maximisation of service delivery
potential, protection of the ability of the infrastructure network(s) to deliver services, cost
effectiveness and efficiency).
(c) Confers a custodianship role on the managers of infrastructure and their political leaders i.e.
that they are the custodians, responsible for the lifelong sustainable operation of the
infrastructure, and for service delivery not only to the current users of the infrastructure, but to
future users as well.
(d) Must take into account both consumer expectations (including levels of service, and cost of
the service) and the legislative environment (e.g. financial and environmental legislation,
including any regulatory regime (e.g. regulation of drinking water quality)).



Another definition is that Infrastructure asset management procedures can be defined as the process
of managing the creation, acquisition, maintenance, operation, rehabilitation, extension and disposal
of the assets of an organisation in order to provide an acceptable level of service in a sustainable and
long-term cost-effective manner.
Thus from these various definitions from different parts and initiatives from around the world, it is
clearly evident that the outcome or the objective is the same being that of integrated approach
towards the maximization of services towards effectiveness and efficiency, as well as meeting
expectations of consumer/customers towards affordable and reliable services.

Concepts used in Infrastructure Asset Management

Similar to the titles and definitions used for Asset Management, the conceptual models of integration
of various activities also tend to differ, however towards the same outcome. A sample of this are
highlighted below.

Source: CDM Consulting Engineers from a Practical Guide

for Utilities and Public Works Directors



Source: United States General Acoounting Office


Source: South African Department of Water Affairs and Forestry Infrastructure Asset
Management Strategy

Source: United States General Accounting Office




Source: Bridging the Gap-An Educational Primer on Sustainable Water

Infrastructure Asset Management by the United States Environmental
Protection Agency and the Pennsylvania State University

It is clear that underlying these different conceptual forms and approaches, shown above and those
which one would encounter in many guidelines are four key elements which are the make-up and the
connection between the elements which then results in an integrated approach which realises
sustainable management of infrastructure assets. These four elements are:
Asset register or inventory.
Condition Assessment.
Risks assessment and determination.
These four elements form the integral parts of any Infrastructure Asset Management Strategy, which
provides a step by step approach towards establishing a good plan and operation which fits into the
sustainable functioning of all water services institutions. Surrounding these four elements will be all
the complementary processes and methodologies which give effect to the strategy, and these will
range from improved data collection techniques, to fixing leaks, to replacement and rehabilitation etc.



Some guiding principles for IAM (sourced from Department of Water affairs
south Africa, integrated asset management strategy)
Systems approach. IAM planning must look at the entire delivery chain (i.e. delivery of water
services), identify the constraints within the system as a whole, and then methodically
address these, prioritising the most serious constraints.
IAM is an integral part of ongoing service delivery. As an integral part, IAM is a
continuous process, not a once-off project or an event. It is a process firstly in the sense that
improvement must be planned, and improvement must be progressive. It is a process
secondly in the sense that improvement is not static demands, performance objectives,
technologies all change with time, and infrastructure is subject to wear and tear and to
obsolescence. And it is a process thirdly in the sense that infrastructure management and
improvement in infrastructure management is, or should be, a day in and day out duty of the
owners of that infrastructure.
Water services focus. This Strategy addresses improvements in the practice of water
services IAM, as opposed to the management of water resource infrastructure or other
municipal infrastructure such as roads and storm water, electricity, solid waste facilities or
public amenities.
IAM focus. Numerous challenges are encountered in IAM, such as the lack of technical
expertise. This Strategy recognises the broad array of challenges with which infrastructure
managers are presented, but concerns itself with the formulation of priority actions to address
IAM-specific issues.
Recognition that water services delivery is both a human right and commodity-based.
Water services infrastructure is utilised to treat, convey or store a commodity i.e. water. The
quality of water services is directly linked to the protection of water as a scarce resource, the
quality of potable water and its impact on health and safety, and the quality of discharge into
river systems.
Outcomes-based. Each priority must be outcomes-based and measurable.
An appropriate mix of short term successes and long term sustainability. Properly
managed infrastructure assets have life spans that can be measured in decades, and thus the
full benefits of IAM are felt over successive generations. Whereas this Strategy recognises
that the full establishment of IAM practices has a medium to long term horizon, it also
recognises that short term successes are not only possible but are required to establish
credibility, harness support and to improve failing service standards.
Promotion of an integrated, inter-disciplinary and inter-sectoral approach. IAM operates
at the interface of several functional disciplines, some of which include accounting and
finance, town and regional planning, and engineering. The role of communities and of political
leadership is also important the latter sometimes of overriding importance. This Strategy
promotes appropriate inter-disciplinary and inter-sectoral alignment, and thus an integrated
approach to IAM.
Focus on the key challenges, and prioritise. Numerous challenges present themselves in
the management of water services infrastructure. The Strategy recognises that only a select
group of challenges can be addressed at any one time, and that the key challenges that
impede the adoption and practice of sound IAM must receive priority attention.



Adoption of the Pareto (80/20) Principle. This Principle states that a small proportion of the
full effort required to achieve a particular result generally achieves close to the desired result.
And that further efforts are often subject to diminishing returns. This is sometimes stated as
80% of the full result from 20% of the full effort, or the 80/20 Principle (or rule). It is usually
valid for IAM. (Extending this thinking, a scan effort, to determine as quickly as possible
where the most critical problems lie, followed by the first steps of what would be a longer
improvement process, would often be worthwhile. This effort can, quickly and cheaply relative
to a more thorough effort, both bring about some rapid incremental improvement and also
ascertain the extent of a problem and how much further effort would be required.)
No one size solution fits all. While the general principles of IAM remain valid for all
institutions, the priorities differ from institution to institution, and also change with time as do
the techniques, the technological and non-infrastructure options and other factors.
Start with the basics, and get them right. The approach must be incremental. Do not
attempt to progress further until the basics are right. Address the weakest links in turn and
as each is improved and is no longer the weakest link, attend to the new weakest link. Where
there is a strength, support it, and build on it.
Political, management and operational focus. All levels must commit to IAM in order for it
to be successful from politicians who ensure political will, legislative compliance and
community requirements, to planning by management, to implementation at the operations

Asset Management around the Water World

In the last decade, there has been considerable amount of attention given to water infrastructure
asset management, and this has led to a large body of literature, publications and articles being
generated (see Resources section of this compendium). Most of the activity seems to be coming
from the industrialised world, and we are noticing greater emphasis to the asset management
challenges d in many developing countries.
According to Schulting (2007), some of the national/regional approaches to infrastructure asset
management like in Australia and New Zealand, Canada, European Union and the UK, have
developed to a level that they could be referred to as AM schools with impact for beyond their initial
range of influence. Many water utilities in Southeast Asia face the challenge of meeting the
increasing demand for better water supply and sanitation coverage. Enormous resource is required
for these utilities to expand, replace, and renew its facilities to increase service efficiency. The
Southeast Asian Water Utilities Network (SEAWAN) with support of the World Bank Institute and the
Asian Development Bank organised a high level attended workshop to discuss how comprehensive
AM can increase productivity and improve efficiency of water utilities in this region. Following a major
review of the water supply services in Kunching City, the Sarawak State Government of Malaysia
extended the study to cover all water supply systems within the state. The development of Strategic
Asset Management Plans was adopted and is used as basis for the water reform in Sarawak. To
resolve a number of the existing problems in the management of water and wastewater infrastructure
in China, the development and implementation of a framework for water asset management (WAM) is
proposed. The study includes a number of concrete recommendations and steps to support the
implementation of the WAM framework.
In his paper Schulting (2007) highlights that one of the most urgent challenges in Japan today is the
replacement of pipelines for the water supply. This is because earthquakes frequently occur in the
country and the aged pipes lack sufficient strength and are not constructed with earthquake-resistant



joint. Performance indicators are used to guide and evaluated the pipeline replacements in an
objective and quantitative matter.
Many developing countries are currently facing challenges related to their water and sewerage
infrastructure assets. Increasing and improving of service coverage is to be achieved through the
decentralisation and division of various tasks and duties from the government to lower role bodies.
Within the 5th Framework Program of the European Union, the twin systems CARE-W (ComputerAided Rehabilitation of Water Networks) and CARE-S (Computer-Aided Rehabilitation of Sewer
Networks) are developed . The CARE approach, which is contributing to a change of culture
regarding network rehabilitation. In Portugal, there are several on-going initiatives in the scope of
R&D projects, utility applications and regulatory policy related to urban water system diagnosis,
rehabilitation planning and investment prioritization. Key organizations are the National Civil
Engineering Laboratory (LNEC) and IRAR, the water regulator. The largest utility in Denmark with
some 500,000 customers CESD started the implementation of AM in 2004. CESD has chosen for a
combination of a top-down approach to start the overall AM process and bottom-up AM activities to
improve the short-term decisions and day-to-day practices. Large cost saving have been achieved. In
the United States, water infrastructure asset management is rapidly developing. Early in 2007, the US
Guidebook Implementing Asset Management: A Practical Guide was released by the Association of
Metropolitan Water Agencies (AMWA), WEF, and National Association of Clean Water Agencies
(NACWA). It is an acknowledgement that utilities and associations in the U.S. are serious about the
topic of asset management and are interested in advancing the science. The joint research bodies of
the water and wastewater sector in the US - AwwaRF and WERF - have started research on asset
management, including development of a tool for asset management and risk management for
wastewater and now drinking water utilities. This tool, entitled Strategic Infrastructure Management
Planned Learning Environment (SIMPLE), provides a structured approach for learning and applying
asset management programs in utilities. Associated training has occurred, particularly in the U.S.,
through seminars and workshops. The US Environmental Protection Agency (USEPA) plays a key
role in this regard.

Challenges and future needs

According to Schulting (2007), strategic asset management is a relatively new topic within the water
community and it should have recognised that different countries and utilities are on different points
on the path of their asset management journey. Various approaches (schools) for the implementation
and application of asset management have been developed over time to address the local situation of
utilities and cultural setting. He highlights the following overall AM and SAM challenges and needs to
be addressed:
Awareness; advocacy and making the case for strategic asset management.
inform/convince external stakeholders of the need of SAM and the associated resources
Common AM framework; although various approaches/schools for SAM exist and are
successfully used, the need for a common understanding of the underling principals and
common terminology/language is widely expressed.
Best practice, case studies, learning tools; there is a clear need of effective and efficient
accumulation, organisation and dissemination of best practice regarding asset management
concepts, processes and practices relevant to the water community.
Tools and models; to support AM in practice; i.e. benefit/cost evaluation, risk assessment,
performance indicators etc.



In addition to the more managerial topics indicated above, a clear need for technical research on i.e.
clever tools for physical condition assessment, cost effective repair and rehabilitation techniques, new
design concepts for the water supply, sewer and urban drainage infrastructure, sensors, etc.

Global cooperation on Strategic Asset Management

In line with a long history of very productive cooperation on a large range of topics within the national
and international water community, the development of strategic infrastructure asset management in
the water sector would profit of a similar approach and attitude. Overall items of cooperative activities
would include:
exchange of information, knowledge and know-how;
research and development on concepts, models and supporting tools;
harmonisation, standardisation and benchmarking.
In addition to and support of the vast amount of national organisations (i.e. AWWA, EWA, WEF,
WSAA, WRC-SA) already active in the AM and SAM area, a number of organisations at a
supranational level are present as platform for collaborative activities, including IWA, Global Water
Research Coalition, ISO and ASTM International, and international operating consultants. The
International Water Association (IWA) plays a role in creating awareness, disseminating the existing
know-how, networking, contributing to the standardisation of concepts and elaborating guidance
documents for the utilities. With the framework of the Global Water Research Coalition a number of
national water research organisations have joint their efforts regarding research on asset
management. The joint activities and projects included a compendium of best AM practises, and tools
and models to support benefit/cost analyses, risk management and data acquisition). Joint research
projects related to AM are part of the research programs of the European Union.
International consultants are playing a relevant role in advising water utilities on the implementation of
AM systems. Within companies quite often sets of AM approaches and tools have been developed to
support the implementation and practical application of asset management. Active at a global level,
these activities also results in exchange of information between their clients and some sort of
harmonisation of the way of working. However, utilities must always be aware that AM is a corporate
approach and not a plug and play system that can be purchased ready to use.

Schulting, FL and Alegre, H. (2007). Global developments of Strategic Asset Management. Paper
presented at the LESAM 2007, IWA




Meter-replacement optimisation at Duinwaterbedrijf Zuid-Holland, The
A B Ramaker*, M Helgers, R de Bont and F Roman
*Duinwaterbedrijf Zuid-Holland, Plein van de Verenigde Naties 11-15,
PO Box 756, Zoetermeer, The Netherlands.

Duinwaterbedrijf Zuid-Holland (Dunewater Company South-Holland) supplies drinking water to 1.2
million inhabitants. An important incentive for cost reduction, better water quality and service, is a
national benchmark that has been introduced in the Dutch drinking water sector in the late 1990s. The
national benchmark played an important role for Duinwaterbedrijf Zuid-Holland in developing its
strategic goals for 2010. Two important pillars have been identified to achieve an excellent ranking in
the benchmark, and thus achieving many of the 2010 company specific strategic goals. The 1 pillar
is cost reduction through process optimisation and operational excellence. One of the more
spectacular results of process optimisation is the reorganisation of the water meter-replacement
department. This example will serve as a case-study in this paper. The 2 pillar for further
improvement has been the introduction of an asset management framework from 2007 onwards.
Duinwaterbedrijf Zuid-Holland regards asset management as an important tool to balance cost and
quality in terms of drinking water and service delivery. Although development of the asset
management framework has been planned for 2007, building blocks have already been available.
507 000 water meters with an average technical life-span of 12 years have been installed to monitor
drinking water use. Duinwaterbedrijf Zuid-Holland developed a business assessment model in 2004 to
choose either to outsource the water meter- replacement department or to maintain existing activities.
Beforehand, it was clear that retaining existing meter-replacement activities would only be feasible if
the water meter-replacement process was competitive with commercial providers that offered such a
service. An explorative study, assessing the opportunities for optimisation, was carried out to
determine the maximum workload per employee. A subsequent step was to assess the current
administrative and logistical process and to develop solutions that optimised the meter-replacement
process. The assessment of the current process showed it to be very labour intensive, with excessive
associated paper work. Identified weaknesses included time-consuming planning procedures that
were customer unfriendly, with many customers failing to honour appointments. Manual methods
generated errors in the system.
The meter-replacement process has since been redesigned and automated in close cooperation with
the employees involved, in order to address these identified weaknesses. A new automated planning
system serves as the heart of the process. A range of options are available because the planning
system is accessible via (mobile) wireless Internet. It is possible to plan appointments in two-hour
time-slots (more customer friendly), with automatic rescheduling when customers are found to be
away from home, using a mobile online wireless connection for rescheduling work assignments,
automatic routing and rerouting via postal codes, information processing, all with a handheld
computer in the field, linked to the Customer Information System (CIF). Customers can reschedule
appointments through telephonic voice-response systems.
The fully automated planning tool has diminished the total costs for water-meter replacement by 30%.
Analyses show that the number of replaced meters has increased from an average 95 meters per
employee per week to 118 meters per employee per week, a productivity increase of 22% The
application was expanded in 2007 to exploit the planning system further, with new functionalities such



as providing customers with the opportunity to change appointments for meter replacement using the

Duinwaterbedrijf Zuid-Holland located in the Western part of The Netherlands (see Figure 1), supplies
1.2 million inhabitants with 74 x 10 m of drinking water per year. The supply area is only 60 km and
is thus densely populated.

Figure 1: Water supply areas in The Netherlands

(Source: VEWIN water supply statistics 2006)

Duinwaterbedrijf Zuid-Holland relies on surface water sources, as the groundwater sources are
brackish and difficult to exploit. The treatment technology of Duinwaterbedrijf Zuid-Holland is
consequently complicated and includes a sophisticated multi-barrier approach. The raw water is
imported from outside the supply area (a side-branch of the river Meuse) by two large transport
mains. The water is pre-treated with iron sulphate at the raw water intake at Brakel in the river Meuse
on a side-branch of the river Meuse. Residence time is approximately 2 months. The pre-treatment
scheme at Brakel also includes micro-sieves for the warmer part of the year. The raw water is



transported after this first purification step to a pumping station Bergambacht (also outside the supply
area). A subsequent second pre-treatment with rapid sand filtration is carried out at Bergambacht.
The pre-treated water is transported from Bergambacht to the dune area near The Hague, situated
along the North Sea coast. Duinwaterbedrijf Zuid-Holland infiltrates the pre-treated water in the dune
area by means of infiltration ponds and the water mixes with precipitation water in the dune aquifer.
The raw water has also been infiltrated directly into the aquifer since the 1990s.
The water is extracted from the dune aquifer after a 2-month period. This natural filtration with dune
infiltration ensures that the raw water has favourable properties for further treatment; stable
temperatures and is almost free of bacteria and viruses. It also ensures a reserve of more than 2
months, should there be a disruption in raw water supply.
As the dunes are an unique and vulnerable nature conservation area (2 400 ha) and at the same
time a precious asset to the drinking water purification process, Duinwaterbedrijf Zuid-Holland puts
great effort in nature conservation measures (deep filtration), maintenance, and promotes
environmental awareness to the many visitors that come to the area every year.

Figure 2: Post-treatment processes after the natural dune filtration

After the natural dune filtration the recovered dune water is post-treated with softening, powdered
carbon, aeration, quick sand-filtration and slow sand-filtration (see Figure 2). The drinking water is
distributed via large storage reservoirs to the customers via a 4 160 km distribution network. Water
use is metered at almost every property by 507 000 water meters in the supply area.

Duties of the asset management department of Duinwaterbedrijf Zuid-Holland

Although Dutch potable water companies are monopolies (with public shareholders such as provinces
and municipalities), they strive continuously for cost reduction and operational excellence. An
important driver for cost reduction is the national benchmark introduced in the Dutch drinking water
sector in the late 1990s. The national benchmark serves to compare the different processes of Dutch
water companies as to costs, environmental impact, water quality and service provision (customer
satisfaction). Drinking water prices (per m ) have dropped by 5.5 % (corrected for inflation) as a
consequence of the introduction of the national benchmark, and this trend is expected to continue.



Many of the strategic goals of Duinwaterbedrijf Zuid-Holland are benchmark driven: cost reduction,
improvement of customer satisfaction, network reliability and drinking water quality.

Pillar for improvement: process optimisation

Two important company pillars have been identified to achieve an excellent benchmark ranking. The
1 pillar is cost reduction through process optimisation to achieve operational excellence. Many
processes have recently been optimised. This has led to a significant cost reduction and better
service, with decreasing water prices and a gradual improvement in benchmark rankings. One of the
significant results is the reorganisation of the water meter-replacement department. This serves as a
case-study in this publication. Pursuant to operational excellence, a dedicated department with
professionals from different expertise areas (process analysts, information analysts, and business
analysts) has been created to support the asset management department in achieving its costreduction goals.

Pillar for improvement: asset management


The 2 pillar for improvement is asset management. Asset management is not new; the distribution
department has managed the distribution network for decades. This has resulted in a ranking of
number 2 on the national benchmark of 2006 for distribution processes. Although there are many
available systems for sound asset management, it is our ambition to introduce an integrated asset
management framework from 2007 onwards so that the first results become available in 2008. We
envisage that by drawing all relevant information into one asset management framework we will attain
a better balance between costs, water quality, and external factors. This is important to prioritise
investment decisions using information analysis, e.g. planning a PVC mains replacement programme
in a part of the supply area. An asset management framework with clearly defined performance
indicators will provide the management team with better decision tools. The challenge is to
continuously re-balance relevant criteria and refine decisions. This balance must ensure an optimal
asset value and concomitant cost reductions, as depicted in Figure 3. The asset management
framework is also a link between strategy and operations.

Figure 3: Asset management will create an optimal asset value (source: The
Woodhouse Partnership)



As mentioned, Duinwaterbedrijf Zuid-Holland will not be starting from scratch. The following building
blocks for an integral asset management framework already exist.
Technical Accountant Report
The technical accountant report assesses the characteristics and performance of all relevant assets
on a regional scale, examples being transport mains, distribution network, water meters and valves.
The technical accountant report is an overview document in which the main outcomes of several
analyses have been summarised. Analyses include:
the length, location, age, materials used in the transport, distribution and house connection
network incidents and service interruptions
lost customer minutes
customer complaint analysis
water quality index.

Customer Minutes Lost (CML)

Duinwaterbedrijf Zuid-Holland uses a nationally developed standard to measure the reliability of the
network. This standard is being applied by all Dutch drinking water companies. The standard includes
all supply interruptions arising from maintenance (planned and unplanned), failures and water meter
replacement. Water quality problems (e.g. water palatability/safety complaints) are included, as well
as customer minutes lost.
Water quality index Aquator and geographical plots
The Aquator is a specifically-developed water quality index for Duinwaterbedrijf Zuid-Holland. It
includes twelve water quality parameters in the distribution network. These parameters can be
monitored by Duinwaterbedrijf Zuid-Holland. The Aquator-index is measured on a regular basis, so
that trends can be analysed. The Aquator is useful for management decisions through setting of
priorities. Decision-making is further optimised by geographical plots in which water quality
parameters and customer complaints are depicted in a pictorial form.
GIS-tools and databases
All relevant data regarding the location of the distribution network are stored on GIS-databases. This
offers the opportunity to relate this data with water quality problems, network interruption incidents
and customer complaints, and provide input to investment and maintenance plans.
Research on water quality and pipe materials
Water quality research and research on pipe materials is carried out to manage water quality
deterioration in the network. Research focuses on sediment and particle accumulation, self cleaning
networks, and biofilm formation. This water quality research is carried out in close relationship with
pipe material research. The pipe material research not only focuses on the interaction with the
drinking water but also on life-span expectancy and reliability of various pipe materials.
Risk assessment
External risks and possible impacts are assessed in a structured approach.



Research as a basis for maintenance decisions

Several tools and models have been developed in the joint research program of Dutch drinking water
companies (BTO), and are being used by drinking water companies. One example is the knowledge
system for determination of remaining asset life-span (kennissysteem levensduur bepaling). In this
model, applied daily by Duinwaterbedrijf Zuid-Holland, maintenance decisions are prioritised on the
basis of a difference between the actual condition of the assets (based on pipe-material research and
data from practice) and the requirements posed by the environment (i.e. where asset failures would
cause the most damage).

Case-study: Optimisation of the water meter-replacement process

507 000 water meters have been installed in the supply area of Duinwaterbedrijf Zuid-Holland to
monitor drinking water use. The measuring device in the water meter is based on an impeller that
transforms flow into rotation. The greater the flow through the water meter, the faster the impeller
rotates. Every rotation equates to a volume of water. Moving parts in a water meter wear with time. As
a consequence some measuring devices may fail to comply with accuracy requirements as defined in
European and national regulations and the water meter should be replaced. The technical life-span of
a water meter is defined as the number of years that a water meter provides meter data of adequate
accuracy, as specified by regulations. The average technical life-span of water meter is 12 years in
the supply area of Duinwaterbedrijf Zuid-Holland. This means that the necessary replacement rate is
approximately 40 000 water meters per year.
In 2004 Duinwaterbedrijf Zuid-Holland put forward a business proposal for a decision as to whether to
outsource water meter-replacement services or to maintain the status quo. Beforehand it was clear
that maintaining existing meter-replacement activities would only be justifiable if the water meterreplacement process was competitive with optionally available outsourcing commercial services.
An explorative study to determine the maximum workload per employee (number of replaced water
meters per day) was carried out to explore areas amenable for work optimisation. A subsequent step
was to assess the current administrative and logistical process, and to construct solutions for
optimisation of the whole meter -replacement process.

Administrative aspects of meter replacement in the current situation

Water meter GPS coordinates are stored in the central Customer Information System (CIS). Queries
using the CIS are performed to locate meters that have reached the end of their technical life-span
and should be replaced. Once identified, work orders for the meter-replacement employees are
processed. Customers are informed by letter and requested to phone the water company to make an
appointment (in 4-hour time-slots) and the appointment information is stored a digital database.
Meter-replacement employees receive their week planning every Monday. If the schedule changes
during the week, meter-replacement employees are informed thereof by telephone. In the event that a
customer is not at home, the meter -replacement employee leaves a letter with a request to make a
new appointment. When several visits (2) and requests by letter (3) remain unanswered, the water
connection to the property is disconnected. If the customer is at home (as planned), the meterreplacement employee completes a paper work-order, and replaces the water meter in accordance
with the normal quality guidelines. At the end, the customer receives a letter with the final meter
reading of the replaced water meter. The meter-replacement employee completes a form to advise
the office that the replacement was carried out.
The whole process, in general, is very labour intensive, with much paper work. An assessment of this
process reveals many bottle-necks, some listed below:
Collating assignments for meter-replacement employees is time-consuming.
Appointments in 4-hour time slots are customer-unfriendly.



Many customers are not at home for an appointment, even though they have the facility
(which they use) to change the date and time themselves.
Many process executions are performed by hand, creating errors.
When there are letter-printing errors it is difficult to identify letters that are missing from a

Solutions for identified bottle-necks

A new process methodology for water-meter replacement has been developed to solve to the
identified bottle-necks. Automation plays an important role in the new procedure, which has to comply
with following criteria:
Appointments must be made in 2-hour time-slots (instead of 4-hour time slots).
The target replacement number is 22-32 per employee per day. This number is based on a
parallel study regarding the physical work load for meter-replacement employees.
The planning system with work-assignments should be continuously updated online, a day
before the actual date for replacement. Work assignments will be base-lined, including a
travelling scheme for the employees. Employees have a hand-held computer (online wireless
connection) to process all relevant information into the planning scheme, and provide access
to the CIS.
The work-status of meter-replacement employees is viewable online.
Changes and mutations should be fed into the CIS without manual executions; the process is
digital and automated.

Figure 4: Meter replacement according the new

procedure (with hand-held computer)



The whole meter-replacement process has been redesigned and automated, based on the criteria
described above, and in close cooperation with the employees involved. A new automated planning
system serves as the heart of the process. The new planning system extracts data from the CIS
concerning the water meters that have reached their technical life-span and automatically notifies
customers by letter that the water meter will be replaced (8 days in advance). The planning system
also calculates an optimised travel route for the meter-replacement employee, based on postal codes.
The planning system is ASP-based (Application Service Provider), which means it is accessible via
wireless Internet. A snapshot of the user interface is depicted in Figure 5.
The meter-replacement employee receives all relevant data on a handheld computer with an online
wireless connection. The planning system reschedules work assignments in the event of
appointments with customers being changed. The meter- replacement employee also uses a
handheld computer to upload information about the replacement into the CIS of Duinwaterbedrijf ZuidHolland. Thus, the new planning system has the following advantages:
Letters are sent automatically to the customer.
If a customer is not at home, a new procedure will be started automatically.
Appointments are in 2-hour time slots.
Employee routing and rerouting is via postal codes.
Replacement information is processed with hand-held computers and uploaded into the
central computer.

Figure 5: A snapshot of the user interface of the new planning tool for meter replacement




Object seeker = by filling in an object number the required address can be found
P O box seeker = by filling in a P O box number the required address can be found
Calendar = this button gives access to the planning on a certain date
Mailbox = this button gives access to the email box of the meter-replacement employee
Name employee = the planning can be found for every employee
Day = day of meter replacement
Time slot (2 hours) = each day has been divided into 4 time slots of 2 hours
Work assignment = work assignment (default: 32 per day). Assignments (addresses) pop up
on the screen once the mouse-pointer touches an assignment

The status of an assignment can differ as follows:

= planned, no confirmation letter yet sent

= planned, letter sent to the customer, appointment not yet confirmed
= appointment confirmed
= meter replaced, assignment complete
= no access to property, customer unwilling to cooperate

Results of the optimised meter-replacement process

The fully automated planning tool has significantly decreased the required number of personnel in the
office, namely from, from 5 to 1.75 FTE. .Additionally, a statistical analysis of the number of replaced
meters has been performed, before and after implementation of the new planning tool. The analysis
shows that the number of replaced meters increased from about 95 meters per employee per week to
118 meters per employee per week, a production increase of 22%. Total costs for water-meter
replacement have dropped by 30%. The production increase is due to better access to properties
(better planning and flexibility for the customer), efficiency in internal planning and fewer
administrative procedures for the meter-replacement employees.

Figure 6: Number of meters replaced per employee per week



Total investment costs for software, hand-held computers and internal hours of Duinwaterbedrijf ZuidHolland are approximately 50 000. Operational costs of the new planning system have already been
included in the costs calculation for the new structure. Currently, with the new planning tool in
operation, the average replacement rate is 23.6 meters/day, whilst under ideal circumstances a
meter- replacement employee should be able to replace 32 water meters per day. In order to increase
the meter-replacement performance rate, some Dutch water companies have decided to replace
water meters on the basis of geographical schemes, and not on the maximum technical life-span of
water meters. Given that some water meters will be replaced before they reach their maximum
technical life-span, they forego this economic loss for other gains. If Duinwaterbedrijf Zuid-Holland
adopted this approach, it would only be economically feasible if meters were replaced within a 2-year
limit of their technical life-span.



Distribution network design at PWN Water Company North-Holland

PWN Water Company North-Holland supplies drinking water to about 1.6 million consumers in the
province of North-Holland in The Netherlands. PWNs yearly water production is about 10 x 10 m .
Water is supplied through over 10 000 km of pipe network and 720 000 connections.

PWNs vision on asset management

PWN realises that through asset management it is possible to have cost-effective management of
assets from source to tap, along the entire life cycle of these assets, aimed at delivering optimised
system performance.
PWN manages its water supply system in one source-to-tap organisation. This offers the best
guarantee for optimal and sustained performance and cost management. This case study attests to
this. One of the primary goals in water supply is managing water quality at the customers tap. To
achieve this, insight is needed in the various contributions to water quality changes throughout the
entire supply system. Possible performance killers need to be identified and can exist at the water
source, during the various water treatment steps, in the transport system and in the distribution
network. Insight in these performance killers can be used to manage water quality effectively.
Successful management means creating the biggest bang for the buck, namely eliminating
performance killers at lowest cost.
A second major element in asset management vision is the life cycle. Water companies are expected
to supply water over many years. A life-cycle approach should therefore prevail over a short-term
management philosophy.
Key elements in the PWN asset management philosophy are the source-to-tap approach and the life
cycle. The importance of these elements can be demonstrated with this case study dealing with the
introduction of new design rules for PWN distribution networks.

Case study: new design rules for distribution networks

Historical distribution network design
Distribution network design was previously a rather
elementary process. Networks were looped and generally
comprised larger pipe diameters to accommodate firefighting demand and extra capacity for projected future need.
The figure shows an example of such a network.
This distribution network was built in 1986 and was last
flushed in 1993. It is a looped network with AC-pipes of
200, 150 and 100 mm in diameter, with some branches
of U-PVC 63 mm.

Generally, larger pipe diameters lead to lower water flow

velocity. This promotes sediment formation in the network.
This material re-suspends with flow increase, creating
consequential water discoloration problems (see figure).



Self cleaning network

v 0,4 m/s (once daily)

Generally, there are three ways to deal with sediment:

Prevention of suspended solids from entering the network.
Emphasis is on reducing the particle load by means of the treatment process. Particles are
inevitably present in conventionally treated water. By taking extra care with treatment
procedures such as filter backwashing, the particle load can be minimised.
Removal of sediment periodically by means of pipe cleaning.
This is the most commonly used approach in existing conventional networks.
Prevention of accumulation of sediment in the network.
Emphasis is on creating higher water velocities in the network, in order to retain particles in

Conventional network
v < 0,4 m/s

Conventional network
(incidental resuspension)
v 0,4 m/s discolored water



New distribution network design rules: self cleaning networks

The latter approach is used in the new PWN network design rules. Pipe diameter is calculated in such
a way that sufficient velocity is created on a daily basis to prevent particles from settling.
In the case of asset drinking water distribution systems the application of self-cleaning networks
appears to be an obvious gain, as it leads to both lower cost and better service levels.
The design rules for self-cleaning networks were developed within the Joint Research Program of the
Dutch water companies. The basic hypothesis is that a regularly-occurring high-velocity will resuspend settled particles, preventing accumulation of sediment quantities that can cause obvious
water discoloration, if subsequently re-suspended. The discoloration risk will thus be decreased
through this self-cleaning process of the networks. In practice the self-cleaning design rules lead to
smaller size pipe diameters in branched terminal networks.



In a distribution network the life cycle, the cost consists of constructing the system (materials and
labour) together with operating the system, i.e. maintenance with respect to water-quality (flushing)
and water quantity (locating and repairing leaks). The service levels demanded by the customers are
related to water quality (most obvious being the colour, taste and odour of the drinking water) and
quantity, expressed as continuity and pressure of supply. With respect to the latter it should be noted
that the fire department is also a drinking water customer. Compared to conventional networks, the
self-cleaning networks are 20% cheaper to build (shorter pipe lengths and smaller diameters) and
cheaper to maintain (no flushing required). Additionally, self-cleaning networks provide better
services, as customers will not receive discoloured water (improvement of water quality), nor are they
inconvenienced by flushing procedures (possibly leading to a temporary loss of supply). Although the
self-cleaning networks are branched end-networks, the continuity of supply is not at risk because the
self-cleaning networks are integrated with looped transport networks. A fire flow of 30 m /h can be
guaranteed (ca. 85%) at the network end (on a 63 mm PVC pipe). Larger upstream flows are
PWN was one of the first water companies in the
Netherlands to introduce these new design rules.
This evolved into a branched network. The first
networks were dimensioned according to more
loosely-interpreted self-cleaning design rules. The
network shown in the figure was built in 2002. It is
a branched network with a main PVC-pipe of
200 mm in diameter and branches of U-PVC
110, 63 mm and some PE 50 mm. It supplies
water to 373 households.

After the first newly designed networks had been

constructed, more research was done on
sediment and cleaning processes and more data
became available from the performance of the
original networks. This work was the basis for a
re-evaluation of the design criteria used by PWN
and led to an even stricter policy with regard to
network design. The third example (called
branched network PLUS) is a network built in
2004 according to these stricter self-cleaning
design rules. It is a branched network with
branches of U-PVC 110, 63, 50 and PE 40
mm. The main pipe is supplying drinking water to
157 households.

Three network designs put to the test

The 3 networks shown in these figures
(conventional, branched and branched PLUS) are
located close to one other and are fed with the
same water quality from a looped structure of
trunk mains of AC 400 mm. This constitutes an
ideal situation for a meaningful comparison of the
performance of these networks. The hypothesis
is that distribution systems that are based on the
self cleaning design principles will be less
networks. This hypothesis is split into three subhypotheses that were tested:




in 3



1. A network with smaller diameter pipes creates higher velocities; the design velocity of 0.4 m/s
will be met frequently in a self-cleaning network.
2. The incoming load of particles (loadin) equals the outgoing volume (loadout) in a self-cleaning
network. In a fouling network the incoming load is larger than the outgoing load due to
sedimentation in the lines. During high demands the outgoing load can exceed the incoming
load due to a contribution from re-suspension from the lines.
3. The quantity of sediment per meter of pipe is lower in self-cleaning networks, compared with
fouling networks. During re-suspension (e.g. while flushing) the emergent turbidity in a selfcleaning network will be less than in a fouling network.
The following measurements were carried out to test these hypotheses:
Flow measurements were carried out on the supply pipes (locations 1 in (locations 1 in
Figure 1) of the branched network and branched network PLUS. There is no discrete point of
supply in the conventional (looped) network so influent flow was not measured. The flow
meter used was a Tokimec UFP 10, measurements being taken every minute on each
location over a period of 11 days.
Particle counters were used to record incoming particles (at supply pipe and locations 1 in
(at supply pipe and locations 1 in Figure 1) and outgoing particles (at pipes downstream from
the supply pipe, (locations 2, 3and 4 in locations 2, 3and 4 in Figure 2.). Particles were
counted with a Pamas Water Viewer using 8 channels, counting particles in diameter ranges
of 1-2 m, 2-3 m, 3-5 m, 5-7 m, 7-10 m, 10-15 m, 15-20 m, > 20 m.
Turbidity during flushing was measured at fire hydrants and service connections using an online Dr. Lange Ultra Turb turbidity meter and through lab analysis of emergent grab-samples
during flushing.
The table shows the results of the velocity calculations, the particle counts (load) and the turbidity
measurements at the measurement locations in the three networks. We draw conclusions from these
data with respect to the relation between maximum velocity and the self-cleaning capacity of a

Results of flow calculations (vmax), particle counting (loadout/loadin) and flushing tests (turbidity) at measurement


Branched Branched




of houses










loadout /

Turbidity of
flush sample


(< 1 FTU)
< 1 FTU
< 1 FTU
< 1 FTU
(< 1 FTU)
< 1 FTU
1,5 2,5 FTU
0,8 3,6 FTU
10 FTU
10 FTU





The branched PLUS network is self-cleaning on a daily basis. The difference between the quantity
(load) and size of particles entering and leaving the network is very small. Also the turbidity increases
only slightly during flushing (being higher than 10 FTU for only a few seconds). The branched
network, given the quality of the incoming water, is self-cleaning on an annual basis. The difference
between quantity (load) and size of particles entering and leaving the network is significant, showing
more and larger particles leaving the network during the high demand measurement period. The
warm summer with its high demands leads to re-suspension and removal of particles from the system,
without receiving discoloration complaints. Most of the particles are being removed during the highdemand period, resulting in only a slight increase in turbidity during flushing (being higher than 10
FTU for only a few seconds).
The looped network is not self-cleaning. The difference between quantity (load) and size of particles
entering and leaving the network is large, with many more and larger particles leaving the network
during flushing.
The new network design criteria now used by PWN demonstrate that benefit objectives are being
achieved. However, pipe cleaning still remains the dominant cure for discoloration problems in the
existing networks. More effort is now being made to reduce the particle load that enters the
distribution system, specifically by optimisation of the conventional treatment process, but also by
considering other techniques, such as ultra filtration. This demonstrates that solving problems related
to discoloured water remains a source-to-tap responsibility in which a life-cycle cost-and-benefit
approach is needed to balance capital and operational expenditure. It also demonstrates that asset
management policy starts at the design process. Incorporating maintainability as a design criterion
makes it possible to gain extra benefits over the life cycle, justifying the extra investment. This case
study demonstrates that investment reductions can be achieved by looking at network design from an




Programme sustainability A case study City of Hamilton, Canada
The City of Hamilton has learned that sustaining an asset management program requires more than
just a good technical program. To be clear, a sound technical program that is founded on quality data
remains the cornerstone for a strong asset management program. However, in order to sustain the
program over a long period of time, the City of Hamilton has learned that keeping the governance of
an organisation educated and informed on the purpose and value of their asset management program
assures its long life.
The asset management team ensures that their recommendations are understood and respected
during council meetings by providing yearly education to new and existing city council members. It
also results in the continuous improvement of the program, as councillors have a better understanding
of the program, and thus can provide better scrutiny. Today the Hamilton asset management program
is able to move toward an integrated program that analyses all city assets for a given location. In
addition, city operations and maintenance now play a vital role in their asset management program.
The asset management team continues to grow and to provide the city with a much clearer outlook for
the future. Most importantly, that outlook is understood by the citys council.

Background Information
Hamilton amalgamates
The City of Hamilton, Ontario, Canada, located on the western shore of Lake Ontario, is known as
Canadas Steeltown, due to the presence of major steelmaking companies. The city was incorporated
in June, 1846. In 2008, this 162-year-old city is now home to a population of more than 500,000 due
in part to an amalgamation of seven municipalities into one single city in 2001. This amalgamation
was one of the catalysts for the Hamilton asset management program, ranking as one of the most
mature programs established in Canada.
In 2001, as an outcome of amalgamation, seven municipalities needed to merge while continuing to
deliver their core services to their citizens. Of the many pains that occur during an amalgamation, the
need to align different organisations with varying service levels and differing delivery mechanisms is
one of the most difficult to execute. As expected, the amalgamation resulted in the upheaval of
programs and service departments, which, in turn, resulted in the creation of an entirely new
organisational structure. It also brought about the merging of assets that ranged in age from well over
100 years to brand new. From an asset perspective, the amalgamation consolidated:
6 200 lane-km of roads in 14 000 segments.
More than 350 bridges and culverts.
2 100 km of water mains in 32 000 segments.
2 500 km of sewer mains in 40 000 segments.
700 facilities.



5 520 acres of parkland and 430 outdoor facilities.

Multiple fleet and transit systems.
There were opportunities amidst the challenges. The amalgamation resulted in the creation of the
transportation, operations and environmental department. At the time, the departmental general
manager was an avid supporter of asset management. As a result, he took this opportunity to create a
whole new service program titled asset management. The new program was formed to help deal
with the amalgamation of a multitude of assets that were previously under seven different
jurisdictions. Additionally, a new position of manager of asset management (asset manager) was
created and an appointee put in place to lead the new program.
We knew we had a problem.
The issue of asset management actually began in the City back in 1998-99 with the water and
wastewater management program. Staff recognised that the infrastructure was aging, little data on
assets were available, and water prices were inadequate to cover actual costs. Realising that they
were facing a potential crisis, city staff developed models for sustainability and resolved to increase
rates by 8 to10%, in order to sustain their services in the long term.
During that period, the Canadian National Research Council (NRC) was carrying out a study on asset
sustainability, named the Municipal Infrastructure Investment Program (MIIP). Seeing an alignment of
our project with their objectives, the Hamilton team negotiated a collaborative status with the MIIP. As
a result, Hamilton went on to participate in this program, meeting with other municipalities and the
NRC to discuss the state of the Citys infrastructure. This participation helped Hamilton better analyse
its own infrastructure and identify the funding gaps.
The result of this effort has been an approval by City Council to move forward with rate increases for
water and wastewater services. In fact, Hamilton has roughly doubled their rates during the last 8
years. Rate increases were approved in the face of opposition from the steel companies in Hamilton.
While the cost was an impact on their bottom line, education and outreach enabled these companies
to understand the importance of good infrastructure and the consequential gains in terms of service
levels provided by the City to their business.







Building an asset management team

A small staff team began developing forecasts for the City of Hamiltons overall asset life through to
2031 (see Figure 1) to assess the needs of the aging infrastructure. These figures were the catalyst
needed to energise the asset management movement in Hamilton.






025% of asset life

Minor Maintenance




2550% of asset life

Major Maintenance




5075% of asset life








75100% of asset life


Figure 1: Hamiltons infrastructure life cycle forecast

The Asset Manager created a team that combined engineering, computer sciences, and financial
skills, and identified it as part of the infrastructure and environmental planning division. One of the
advantages of starting an entirely new department during the amalgamation process has been that all
people recruited to the asset management team actually wanted to be there. None were coerced to fit
into this team. It was new, and had no precedent within the organisation. This alone allowed for the
creation of a strong team with members that had been recruited from other business units on the
basis of their achievements record and suitable skills suitability.
The Asset Management team began as a smaller group, quickly proving that it could deliver promises.
The team was able to prioritise capital improvement programs and minimise political influence, using
reliable data as evidence of tangible results. This success allowed the asset management team to
address the capital planning function for rehabilitation and reconstruction of capital projects. The
asset management team created an engineering service and project management component during
the ensuing reorganisation, and included a strategic and environmental planning function.
As of June 2007, Hamiltons asset management team established a specifier/provider organisation.
The asset management team would thus not only define the projects for delivery, but would also
define the manner in which projects should be delivered. Under the Capital Planning and
Implementation division (CPI), as senior director, the asset manager now oversees engineering
services, including asset management, survey and technical services, design and development, and
construction and inspection Services. Also part of the portfolio is strategic and environmental planning
services, including strategic planning, environmental planning, open space development and park
planning, and business support services.

Getting traction
Many challenges arose as the asset management team tried to launch the initial programs. For
example, the asset management team had inherited seven differing road and bridge management
systems as a result of the amalgamation, all with various levels of data, including core data that
Asset data (including attributes).



Location data.
Condition data (physical and functional).
Historical maintenance data.
The quality of information for each system ranged from good to very poor, and each system used
differing data standards with varying levels of detail. Additionally, the information was distributed
across various different types of systems and/or databases. Each jurisdiction had ownership of their
information, and the systems used for processing, and on the basis of familiarity, they were thus
inclined to show preference for their own familiar choices, and motivate them for future use.
This was resolved by scrapping all existing road management systems and proceeding with the
implementation of the Hansen pavement management system. While this ultimately ended up serving
the City well, it was a difficult change to implement.

Creation of an integrated AM program

Having moved forward with a coherent new roads asset management program, the City was able to
develop an integrated asset management program, focusing on layering the asset condition data for
multiple types of assets in a given location in order to better understand the risks and value of
investments in a single location. The purpose was to:
Minimise the number of asset closures in a given area.
Minimise the impact of the destroying a relatively new asset to restore an old one in the same
Maximise effective use of rehabilitation funds.
Minimise the interruption to residents and businesses within the City of Hamilton.
Some of the programs to integrate asset management that are in progress include the following:
Ongoing asset inventory data capture and initial assessment.
o Hamilton has began utilising a Global Positioning System (GPS) to determine the
exact location of sewer lines during maintenance hole inspections.

Carrying out ongoing data capture from as-built drawings.

Road inspection programs to monitor the condition and commission repairs or rehabilitation of
all paved surfaces.
Implementation of a bridge management system that includes the inspection of bridges every
two years.
Implementation of a pothole inspection and maintenance program that includes full system
inspection on a three-year cycle, including development of a condition assessment rating
system to support this program.
Delivery of the sewer relining program (carried out at night).
Linking construction to the asset management program so that an asset condition
assessment can be updated when construction is completed.



Ongoing capture of water pipe condition assessment, along with asset attributes such as soil
conditions, pipe material, and installation date.
Ongoing data analysis of assets such as tracking breaks per km, flow problems, odour
complaints, back-ups, flows, pressures, etc.
Hydrant inspection program performed by the city fire department.
Ongoing geospatial data capture using Intergraph and Hansen.

Developing buy-in to an integrated asset management program

An initial challenge that Hamilton has faced early in the creation of its integrated asset management
program has been to get the different departments/disciplines to participate in an integrated AM
program. The solution came in the form of an infrastructure project coordinating committee,
developed to provide a forum for all business units to jointly review the infrastructure projects planned
by the asset management team, and be able to provide feedback and raise issues before finalisation
of any plan.
The initial infrastructure project coordinating committee was not fully supported through meeting
attendance by all departments. Some managers thought they would voice their issues as it suited
them. Knowing the importance of structured and consistent collection of participant input, the asset
management team was able, with the support of senior management, to institute a very simple but
powerful rule: If you dont show up, you dont get your say on the program, and there will be no further
opportunities for your input on the matter. This new rule has every department attending meetings
regularly, with appropriate representation. While this effort was not without its bumps and bruises,
departmental intercommunication has improved, infighting has greatly diminished, and the objectives
of the integrated asset management plan have come to fruition.

Integration with maintenance and operations

Another key aspect of Hamiltons success was the integration of maintenance and operations into the
asset management decision making process. These valuable resources are ignored only too often. At
Hamilton, they contribute to two key elements in the decision making process: they
Confirm the need for capital investments based on their understanding of asset condition.
Provide maintenance and operational alternatives to capital investments in order to mitigate
existing risks.
Maintenance and Operations division (M&O) representatives now participate in infrastructure project
coordinating committee meetings as part of the integration process. Having the maintenance teams in
the meetings allows them to confirm, with detail, why work needs to be done, provide feedback on
their inspections of infrastructure, and clarify their ratings.
Part of capital planning is to identify supplemental projects that can be done if money happens to be
available. Favourable bids below budget on approved projects render funds available. Because of
their close contact with the assets, the M&O team play an important role in identifying and prioritising
the mid-year projects.

Importance of information technology

A recent trend is the growth in the implementation of Integrated Decision Support Systems (IDSS)
information technology as a tool to support an integrated asset management system. The purpose of
these tools is to help cities create and analyse scenarios that allow for the optimisation of their asset



life cycles, based on condition and risk rankings. These tools take management system data (in
Hamiltons case, data captured in CMMS and GIS) and provide a prioritised list of capital projects
based on asset condition deterioration, project cost benefits, and risk mitigation. The tools also can
identify the projected financial benefits to help justify alternate sources of funding.
Based on the projected benefits, Hamilton has moved ahead with the piloting of IDSS. The intent has
been to eliminate having to analyse the different asset types and instead, collate all data for all assets
for a final integrated view. The purpose has been that it would be better to analyse them together as
an intergraded system. to avoid premature preferences and prejudices.
However, the City has decided for the time being that it will not proceed immediately with the full
implementation of an IDSS. Two factors have weighed heavily on their decision. The first has been
that the effort was not producing the expected results. The second is that the City is having acute
problems as a consequence of long term under-funding and therefore knows without doubt as to
which projects need priority over the next three years.

Description of best practices, including institutions (personnel/departments) involved

(Importance of governance understanding and buy-in of the AM program)
How has Hamilton been able to sustain and grow their asset management program? How was the
asset management team able to develop support for a long-term infrastructure investment program
with elected officials who often serve short terms of 3 to 4 years in a climate of severe funding limits?
The simple answer has been through education and communication.
The asset management team realised early on in the program that their efforts could easily be
abandoned as soon as the following city election. Councillors are elected on the basis of promises to
deal with perceived needs within their wards. The problems are generally very visible, as are their
solutions. The prospect existed that the Asset Management team funding could be reduced, or even
curtailed. Having realised real benefits from the earlier efforts with their water and wastewater
program, the asset management team saw the value in proactive dealings with council to ensure their
Councillors' asset management education workshops
A Council education workshop is held every spring by our asset management team. New councillors
are educated during these workshops on the internal workings of the asset management and capital
planning processes. Because many councillors do not have a city services background, the intent is
to help them understand the steps needed to deliver capital projects. They learn, for example, that
should they want to build a park in their ward, they first need a master plan, a process that alone can
take about two years. Experienced councillors receive updates on the latest developments. The asset
management team has found that these educational workshops have resulted in much better working
relationships and communication with council, along with a better mutual understanding of each
others needs.
As part of the education, there is a review of the time-related value of money, as well as concepts
dealing with asset deterioration over time.
There is an update on asset conditions during these workshops, and it is conducted through the use
of report cards, with red down-pointing arrows, or green horizontal-forward-pointing arrows that
indicate current asset conditions, as well as predicted status in 15-20 years. This is done to create an
awareness of asset condition beyond those features that are openly visible such as streets and
sidewalks. Councillors dislike seeing a poor condition rating of their assets, and are consequently
much more willing to pay attention to those assets that are not as obvious.



One-on-one meetings with councillors

The asset manager meets every year with every councillor to review the latest three-year capital plan.
The meetings cover all assets including:
Streets and sidewalks.
Water, wastewater, and storm drain pipes.
Parks and green lands.
Public facilities including municipal buildings, arenas, clinics, etc.
Transit system infrastructure.
Forestry, bridges, and culverts.
Traffic signals.
The purpose of the meetings is to ensure that all councillors understand the overall investment
program, where funds are committed, with the reasons, and where councillors become informed of
what is planned in their wards, and have the opportunity to comment on their priorities.
Detailed city tours
The asset management team takes the mayor and all city councillors on a joint tour of the city
once a year. The purpose is to explain the investment plan while on the tour, and to provide a
visual in situ support of the plan. The tour begins at the city hall, moving from one
neighbourhood to another, going as far as the most remote communities. Providing visual
proof and comparisons between neighbourhoods and wards has resulted in reduced backlash from individual council members.
Ward-by-ward breakdown of all proposed projects (mapped)
While the education, one-on-one meetings, and tours greatly help in creating a big-picture
grasp of matters for council members, they still rightfully remain advocates for their wards.
The asset management team, to ensure that levels of equity and visibility remain, provides
each councillor with maps of each ward to show all work done as well as work to be done.
This also serves as a tool to help the council better understand the work that still has to be
undertaken. They then have the opportunity to match their wish list to project work that has
been planned by the City.

Summary of key lessons and experiences (results and findings)

The need to develop a strong technical asset management program, capable of optimising asset life
cycles, is critical in the quest for sustainability. The City of Hamiltons in-depth asset management
program demonstrates the value of such a program. However, a strong program that is established
within a volatile environment, including city politics, will always be vulnerable to the winds of political
The success of City of Hamiltons unique approach to securing a proactive buy-in from their council
members must not be underestimated. This approach should provide a blueprint to all asset
management teams that ultimately report to a political body. City department managers too often
blame the council for poor funding decisions. The City of Hamilton model clearly demonstrates that
the burden of proof rests squarely on that of the city department managers. The City of Hamiltons
approach has demonstrated that most council members, even those with limited understanding of city
operations, can make the right decisions, when equipped with education and understanding of a



Application of practice of understanding the condition of buried assets City

of Atlanta, Georgia (Atlanta Department of Watershed Management)
The City of Atlanta, located in the northern part of Georgia, has a sewer system with parts that date
back to the mid-1860s. Combined Sewer Overflows (CSOs) and Separate Sewer Overflows (SS0s)
consent decrees has required the City to make aggressive progress in dealing with rehabilitation and
repair of the sewer system. The City of Atlanta has embraced an asset management philosophy as
part of the overall management and implementation of the consent decree. To date, the City has
spent approximately US$ 900 million on collection system improvements, including sewer separation,
Gunnite improvements, and more than 250 miles of rehabilitation. A total of US$ 2 billion to 2.5 billion
will be spent for the entire Sanitary Sewer Evaluation Survey (SSES), as well as rehabilitation and
capacity relief projects. The SSES cost is US$ 130 million, or approximately 5% of the entire
investment. The City examined spills early on in the program and found that they were occurring so
randomly that inspecting the entire collection system would be necessary. The program has focused
on three components: the Sanitary Sewer Evaluation Study (SSES) and rehabilitation selection
process, completely updating the Geographic Information System (GIS), and constructing a hydraulic
City collection system maintenance crews clean and maintain the system. Fully 25% of the entire
system is being cleaned every year, but it is focused in specific areas that have the greatest need,
rather than being rotated throughout the City. The cleaning program is closely integrated with the
Citys grease management program. Atlantas water and sewer rates have increased dramatically
during the past five years to support these compliance programs. However, a 1 cent sales tax was
approved, and this also generates revenue for the system and reduces the amount directly paid by
customers. This has required strong leadership which the City has enjoyed under Mayor Franklin.

The City of Atlanta is located in the northern part of the state of Georgia. The Citys motto is
Resurgens, Latin for Rising Again. It reflects the Citys resurgence as the economic centre of the
New South. The City is the hub of north Georgias explosive growth, home to the worlds busiest
commercial airport, and location of the 1996 summer Olympic Games.
The Citys sewer system was an entity of the general fund for 85 years, and was moved into an
enterprise fund in the mid-1960s, providing the system with dedicated funding. Some of the Citys
infrastructure dates from Atlantas first period of rapid growth in the 1860s. The age of Atlantas
infrastructure and extensive combined sewer system, along with a hilly topography and upstream
location in the sensitive Chattahoochee River watershed has meant that the City faces unique
challenges. The City is the poster child for Combined Sewer Overflows (CSOs) and Separate Sewer
Overflows (SS0s) consent decrees.
CSO and SSO Consent Decrees Require Compliance
The Clean Water Act passed in 1972 prohibits unpermitted discharges of wastewater to receiving
waters of the United States. Combined sewers, such as those in Atlanta, were originally designed and
constructed prior to 1972 to overflow into receiving waters during periods of heavier rainfall.
The City of Atlanta entered into a consent decree with the U.S. Environmental Protection Agency and
the Georgia Environmental Protection Division to reduce or eliminate CSOs. A consent decree is a
legal document that is signed by both the community and an enforcement agency that binds the
community to complete specific activities that will lead to the elimination or reduction of sewer
overflows in their municipal sewer system. For the City of Atlantas CSO consent decree, the



compliance period was 1998 to 2008. The CSO remedial actions included: evaluation programs;
approximately US$1 billion in capital improvements to the wastewater system; management,
operations and maintenance programs, and supplemental environmental programs such as the
acquisition of greenways.
The City also has an SSO consent decree. SSOs may occur when there is a constriction in the sewer,
such as a blockage from tree roots, grease, or a sewer collapse. SSOs also may occur due to
insufficient capacity or infiltration or inflow of rainwater. The Citys compliance period for the SSO
consent decree is 1999 to 2014. The SSO remedial actions include: management, operations and
maintenance programs; US$1.7 billion in capital improvements; and evaluation, rehabilitation and
capacity relief in the collection and transmission systems.

Background Information
Development of a Plan
In 2002, Atlantas newly elected Mayor Shirley Franklin unveiled a five-point plan in response to the
requirements of both consent decrees. The plan, which became known as the Clean Water Atlanta
Program, has focused on:
Professional management of the consent decrees.
Compliance with SSO consent decree.
Compliance with the CSO consent decree.
Water quality monitoring.
The Mayor formed the Department of Watershed Management (DWM) in September 2002, which
includes drinking water, wastewater, and storm-water management functions. The annual operating
budget is US$468 million, and the 10-year capital improvement program is US$3.9 billion. The DWM
has more than 1 800 authorised full-time positions. While the City of Atlanta includes only one-tenth
of the over five million population of metro Atlanta, Atlanta is a regional service provider. The City
provides treated water to more than 1 million people in a 650 square mile area, and wastewater
treatment for 1.2 million. The system comprises 1 900 miles of sewers (300 miles of which are
combined), three wastewater plants with an average daily flow of 220 million gallons per day, and 16
pump stations. In addition, within the City, lateral sewers from the right-of-way to the main line
comprise 284 miles.
The City of Atlanta has embraced an asset management philosophy as part of the overall
management and implementation of the consent decree. Atlanta did not have the tools to make a
case for asset management in previous years. The City did not have analytical capabilities it needed,
such as IT tools, which are critical to make informed strategic asset management decisions.

Description of best practices including institutions (personnel/departments) involved

Timetable for improvements
To date, the City has spent approximately US$ 900 million on collection system improvements,
including sewer separation, Gunnite improvements, and more than 250 miles of rehabilitation. A total
of US$ 2 billion to 2.5 billion will be spent for the entire Sanitary Sewer Evaluation Survey (SSES), as
well as rehabilitation and capacity relief projects. The SSES cost is US$ 130 million, or approximately
5% of the entire investment.



The City has an aggressive program to complete sewer rehabilitation in advance of consent decree

Sewer Group

lengths of























The First Amended SSO Consent Decree (FACD) originally negotiated a staggered schedule for
completion of the six sewer system collection groups. A sewer-shed ranges between 10 to 50 000
linear feet of mainline sewer, with an average of 33 500 feet. There are 40 to 50 sewer-sheds per
sewer group. A separate schedule for analysis and implementation of capacity relief independent from
sewer rehabilitation has been established in order to deal with parts of the system with capacity
constraints. Every sewer group always has evaluation, rehabilitation, and maintenance work in
progress in order to spread the benefits from this work throughout the City.
Planning the program
The City looked at spills early on in the program and found that they were occurring so randomly that
inspecting the entire collection system would be necessary. The collection system maintenance staff
identified problem areas throughout system in several work sessions. Subsequently, the prioritisation
of the sewer groups took into account the following criteria:
Frequency of overflows.
Severity of rainfall derived infiltration and inflow (RDI/I).
Risk to surface waters (creeks).
Impact of failure.
Status of any ongoing rehabilitation or renewal.
Available capacity of sewers.
Judgment of sewer operation and maintenance division.
Relative Impact of RDI/I from jurisdictions outside the Citys control.
Proposed development intensity.
Location of sewer within the combined system.



Sewer group boundaries were adjusted in order to deal with identified problems, especially upstream
issues, in order to have a true system perspective.
Components of the plan
The program has focused on three components: the Sanitary Sewer Evaluation Study (SSES) and
rehabilitation selection process, completely updating the Geographic Information System (GIS), and
constructing a hydraulic model. Data from the SSES are synthesised through an in-depth QA/QC
process. This is then uploaded to the GIS hub, following which the hydraulic modelling group
performs analysis in support of the rehabilitation selection process.
The City felt it had to inspect the entire system to establish credibility with the regulator.. The SSES is
focused on establishing a comprehensive and accurate inventory of the system, and developing data
for making decisions. The system is inspected for overall condition, and sources of inflow and
infiltration to the system are identified. Cost effective rehabilitation solutions are developed and
incorporated into capacity upgrades, where necessary, within the rehabilitation proposals.
Work has been prioritised throughout the six sewer groups in a manner that the areas that are
assessed to be in the worst condition are evaluation first. SSES inspection activities include: closed
circuit televising (CCTV); manhole inspections, including GPS location; smoke and dye testing of the
lateral; building plumbing location and inspection; flow isolation, service lateral inspection and
temporary flow monitoring, as appropriate.
The City uses up to four contractors under a detailed specification to perform the SSES work. Each
contractor must follow a well-defined data/report protocol. Primavera Expedition is used for
submittals and payment processing. The typical field inspection cycle includes manhole data
collection and verification of connectivity, manual and digital CCTV data review, and then merging
and uploading data to GIS.
The City has implemented an enterprise-wide GIS. GIS information can be accessed via the Internet
using web-based tools and ESRIs ArcIMS (Internet Mapping Service), ArcGIS Server, an Oracle
database and ESRIs Spatial Database Engine. As part of this project, CAD-based drawings had to
be converted to a geodatabase format.
The GIS provides many benefits. For example, the City has placed great emphasis on capacity
management, taking RDI/I (Rainfall Derived Infiltration and Inflow) out of the system. Many areas of
the city are capacity limited, so the City has focused on large-diameter main lines with larger carrying
capacity, believing that they provide the most benefit. The City has given special attention to trunk
sewers that have been constructed in flood plains and combined sewers in major creeks that are
often under water tables. In rain events, with stream levels rising, and when certain conditions exist,
trunk sewers essentially drain surface water.
In Atlanta, if an engineering analysis determines that a sewer basin has a capacity limitation, then
from that time on new wastewater flow is not allowed to enter the system until additional capacity is
provided by the City. The FACD allows new flows, but measures must be taken to insure that the
new flow will not worsen conditions. This means that new flows can be allowed to enter the collection
system, but only after the City has eliminated RDI/I through a find and fix process. At first, five
gallons would have to be removed for every one gallon allowed into the system. Currently three
gallons have to be removed for every gallon allowed. Due to the RDI/I reductions and credits gained,



the City can continue to develop and add capacity, and has been able to approve about 15 000
capacity permits.
The GIS develops automated RDI/I credit reports by sewer-shed that are created from calculations
from CCTV condition information and completed rehabilitation.
The capacity credit system
requirement does not exist when sufficient permanent capacity is in place.
Another benefit of the GIS is traffic coordination. Traffic coordination is managed through a GIS
interface with the Work Order Management System to deliver public information about projects in the
right-of-way that may disrupt traffic. Also, the City has developed real-time mobile applications.
Hydraulic Modelling
There is a GIS link to the hydraulic model. For the hydraulic model, the City has a macro model of the
system for sewers that are 12 or more inches in diameter. This system model is always being
updated. In addition, a series of micro models are built, to evaluate structural and service defects that
have been identified in the SSES work.
The City makes decisions on where capacity and structural rehabilitation is needed, using the results
of the SSES and hydraulic models. The overriding goal is system reliability and elimination of
overflows. The City strives to identify where they can most cost-effectively rehabilitate.
Rehabilitation contracts are defined in two categories. Defined contracts, comprising 75% of the
value of the work, are used to implement the designs of the Program Management Team (PMT).
Undefined contracts have a rehabilitation method specificity only, and do not include design. They
are used to correct problems that must be addressed immediately, and comprise the remaining 25%.
The PMT has developed a Rehabilitation Selection Tool (RST), which is a web-based application that
uses SSES and GIS data to help make decisions about rehabilitation technology to be used, together
with the associated costs. The RST helps guide the selection process but does not replace
engineering judgment in the process. The RST helps the City make efficient, timely decisions. City
staff and contractors use the NASSCO pipeline assessment condition protocol grade system for
categorising sewer conditions, based on the internal condition determined during the CCTV inspection
process. If the City identifies a sewer segment with an Internal Condition Grade (ICG) of 5 (worst
condition, indicating actual or imminent collapse), the contractor takes immediate action in terms of an
undefined contract. This activity includes point repairs of potential cave-ins. Parts of the system with I
C G of 4 and 5 form the bulk of the rehabilitation work, and are typically where structural and service
defects are identified. Rehabilitation may include remedial work on pipe bursts, if capacity
management indicates the need for capacity relief in trunk lines.



The SSES Team works in close coordination with the modelling team. The modelling Team takes the
SSES data, and uses it to update, change and refine the hydraulic model. Different teams use
extensive triage and interaction. The intent is to move into more structured, formal decision-making
over time. However, the overarching finding and big lesson learned from SSES is that cleaning and
CCTV have helped Atlanta get to know their system and has generated data for making decisions.
Data are revised and acted upon in accordance with the flow chart shown in Figure 1.

Figure 1: Data Management Process



Early on, City contractors were inspecting both main lines and laterals through CCTV. While all major
camera vendors say that they can inspect a lateral from the main line by launching spirals, this
technology failed to provide the results that the City needed. The City has a program to put cleanouts at the demarcation point between public and private parts of each lateral.
CCTV tools are used to analyse past and current shots of the sewer system. Eighty percent of
Atlantas system is small-diameter line under 18 inches. The assessment of these lines is only
structural. The City rehabilitates the lines where there are ICGs of ratings 4 and 5. Everything else
has been deferred until 2014, after the consent decree period. Trunk sewers are treated differently, as
capacity relief and RDI/I drive the decisions.
The City anticipates rehabilitating 30% of the entire system (similar to the results seen in Sewer
Group 1) during the consent decree period.
Other Initiatives
The program has focused on three components: the Sanitary Sewer Evaluation Study (SSES) and
rehabilitation selection process, completely updating the Geographic Information System (GIS), and
constructing a hydraulic model. Data from the SSES are synthesised through an in-depth QA/QC
process. This is then uploaded to the GIS hub, following which the hydraulic modelling group then
performs analysis in support of the rehabilitation selection process.
The City contractors that are undertaking CCTV of the system do an initial cleaning that is sufficient to
permit CCTV inspection. City collection system maintenance crews clean and maintain the system.
Fully 25% of the entire system is being cleaned every year, but this is focused in specific areas that
have the greatest need, rather than being rotated throughout the City.
The cleaning program is closely integrated with the Citys grease management program. Because
70% of spills have been blockage related, and the majority of the blockages have been grease
related, the City now issues annual permits to all food service establishments, and inspects their
grease traps three times per year. This level of inspection for grease traps represents doubling that
specified in the consent decree, but is carried out because the City has had better results from three
inspections per annum.
The City has reduced spills from 1 000 to 300 per year since 1999 through rehabilitation, capacity
relief, and increased O&M for blockages.
Interim results
Some interesting findings and lessons learned have emerged during the initial phases of the Citys
Of approximately 40 000 manholes in the system, 1 400 manholes, or 4%, were buried and
had to be raised.
There are typically more trees, and as a result, more root issues, where development is less
dense. Vandalism is another significant contribution to blockages.
Sewer Groups 5 and 6 are the combined sewer areas that require the greatest amount of
work, and will be done last.
Only a very few miles of force or rising/pumped mains have been inspected to date, given the
challenge of inspecting these lines.



Going forward, a significant benefit to the system is the requirement that new development
must reduce runoff peak by 30% for storm-water. Wastewater flows for the newer
developments are a much smaller issue.
The City also has a very significant water main replacement program, with its own system of
prioritising replacements. The City coordinates sewer, water and gas work to coordinate and
lessen the impacts on the neighbourhood, as well as paving after all the work has been
Program Funding
Atlantas water and sewer rates have increased dramatically during the past five years to support their
compliance programs. However, a 1 US-cent sales tax was approved that also generates revenue for
the system and reduces the direct amount paid by customers. This has required strong leadership
which the City has enjoyed under Mayor Franklin. The Mayor is a former chief operating officer and
bureau director under former mayors Maynard and Jackson. She has been open and transparent, and
very direct in reaching out to the public, making clean water the focus of her tenure. She realises that
rigid enforcement of the consent decree would shut down the City. To keep processes going, she
must have the necessary revenue, including a long-term financial plan with continued rate increases.
The business community is behind the mayor, as well as the financial plan, because without investing
in the water and sewer system, they recognise that the City will be shut down.
The average water and sewer bill is US$ 65/month. The City does not advise customers as to how
this service compares with water utilities in other parts of the country, but rather how the bill compares
with the other utilities, such as electricity and gas. The City has the evidence that water conservation
is happening as a result of a tiered rate structure, with increasing rates. The City has some
affordability programs for elderly customers.

Summary of key lessons and experiences (results and findings)

1. Knowledge is all about data (text, images, etc.). The City now has 13 terabytes of information
on its system. Data are a key element of decisions in any strategic asset management
2. Link the GIS to a hydraulic model to facilitate decision-making.
3. Minimise the cost of construction with contractors, through strategic ways to optimise
decision-making and results.
4. Conduct collection systems, operations, and maintenance activities on a regular, strategic
and predictive basis.
5. Engage the support of the public to make the necessary investments.

Bibliographic information
Hunter, R.J. and Sukenik, W.H. (2007) Atlantas Consent Decrees Drive a Substantial Commitment to
Trenchless Sewer Rehabilitation.
Hutchinson, R.El. El-Sayegh, H.K., and Chambers, L. (2007) Atlantas SSES & Integrated Sewer
Rehabilitation Selection Process.
Brown, C. and Toomer, K. (2007) Clean Water Atlanta Enterprise GIS.
Bechara, A., Brewer, B., Clark, L., and Iaukea, K. (2007) Hydraulic Modeling A Tool for Addressing
the Consent Decree.



Application of practice of setting the required level of service - City of

Columbus Water Works (CWW), GA.
Columbus Water Works (CWW) is a public agency that supplies drinking water and provides
wastewater treatment services to the Columbus, Ga. area, in west central Georgia, about 100 miles
south-west of Atlanta. Faced with a number of challenges as it transitioned to being a regional water
provider, CWW embarked on strategic asset management, as an outcome to its strategic plan. In
developing the strategic plan, CWW had undertaken an analysis of its strengths, weaknesses,
opportunities and threats, both internally and externally. Externally, CWW began a focused effort to
reach out to its customers and stakeholders, to better understand the expectations of its rate-paying
customers, business owners, developers, regulators and others. Since addressing strategic asset
management as part of the ongoing Strategic Plan implementation, CWW has undertaken a
formalised process for regular monthly surveys and other ongoing communication with its customers.
This process has helped CWW better understand the needs and expectations of customers.
Columbus Water Works has been able to gain a comprehensive perspective to setting service levels
by actively communicating with their customers through the use of surveys, focus groups and
interviews. Service levels have been established using industry (QualServe) standards, these being
the drivers of their asset management program at the strategic, operational, and tactical levels.

Columbus Water Works is a public agency that supplies drinking water and provides wastewater
treatment to the Columbus, Ga. area, in west central Georgia, about 100 miles south-west of Atlanta.
CWW serves more than 230 000 people in Columbus, Fort Benning and parts of Harris and Talbot
counties, providing water to 65 000 customer accounts and collection and treatment of wastewater to
62 000 customer locations. The customer base is mostly residential, as the textile industry of the past
has been replaced with service industries such as AFLAC, Total Systems Financial Services, and
Carmike Cinemas. Columbus Water Works recently obtained a 50-year contract to supply the
adjacent Fort Benning army base with water and wastewater treatment services. Recent base
closings and consolidation will result in the relocation of tens of thousands of additional troops to Fort
Benning. The addition of service to Fort Benning has resulted in a 20% increase in size of the
Columbus system. Thus, CWW is becoming a regional water and wastewater treatment service
provider for west central Georgia. CWW places great emphasis on its role of being a responsible
environmental steward of the Chattahoochee River watershed for future generations. As an
organization, CWW has a clear vision and mission to not only sustain the current level of service but
also to improve it for future generations. Furthermore, CWW has a goal of being the best provider of
water and wastewater utility services in the United States. For these reasons, CWW has embarked
on strategic asset management, as part of its Strategic Plan.

Background information
In 2004, CWW embarked on an effort to update its strategic plan, in conjunction with an update of its
facilities master plan, both of which were originally created in 1999. CWW faced, and continues to
face, many challenges in terms of being a regional service provider, including the impact of
regulations, changing water supply and demand, and rising costs. At the core of these challenges
was the need to build and maintain strong relationships with customers, while attempting to meet and
anticipate their expectations in the adverse situation of aging infrastructure. While it was recognised
that it was possible to make incremental improvements without a strategic plan, CWW wanted to take
full advantage of opportunities for significant change. CWW embarked on a strategic journey by
undertaking a comprehensive, one-year strategic planning process. The process was carried out by



senior- and middle-level managers as a strategic planning group. Employee sub-team members
added additional expertise and involvement across the organisation.
CWW used a strategic planning approach based on methodology developed by the AwwaRF
(American Water Works Association Research Foundation, subsequently renamed the Water
Research Foundation) project, Strategic Planning and Organizational Development for Water
Utilities in which they participated. AWWARFS "scan, plan, do" approach includes many proven and
innovative methods and tools to develop and implement a comprehensive set of business strategies.
The AWWARF project included the process for strategy development and addressed organisational
needs such as developing new skills, leadership, dealing with tough issues, adapting to uncertainty,
handling emerging goals, and managing communications. Six key CWW business strategies emerged
from this comprehensive planning process:
Strategy 1 - Enhance customer satisfaction
Strategy 2 - Strengthen regional economic potential
Strategy 3 - Leverage information technology
Strategy 4 - Optimise infrastructural performance
Strategy 5 - Develop a sustainable workforce
Strategy 6 - Maintain financial stability

The value of feedback

CWW reached out to its customers and stakeholders for input on key issues via in-person interviews
and interactive focus groups during the strategic planning process. Industry trends and significant
business drivers were evaluated to develop an appropriate, exciting future direction for the
CWW conducted regular assessments of customer satisfaction and had measured an overall
customer satisfaction rating of 90%. However, CWW desired more detailed feedback that would fully
address needs for improvement and desired customer service levels. CWW defined customers as
anyone who pays fees for receiving products and services on a regular or requested basis, is directly
impacted by the performance of CWW and/or is in need of information.
CWW conducted customer focus groups to bring together similar customer groups for discussion and
input to begin a dialogue on desired service levels. Groups included private residents, small and
large businesses, builders and developers, and top appointed and elected officials.
Meetings were held in a convenient location. Ice breaker types of questions were developed to
encourage conversation. Non-residential customer participants were asked to describe their
organisation (if applicable) and role, as well as their role in the community and interaction with CWW.
Residential customers were asked to address their perception and needs responding to the following
What is your perception of CWW?
Based on your awareness of CWW services, what is beneficial to you now and what would be
more helpful?
What additional services or products would you like to see CWW provide?
Rate your understanding of your current water/sewer bill. What do you like/dislike about it,
and why?



What is the perceived value received for the amount you have paid for water and sewage
Compared with other utilities and services (refuse/solid waste removal, electricity, gas,
telephone, cable) how would you rate CWWs level of service, and why?
In order to correlate with previous survey results, residential customers were also asked the following:
Recent customer surveys told us that the most important value to our customers is that we
have your best interests at heart. How would you define best interests?
What type of information and topics do you feel are most important to reach the public? What
channels are most effective for CWW to get information out and to receive input?
The feedback from customers has indicated that the bill insert is a good way to disseminate
information. Survey results for 2007 showed that 72% of CWW customers remember seeing the
inserts. Surveys, focus groups, and stakeholder interviews have been found to be effective ways to
get input.
Commercial and industrial customers were asked:
How could CWWs future direction be more beneficial to your organisation?
Based on CWWs major goals, which do you feel are most important matters; to the general
public; to your organisation or business? Give reasons? Are there other considerations?
What type of information and topics do you feel are most important to reach your company?
What techniques are most effective for CWW to get information out and to receive input?
Developers and builders were asked:
What is your perception of CWWs process for obtaining new services? What should we
change or leave as is? Why?
How can we better coordinate our expansion plans with the developer communitys plans?
Should CWWs growth strategy include expanding to a regional entity? Why or why not?
How could CWWs future direction be more beneficial to your organisation?
CWW used this customer focus group process as an opportunity to educate the public on what CWW
does, to continue to explore different communication avenues, and to counter media bias towards
reporting only the bad news. For example, CWW was able to clarify on methods of water use
estimation, and raise awareness of the advantages of multiple payment options.
CWW determined that its customers would like to be consulted occasionally and receive updates on
topics of interest from in-house utility experts. Interestingly, CWW found that its customers preferred
smaller regular rate increases, rather than larger and less frequent ones, like every five years, as
CWW used to do.
Some key outcomes that resulted from the focus group discussions included:
Standardise times for routine response.
Provide additional customer service training for all employees.
Expand technology-based customer service capabilities.



CWW looked at their most common service request types, and from these records, determined their
average response times. From QualServe and other utilities, they then established standardised
response times and standards for CWW. These focus-group discussions were direct input to the
Strategic Plan (2005-2010). Currently, CWW is implementing the strategies that emerged from the
planning process. Six strategy teams, one for each strategy listed above, are taking action,
evaluating the results, and making adjustments as necessary.

Assessing asset management practices

As a start to implementing Strategy 4, which has been to optimise infrastructure performance, CWW
strove to understand its current state of strategic asset management as compared to best practices,
identify areas for improvement, and develop an improvement plan to take its strategic asset
management program to the next level of competency. CWWs ultimate goal was a cost-effective,
results-oriented asset management program.
Consequently, they decided to undertake an
assessment of its practices.
The assessment process, guided by the six elements shown in Table 1 below, measured the degree
that a utility can readily answer the following core asset management plan questions.
1. What is the current state of my assets?
2. What is my required sustainable level of service?
3. Which assets are critical to sustained performance?
4. What is my minimum life-cycle costs, related to CIP and O&M strategies?
5. Given the above, what is my best long-term funding strategy?


Establish asset hierarchy, collect asset data and manage asset inventory

Identify critical assets having the highest consequence of failure

Determine service levels for critical assets including likelihood of failure

Assess condition of the critical assets and develop associated rating

Develop planned maintenance requirements for critical assets

Use life-cycle cost analysis for asset repair/replace/retire/renew decision-making

Table 1: Key elements of asset management used as an assessment framework

The concept of the interdependency of practices, technology and organisation that comprise any
business system or process was used when conducting this assessment. In the area of technology,
CWW has developed systems populated with data related to its assets in various sub-systems
including Customer Information (Avenir), Geographic Information, (ESRI), Financial Information
(Lawson), and Work Management (Maximo). While these systems were not implemented with a
comprehensive asset management program as a specific goal, they collectively serve as a backbone
for the asset register, condition assessment, customer service delivery and support, maintenance
planning and execution, and financial planning. They holistically monitor, project, and thus manage
asset life-cycle related information.



Description of best practice, including institutions (personnel/departments) involved

Service level agreements
A service level agreement, in its most explicit form,, is a contract that exists between consumers of
goods or services, or an agency that represents consumers and providers of goods or services. It
identifies the services to be provided, the required quality of the service, the timeframe within which
that service is to be provided, and the pricing of these services.
In practice, within Canada and the United States, service level requirements for most public water and
wastewater agencies are expressed at the provincial, state or national level as regulatory
requirements for drinking water quality standards, and discharged treated wastewater permit levels.
Service levels may be stated in policies such as standards for connection and for termination of
service in the event of non-payment. Other service levels, such as service interruptions arising from
main breaks, or "boil water" notices, are more often implied or incorporated in performance measures
that the utility strives to meet.
Some utilities may be subject to regulation by a state public utilities commission or public service
commission. In general, these utilities are private drinking water utilities although there are some
notable exceptions. One example is the Narragansett Bay Commission in Rhode Island, a wastewater
treatment agency that is regulated by Rhode Islands Division of Public Utilities and Carriers and
Public Utilities Commission. This public utilities commission also regulates for-profit water supply
utilities, as well as municipal water utility boards that serve areas outside the boundaries of their
municipality. This, and similar commissions provide monitoring of the provision of services, including
service terms and regulations, as well as monitoring of user-charges or fees, to prevent monopolistic
Utilities informally benchmark past performance, and make good use of this information. Progressive
utilities use accepted measures such as found in the QualServe Performance Indicators to trend and
compare performance.
Figure 1 provides a practical approach that embodies regulatory
requirements, customer feedback and periodic performance measurement and comparisons. High
performance utilities that value frequent customer feedback, consciously or intuitively practice the
cycle as depicted in Figure 1.

Set (Initial) Levels

of Service

Define Performance
Measures & Targets

and Judgment
Use Qualserve PIs
e.g. interrupts/ 1000
Baseline Customer
Satisfaction Indices

Communicate to
Measure Performance
and Cost

Sample Service
Focus Groups

Solicit Customer

Satisfaction Index




Figure 1: An approach to defining and monitoring asset service levels



CWW has adopted the following practices for service levels:

(a) Service levels are to be well defined, based on measurable performance targets affecting the
(b) Desired service level has been defined for asset service types.
(c) Customers are involved in feedback for setting targets for service levels.
Using QualServe data, CWW sets benchmarks for the number and duration of service interruptions.
The percentage of company benchmarks met each year is a factor in evaluating employee bonuses.
CWW has evolved its focus on levels of service. CWWs customer service department takes the lead
on this area. CWW continues to survey customer opinion to obtain input into its asset management
program. Currently, CWW is contracting with an independent consultant that specialises in consumer
surveys for water utilities, to administer ongoing surveys. The target is to obtain 1 200 survey
responses per year, representing approximately 1.8% of CWWs customer base. Accordingly, 100
surveys are conducted each month by telephone. The customers to be contacted are selected from a
list provided by CWW. These have accordingly received service and are supplemented by others
selected randomly from the general customer base to attain a number of 100 surveys. For example, if
the number of customers who had received service is 650, and this represents 1% of the whole
customer base, then 1% of those surveyed are selected from this list and the remainder are randomly
selected. Customers answer more than 100 questions, and the questionnaire takes approximately 20
min to complete. There are five categories of questions:
Water quality (taste, odour, colour, clarity).
Customer service.
Information factor.
Reliability factor.
Value-for-money in terms of service received.
The five categories are combined into an overall satisfaction rating. CWW receives a quarterly report
based on 300 responses that provides a regular snapshot of feedback from customers. CWW also
receives an annual report on results. CWW generally receives good to very good overall satisfaction
scores. Surprisingly, the "value-for-money" scores were found to increase immediately after each rate
Currently, CWW is testing customer knowledge of the Water Environment Federations Water is Life
and the American Water Works Associations Only Tap Water Delivers programs to educate
customers on the investment that is needed in infrastructure. For example, CWW is using
communication tools such as bill inserts, logo, and messages. CWW is also broadcasting 30-second
television advertisements. They are finding that customer recall of the Water is Life literature is quite
high. As CWW continues to use this survey method to obtain customer input, they are planning to ask
customer questions about their willingness to support continued investment in infrastructure in mid2008. These questions will be incorporated into the strategic asset management program by
considering long term investment in infrastructure, and the associated service levels. As a result,
CWW is evolving toward the model that has been depicted in Figure 1 (above.)

Summary of key lessons and experiences (results and findings)

Columbus Water Works has been able to gain a comprehensive perspective to setting service levels
by actively communicating with their customers through the use of surveys, focus groups and
interviews. Service levels have been established using industry (QualServe) standards, which drives
their asset management program at the strategic, operational, and tactical levels.



Application of reliability centred maintenance at their Deer Island advanced

wastewater treatment plant as a best practice O&M strategy Massachusetts Water Resources Authority (MWRA), MASS.
MWRAs focus on asset management has been on operation and maintenance improvements. The
focus is to employ industry best practices to extend equipment life and protect the ratepayer
investment in the utility. MWRA completed extensive benchmarking to determine the best practices to
adopt from both outside and inside the water and wastewater sectors. An overall facilities asset
management program plan was prepared and implemented using task teams, extensive training and
application of best practices. This work on best practices focused on implementation of reliabilitycentred maintenance at MWRAs Deer Island wastewater treatment plant. A focus on identifying
critical assets and implementing predictive and reliability-centred maintenance has saved MWRA
money as well as enhanced plant reliability. Such approaches that bring change to an organisation
will take time to implement, but will be well worth the effort.

MWRA provides wholesale potable water and wastewater services to 2.5 million people and more
than 5 500 businesses in 61 communities in central and eastern Massachusetts. MWRA was created
as an authority in 1985. MWRA supplies wholesale water to local water departments in 48
communities: 42 in greater Boston and the MetroWest areas and three in central Massachusetts.
MWRA also provides a back-up water supply in three other communities. MWRA's water comes from
the Quabbin Reservoir, about 65 miles west of Boston, and the Wachusett Reservoir, about 35 miles
west of Boston. Water is treated at the John J. Carroll water treatment plant at Walnut Hill in
Marlborough, Mass., and at the Ware water treatment facility in Ware, Mass., from where it is then
stored, conveyed, and distributed.
Wastewater from 5 400 miles of local sewers is transported into 228 miles of MWRA interceptor
sewers. The interceptor sewers, ranging from 8 inches to 11 feet in diameter, carry the region's
wastewater to two MWRA treatment plants: the Deer Island wastewater treatment plant and the
Clinton wastewater treatment plant. Though most of the wastewater flows by gravity, some low-lying
areas require pumping.
Since the time of formation of the Authority, MWRA has constructed over $6 billion of new potable
water and wastewater assets, with another US$1.6 billion planned for the near future (all costs in this
paper are quoted in US-Dollars). These facilities are operated by union staff, and have helped to
clean up Boston Harbour and Massachusetts Bay.
MWRA has a goal of limiting rate increases to its customer communities, which translates into
keeping operating expenses as low as possible, without compromising services vital to public health,
environmental protection, and the economy. To meet this goal, and as part of its investment in
assets, MWRA has established progressive asset management programs to ensure that facilities do
not fall into a cycle of disrepair.
In particular, MWRA has implemented a state-of-the-practice
reliability-centred maintenance program at its biggest new investment, the Deer Island Wastewater
Treatment Plant.

Background Information
MWRA completed the Deer Island Wastewater Treatment Plant in November 2001. Designed to treat
1.2 billion gallons/d, the plant provides preliminary, primary, and secondary treatment to wastewater



flows. The first phase of secondary treatment began operating in July 1997. Figure 1 shows the
major plant process components.

Figure 1: MWRA Deer Island wastewater treatment plant major processes

(copyright MWRA; used with permission)

After preliminary screening and primary sedimentation, pure oxygen from on-site plant is sparged into
the wastewater to enhance secondary treatment. Effluent is disinfected before it is discharged to the
receiving waters (Massachusetts Bay) through a 9.5-mile outfall tunnel bored through solid rock more
than 250 feet below the ocean floor. The tunnel's last 1.25 miles includes 54 separate release points,
being "diffusers. By extending into an area with water depths up to 120 feet, this outfall provides a
much higher rate of mixing and/or dilution than is possible with previous discharge locations in the
shallow waters of Boston Harbour.
Sludge from primary and secondary treatment is processed further in egg-shaped anaerobic
digesters, where it is mixed and heated, reducing sludge bulk and killing pathogens. It is then
pumped 7.5 miles to a pelletizing plant in Quincy, Mass., where it is dewatered, heat-dried, and
converted to a pellet fertiliser for use in agriculture, forestry, and land reclamation.
The Deer Island plant uses computerised systems to assist operations and maintenance
management, including a process information control system (PICS) and an operation management
system (OMS). PICS provides real-time operations data from systems throughout the plant (including



system status, flow, etc.). OMS correlates PICS data with laboratory analyses to monitor plant
process performance in meeting its discharge permit specification, issued by the U.S. Environmental
Protection Agency and the Massachusetts Department of Environmental Protection.

Description of best practices, including institutions (personnel/departments) involved

In 2000, MWRA commenced the Facility Asset Management Program (FAMP) for the Deer Island
treatment plant and related residuals and transport (pump stations) facilities to plan, manage, and
coordinate the engineering, maintenance, operation, and financing required to maintain these facilities
in accordance with regulatory requirements. The FAMP can be further described as having two
To cost-effectively replace the less durable capital components of the facilities at the
appropriate time to ensure reliable plant operation and preserve the value of the original
To prolong the equipment life and control the rate of replacement of major asset systems (i.e.
to avoid large spending spikes for consolidated retrofit or rehabilitation).
The FAMP concentrated on four areas: Maximo (a computerised maintenance management system),
maintenance strategies (that include the reliability-centred maintenance program), condition
monitoring, and business practices. The best practices for each area are listed below.

Maximo (computerized maintenance management system)

Deer Island has been using Maximo as its computerised maintenance management software since
1995. This software package is a powerful maintenance management tool that is used by the work
coordination group to manage all aspects of the Deer Island maintenance program. MWRA notes
that, Simply buying a CMMS does not constitute an asset management program. (Maintenance
Technology, March 2004)
MWRA conducted a post-implementation audit of Deer Island's Maximo database that included a
review of its data quality and its present utilisation.
Data quality
The audit concentrated on 1 250 pieces of plant equipment in a specific section of the plant, namely,
primary clarifier battery A. The review included a three-pronged approach where data were crosscompared between the Maximo database (equipment, inventory, preventive maintenance, and work
orders modules), field nameplate data, and technical information (operation and maintenance
manuals and process and instrumentation diagrams) located at the on-site technical library. As
expected, the audit concluded that the quality of data in Maximo needed to be improved. Equipment
records were missing from the database, not all equipment data was completed, and data conflicted
between Maximo, the field, and the library. A corrective action plan was initiated. To date, 95% of the
Maximo data has been corrected. New procedures have been developed and implemented to avoid
future data quality issues.
Utilisation of the Maximo System
The review revealed that Deer Island was utilising approximately two-thirds of the available Maximo
features (termed functionality) which reportedly is the case with most other maintenance
organisations. To enhance the data quality and support a new asset management initiative, a
consultant recommended the use of additional specific Maximo modules and programming
enhancements. A corrective action plan was developed as a result of the audit and the following
changes have been made to improve Maximo use:



Required fields for work orders were established to ensure data quality.
Maximo link to the financial software (Lawson) allows equipment costs to be determined,
spare part inventory to be viewed in Maximo, kitting of parts for preventive maintenance work
orders, and electronic requests made for non stock parts.
Maximo link to distributed control system allows electronic generation of preventive
maintenance tasks based on run-hours.
New reports were prepared to generate metrics.
New reports were generated to support scheduling work one week in advance. This effort
has been piloted in one area with success and is planned for application at the rest of the
facility in the next two years.

Maintenance strategies
During the construction of the new Deer Island treatment plant, each construction contract was
required to review the equipment operation and instruction manuals to identify the preventive
maintenance tasks to be completed, to protect equipment warranties. These preventive maintenance
tasks were entered into Maximo and then issued and completed by in-house staff. As a result of this
effort, approximately 39 800 preventive maintenance work orders were completed, and 84 000 h
expended during the 1999 financial year. Completion of preventive maintenance work accounted for
37% of all work hours.
Two maintenance initiatives were undertaken to improve the preventive maintenance program.
Reliability-centred maintenance was selected to review the critical plant systems, and a preventive
maintenance optimisation program was undertaken for all other plant components.
Reliability-Centred Maintenance (RCM)
A reliability engineer was hired to initiate a reliability-centred maintenance (RCM) program in order to
ensure that the correct maintenance was being completed for the Deer Island assets. The RCM
program concentrated on the most critical systems, considering both probability and consequence of
failure. A criticality analysis was completed that reviewed all systems on Deer Island, and this
identified approximately 100 systems that should have a complete RCM analysis.
The RCM program was initially piloted on 12 systems in the primary clarifier battery A. These
systems included the primary sludge pumps, long collectors, cross collectors, primary scum, hot water
system, sampling system, electrical power supply, fire protection, influent channel aeration, HVAC,
chlorine gas detectors, and sump pumps. The result of the initial pilot analysis was that preventive
maintenance hours were reduced by 25% through elimination of low-value preventive maintenance.
In addition, preventive maintenance activities increased for more critical equipment, and additional
condition monitoring techniques were introduced.
Implementation of asset management involved change, which met initially with resistance from staff.
Some employees feared that this could lead to privatisation or staff reduction. MWRA took the
approach of selecting volunteers to train and implement the RCM program during its development.
One mechanic volunteered, and he was given two weeks of facilitation training as well mentoring on
his first RCM analysis. As an RCM facilitator, his writing skills, presentation skills, and use of Maximo
increased dramatically. Consequently, he was first promoted to work coordination planner and
thereafter to maintenance manager. Others, seeing this opportunity to advance, have also
volunteered to be involved.



MWRA had completed 67 RCM analyses by January 2008, with another 30 planned for completion
during the next three years. The RCM program has not only been successful in improving the
preventive maintenance program for critical assets but also has improved staff sense of ownership of
equipment, improved operations and maintenance interrelations, improved staff knowledge of
systems, and increased staff skills.
Preventive maintenance optimisation
Preventive Maintenance Optimisation (PMO) is an effort to reduce the workload associated with
preventive maintenance of plant equipment without increasing the risk of failure to an unacceptable
level. Failure is defined as the inability of a component to perform its function(s) as defined by the
users of the equipment.
Machinery will always carry some risk of failure, regardless of the amount and frequency of preventive
maintenance devoted to it. In many cases, performing PM tasks, where not really needed, can
actually increase the chance of failure. A PM program is optimised when the least of amount of
resources are devoted to PM that will provide an acceptable level of risk of failure. If acceptable is
defined as the lowest overall costs associated with failure (to include corrective maintenance, safety
consequences, loss of operations, environmental damages/fines, etc.), then the PM program is
optimised when the total costs (PM costs + failures/CM costs) are minimised. This is depicted in the
following chart:

Figure 2: Optimisation of preventive maintenance (PM) costs

In many plants, Deer Island Treatment Plant included, the preventive maintenance program has been
developed using the recommendations issued by the equipment manufacturer.
Since the
manufacturer usually issues a warranty with the equipment purchase, the manufacturer benefits by
making conservative recommendations for maintenance of its product.
To implement the
manufacturers recommended maintenance to all equipment in a working plant, without consideration
of its working environment, operational frequency, available work resources, etc., would be extremely
taxing on the maintenance department and would waste valuable resources.



The intent of PMO is to review the existing maintenance program to:

Delete maintenance tasks that add little value.
Extend or reduce PM interval based on type of task, equipment usage, environment, and
industry guidelines.
Reassign tasks to operations that would be better served by having the operations
department perform the task during routine surveillance.
A PMO program was completed for all of the equipment that had not had an RCM analysis that had
previously been completed. The result was a significant reduction in the preventive maintenance
hours expended. There have been no adverse impacts on plant operation during the four-year period
following the PMO implementation. The plant equipment availability has increased, and the
maintenance backlog has remained within industry guidelines.
Maintenance resources that were freed up from performing low-value PM work orders were
reallocated to complete the more valuable Predictive Maintenance (PDM) tasks, supporting the RCM
effort, and completing corrective maintenance.
The implementation of the RCM and PMO programs resulted in a reduction of approximately 21 500
PM work orders per year and a reduction of 37 000 hours per year completed by maintenance staff. It
should be noted that some of the reduction in PM work orders and hours came from operators who
were performing light maintenance. Approximately 6 000 h per year of the reduction is attributed to
light maintenance being completed by operations staff.

Condition monitoring
Two full-time staff completed condition monitoring of critical assets, and more than 50 maintenance
technicians have been trained in vibration analysis, lubrication technology, precision laser alignment,
and acoustic ultrasonics. Condition monitoring techniques used include vibration monitoring and
acoustic spectral analysis, lubricating oil sampling, acoustic ultrasonic detection, ultrasonic thickness
testing, laser alignment, and infrared thermography.
Acoustic ultrasonic detection provides a prior indication of a potential equipment problem before
detectability by traditional condition monitoring techniques (vibration, oil analysis, heat). The potential
failure to function failure graph in Figure 3 illustrates this point.

Ball Bearing P - F Interval



Audible Noise
Failure (F)


Figure 3: Ball bearing potential failure to function failure graph

Normal corrective actions when vibration or ultrasonic acoustic readings are high include greasing,
alignment, or follow-up oil analysis. In some cases, the corrective actions do not result in lowered
levels, and so planning for equipment or component replacement is then recommended. The



acoustic ultrasonic and vibration results have been very important in keeping equipment availability
high and extending equipment life. In most cases, the corrective actions extend equipment life by
removing the causes of a bearing problem. If unacceptable symptoms remain after corrective actions
have been made, sufficient time is available to plan for equipment repair or replacement.
Acoustic ultrasonic case study
The primary scum pumps, motor, and bearings are monitored using acoustic ultrasonic detection to
provide advance warning of imminent failures. Using an SDT 170 ultrasonic detector, engineering
staff found unacceptable noises and noise levels in 10 of the 14 primary scum pump bearings,
indicating potential problems. As a result, samples of lubricating oil were taken from six of the pumps
gearboxes. Results from National Tribology laboratory tests showed viscosities that were much too
high, indicating that the wrong lubricating oil had been used. Oil in all the pumps was changed to the
specified oil, and ultrasonic monitoring was repeated. Thereafter, only two pumps had unacceptable
noise levels. Maintenance staff conducted an alignment check before considering replacement of the
bearings. Laser alignment checking of a machine with an Optalign Plus Laser revealed that a
coupling was in poor condition, and the machine was badly misaligned. A new coupling was installed
and aligned, and re-testing confirmed that the problem had been solved. These types of analyses
and troubleshooting are much less expensive than alternatives such as run-to-failure reactive
maintenance, and even more intrusive maintenance such as bearing replacement when a bearing, on
analysis, does not give any evidence of requiring replacement. In fact, full analysis showed that,
when applied to all 14 primary scum pumps, expending $ 3 920 in preventive maintenance resulted in
cost savings of $ 43 680 to $ 74 480. At the same time, equipment availability and reliability was
increased. Condition monitoring techniques used at Deer Island have provided many such examples
of the benefits and return on investment through proactive maintenance.
Lubrication savings case study
The secondary reactors have 36 mixers and 36 aerators that have Lightning triple-reduction
gearboxes. Twenty gearboxes have failed during the past 10 years, each requiring a rebuild. The
cost to repair each gearbox is $ 25 000 to $ 40 000, depending on the extent of gear damage. Thus,
the oil sampling program is important to extend this equipment's life. An oil sampling program was
initiated in 2002 for these components. Savings have been made by skipping unnecessary oil
changes that were previously scheduled and completed yearly for all gearboxes. The results of the
sampling program have been:
First year of sampling (2002)
The first round of sampling identified 48 of the 72 gearboxes with oil in good condition, which did not
require an oil change. This resulted in cost savings of $ 39 610.
Second year of sampling (2003) Introduction of oil filtering
The second year of sampling identified 39 gearboxes with oil in good condition and 20 gearboxes that
required oil filtering to remove contaminants and wear particles, for a total of 59 avoided oil changes.
This resulted in cost savings of $ 42 545 (~ $16 000 through oil filtering).
Third year of sampling (2004)
The third year of sampling identified 43 gearboxes with oil in good condition and 13 gearboxes that
required oil filtering to remove contaminants and wear particles for a total of 56 avoided oil changes.
This resulted in cost savings of $ 41 190 (~ $ 10 000 through oil filtering).
Fourth year of sampling (2005) Commenced laser alignment of gearboxes
The fourth year of sampling identified 45 gearboxes with oil in good condition and 20 gearboxes that
required oil filtering to remove contaminants and wear particles for a total of 65 avoided oil changes.
This resulted in cost savings of $ 45 225 (~ $ 16 000 through oil filtering).



Fifth year of sampling (2006) Improved breathers installed to prevent contamination

The fifth year of sampling identified 34 gearboxes with oil in good condition and 25 gearboxes that
required oil filtering to remove contaminants and wear particles for a total of 59 avoided oil changes.
This resulted in cost savings of $ 40 876 (~ $ 19 000 through oil filtering).
Sixth year of sampling (2007) Sampling ports installed to improve sampling representivity
The sixth year of sampling identified 45 gearboxes with oil in good condition and 20 gearboxes that
required oil filtering to remove contaminants and wear particles for a total of 65 avoided oil changes.
This resulted in cost savings of:
Oil savings = 65 x 32 gallons/gearbox x $ 23.44/gallon = $ 48 755
Saved labour = 4 hours/oil change x 45 x $ 30/hour = $ 5 400
Sampling Cost = 124 samples x $ 30.15/sample = ($ 3 739)
Sampling Labour = 124 samples x 1 h/sample x $ 30/h = ($ 3 720)
Total Savings = $ 46 696 (~ $ 16 000 through oil filtering)
Consequently, since December 2002, savings of $ 256 142 have been realised from reduced oil
usage and saved labour required for oil changes. In addition and more importantly, oil was changed
on the basis of the oil condition as indicated by oil sampling and analysis programmes since 2002,
resulting in extended gearbox life.
Infrared case study
The condition monitoring group tested 21 steam traps in the Deer Island Thermal Plant. Testing of
steam traps by the use of an infrared camera (shown in Figure 5) can show if the steam trap is cycling
on and off properly, or if the steam trap is defective.

Figure 4: Infrared camera in use at Deer Island

(FLIR Model E-4)

The initial testing of the steam traps found 19 steam traps operating properly and two malfunctioning
bucket-type steam traps. Figure 5 shows an infrared image of the bucket trap showing a hot
temperature on both the inlet and outlet of the steam trap. A bucket-type steam traps function is to
drain condensate formed in steam lines. When sufficient condensate forms in the trap, the bucket
lifts, condensate is drained, and the bucket lowers to isolate the steam from the drainage port. In the
case shown below, the trap remained continuously open and steam was continuously passing
through the trap. Both bucket-type steam traps were replaced, resulting in reduced steam leakage,
with concomitant energy savings.



Figure 5: Infrared camera images

Vibration case study

In October 2004, a lubrication oil pump for the steam-turbine generator began to show signs of
imbalance at the free end of the motor. A work order was created to clean the cooling fan. The
vibration decreased slightly, before increasing again. The vibration spectrum for the motor is shown
in Figure 6.

1X peak means imbalance

Figure 6: Motor vibration spectrum sample

The condition monitoring staff then examined the motor cooling fan as shown below and found drywall screws had been used to balance the motor, and it appeared that one screw had dislodged as
shown in Figure 7.



Cooling fan
This arrow
points at the
wrong screw

Free end of the


Figure 7: Cooling fan motor

The cooling fan was replaced, and vibration returned to acceptable levels. This corrective action
extended equipment life by relieving the imbalance from the bearing load.
Overall, the condition monitoring program has been very successful in locating potential problems
with equipment and identifying corrective actions. The program has contributed to improving the plant
equipment availability and longer equipment life.

Business practices
Productivity improvement program
Several changes to all trades and operations job descriptions were negotiated with the unions in order
to improve the productivity of the maintenance program. One goal was to implement cross-functional
crews to break down trade silos, improving teamwork and speed-work completion. In addition, light
maintenance tasks were included in the job descriptions of all trades and operation staff. Pay
increases over a three-year period were provided along with training and testing of staff in the new
skills necessary to perform these light maintenance tasks.
Cross-functional crews
Mixed crews were implemented in all plant areas In March of 2002. A mixed crew includes
electricians, I&C, M&O Specialists, and plumbers, all lead by a single unit supervisor. The mixed
crew format increases efficiency by reducing downtime in waiting for specific trades to support a multidiscipline work order. In addition, light maintenance such as HVAC filter changes, light bulb
replacements, and lubrication are now assigned to an area-maintenance team and/or operations, and
not to a specific trade, This has allowed the unit supervisor in the maintenance group a flexibility to
assign these tasks to any one of a number of staff, based upon their availability, the priorities of the
week, and the day. Non-licensed piping and HVAC equipment repairs are not only assigned to the
plumbers and HVAC technicians but also to M&O specialists as well. As a result of this flexibility, over
20% of work is now accomplished by an alternative trade.
Operations light maintenance
Preventive maintenance (PM) tasks prior to this program were all completed by maintenance staff.
Operations were assigned, and Maximo work orders provided for inspection and lubrication tasks.



Operations currently completes >16% of all PM hours (>500 hours/month). The best in class goal
accepted by the Society of Maintenance and Reliability Professionals (SMRP) is 10-15% of all PM
hours, and this goal has been met or exceeded at Deer Island since 2003.
As part of the overall program, metrics were developed to promote change through tracking the
implementation of the new maintenance initiatives, and to monitor the impact of the new programs on
maintenance performance. Monthly and yearly metrics were developed.
A task team was formed to review the existing maintenance metrics and recommend new metrics.
Metrics from maintenance text books, papers, and the internet were researched and reviewed for the
Deer Island Treatment Plant.
Some key metrics that are used to determine the impact of the maintenance program are
maintenance backlog and availability.
Maintenance backlog

Backlog is determined by totalling the estimated work hours for "in progress" work
orders and dividing this by the available staff hours each week. To determine this
metric, estimated hours are up-loaded in Maximo by planners for all work orders.


This metric is helpful to gauge whether there are adequate staff to complete the
maintenance work. Increasing backlog can be attributed to increased equipment
problems, or a decline in maintenance efficiency. The industry guideline is to have a
backlog of 3 to 6 weeks.


Availability is the amount of critical plant equipment that is available for service. At
Deer Island, operations issues a daily list of approximately 350 equipment items that
are available to treat the maximum plant capacity of 1.27 billion gallons per day. The
normal plant flow is 360 million gallons per day.


This metric is used to determine if plant equipment is being properly maintained and
that operations are not impacted from equipment being out of service. The industry
guideline is 97%. Deer Island has met or exceeded this goal since 2003, and in
2007, reached 99.2% availability.

Summary of key lessons and experiences (results and findings)

A focus on identifying critical assets and implementing predictive and reliability-centred maintenance
has saved MWRA money as well as enhanced plant reliability. Approaches of this kind that result in
change in an organisation will take time to implement, but the returns make the effort worthwhile.



Seattle Public Utilities (SPU) takes aim at critical assets using risk-based
decision-making and strategic asset management plans - City of Seattle
(Washington State)
What does asset failure mean to your utility? For Seattle Public Utilities (SPU), unanticipated asset
failure is a problem that they hope will be a thing of the past. A sudden sewer collapse back in 2002
has been one of the motivators for organisational and practice changes that have led to the
development of a world-class asset management program.
Terry Martin, wastewater and drainage lead in the office of strategic asset management at SPU, notes
that not only was the repair costly, but a nearby hospital was also entitled to financial compensation
(Martin, 2005, p.14). The last time the pipe had been inspected was in 1992. Since SPU used a timeinterval based condition inspection program, the routines in place at SPU did not have the pipe
scheduled for re-inspection for another 20 years. In retrospect, the long period between pipe
inspections made little sense, as this particular pipe formed a portion of one of the oldest sets of
infrastructure in the city of Seattle, was subject to high combined sewer and storm-water flows, and
was adjacent to a hospital.
SPU has since implemented a comprehensive asset management program to help pre-empt future
failures and limit the degree of consequential damage. The program has focused on establishing
levels of service, understanding business risk, and developing business cases for asset investment.
They have looked to some of the worlds best asset managers to help develop their asset
management program. This case study describes some of the key elements of the organisational and
practice changes that have taken place at SPU in support of their chosen asset management path.

Seattle, WA, the Emerald City

Located in the State of Washington, approximately 113 miles
south of the US-Canadian border, the city of Seattle is situated
at sea level between Puget Sound and Lake Washington. As
the largest city in the Pacific Northwest, Seattle is often called
the Emerald City, referring to the lush trees found within the
city. Surrounded by water, Seattle is 142 square miles, of
which 40%, or 58.6 square miles, is water. The highest point of
elevation in the city is 520 feet.
The city of Seattle has two utility departments. Seattle City
Light provides electricity, and Seattle Public Utilities (SPU) provides water, sewer, storm-water, and
solid waste services.
SPU also provides drinking water services to several neighbouring
communities, and manages its own internal engineering services to support SPU-related projects and
SPU operates and maintains a variety of assets including, but not limited to, the following for provision
of clean water (Kelly, January, 2008):
Four dams and two head works facilities.
175 miles of large diameter transmission pipelines.
1 700 miles of water mains.
175 000 metered services.



17 000 valves.
29 water pumping plants.
16 reservoirs.
16 elevated tanks/standpipes.
SPU also owns two major drinking water treatment facilities. These are operated and maintained by
private contractors. In addition, there are more than a dozen small water chlorination facilities
throughout the city.
In the storm-water and wastewater arenas, SPU is responsible for maintaining the assets listed below
(2005-2010 Adopted Capital Improvement Program, p.457 and Input from Elizabeth Kelly, January,
2008). Note that wastewater treatment services are provided by King County:
582 miles of sewage system pipelines.
482 miles of drainage system pipelines.
905 miles of combined system pipelines.
68 pump stations.
92 combined sewer and pump station outfalls.
38 combined sewer overflow control detention tanks/pipes.
140 miles of ditches and culverts
SPU also manages assets related to providing solid waste management services.
In total, the annual operating budget of SPU is approximately US$ 600 million with assets totalling
approximately US$ 4.5 billion (Kelly, Commitment to Customer Service January/February, 2008).
(Currency in this paper is US$)
The main goal of SPU is to meet the service expectations of their customers (Seattle Public Utilities
2007-2008 Strategic Business Plan, p.2, 2007). At the same time, the utility strives to balance the
provision of value with life-cycle costs, while considering financial, environmental, and social factors.
These three factors form the basis of what is known as triple bottom line, a concept that is gaining
popularity due to public interest in sustainable practices. In order to achieve their vision, SPU
recognises the importance of a world-class asset management (AM) program and has gone to great
lengths to establish a program that incorporates the values and goals of the organisation as part of
the decision-making process.
This case study focuses on Seattle's journey in asset management towards a more strategic focus,
with particular emphasis on the incorporation of a risk management-based approach. Using the
management of sewer pipe assets as an example, the case study illustrates how SPU uses a riskbased decision process to determine the assets that are the most important in sustaining levels of

Asset management background

Providing water utility services is very infrastructure intensive. The challenges include rate pressure,
public scrutiny, an aging infrastructure, and environmental sensitivity, which all become part of public
service delivery (SPU Asset Management Framework, 2007, p.3). SPU began planning and



implementing a strategic asset management program in 2002. As explained in SPUs Asset

Management Framework, quick wins are attributed to the initial approach that SPU chose for asset
management development. Instead of implementing a complex asset management program at the
outset, executives at SPU instead decided to adopt an early gains approach. This decision resulted
in the development of a core AM philosophy, implementation of key AM elements, and a focus on
early gains, to build staff understanding and confidence. This has proven to work very well.

Asset management an evolutionary path

Development of the asset management program in use today by SPU represents five years of effort,
including realignment, modifications, trial and error, piloting, review, data acquisition and refinement,
analysis, collaboration, and more. A great degree of effort has been devoted to:
Creation of a formal asset management document that identifies the overall framework.
Implementation of an organisational model that distinguishes specifiers from service providers
(see below for explanation).
Establishment of an Asset Management Committee (AMC) and other decision-making
Development of a process for analysing and documenting strategies for management of
assets (Strategic Asset Management Plans, or SAMPs).
Development of the Project Development Plan (PDP) process and the Project Development
Plan Guidebook called the Quick Start Guide to explain the PDP process.
Documentation of the triple bottom line process (wherein they consider social, environmental,
as well as financial costs, benefits, and risks) in a Triple Bottom Line (TBL) Guidebook.
Establishment of agreed-upon service levels.
Establishment of a corporate wide-risk management framework.
Establishment of a quarterly performance indicator report that helps SPU understand how
well theyre doing in the provision of service, and to help focus the improvement efforts.
As part of the asset management development process, concepts were transferred from Australia to
SPU by having utility managers from these locations work directly with SPU staff, as part of a staff
exchange program. With the Australians assisting, SPU was challenged to think outside the box
and develop a unique program that would better prepare SPU for the future. A component of this was
establishment of the specifier-provider model.
Specifiers at SPU are accountable for proposing service levels, ensuring that the organisation meets
service levels, negotiating and meeting regulatory requirements, developing business cases for
investments, and determining the best balance between capital expenditures and maintenance
activities. Service providers are responsible for implementation of various capital, operational, and
maintenance programs, including the day-to-day activities of equipment maintenance and operations.
Service providers become very good at achieving results in their specific area, and their opinions and
perspectives are very valuable to the specifiers. Service providers also have a critical role at the
customer interface.



SPU Strives for a customer-centric approach

There has been even greater awareness in recent times at SPU to adopt a customer-centric approach
to asset management, rather than focusing totally on asset condition. Liz Kelly, director of the office of
strategic asset management for SPU, explains:
SPU is creating a new paradigm for how staff and management make decisions. SPU
employees routinely ask: What is the best way we can meet the needs of our customers?
What would our customers think about this decision? Are we providing equitable services
across all of our communities?
Commitment to Customer Service January/February, 2008
The models and guides that form part of SPUs asset management program encourage these types of
customer considerations when making decisions (see Project Development Process Quick Start
Guide, p.14). The driver for project initiation and implementation for SPU and many other utilities
hinges on maintaining customer service levels and keeping within regulations. It is important,
therefore, to ensure customer service levels are set accurately and with care., SPU created an interim
service level project team and a service level steering committee to facilitate the proper assessment
and determination of service levels. Members of these teams met regularly to review near-term and
long-term service level and performance indicator work. The results of the service level project team
and steering committee not only provided information on customer satisfaction but also provided data
regarding customer willingness to pay.

The Strategic Asset Management Plan (SAMP)

SPUs approach to determining which assets are critical to sustain performance is very structured. It is
planned, documented, and put through an iterative process of refinement. There is less emphasis on
critical assets and more emphasis on risk-based decision-making. SPU has a specialised team
responsible for crafting Strategic Asset Management Plans (a.k.a. SAMPs) for various categories of
assets,. for the purpose of consolidating the asset management information, strategy and approach
for a group of assets,. SAMPs are created by a small group of about six experienced individuals who
have been practicing SAMP development for several years. The current definition of a SAMP as
approved by the SPU Asset Management Committee (AMC) is as follows:
A Strategic Asset Management Plan (or, SAMP) is a short-term planning document that
guides our management of a category of assets to meet defined objectives.
Management of assets in this sense incorporates operations, maintenance, and capital
activities and investments. SAMPs assess the current status of a category of assets by 1)
characterising asset attributes such as type, size, material, and age, 2) listing relative service
levels and regulations, 3) describing current operations, maintenance, and
rehabilitation/replacement strategies, and 4) characterising the risk associated with SPUs
operation and ownership of the assets. SAMPs then characterise our risk tolerance for the
asset category and define our program for controlling risks associated with ownership and
operation of assets. Once approved by the Asset Management Committee (AMC), SAMPs
become guiding documents that may feed into our data management strategies, our condition
assessment strategies, our operations strategies, our maintenance plans and strategies, our
renewal and replacement strategies, and our CIP planning processes. Such strategies are
referred to as tactical plans and are an integral component of SAMPs. SAMPs also have a
secondary value in centralising information related to an asset category and creating
transparency regarding maintenance, operations, and renewal/rehabilitation strategies.
SAMPs have owners (i.e., individuals responsible for their preparation and updating), but
are developed in a collaborative manner among Corporate Asset Management (CAM),
resource planning, strategic Operations, and others, as is appropriate. SAMPs and
comprehensive plans function collectively to set the strategic framework for SPUs
management of assets.
SAMP Definition, Approved by AMC on Dec 15, 2004



A list of SAMPs that have been completed by SPU includes the following asset categories:
Wastewater and water pump stations.
Wastewater collection pipes.
Wastewater structures including:
o Combined overflow structures.

Underground detention valves.


Tide gates.

Drainage collection pipes.

Water distribution pipes.
Water service pipes.
Water pump stations.
Water reservoirs and tanks.
Water valves.
Water hydrants.
Water treatment facilities.

How SPU makes well-informed, risk-based decisions

SPU calculates the annual risk cost of failure in order to make informed risk-based decisions. The
formula used to determine this is:
Risk cost of failure = consequence of failure x probability of failure
Terry Martin of SPU describes the process of determining both sides of this equation in detail
as it relates to the practices followed by SPU for wastewater collection pipes (2005, p.14 &
16). He also explains how SPU identifies pipes of greatest concern with the use of risk-based
decision-making tools.
Consequence of failure
SPU has divided the consequence of failure into three main subgroups:
1. Baseline generic financial costs for repair of an asset failure: This refers to the
hard cost of repairing an asset. In order to arrive at these estimated costs, SPU
uses historical cost information for similar assets and asset attributes (i.e. diameter of
pipe, material of construction, and depth to pipe) to extrapolate to an approximate
2. Location-specific factors that increase the financial cost of an asset failure:
Depending where the asset is located, specific factors can be accounted for that will



increase the financial cost of performing repair work for a given asset failure. This
might include location under a bridge, location under railroad tracks, location under a
body of water, etc.
3. Location-specific factors which increase the environmental and social costs of
an asset failure: Location-specific factors can not only increase the financial cost of
an asset failure but can also increase the environmental and social costs. For
instance, an asset failure located in the heart of the city will not only increase the
repair cost, due to necessary traffic control, but will also cause social disruption.
Martin discusses other intangible (i.e. non-financial) costs that come into
consideration including potential damage to public health, environmental damage
potential, unfavourable publicity potential, and property damage potential.

Probability of failure
SPU has developed a model for predicting the probability of failure that uses a two-pronged approach.
The first consideration is location-specific cost multipliers that will increase the probability of failure.
For wastewater collection pipes, such factors might include:
Steep slope (H2S generation, causing material degradation).
Proximity to industrial, commercial and institutional (ICI) facilities that release substances that
may affect the life of a pipe.
SPU has taken the initiative to create a program that interfaces with the citys GIS system to apply an
appropriate factor of failure probability, based on such location-specific cost multipliers.
The second way the probability of failure is determined is based on the age of each asset and
susceptibility to material degradation. SPU has created predictive failure curves that identify the
expected failure rate of a pipe, for instance, based on its material composition and age. Martin
explains that this in-house model is capable of applying a unique failure curve to each pipe that is
identified in the system. SPU continues to update the model, as more empirical information becomes

Calculating annual risk cost

Once the unique consequence of failure and the unique probability of failure are known for a given
asset, SPU defines the product of these as the total risk cost of failure. When considering which
assets are of highest risk, there are a few other factors that need to be considered, according to
Martin. One factor is the cost to perform a condition assessment. Therefore, the life-cycle cost of the
condition assessment is calculated and taken into account. Another factor might be the cycle time
required to perform a condition assessment. For pipe assets, SPU has estimated the cycle time for a
CCTV assessment to be approximately five years.
Determining which assets are high risk is done by comparing the calculated risk cost of an emergency
failure repair for a given asset versus the Net Present Value (NPV) of a scheduled repair, plus the life
cycle cost to perform condition assessment. If the risk cost of an emergency failure repair is greater
(i.e. the benefit/cost ratio is greater than one), then the asset is deemed high risk and an initial
condition assessment (i.e. CCTV video inspection) is performed on the asset.

Advice to other utilities seeking to implement an asset management program

When asked what advice he would pass along to other utilities that are trying to develop an asset
management program, Terry Martin offers the following thoughts (phone interview held 18 February,



The first relates to data management. Asset management helps to obtain better data.
Strategic Asset Management requires reliable asset data. In order to implement a good asset
management program, a great deal of effort must be put forth to get existing data organised
and accurate, and to ensure that new data are accurate, and correctly captured. A GIS
system has proven to be invaluable for SPU in acquiring accurate asset data. However,
existing GIS data at the outset had numerous errors which required correction.
An important aspect of asset management is change. Martin specifically addressed the
difficulty associated with encouraging a culture of change within SPU. According to Martin,
asset management will challenge many people within the organisation to do things differently
from what they are used to doing, and it is likely that some individuals in the organization will
not be on-board with the changes that need to happen. Some just do not want to change. An
example is documentation. Asset management involves documenting how things are done.
This is seen as a threat to job security by some employees.
Finally, Martin suggests that utilities consider asset risk rather than criticality. By shifting ones way of
thinking more to the apparent level of risk, rather than criticality, Martin has found that some sneaky
things tend to become apparent that would otherwise not be realised. A very simple example of this
is as follows:
A fire hydrant is a critical asset. It must be located in certain areas within the city in case of
fire. From a criticality perspective, you will consider the locations where the hydrant needs to
be located to mitigate risk. However, from the same perspective, you may not consider the
risk associated with a fire hydrant that is not painted correctly. Risk-based decisions may
take into account that the hydrant will degrade sooner, may not be suitable for its intended
use, and may need to be replaced earlier than the anticipated life expectancy period dictates.
It will also address the importance of repainting the hydrants over time.

Conclusion: The benefits of asset management for SPU

After reviewing the information presented by Seattle Public Utilities with regard to the overall asset
management program, it is clear that a very concise and well thought-out program is taking shape. In
fact, as an exercise, it is interesting to note the many advantages that arise from the program in use
today. These include the following:
1. A better understanding of the cost of doing business. This results in more informed capital
improvement project (CIP) decisions.
2. An opportunity to include rate-payer input when making CIP decisions, as the cost of
business is better defined. Approximate rate increases for a given level of service can then
be proposed to rate-payers, for instance.
3. Assets of greatest risk become better defined and targeted as the main areas of focus. This
allows for efficient allocation of asset management resources (i.e. maintenance staff).
4. As more empirical data become available in future, SPU continually improves and refines
assumptions, information, models and data that form part of the decision making process.
5. An asset management program that encourages a customer-centric mentality and approach
to decision-making with the use of models, guides, agreed-to service levels and culture
6. Greater accuracy in predicting asset failure before failure occurs.
7. Decisions that are made with consideration of the triple bottom line (i.e. considering social,
environmental, and economic factors). Many utilities have still not formally embraced TBL as
part of their decision making process.



8. Various initiatives that can be integrated to foster sound decision making. As an example,
SPU leverages its geospatial data to produce risk-based charts and graphs for assets
management purposes.
9. Information and processes required to make a decision are well-documented and available to
all staff as user-friendly models, often with associated templates and guidebooks.

Martin, T. (2005) Modeling System Leverages GIS to Assess Critical Assets. Waterworld Magazine,
April 2005, pp. 14 and 16.
Kelly, E. (2008) Commitment to Customer Service Setting Service Levels for Asset Management
Decisions. Underground Infrastructure Management, January/February 2008, [online] Available from:
www.uimonline.com/index/cover [Accessed 3 March 2008].
UNITED STATES OF AMERICA Seattle Public Utilities Asset Management Committee (2004) SAMP
Definition, (Approved by AMC on Dec 15, 2004), Seattle, Washington.
UNITED STATES OF AMERICA Seattle Public Utilities Asset Management Division (2007) Seattle
Public Utilities Asset Management Framework, February 2007, Seattle, Washington.
UNITED STATES OF AMERICA Clarke, C. & White, S. (2007) Essential Elements of the Strategic
Business Plan. In: Seattle Public Utilities 2007-2008 Strategic Business Plan, Seattle, Washington.




Decision-making for capital investments - Seattle public utilities case study:
Seattle Public Utilities operates water supply, wastewater collection, drainage and solid waste
collection utilities for the City of Seattle and surrounding communities in King County, Washington
State, USA. The organisation collectively serves a population of approximately 1.5 million people.
Seattle Public Utilities owns assets valued at over US$ 4.5 billion (year 2007), including piping
systems, large water reservoirs, water treatment plants, pump stations, roads and buildings.
Seattle Public Utilities has put in place an ambitious program to manage these assets since 2002.
The program now permeates the organisations decision-making processes. Although Seattle Public
Utilities program has many noteworthy features, this case study will focus on the utilitys
organisational structure and on their process for defining, reviewing and approving capital projects,
including application of a procedure for triple bottom line evaluation of costs and benefits. The triple
bottom line addresses financial, social and environmental considerations in decision-making.

Seattle Public Utilities launched a major effort to apply asset management principles, beginning in
2002. There were several drivers for this program, including:
Concerns over the organisations financial position.
Increased needs for spending on capital projects, operations and maintenance.
Significant regulatory requirements in each of the utilitys service lines.
Continued aging of infrastructure.
Public interest in environmental protection.
Seattle Public Utilities forged contacts with several utilities that had developed extensive asset
management programs. These programs were adapted to meet Seattle Public Utilities own needs,
regulatory environment and administrative context. In order to develop its program, the utility
engaged in exchanges with Hunter Water Corporation in Australia that enabled staff from each utility
to cross the Pacific for extended on-site interactions.
Seattle Public Utilities developed a number of guidebooks and other materials posted on its intranet
system to support the decision-making process. Those staff members charged with contributing to
the process have access to detailed procedures, examples and information.

Organization and Management

Seattle Public Utilities gave careful consideration as to how the organisational structure and
management processes could support the program effectively in building its asset management
program. A reorganisation was undertaken under the leadership of the utilitys director, in part, to
advance this objective. One of the basic organising principles was the definition of two distinct roles:
specifiers and service providers.



Specifiers plan, specify and are accountable for the delivery of utility and corporate services. They
are responsible for making sure Seattle Public Utilities establishes and meets service levels,
consistent with financial constraints and life cycle principles. In addition, specifiers are responsible for
ensuring that asset management principles are applied in making or recommending resource
allocation decisions.
Service providers deliver the services defined in negotiated service agreements. They are
accountable to the specifier for producing all agreed deliverables and meeting the agreed-upon
scope, schedule, budget and performance requirements. Service providers also work with specifiers
to determine appropriate work objectives, outcomes and/or options.

Table 1.1: Roles of specifiers and service providers

Table 1.1 further describes these roles.

An Asset Management Committee (AMC) was formed, comprised of senior management from across
the organisation. The AMC has review authority over service levels that are formally defined, and
monitors key performance indicators on a quarterly basis. The AMC also provides funding approval
for all capital investments, and provides direct approval for all projects costing over US$250,000. The
AMC has delegated funding approval to other committees and individuals for projects less than this
An asset management group was also formed within the directors office to support the program,
provide cross-functional integration and assist with translating concepts into actual operations. This
group provides the following functions:
Reinforces and institutionalises asset management principles and advises senior



Creates consistent processes and accountabilities.

Supports specifiers in developing business cases and other products.
Contributes analysis to optimise management of assets.
Over the course of the past several years, the group has also had other functions, including:
Conducting economic analysis.
Developing performance management systems.
Assisting with development of service levels and service agreements.
Coordinating between utility lines of business and other branches of the organization.
Establishing a facilities assets management group.
These functions support asset-management planning and decision-making throughout the
organisation, including assistance in developing the project development plans described below.
Seattle Public Utilities views this as a process of culture change within the organisation. Leadership
from the director and senior management has been essential in accomplishing this change. It is a
multiple year process that hinges on effective communication of objectives and clear definition of roles
and expectations throughout the entire organisation.

Capital decision-making at Seattle Public Utilities

The capital decision-making process includes a well-defined format for developing and presenting
information, a structured review process involving the organisations senior management, explicit
consideration of risk factors, and consistent application of triple-bottom line methodology. This
section presents those elements and provides an example of how a project was analysed.

The project development plan

The capital decision making process is highly structured to ensure sound decisions are made using
the best available information, consistent with Seattle Public Utilities policies. At the heart of this
process is the Project Development Plan, recorded in an undated and unpublished internal document,
"Seattle Public Utilities, Quick Start Guide, Project Development Plans". A project development plan
contains a number of elements including:
Project background and objective of the capital project. The objective includes two elements:
1) the function of the project in terms of the problem or opportunity that it will address; 2) the
type of value created by solving that problem or realising the opportunity.
Discussion of the base case without the proposed project, and of other options for achieving
the objective. Among other purposes, this discussion enables comparison of more capitalintensive solutions with solutions that have higher operational or maintenance costs.
Economic analysis of the project options. Cost-effectiveness analysis is sufficient if the
function is an absolute requirement (e.g. for regulatory compliance). For all other projects, a
more extensive analysis is required to compare costs and benefits. Both types of analysis
utilise full life-cycle costs. In addition, the cost-benefit analysis uses a triple bottom line
evaluation and a review of how sensitive the results are to differing assumptions.



Identification and characterisation of risk factors for each option, including the no-action
option. Risk factors are quantified where possible (e.g. range of effects, number of people
affected, etc.). In addition a risk cost is estimated and incorporated in the cost-benefit
Recommendation of a single project option, together with budget impacts and recommended
implementation schedule.
Seattle Public Utilities has developed standard templates for the analysis and Project Development
Plan document to provide a common basis for the many staff involved in preparing and reviewing

Participants in capital project decision-making

The process for developing and reviewing project development plans is also highly structured. The
Project Development Plan (PDP) is initiated and directed by a specifier. Each PDP is assigned an
executive sponsor. The written PDP records the problem statement and business case for the
proposed investment. In the process of developing the problem statement, the business case, and
the PDP, the specifier receives assistance from staff representing different functions within the
organisation. These include a financial performance manager, field operations liaison, economist and
a corporate asset management reviewer. Other reviewers may also play a part, such as scientists,
security specialists and risk analysts. Each of these contributors has a defined role in shaping the
Once prepared, all project development plans for investments exceeding US$ 250 000 are reviewed
by Seattle Public Utilities Asset Management Committee, comprised of senior management within the
organisation. The committee first reviews the PDP prior to the preliminary engineering stage. If it is
approved, the committee reviews an updated and more detailed version of the PDP again prior to the
design and construction stage.

Triple bottom line evaluation

Economic analysis of proposed projects is structured to address three categories of costs and
benefits: financial, social, and environmental. These three categories have become known as the
triple bottom line.
Traditional analysis of capital projects by public agencies previously focused primarily on the financial
component and functional benefits internal to the utility. In contrast, the triple bottom line evaluation
incorporates social and environmental effects in the project evaluation. This can be challenging, as
costs and benefits in the social and environmental arenas are inherently hard to define and quantify.
At Seattle Public Utilities the economist supporting each PDP assists the project specifier to identify
and evaluate these effects.
Guidance on cost/benefit analysis at Seattle Public Utilities directs staff to include both internal and
external costs and benefits. Internal costs are those incurred by ratepayers of Seattle Public Utilities
in carrying out the project. Internal benefits are those achieved for Seattle Public Utilities customers.
External costs and benefits are those affecting other parties, such as the public at large, other city
departments, other jurisdictions or Indian tribes, as well as the natural environment. This
comprehensive analysis enables Seattle Public Utilities to make well-informed choices that support
efficient allocation of both its own resources and societal resources, and also to address equity issues
in its decisions.
Seattle Public Utilities develops a risk signature for each project, which is also incorporated in the
evaluation. Risk analysis may be used to quantify risks in any of the three categories: financial,
social, or environmental. A risk cost is developed for all risk signatures and included in the



cost/benefit comparison. Problems assigned a high or critical risk signature will require more
thorough analysis, as well as consideration of risk mitigation strategies.
Seattle Public Utilities uses various techniques to quantify costs and benefits, and to translate them
into monetary values. Monetisation techniques must be matched with the type of cost or benefit
involved. Techniques include direct market valuation; indirect market valuation, such as hedonic
estimation, travel-cost methodology and other techniques; and contingent valuation using surveys for
non-market values. Risk cost is also monetised using an analysis of possible outcomes and the
probabilities of those outcomes.
Not every project warrants microscopic examination of all costs and benefits. The level of analysis
must be matched with the cost of the investment, and the value of the information needed in the
project evaluation. For example, the analysis in some cases can be completed, and decisions made
without converting particular costs and benefits to monetary values. This simplifies the procedure

Project example
One of the many projects analysed in recent years using this procedure was the proposal to modify
temporary pumping facilities, used in dry years, to utilise inaccessible storage capacity within Seattle
Public Utilities Morse Lake impoundment. Under normal operating conditions, the water level in
Morse Lake fluctuates between 1 532 feet (466.9 m) and 1 563 feet (476.4 m) above sea level. Water
stored below 1 532 feet (466.9 m) cannot be accessed using gravity flow. Seattle Public Utilities
operates two sets of pumps mounted on barges to draw water below 1 532 feet (466.9 m). However,
this practice raises problems for a number of reasons. Commissioning of the pumps requires
significant lead time, and it is difficult to know in any given year whether they will actually be needed.
As a result, mobilisation of the pumps, with significant costs, is necessary, even in those many years
when final water supply conditions have not required the use of the inaccessible storage capacity in
Morse Lake (Figure 1.1).

Morse Lake Minimum Elevation (Feet)

Notice to

1544 1.1: Expected frequency of minimum water surface elevations, Morse Lake Reservoir
Minimum Water Surface Elevations and

Their Probabilities Given 150 MGD

Average Annual Demand




0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 24% 26% 28% 30% 32% 34% 36% 38% 40%

Note: 150 MGD (million gallons per day) = 568 M/d(megalitres per day)
Figure 1.1: Expected frequency of minimum water surface elevations, Morse Lake Reservoir



There are several other risks associated with the current system for tapping inaccessible storage
capacity. Fuel storage needed for pump generators poses risks to water quality. The use of bargemounted pumps poses inherent hazards on a large lake situated in a mountainous area where high
winds can occur. Moreover, the pumps themselves are aging and could fail at a time when they are
Given this situation, Seattle Public Utilities has examined a range of alternatives.
consideration of nine options, the following four were analysed in the most recent PDP:
Option 0:
Option 1:
Option 5:
Option 7:

After initial

Retain the status quo

Improve the existing system
Land-based pump station and land discharge
Submersible pumps with underwater discharges to existing dyke.

Costs and benefits of each option were analysed, and net present value was calculated to allow for
comparisons. Benefits that were key to making a choice among options included the avoidance of
false mobilisation costs, and reductions in risks. In addition, as a result of the risks associated with
the current system, Seattle Public Utilities water managers are forced to call for voluntary curtailment
of water usage much more frequently than they would with more reliable pumping facilities.
Therefore, consideration of the value of reducing unnecessary curtailments became a key factor in
the cost/benefit analysis. This value hinges largely on social costs imposed on the public during
curtailments, a consideration that is captured in the triple bottom line methodology.
Figure 1.2 shows how costs and benefits of the three action options (excluding status quo) compare,
without considering the social costs of curtailment. When this factor is excluded, the capital costs
associated with improving the facilities outweigh the calculated benefits by US$ 9 to 16 million (Table

Present value CIP costs of pumping


Present value benefits of pumping alternatives

(Over and above status quo CIP costs)


(Benefits = avoided costs)





consist of


Failure Risk












O&M Costs


Option 1

Option 5

Option 7

Option 1

Option 5

Option 7

Costs in US$
CIP Capital improvement program
Figure 1.2: Cost and benefits of the 3 options (without social benefits)



However, incorporation of social benefits yields a substantially different result. Curtailments require
customers to make sacrifices such as accepting brown lawns at residences and public parks, not
washing cars at the desired frequency, reducing showering, etc. New landscaping projects are
deferred and the landscaping industry experiences economic losses. SPUs economists analysed
this in terms of loss of consumer surplus, based on the utilitys demand-curve for water. Taking into
account the expected frequency of curtailments (one year in eight) the analysis estimated that Seattle
Public Utilities customers faced an annualised cost of US$ 2.7 million from curtailments. Projected
over 50 years and discounted at five percent, this translates into a present value cost of US$53
million. This factor alone was enough to offset the apparent differential between costs and benefits
discussed above. With this factor included, the net present values of both Options 5 and 7 are
approximately US$37 million (Table 1.2).

Status quo

Option 1

Option 5

$63 328 186

$72 777 109

$26 277 097

$26 640 209


$3 363 624

$15 838 905

$25 166 493

$24 887 815


$4 784 584

$1 782 536

$588 150

$775 899


$1 240 097

$1 240 097

$194 123

$416 744


$285 138

$285 138

$192 041

$228 749

Risk of component

$581 431

$557 121

$136 291

$331 001

$53 073 312

$53 073 312




$12 475 281

$21 802 869

$21 524 191


$12 475 281

$21 802 869

$21 524 191

PV of costs*

Costs net of status quo
Benefits (avoided costs)

Option 7


$3 026 359

$58 853 957

$58 212 168

Reduced O&M


$3 002 048

$4 196 434

$4 008 685

Reduced mobilization



$1 045 974

$823 354

Reduced pumping costs



$93 097

$56 388

Reduced failure risks


$24 310

$445 140

$250 430

Eliminate curtailments



$53 073 312

$53 073 312



-$9 448 923

$37 051 089

$36 687 977

NPV excluding curtailment**


-$9 448 923

-$16 022 223

-$16 385 335


Includes the cost of unnecessary curtailments.

i.e., the net present value (NPV) excluding the benefits of eliminating unnecessary curtailments.
Not applicable
Table 1.2 Results of cost/benefit analysis, US$

The PDP summarising this analysis has also provided a discussion of unquantified risk costs. If
curtailment was not considered in the analysis, decision makers could see that the value of avoided
risk would need to be at least US$9 to 16 million in order to justify the investment (Table 1.2). The



Project Development Plan also includes a sensitivity analysis that shows that the results are robust to
changes in assumptions regarding discount rates, the frequency of curtailments, and other factors.
Option 5 was selected for implementation as a result of this analysis.

Summary of results
The procedure outlined in this case study offers Seattle Public Utilities a consistent and transparent
process for making decisions on capital projects. All capital investments greater than US$ 250,000
now are required to go through this process prior to approval. The AMC, comprised of senior
management at Seattle Public Utilities, is able to make sound decisions based on a well documented
business case. Project expectations are well defined and alternatives are thoroughly examined.
Consideration of the full range of costs and benefits using triple bottom line concepts leads to better
decisions that are fully supported with facts and analysis, as is demonstrated by the example above.
Putting this program in place has not been without challenges. It has taken considerable effort to train
staff in multiple branches of the organisation in the techniques described here. However, Seattle
Public Utilities management believes the benefits from this approach are worth the effort.
Together with other aspects of the utilitys asset management program, this procedure has enabled
Seattle Public Utilities to reduce capital costs. Capital spending needs covering the period from 2003
to 2008 were reduced approximately 20%, while remaining within acceptable risk tolerances.
Together with reductions in costs of operations and maintenance, this has also reduced the predicted
growth in customer utility rates. In 2002 the monthly bill for all four utility services for a typical singlefamily residence was projected to reach US$127 by year 2010. A new forecast compiled in 2004,
when the asset management program had been developed, showed the 2010 average residential rate
to be US$120, or 5.5 percent less than originally forecast (Martin, Terry, May 2006, Asset
Management at Seattle Public Utilities, the Australian Approach (unpublished presentation).
For more information on triple bottom line, consult the following document sources:
Report 91179 sponsored by Water Research Foundation (formerly AWWA Research Foundation) and
Commonwealth Scientific and Industrial Research Organisation (CSIRO) at the Foundation website
Kenway, S., C. Howe, and S. Maheepala (2007) Triple Bottom Line Reporting of Sustainable Water
Utility Performance (Project #3125, Report 91179). USA: AWWA Research Foundation and
American Water Works Association and United Kingdom: International Water Association.



Continuous leak detection to monitor condition of water distribution pipes American Water case study
This case study involves American Water, a private company providing water and wastewater
services to over 300 water systems throughout the USA. American Water has been pilot-testing a
new approach to monitoring the condition of buried water distribution piping. The practice utilises
acoustic technology to detect leakage, coupled with daily data transmission using a fixed-network
automatic meter reading (AMR) system. This enables continuous data collection and immediate
detection of small leaks in system piping. Continuous leak detection enables American Water to
identify small leaks before they become major main breaks, and also enables proactive scheduling of
repair or replacement of problem mains. This reduces the cost of managing water mains as they age.
This case study reviews American Waters experience in Connellsville, Pennsylvania, where the
demonstration of technology took place. While initiated independently by American Water, the work
documented in this case study is also being utilised in a tailored collaboration project funded by Water
Research Foundation in partnership with the National Research Council, Canada (Foundation Project

Searching for methods to find leaks
Main breaks drive up maintenance costs, disrupt customer service, and waste water. A typical
practice in most water systems is to react to main breaks when water is noticed at the surface, or
becomes evident in some other way. However, leaks can occur for a long period of time before being
detected. Continuous low level water leakage often erodes adjacent soils and damages utilities and
roads, raising the cost of repair and restoration once the leak becomes large enough to be found. In
some instances leaks also lead to soil movements that damage property and present public safety
hazards. In addition, main breaks identified through traditional methods can require an immediate
response, with repair work performed after-hours or on weekends or holidays when labour costs are
higher than during normal working hours.
Some utilities conduct periodic leak surveys. However American Water has found these programs to
be expensive, often yielding poor results, and providing only a snapshot of leaks identified at that
time. Asset managers at American Water became interested in developing an improved predictive
approach because this offers a potential to achieve several benefits:
Reduce the damage caused by main breaks, thereby reducing costs of repair and restoration.
Permit early action to repair failing pipes and extend their lifespan.
Allow planned scheduling of repairs to failing mains, reducing labour costs for these repairs.
Reduce unplanned water supply interruptions, improving customer service and fire protection.
Reduce water losses, reducing the cost of supply and supporting overall resource
management objectives.
Various proactive approaches to predicting main breaks can be applied. However, there is no system
that can really predict where and when a given pipe will fail. Predictive approaches depend on a
variety of local factors such as soil type, pipe material, water chemistry and temperature. These



factors hamper the prediction accuracy of available techniques.

considerable data and are difficult to transfer from one locale to another.

Statistical models require

A monitoring program for pipe leaks was selected for pilot demonstration testing at full scale because
of continuing improvements in the necessary technologies and the strong relationship between pipe
leaks and complete pipe failure.

Practice demonstrated in Connellsville, Pennsylvania, USA

Connellsville is located in the south-western corner of Pennsylvania in a steep valley. Water supply
for Connellsville is purchased from an adjacent system at a cost that is relatively high in the US
context: US$ 1.94 per thousand gallons (US$ 0.51/k). The water source is the Youghiogheny River.
This source can experience sharp changes in temperature throughout the year. Rapidly falling
temperatures in the source water can introduce stresses in the metal distribution pipes, thus leading
to a higher incidence of water main leaks in the autumn and winter months.
American Water serves approximately 12 000 people in Connellsville. The distribution system has
about 57 miles (91.7 km) of water main. The town was established approximately 200 years ago and
has a history of coal mining and related industries. Two-thirds of the pipe network is over 100 years
old and includes a high proportion of galvanized, two-inch diameter piping.
Prior to the pilot demonstration program described in this case study, non-revenue water was
approximately 27% of total water purchased. Non-revenue water included leakage as well as
intentional uses such as fire flows, flushing and blow-offs used to alleviate aesthetic problems with
water quality.

Continuous leak detection program

There are two key elements to American Waters continuous leak detection program in Connellsville.
First, the program uses continuous acoustic monitoring devices and associated data processing
software. Second, the program uses an AMR (automatic meter reading) system to relay data from
the field to a central computer where analysts can process and interpret the acoustic data. Both of
these elements are described here.

Acoustic monitoring

The Connellsville pilot system uses an acoustic monitoring system called MLOG , provided by Flow

Metrix of Maynard, Massachusetts. While this case study describes MLOG , other acoustic

monitoring systems can also be tailored to this type of application. Figure 2.1 displays an MLOG



Figure 2.1: MLOG Sensor (at left) Adjacent to a Water Meter

The acoustic sensors were attached to a subset of service lines in close proximity to service meters.
Approximately 500 sensors were placed at regular intervals throughout the Connellsville distribution
system, which has about 5,000 service connections. With a listening range of 300 to 500 feet (about
91 to 153 m), this enabled acoustic coverage of most of the distribution system. The sensors record
acoustic data each night. Data from the sensor network are compiled daily and analysed at a central
computer to identify possible leaks and assist field crews in determining locations for site visits.
A crew comprised of an analyst and two field staff was formed to investigate suspected leaks. Careful
analysis is required to distinguish noise emitted from leaks and noise emitted by water uses, leaking
plumbing fixtures on the customers property, and background noise from non-water equipment such
as power transformers, heaters, compressors and air conditioners. However, American Water
operations staff has been able to learn quickly how to distinguish acoustic characteristics that
represented leaks, and no formal training on this was needed. A leak investigator was assigned to
distinguish actual leaks from false positives through a data correlation procedure involving data from
several acoustic sensors, as well as AMR data that helped distinguish leaks on the customer
property. A brief site visit for remaining candidate leaks serves to confirm whether excavation is
warranted. Once a leak is confirmed, a repair crew excavates and fixes the leaking pipe, usually
within 24 to 72 h of initial detection.
Figure 2.2 displays typical data from the acoustic sensor. The graph shows a leak detected on
December 7 and repaired December 19. The yellow line represents a full range of frequencies
detected by the sensors, while the gray and blue lines depict frequency ranges that specifically
correlate with different types of sounds detected. A dotted red line displays background noise. These
patterns help the analyst distinguish leaks from other noises, and determine what type of leak to look



Figure 2.2: Graphic display of acoustic data

The sensors are capable of detecting leaks as low as one gallon per minute (3.8 /min). Batteries are
used to power the sensors. Sensor life is limited by battery life, expected to be 15 years.

Automatic meter reading (AMR) System

The Connellsville pilot demonstration uses an AMR system to transmit acoustic data to a central
computer. The Connellsville system is a fixed network AMR provided by Hexagram (Cleveland, OH).
Mobile radio-read systems can also be deployed.
One advantage of coupling the acoustic system with an AMR system is that metered flows related to
on-site water uses and plumbing fixture leaks can be readily separated from other acoustic data. This
is because high-resolution data on customer use (flow through the meter) can be compared directly
with the acoustic record.

Results from Connellsville

The number of leaks identified for repair during the pilot demonstration increased by 83%, compared
with a pre-pilot period. There were 119 leaks detected during the pilot period, compared with 65 over
the pre-pilot period. This comparison covers the same number of months and same seasonal
More than half of the leaks identified in 2005 (24 of 46) were repaired before leaking water surfaced.
Another 10 leaks were identified by the acoustic sensors, but surfaced before repairs were made.
Twelve leaks surfaced without being identified by the acoustic sensors. Some of these apparently
surfaced almost immediately after the leak occurred. Similar results were experienced in 2006.



Non-revenue water has been reduced by 16% (from approximately 27% before the pilot began, and
currently is only 11%). This has resulted from a combination of leak reduction (13.5%) and an effort
to reduce system-flushing for water quality purposes (2.5%).This has reduced the operational cost of
purchasing water from an adjacent water authority by nearly US$180,000 used per year.
One important advantage of the program is that leak repairs can be scheduled to be performed while
leaks remain small. This reduces damage to road substrate, adjoining utilities, and properties, which
reduces restoration costs. Scheduling repairs also reduces overtime labour and helps to control
disruption experienced by water consumers. In addition, by permitting earlier, identification and repair
of leaks field repairs during the coldest months can be reduced. This helps to reduce repair costs that
tend to be higher under freezing conditions and more limited daylight hours.
As a side benefit, the acoustic monitoring has also been useful in identifying leaks occurring on the
consumers property. This can allow notification to customers of leaks they are experiencing, prior to
the customer receiving a large water bill. This improved customer service is expected to reduce
customer complaints related to unexpectedly high water bills.
There are limitations to this program. Even with careful analysis, false positives can still occur.
However, virtually all false positives are eliminated from consideration, based on either careful data
analysis, or a brief site visit. In addition, not every leak that surfaced in Connellsville was detected by
the acoustic system. Also, the acoustic signal is less robust when plastic piping is present.
More intensive analysis of main breaks is being explored using the detailed leak information available
from the acoustic monitoring system,. Since leaks can be identified as soon as they occur, the timing
and location of leaks can be correlated with transient factors such as changes in water temperature
and surge conditions, as well as static factors such as pipe material, age and soil type. With more
leak repairs, data can also be gathered on the physical characteristics of the leaking pipe. These
data also permit more in-depth analysis of the economics of allowing leaks to continue, versus
carrying out immediate repairs. All of these elements contribute to improved management of aging
water mains within American Waters system.
The utility plans to continue evaluating competing equipment and vendors, for both the acoustic
sensors and AMR system.

Summary of results
The pilot demonstration project in Connellsville has been quite successful in improving condition
monitoring of water mains. American Water has reduced its unit cost for repairing main breaks by an
estimated US$ 400 for each break repaired. In addition, the utilitys estimates show costs associated
with expensive purchased water have been cut by nearly US$ 180 000 per year by reducing water
The utility estimates that a full system for AMR and acoustic monitoring would cost approximately US$
750 000. This includes the cost of meter replacement. The savings in purchased water resulting from
reduced water losses provide a financial benefit, and lowers the payback period for installing AMR in
the Connellsville system from approximately twelve years to three years (the payback calculation
incorporates additional cost-savings from reduced meter-reading and customer service-needs). In
addition, the program now provides an improved understanding of system-wide main condition and
the specific static and transient conditions that cause main breaks.
The system is not 100% accurate and undetected main breaks still occur. Research indicates that
about 25% of the leaks in Connellsville come to the surface too quickly for acoustic detection to permit
advance warning. This is especially true of circumferential breaks in cast iron water mains.



Despite these limitations, American Water sees considerable advantages to using continuous
acoustic monitoring as an element of its condition monitoring program for buried water mains. The
utility is now experimenting with transferring this technology to larger water systems, and will continue
evaluating financial payoffs and other benefits for systems with different piping and service area
For more information on this project by American Water, Water Research Foundation, and National
Research Centre (Canada), refer to:
Water Research Foundation (2009) Project Snapshot: Continuous System Leak Monitoring - From
Start To Repair #3183. [Online]. Available:
www.waterresearchfoundation.org/research/TopicsAndProjects/projectSnapshot.aspx. [Cited
February 11, 2009]



Use of electronic mobile and field solutions by Las Vegas Valley Water District
- Las Vegas Valley Water District
The Las Vegas Valley Water District (referred to as district) was formed as a non-profit governmental
subdivision of the State of Nevada, USA, and is a quasi-municipal corporation that was created by
special act of the Nevada Legislature in 1947. The district was established to acquire and distribute
water to customers in the Las Vegas Valley, including the unincorporated metropolitan area of Clark
County and the City of Las Vegas. The district began operations on July 1, 1954 and helped build the
City of Las Vegas water delivery system. The district now provides water to more than one million
people in southern Nevada. This area is growing rapidly, and responding efficiently to development
interests is an important aspect of the districts day-to-day operations and long- term asset
management program.

The district formed their asset management (AM) group in early 2003. The AM group is a stand-alone
entity that resides within the districts operations department. The departments strategic plan states
that the asset management goal is:
To develop, communicate, and integrate a management strategy for assets and maintenance that
emphasises return on investment and sustainability of infrastructure while achieving desired levels of
service to our customers over the lifecycle of district assets.
A successful AM program relies on efficient data collection during the entire lifecycles of assets. The
district has initiated numerous mobile and field initiatives that use both wireless and docking
technologies to assist with AM and customer service. The focus of this discussion is on the Mobile
Inspection Data Acquisition System (MIDAS) and the ViryaNet System, which are an integral part of
the daily work of district field staff. Four field applications are briefly discussed, in addition to these
two mobile solutions.
The districts mobile and field solution systems described in this case study are used to collect many
sets of data that support related utility functions. These include acceptance of developer construction
improvements, inspection processes, meter reading and services, management information, risk
assessment and risk management.

MIDAS Mobile inspection data acquisition system

MIDAS was jointly developed by the district and the private company, Gatekeeper Systems of
Pasadena, California, using todays technology for both system development and system use.
MIDAS facilitates the documentation required as district inspectors perform more than 100 field
inspections through the development life of a single, new subdivision or commercial property. On
average, utility staff conducts 300 inspections per day. At any given time, the district may track about
1,150 developer projects. Thus, this is a substantial workload, requiring rapid response and accurate
The district has deployed 30 mobile units on Panasonic Toughbook laptop computers. The district
selected this ruggedized computer because inspectors spend a substantial amount of time on
construction sites and need equipment that can endure tough environmental conditions. Each truck is
equipped for wireless connectivity (via cell phone towers, using mobile wireless broadband services
provided by telephone companies) and is capable of determining exact location using a Global
Positioning System (GPS).



The number of inspections also increases as the district service area rapidly expands. The MIDAS
system enables each inspector to perform more inspections daily than they previously could, as they
do not need to spend as much time in the office at the beginning and the end of each day. The
system did not eliminate the need to hire more inspectors, but it did absorb the existing number of
new inspectors. Also, MIDAS helps ensure relevant inspection data are collected and documented in
the field. This is accomplished by structured inspection forms that require completion of relevant
inspection data fields before transmittal. MIDAS contains information on pass-fail status, alerts an
inspector about the need for an inspection, and is used to collect related additional data regarding the
project such as the results of pressure and chlorine tests. MIDAS also enables the inspector to obtain
information on other projects, allowing staff to stand in for absent inspectors, when needed, or
address the needs of customers more responsively.
MIDAS is also used by inspectors to schedule GPS data collection by the GPS/survey staff. GPS
data are collected the day after the inspection and are available for interested technical staff within 24
hours. As a result of having this timely-collected GPS data, appurtenances can be located in areas of
heavy construction where there are few structural features such as a curb or gutter from which
facilities can be located.
Inspection data are evaluated to determine trends that may be associated with one or more
contractors or developers. A trend in the frequency of failed inspections relating to some aspect of a
construction project may indicate that a contractor or subcontractor needs to change the way work is
done in future.
Development of MIDAS followed an initial project phase where system requirements were
documented. Requirements recognised both technical considerations (e.g., interface with existing
legacy system) and users input. , The district and gatekeeper developed MIDAS using industrystandard software to ensure system longevity, scalability and adaptability to future needs. Some of
the advanced technical solutions employed during MIDAS development and use are presented as
The system leverages wireless and store-forward technologies that accommodate potential loss of
coverage. If the wireless connection is lost, the system retains the gathered data while the inspector
continues work. When the wireless is reconnected, data are transferred between the laptop and the
office system asynchronously, without inspector intervention.
The system has been built with user-definable validation data-fields (drop-down field selection tables),
allowing for validation of data entered by field inspectors
When a particular inspection type is selected on the laptop, only the relevant fields are displayed on
screen. This dynamic data field display means that inspectors need only view and complete fields
required for the inspection that they are currently performing.
The system allows inspectors to download and view any inspection in the system and transfer any
inspection data to another inspector, as required. For example, an inspection that has not
commenced can be reassigned to another inspector who will be on site another day.
The system is easy to use, as reported by the inspectors
The Infor Public Sector Essentials system (product name) supports the developer permit and
inspection process. Formerly, this system was called the Hansen Technologies Permitting System.
Infor acquired Hansen Technologies in June 2007.
When an inspection results in a failure or non-approval, the system automatically issues a follow-up
quality assurance inspection. When the cause of a failed inspection is corrected, the system shows
that the quality assurance inspection has been resolved. The Infor Public Sector Essentials system



helps ensure that all of the projects quality assurance exceptions are settled, prior to a project
Facility View, the districts mapping application associated with MIDAS, provides infrastructure and
facility layers for use in the field. The system uses Autodesk Map Guide (product name) and
Gatekeepers Navigate Software (product name) as the user interface.
Inspectors may access digital drawings through an encrypted system tied to the Facility View system.
This includes approved developer drawings and other drawings associated with the inspection.
Contractors or developers can log into an Internet site and request inspections that are complimentary
to the MIDAS and Infor systems. In the month of April 2007 alone, there were approximately 1 000
developer-hits on the system. Before this system was automated, developers had to schedule a
request before 14:00 on the day before the inspection. The new system has no cut-off time. The
district estimates that the return on their investment has been achieved in six months.
New inspectors now receive their training on the MIDAS system from the better-experienced
inspectors who already use MIDAS.

ViryaNet ServiceHub mobile application mobile work order management

Approximately 200 users, from eight work areas, currently utilise the ViryaNet ServiceHub Mobile
Application (product name) in their day-to-day operations. ViryaNet is a customisable off-the-shelf
application. Meter Field Services uses ViryaNet for meter reading and field activity work sent from the
customer care and billing system. Completion data are captured and the appropriate next action is
queued in the customer care system. The meter shop uses ViryaNet to dispatch and provide
completion data for preventative maintenance and unscheduled work orders sent from Avantis
(product name), the districts CMMS (Computerised Maintenance Management System) and asset
management system. The facilities maintenance and grounds maintenance area uses ViryaNet to
dispatch and provide completion data for work orders created in the Avantis system. Fleet Services
uses ViryaNet to dispatch and provide completion data for preventative maintenance and
unscheduled work orders sent from the Avantis system. Distribution field crews and preventative
maintenance groups utilise ViryaNet to report completion information which is routed through Avantis
or through the customer care and billing system. Customer Service uses ViryaNet for investigations,
bench tests and high-consumption work-types originating from the customer care and billing system.
Water wastage staff, who address conservation measures and enforcement, uses ViryaNet for field
activity work sent from the customer care and billing system as well as to create water wastage
investigation records that arise from their patrols.
Links provided within ViryaNet to the in-house-developed applications provide mobile field employees
with the ability to research historical information fully on a real-time basis in the field. District digital
assets have been leveraged by integrating ViryaNet into host applications such as the customer care
and billing system, Avantis and the time-entry/time-keeping system.
The hardware used is a combination of Panasonic Toughbooks and the General Dynamics GoBook
XR-1 rugged wireless notebooks. These units are mounted in the trucks and connected to the
network through SecurID (product name) over a telephone companys wireless network with internal
aircards or mounted brick modems.

Other field solution systems used at the district

In addition to the MIDAS and ViryaNet mobile solutions, the district uses a number of field solution
systems which are briefly described below:

Locator system Call USA



The district uses Call USA (product name) to automate the location and marking of buried assets, as
is required by Nevada revised statutes (NRS) 455.080 455.180. The system automatically receives,
maps and dispatches call before you dig tickets (requests for information) to facility locators in the
field for disposition. FacilityView is also integrated with this application for access to maps and
engineering record drawings in the field.

Firefly AMR mobile solution

The Automatic Meter Reading (AMR) (product name) system is used to manage and read water
meters. The district is currently deploying the Datamatic AMR radio system solution. Datamatic
optical sensor fireflies are being attached to all of the district's existing straight-reading water meters.
The AMR system was piloted for nearly two years before its full deployment. The district's 352 000
residential, commercial and industrial meters should all be converted to mobile radio-read capability
by June 2008. The fireflies not only enable the district to collect monthly meter readings for billing
purposes, but maintain 74 d of actual water consumption data, on site, for use in customer service
and conservation efforts. Nine mobile receivers now perform the previous functions of 25 manual
meter readers, and the new system has eliminated all safety issues related to confined-space entry.
The system also notifies applicable staff of potential on-site consumer leaks by identifying all accounts
that record a minimum 10 gallons (37.9 ) per hour usage for 24 h prior to reading. The district
indicates that a cost-benefit analysis identified some US$ 31 million in direct hard-cost savings over
the life of the hardware, with an additional US$ 44 million in related benefits to its customer service
and conservation divisions.

Distribution Permalog logger system a preventive asset management tool

The Permalog (product name) units are deployed in areas of the distribution system to provide
continuous monitoring of leakage. Easily installed onto a valves operating nut inside a valve
chamber, they are retained in place by a strong magnet. Each Permalog unit adapts itself
automatically to its environment. As soon as a possible leak is detected, the Permalog unit enters an
alarm state and transmits a radio signal to indicate a potential "leak" condition. These radio signals
are collected using specially equipped vehicles and then uploaded into network applications for
analysis. From there, work orders are created for suspected leaks and sent electronically to the
ViryaNet mobile application to be assigned for investigation.
Since 2004, when the city officially went on a Stage 2 drought alert, the district detected just over 1
000 underground leaks using Permalog and adopted conservative methods through mid 2007.
Assuming an average leak rate of 2 gallons (7.6 ) per minute, the district estimates that it saved 282
acre feet (347 706 m3) of water during that period.

Wachs system preventive asset management tool for valves

The Wachs system (product name) is used to exercise valves by actuation, and detect variability in
torque during the exercising of the valves throughout the districts distribution system. A valve turning
crank, mounted on a truck, senses the torque at 1/5 of a turn and records the torque in a data box.
Readings are used to predict a valve failure, enabling the district to be proactive in valve repair and
In more detail, the valves are exercised (turned) on a regular basis throughout the distribution system.
The district implements this proactive and preventative maintenance program to help ensure that the
valves will function to isolate a water main if it is breached. District valves are exercised at least once
a year for critical valves, at least once every three years for semi-critical valves, and at least once
every five years for non-critical valves, to keep them from seizing. Several things can go wrong with a
valve, including bent shafts, worn seals, and unexpected corrosion, all impairing proper operation. As
the field crew exercises the valve, they will open and close it several times, counting the turns. The
correct number of turns for each valve is specified in the valve maintenance literature. As the valve



turns, the required torque is measured at five equally-spaced points on a circumference. The torque
is measured when opening and closing. Engineers can then analyse the turning torque data, to
determine whether there are any malfunctions with a valve. All of this information leads to a decision
on whether to replace or repair a valve. Torque patterns are often consistent with specific problems.
A fairly reliable diagnosis of a problem can be made by an engineer when he recognises one or more
of these patterns in the data.

Summary of results
The district has invested in and has applied cutting-edge, state-of-the-art technology in both its dayto-day operations, as well as its AM program. This investment, coupled with the wide-spread
acceptance and use by district staff, has increased efficiency, will save millions of dollars in operating
costs, and facilitates the collection and integration of data for use in improving AM programs.
MIDAS enables district inspectors to manage their workloads efficiently and effectively in the field
where they spend most of their time, as well as engender more effective business processes. The
many benefits of MIDAS include:
Developers can enter MIDAS from their own office or remote locations and request
inspections automatically one day in advance.
Dispatchers send inspection related information directly to inspectors in the field based on the
location assignments of the inspector. GPS modem boxes in the inspectors trucks identify
inspectors locations.
The use of data field validation means that inspectors reports contain more relevant details,
are easier to understand (not subject to illegible hand-writing), and are completed on time.
Inspection results are stored on each inspectors laptop, as well as the Infor system, so that
the results are always available to the inspector. This saves time, as inspectors do not have
to return to the office to find the required results or forms.
MIDAS built-in business rules ensure that failed inspections requiring additional work are
Developers must pass a prerequisite inspection before a successor inspection can be
Real-time inspections are available for developer viewing almost as soon as the inspection is
complete, because completed inspections are immediately uploaded from the laptop to the
office system.
While the inspection quality assurance team executes quality control checks on completed
inspections, they also gather GPS data points for each facility. The inspection quality
assurance team has been able complete many more quality checks after using the new
system, and the data are more quickly available using wireless communications.
The field mapping integration enables inspectors to know the field location of all infrastructure
nodes for a safer inspection process.
Using GPS allows all pipes, valves, vaults and hydrants can be positioned for each project.
The system also allows verification and red-lining (corrections) to be performed in the field, to
reduce errors on the as-built drawings.



Before MIDAS was automated, developers had to schedule a request before 14:00. The new
system has no cut-off time. The district estimates that the return on their investment has
been achieved in six months.
ViryaNet connects approximately 200 users from eight work areas with host system data and realtime access to the districts digital assets. Dispatchers, crew leads, supervisors and managers have
access to real-time meaningful data on work being performed, as well as on the workforce while
working on site. The benefits of ViryaNet include:
The field crews can stay connected to the mobile system to receive newly dispatched work
and provide updates on completed work throughout the day without having to receive phone
calls or return to the office in the middle of the day.
The mobile user has control over how to receive and process work throughout the day. New
work can be created and automatically dispatched by mobile field employees, or can be
created by office employees in the host system.
The user also can download work, process the orders offline then reconnect to upload
completed work and receive new orders.

The field solution systems used by the district have provided the following positive results:
The Locator System has reduced the work time by two to three hours per location, based on
the ability to pre-sort drawings and display them in the field.
The Wachs and Permalog logger activities are operations and maintenance practices oriented
towards reducing the risk of system failure. The Wachs system is a tool used in preventive
maintenance on the water main valves in the water system. The Permalog logger system is a
tool to help identify leaks in the water mains while they are small. Both the Permalog logger
and Wachs systems provide preventive asset management information that has enabled the
district to resolve leaks and valve problems in a proactive, rather than reactive manner.
The Firefly AMR system provides early detection of water meter problems, and enables the
district to repair or replace the suspect meter proactively.
For more information on the use of field computing systems in the USA, refer to:
Report No. 91224 funded by Water Research Foundation (formerly AWWA Research Foundation)
and the San Francisco Public Utilities Commission, on the Foundation website,
Stern, C., K. Mallakis, M. Hernandez, B. Iadarola, U. Srinivasan and S. Sakpal. (2008) Field
Computing Applications and Wireless Technologies for Water Utilities (Project No. 3178, Report No.
91224). USA: AWWA Research Foundation.



The use of GIS to support EPCORs business processes - EPCOR Water

EPCOR Water Services provides water and wastewater services to over one million people in the
more than 50 communities around Edmonton, Alberta, Canada and across Western Canada.
EPCORs 450 water professionals manage treatment plants and a network of pipelines that span
more than 3 400 km (2 112 miles). In addition to their Edmonton service area, EPCOR also serves:
British Columbia, Canada: communities of Port Hardy, Sooke, White Rock, the French Creek
area of the Regional District of Nanaimo, and the Britannia Mine.
Alberta, Canada: communities of Strathmore, Canmore, Okotoks, Chestermere, Picture Butte,
Red Deer County, Wetaskiwin and the oil sands plants in Fort McMurray.

This case study provides an overview of the development, key components and principles of
EPCORs Geographic Information System (GIS), as well as their key business processes and
application areas that support effective asset management. EPCOR was a key participant in Water
Research Foundations Report 91164, Building a Business Case for Geospacial Information
Technology A Practitioners Guide to Financial and Strategic Analysis, 2007, where more detailed
information is available. Table 4.1 provides a timeline of the technology deployment and asset
management activities from EPCORs initial installation in 1978 to their third major platform and tool
upgrade designed to leverage recent advances in geospatial technologies.

GIS strategy in EPCOR

EPCORs enterprise GIS deployment is based on a GIS strategic plan that includes two key
GIS technology improves efficiency and/or effectiveness
Accurate information should be available to utility employees when they need it, where they need it,
and in a format that meets their needs.
All potential system additions and modifications are evaluated against EPCORs defined GIS strategy
and these two principles. As such, these two principles drive EPCOR to excel in the following areas:
Intuitive interfaces support users in doing their jobs. While users have a great deal of
experience and expertise in their business areas, the users may not be knowledgeable of GIS
concepts and existing system data structures.
Best business practices and corporate standards ensure accurate and complete data, timely
updates, secure data, and security in a users data use.
Users resolve problems in real-time.
advanced usage.

The structure supports enhancement requests and

GIS data is shared with the GeoEdmonton Right of Way (ROW) alliance partners.




1st Generation GIS CAD based. All users
have the same viewing tools
Limited querying capabilities to locate
addresses and assets. No analysis








2nd Generation GIS CAD and database

Ability to access information via CAD
interface and export tables to other tools
for analysis reviews
Thematic mapping available to assist with
analysis of data
Water main renewal program candidate
selection process automated
Limited work management functionality
built into GIS for hydrants and valve
All mains hydraulic model developed from
GIS, SCADA, and Customer Information
Systems (CIS).

Asset management activity

GeoEdmonton Alliance is established.

Formal water main renewal program

established based on break frequencies.
Neighbourhood Improvement Program
(NIP) initiated coordinating water main
renewal construction with sewer main
and capital roadway improvement
program based on an overall NIP ranking.
Water main renewal program expanded
into two programs:
1. Reactive renewal (replacement based
on higher break frequency);
2. Cathodic protection installation (based
on low break frequencies).

preparation tools developed
Field computers deployed with view
access only via CAD interface.
Work management systems upgraded to
open database architecture to allow
linkage to GIS for analysis.

3rd Generation GIS Vendor neutral based

approach for data entry and query tools
built on an Oracle Spatial database
Live access and update capabilities for
field staff to GIS and work management
System (LIMS) upgraded to link with GIS
to allow additional analysis capabilities.

Hydraulic model used to optimise longrange plans, water transmission main

shut downs, and optimal pumping
strategies to reduce power costs.

Proactive water main renewal program

initiated focusing on hydraulic and water
quality deficiencies
Water main lining program initiated for
areas with good hydraulics, low break
frequencies, but poor water quality.
Modified duty program initiated for cross
training of injured employees on GIS.

Table 4.1 Timeline of the technology deployment and asset management activities



For example, EPCORs previous approach to GIS was based on a single system with extensive
functionality to meet all user needs. The previous approach resulted in an overly complex system that
was difficult to maintain and ineffective in training individual users in proper usage. Whereas
EPCORs current approach, aligned with the two key principles, has resulted in a structure that
focuses on a core Oracle Spatial database and associated tools specifically designed for each user
category that adds and accesses data from the core database. The new approach allows
modifications of individual tools, if there is a change in business requirements, without having to
modify the entire system.

GIS in support of asset management at EPCOR

Since EPCOR employed GIS almost 30 years ago, the use of GIS to support asset management is
intrinsic in almost all of EPCORs water distribution activities. EPCORs staff has access to the full
suite of GIS tools from their desktops and receive regular training on the use of these tools to meet
their specific asset management needs. EPCOR is constantly evaluating advances in technologies
and changing business needs to ensure that the optimum GIS tools are available to enable effective
and efficient asset management decision-making. For example, having GIS maps available in the
field and linked with the work management system, support EPCORs field activities in several areas
including water distribution operations, maintenance and inspection. The following sections describe
several of the asset management programs within EPCOR, and how GIS beneficially supports these

Water main renewal programs

In 1985, EPCOR instituted an extensive water main Renewal and Replacement (R&R) Program.
EPCOR leveraged this program to reduce annual water main breaks from a high of 1 600 to a current
level of less than 400 per year. Since inception, approximately 490 km (304 miles) of water main
have been replaced or refurbished (e.g., cathodic protection or epoxy lining). The information
contained within EPCORs GIS plays an essential role in their decision process for correctly
prioritising and allocating the CAN$24 million annual R&R budget.
Historically, EPCORs R&R program focused on renewal of cast iron water mains. Recently,
EPCORs R&R program has expanded to other pipe materials such as asbestos cement. The GIS
model has been extended to also address these other materials as they age, and their characteristics,
together with impact from environmental change. The following sections describe EPCORs four main
renewal programs and how GIS supports the candidate selection process.

Reactive renewal program

The on-going reactive renewal program involves replacing deteriorating distribution water mains with
PVC pipe. Specific locations requiring replacement are identified in order of decreasing priority
through a GIS application that calculates break frequencies for candidate stretches between valves.
All candidates with frequencies of 5 or more breaks/kmyr (8 breaks/mileyr) are identified for
replacement. Additional criteria include a minimum of at least 2 breaks in 5 years, more than 6
breaks in 5 years, and 12 or more breaks since 1982.
After candidate water mains are selected, GIS is utilised to optimise the scope of each renewal
project. For example, a candidate section may need to be extended due to the presence of a critical
customer in the immediate vicinity, who is located on the side of the candidate valve that is not
identified as a replacement candidate. Also, an R&R project may need to occur concurrently with
adding a loop to improve fire flows, thereby improving service to areas projected for increased



development following land rezoning. If road improvements are planned in an area, consideration is
given to advancing the water main replacement program, to take advantage of economies of scale
from doing two infrastructure upgrades simultaneously, and to minimise impacts on the community by
coordinating construction programs.
Candidates selected for replacement are evaluated using the all mains hydraulic model (discussed
below). Moreover, a hydraulic analysis is completed to verify fire flows in areas affected by the
construction and to provide advance notification to the fire department regarding potential problem

Proactive renewal program

Standards have changed since much of the cast-iron pipe was installed in Edmonton. EPCORs
proactive renewal program relies on both automatic and manual database GIS queries and data input.
For example, data from numerous sources (work management system, hydraulic model, LIMS, land
use rezoning plans) is reviewed within EPCORs GIS to identify and rank water mains that do not
meet current design standards. The recommended renewal solution in these instances often requires
replacement of water mains with larger pipe sizes, the addition of pipe looping in the area to increase
the ability to deliver fire flows and eliminate dead-end mains, or some combination of both. The
identified projects are then grouped into geographic areas to reduce costs for mobilisation and
demobilisation of the water main renewal contractors.
Some factors used in ranking candidate areas for proactive renewal include land use, road type,
number of customers, fire flow availability and requirements, water quality, and uni-directional flushing
times (an indicator of stagnant water problems).

Water distribution main cathodic protection

EPCORs water distribution main cathodic protection project provides for the installation of sacrificial
anodes on cast iron water distribution piping to reduce or eliminate corrosion of this metallic pipe.
Cast iron pipe, like all other metals, deteriorates when buried. The soil conditions in Edmonton are
conducive to external corrosion of buried metals. Application of cathodic protection extends the
service life of the cast iron pipe network, thus maintaining an acceptable break rate and stabilising the
rate of future pipe replacement.
Projects are selected based on the 5-year average break frequency value of between 1.0 and 2.0
breaks/kmyr (1.6 to 3.2 breaks/mileyr). EPCOR utilises its GIS to review this program with the goal
of further identifying any geographic factors that could help refine the selection criteria for installation
of additional anodes.

Water main internal lining program

In 2001, Edmonton initiated a pilot project to line cast iron pipe with epoxy. This program continues
today and is focused on areas where the primary deficiency is caused by water quality concerns due
to low flow. These areas have no water main breaks and have sufficient hydraulic capacity.
However, GIS and hydraulic models are utilised to plan each lining season considering adequate
flows for fire protection and customer access during the program. The associated construction
causes significant street disruption due to the number of access holes required for the lining machine
and the number of streets impacted. As such, the program considers the efficient deployment of the
lining rig in a short time span (i.e. moving from stretch to stretch in one geographic area).

Hydraulic modelling
EPCOR effectively utilises GIS to support hydraulically modelling of the performance of its water
distribution system. As part of its asset management program, EPCOR uses its hydraulic model to:



Analyse peak hour and fire flow pressures to ensure that adequate assets are constructed to
support future development activities.
Analyse the effects of planned shutdowns for any portion of the water system related to:
o Neighbourhood Improvement Program initiatives that require EPCOR to ensure that
there is minimum disruption to the City of Edmontons customers.

Proposed transmission main shutdowns for asset management programs, including

valve chamber refurbishment.

Identify Cast Iron Renewal Program improvement areas by upsizing or redesigning the water
main configuration.
Analyse inquires from customers on low pressure concerns and the applicability or the effect
of potential solutions.
Conduct special studies to assess and plan the water distribution system.
EPCORs hydraulic models are based upon and constructed largely on the information contained
within its GIS, as well other key data contained within a number of other information systems.

Modified duties for injured staff

EPCOR has modified duty business processes in place for enabling people who have been injured.
As required, the injured persons are placed on modified duty until they are cleared to resume their
normal work. However, during this period, many of these employees are able to work effectively at
less physically-demanding positions, such as advancing EPCORs GIS and asset management
For example, field crews are large users of the GIS systems. GIS input tools are designed so that
EPCORs personnel on modified duty can help update or record asset information. Some activities
that can be completed in this manner include:
Updating attribute data for assets (i.e. hydrants, values or water mains).
Recording areas for proposed development.
Recording locations of main breaks or location of service lines.
Drawings/plans/details get linked to GIS.
General asset records filing.
GIS data checking and field verification.
Among the soft benefits achieved are, i.e. the interaction between the drafting office and field staff.
Any employee on modified duty due to injury acts as a conduit, as they are able to effectively
communicate with field crews to verify and validate key information. Concurrently, drafting
technologists learn more about field activities and injured field staff members develop additional
computer skills. Since employing this program, some field crew members have even expressed
interest in exploring a GIS recording position as part of advancing their career path.



GeoEdmonton alliance
One unique aspect of the EPCORs GIS is the association with the GeoEdmonton ROW alliance. By
its nature, ROW facilitates extensive data- sharing between municipal departments and the utilities in
the region. The alliance includes the City of Edmonton, Telus Communications, EPCOR, Atco Gas,
AT&T Canada, Bell West, Shaw Cablesystems, NAIT and the Province of Alberta - Alberta Registries
and Alberta Infrastructure and is regarded as one of the premiere municipal GIS systems in the world.
ROW was established in 1978 when groups in the region were required to update their mapping
systems to the metric system. Rather than updating records in isolation, ROW was formed to build
and maintain a coordinated system. Initially, ROW members each provided a data layer to a common
system that was then made available to all alliance partners. Today, the alliance mandates data
standards, procedures, and compatible technologies to facilitate effective information sharing for an
annual fee to support continued system development.
The City of Edmonton maintains over 200 data sets that are available to ROW partners. Partners
reciprocate by sharing their data sets, thereby reducing duplicate efforts, increasing efficiencies and
lowering costs. Other benefits realised by the partners of the GeoEdmonton Alliance include:
Access to a cadastral map that is highly accurate, updated nightly, and common to all users.
Oil, gas and related underground utilities data is portrayed in the GIS for infrastructure
planning and design.
Aerial photos assist planning and design functions by providing a realistic view of site
Topographic features such as bridges/overpasses, trees/groundcover and civic buildings are
useful to planners, designers and those concerned with routing along the street network.
Ground elevations required to plan drainage are available to planners and civil works
Locations of hazardous materials are available to emergency responders and those
responsible for protecting the environment.
Addressing legal descriptions and statutory plans within those areas involved in land
development and those who provide services to properties.

Financial benefits
Financial benefits are directly attributable to a mature GIS program and participation in the
GeoEdmonton Alliance. An example of this financial benefit is found in an application (ADAPT
Automated Drawing and Preparation Tool) that automated the merging of GIS data layers and the
addition of standard notes to create an initial construction drawing. This reduced the amount of time
needed to prepare construction drawings by gathering and merging utility and base data layers,
standardising the graphic elements used on construction drawings, providing faster digital input and
editing of design information. Implementation of ADAPT achieved a pay-back in the first year, and
had an annualised return on investment of over 700%. It positioned EPCOR to absorb a workload
increase of close to 50% without adding any new positions or compromising work quality.

Summary of results
EPCORs deployment and use of GIS over the past 30 years has been extremely successful in
supporting its business processes and application areas that support effective asset management.



EPCOR constantly evaluates advances in technologies and changing business needs to ensure that
the optimum GIS tools are available to enable effective and efficient asset management decision
making and for correctly prioritising and allocating its CAN$ 24M annual renewal and replacement
budget. GIS tools are utilised to optimise the scope of each renewal project. EPCORs proactive
renewal program relies on both automatic and manual database GIS queries and data input to identify
and rank water mains that do not meet current design standards. The identified projects are then
grouped into geographic areas to reduce overall costs of the water main renewal contractors.
Application of EPCORs cathodic protection program extends the service life of its cast iron pipe
network, thus stabilising the rate of future pipe replacement. Water main breaks are now at their
lowest level since the 1960s and the water-loss rate is less than one half of the national average.
There were 385 main breaks in 2005, down from over 1 500 main breaks in 1986. EPCOR also
utilises GIS to support hydraulic modelling of the performance of its water distribution systems for a
variety of beneficial uses, including fire flows, planned shutdowns, customer low pressure concerns,
coordination with roadway capital improvement programs, and water distribution system master
planning. Distribution services staff have increased efficiency, and work to a performance target of
repairing 94% of all water main breaks within 24 h of detection. EPCORs modified duty processes
allow personnel who are injured to help update or record asset information within the GIS until they
are cleared to resume their normal work. EPCORs association with the GeoEdmonton Right of Way
Alliance further helps to eliminate duplicate efforts by sharing their data sets with the members,
thereby increasing efficiencies and lowering total costs.
For more information on GIS in N. America (especially EPCOR) see the following:
Report No. 91164 sponsored by Water Research Foundation (formerly AWWA Research
Foundation), Geospatial Information & Technology Association (GITA), GeoConnections Canada, and
U.S. Federal Geographic Data Committee on the Foundation website:
Lerner, N., S. Ancel, M. Stewart, and D. DiSera (2007) Building a Business Case for Geospatial
Information Technology: A Practitioner's Guide to Financial and Strategic Analysis (Project #3051,
Report 91164). USA: AWWA Research Foundation and American Water Works Association and
United Kingdom: International Water Association.



Main replacement and rehabilitation program Louisville Water Company

This case study involves the Louisville Water Company, which provides water treatment and
distribution to a population of approximately 810 000 in the metropolitan area of Louisville, Kentucky,
USA. Louisville Water Company utilises an extensive distribution system comprised of approximately
3 750 miles (6 034 km) of mains to deliver Ohio River water to 271 900 service connections. In
addition to serving retail customers, Louisville Water Company wholesales water to several nearby
communities in Oldham, Bullitt, Shelby, and Spencer counties.
Louisville Water Company has implemented an aggressive Main Replacement and Rehabilitation
Program (MRRP) over the past 15 years. The purpose of the MRRP is to prioritise replacement and
rehabilitation decisions on ageing cast-iron water mains that exhibit high failure frequencies and have
characteristics that contribute to poor water quality and reduced hydraulic capacity.
Over the course of this long-term program, Louisville Water Company has compiled a store of
information that illustrates the costs and results of its efforts. In addition, the long history of the
program contains highlights regarding challenges and successes that are likely of interest to other
utilities considering similar strategies.

Ageing and failing water mains pose a significant challenge to utilities, particularly those that serve
communities that have been established for over 100 years. Main breaks and leaks place a strain on
utility staff and financial resources. Historically, water systems have addressed main failures as they
occur. Such was the case with Louisville Water Company prior to the late 1980s. Realising the
difficulties in continuing to keep up with increasing rates of main breaks, the utility began to develop
long-term proactive strategies for asset management. A key program has been the MRRP. The
stated goals for the main replacement and rehabilitation program are to:
Improve water quality and customer service.
Reduce maintenance.
Improve hydraulic capacity and fire flow.
Improve coordination with economic development and paving programs.
These goals are extensions of the utilitys mission, which is founded on providing quality, service, and
value to its shareholders. While Louisville Water Company has implemented a variety of programs
related to asset management, the MRRP remains the most developed and mature of these efforts
and is the focus of this case study.

Main Replacement and Rehabilitation Program (MRRP)

History of MRRP
The MRRP had its beginnings with a knowledge building phase that took place between 1985 and
1990. During this time, Louisville Water Company conducted extensive data and records analyses. A
database was developed, containing historical information regarding installation and maintenance
histories for pipes, valves, hydrants, and service lines. This information was integrated into a facility



management system. With the addition of an AutoCAD mapping system and upgraded work order
system, the utility began to track closely the breaks, leaks, repairs, and testing of distribution system
components. These databases were then used to develop a vintage/cohort analysis, in which the
break and leak frequencies of different ages of pipes were compared over time. This provided the
data needed to prioritise types and ages of pipe for replacement versus rehabilitation.
This wealth of information made possible a detailed plan for future activities. In the early 1990s, main
break modelling was conducted and resulted in recommendations for a 15-year infrastructure renewal
and rehabilitation program. In 1993, the program was launched and a pipe evaluation model was
developed to support decision-making processes. This enabled multi-year planning and coordination
of projects, which resulted in efficient implementation of the program.

Key elements of the MRRP

The primary focus of the annual MRRP has been the replacement and rehabilitation of ageing,
unlined cast-iron water mains installed prior to 1937, and ranging in diameters from 4 to 20 inches
(101 to 508 mm). Pipe segments installed from 1862-1865 (typically unlined pit cast-ironcast-iron
pipe) and 1926-1931 (typically unlined deLavaud cast-iron pipe) were identified for removal from the
system because of their high failure frequencies in excess of system-wide averages. These vintage
pipes also exhibit characteristics that contribute to poor water quality (namely associated with high
levels of red or rusty water complaints) and a reduction in hydraulic capacity, pressure and flow due to
internal tuberculation. Thus, these pipes were scheduled for replacement. As of early 2007, the
average cost to replace piping was estimated to be US$ 380 000/mile (US$ 237 500/km) or US$
72.00/ linear foot (US$ 237.50/m).
By contrast, pipelines installed from 1866 to 1925 (typically unlined sand-cast-iron pipe) have proved
to be very reliable, and have been the focus of the rehabilitation portion of the program, with
approximately 80% of those water mains targeted for rehabilitation, and the remaining 20% for
replacement. Rehabilitation efforts have focused on cleaning and lining the unlined pipes from 1866
to 1925, most often via the Perkins Process, which involves in-place cleaning and cement mortar
lining. In early 2007, the estimated cost to rehabilitate pipe was US$ 225 000/mile (US$140 625/km),
or US$ 42.60/ linear foot (US$ 140.63/m).
The MRRP project planning and estimating are directed by a planning engineer, whose primary work
focus includes asset management for buried infrastructure. A team of design engineers works in a
rotational manner on the design, bidding, inspection, and construction elements of the program, so
that each staff member becomes familiar with all project management aspects of the MRRP.

Selection methodology
The pipe evaluation model created in 1993 considers 23 criteria to aid in evaluating and justifying
water main projects. The categories of criteria are provided below.






Central business district, redevelopment areas, and roadway



Main size, fire flow availability, number of parallel mains, high

pressure frequency, and low pressure frequency


Main break frequency, joint leak frequency, material samples,

corrosive soil data, installation date, pipe type, joint type, and
maintenance record

Quality of service

Taste and odour complaints, discoloured water complaints, water

quality data, number of domestic/fire services, lead service
frequency, dead-end water mains, and paving age.
Table 5.1: Categories of criteria

Projects are scored according to all criteria, which in turn are weighted based upon degree of
importance. Criteria with the highest weighting with regard to pipe replacement projects include
location in the central business district, main size, and main break frequency. By contrast, criteria
with high weightings with respect to pipe rehabilitation projects include type and age of pipe, and
frequency of water quality complaints.
The result of the scoring process is a prioritised list of rehabilitation and replacement projects. For
example, a segment of pipe that is located in the central business district and that scores high in
terms of main break frequency would typically be prioritised for replacement. A small diameter main
that is located in a residential area and scores high in terms of water quality complaints would likely
be identified for rehabilitation. Currently Louisville Water Company uses a criterion of 2 breaks per
mile per year (1.2 breaks per kmyr) as the primary justification for replacement of pipeline segments.
This threshold is approximately a factor of 10 times greater than the utilitys average break rate of
0.23 breaks/mileyr (0.14 breaks/kmyr).

Results of MRRP
Louisville Water Company has maintained extensive data records regarding main breaks and joint
leaks throughout the course of the MRRP. As shown in Figure 5.1, the combined 10-year moving
average of main break and joint leak frequency has decreased over the past 20 years, from
approximately 34 annual breaks/leaks per 100 miles (21 annual breaks/leaks per 100 km) in 1986, to
24 annual breaks/leaks per 100 miles (15 annual breaks/leaks per 100 km) in 2006. The MRRPs
contribution to this improvement is illustrated by the downward trend of main break frequency
between 1994 and 2006. The number of main breaks experienced annually by the system decreased
by 25%, from 927 in 1994 to an average of 700 during 2001-2006. During this same period of time,
the amount of piping in the system increased by 25%, from about 3 000 miles (4 827 km) to 3 750
miles (6 034 km).



(Breaks or Leaks per 100 miles of Main)








































Combined MB/JL Freq. (10yr)


Main Break Frequency (10yr)

Joint Leak Frequency (10yr)























1986 1987 1988 1989

1990 1991 1992 1993

1994 1995 1996 1997 1998

1999 2000 2001 2002

2003 2004 2005 2006



MB/JL indicates main break/joint leak.

100 miles = 161 km.
Figure 5.1: Louisville Water Company main break and joint leak frequency
(1986-2006, 10-year moving average)

Other benefits of the MRRP include a reduction in red water complaints, upsizing of mains in fire
flow critical areas as a part of pipe replacement, and full conversion to lined pipes for the cast-iron
portions of the distribution system, thus restoring hydraulic capacities in these older mains.
During the 15-year program, 172 miles (277 km) of main have been replaced, and 251 miles (404 km)
have been cleaned and lined. In total, the 423 miles (681 km) addressed translates to an average of
28 miles (45 km) per year. At the conclusion of the MRRP, all recorded or known unlined cast-iron
pipe will have been replaced or rehabilitated. The total cost of the program, through the early part of
2007, has been US$132 million. Figure 5.2 illustrates the investments in the program over time, in
terms of mileage completed and costs per year. The dramatic decrease in 2004 in annual mileage
completed arises from the program nearing completion.























Annual Funding
$6 to $10 million per year





Cumlative Funding
$132 million spent



Annual Mileage
423 miles completed







































Figure 5.2: Louisville Water Company MRRP historical mileage and costs



Other related programs and activities

Many other asset management programs are closely tied to Louisville Water Companys MRRP.
Condition assessment, in the form of visual inspection of above-ground facilities and periodic leak
detection surveys, provides information for use in the pipe evaluation model. The utility has
developed a robust GIS platform that contains information on the entire distribution system, which is
used for a variety of analyses and planning purposes.
Louisville Water Company maintains a separate renewal program for components of the distribution
system other than pipelines. The point capital program involves rehabilitation and replacement of fire
hydrants, large meters, services, blow-offs, valves, and some discrete short sections of transmission
and distribution mains. Particular attention has been paid to fire hydrants, the maintenance and
renewal of which is the responsibility of a specific work team. The benefit of this program is that the
utility reports a hydrant availability measurement of 99.9%, meaning that 99.9% of all hydrants are in
good, operating condition. The hydrant program is coordinated with the MRRP to ensure construction
efficiency, as is the replacement of lead service lines as described below.
Louisville Water Company also continues to systematically remove the utility portion of leadcontaining services from the distribution system in order to reduce the potential for elemental lead
entering the tap water, and because of increasing awareness regarding public health issues and
future regulatory compliance. So far, 20,000 lead-containing service lines have been removed. The
utility encourages the homeowner to remove their portion at the same time. The typical cost to
remove the utility portion of a lead-containing service line and replace it with an alternative material is
estimated to be US$ 1 250 to an average homeowner. The Lead Service Renewal Program aims to
remove the remaining 15 000 lead-containing services from the distribution system by the year 2015.
This effort is being conducted by Louisville Water Company crews and contractors with an emphasis
on coordination with local paving and construction programs.

Summary of results
Louisville Water Companys Main Replacement and Rehabilitation Program has been successful in
reducing the frequency of main breaks and joint leaks throughout the utilitys distribution system. The
cumulative investment of US$ 132 million over the 15 years of the program life has reduced the
annual number of breaks by 25%, resulting in substantial cost savings related to unplanned
maintenance and water loss. Some of the lessons that Louisville Water Company learned throughout
the development of the MRRP include:
Data collection and management is foundational to the success of the program.
Quantitative data can be used to build long term goals and objectives.
Pipe age should not necessarily be the primary decision-making tool. Some of Louisville
Water Companys older pipes are in sound structural condition and do not warrant
replacement, unlike other mains installed more recently.
Once measures and procedures are established, they must be maintained. This allows
analyses and results to be generated in a consistent manner, which in turn allows success (or
failure) to be readily tracked. Program accountability is key to maintaining relationships with
utility management and investors.
As the first 15-year MRRP comes to a close, Louisville Water Company is actively looking to the
future and developing the next generation of the MRRP. The focus of this next phase will likely be
replacement and/or rehabilitation of lined cast-iron pipe and unwrapped ductile iron pipe installed prior
to 1981, as these materials do not have protection from possible external corrosion. There is
approximately 2 000 miles (3 218 km) of such pipe, comprising more than half the distribution system.
Louisville Water Company plans to incorporate risk assessment into its decision making process as it
prioritises projects and weighs rehabilitation options, including cathodic protection, against
replacement. This will broaden the utilitys current approach by allowing it to better capture concepts
such as consequence of failure, in addition to the likelihood and frequency of failure.
For more information on repair and rehabilitation in N. America, search the reports and projects funded by Water
Research Foundation at www.waterresearchfoundation.org.




Effective risk modelling of water infrastructure assets in the United Kingdom
M Randall-Smith
Technical Director, Mouchel Parkman (Formally Ewan Group)
(Acknowledgement First published and presented at the 4th CIWEM Annual Conference, Newcastle
Upon Tyne, UK, September 2006)

Privatisation of the UKs water companies has resulted in a significant evolution in the industrys
approach to water and wastewater infrastructure maintenance. Regulators have increasingly required
companies to understand assets, their performance and their deterioration far better than ever before.
Combined with pressures to reduce operational expenditure linked to efficiency targets, this has led
companies to look for methodologies that support their decision-making processes when targeting
maintenance investment. An increased need to better manage how distribution systems are operated
is adding to this need.
During AMP3 UKWIR (United Kingdom Water Information Research) published a Capital
Maintenance Planning Common Framework (CMPCF) (UKWIR 2002). This set out the need for riskbased methodologies, based on the analysis of past performance and taking into account future
differences, from which conclusions could be drawn about the extent of maintenance investment
needed and the assets which should be targeted. The Water Services Regulatory Authority (WSWA,
formerly Ofwat), is now expecting water companies to demonstrate increasing levels of understanding
of service consequence risks in their business planning activities and when prioritising assets for
rehabilitation when delivering their programmes.
The Drinking Water Inspectorate (DWI) has also required water companies to develop proactive water
distribution system operational and maintenance strategies (DOMS), and that these should be
demonstrated through the submission of documents setting out these strategies (Husband, S and
Boxall, J (2006) Prediction and Control of Discolouration in Distribution Systems (PODDS).
Engineering and Physical Sciences Research Council (EPSRC)/ University of Sheffield.
http://www.podds.co.uk). Key to DOMS is the need to understand and manage the risks to customer
service associated with network operational activity.
Both the asset planning and the distribution system operational considerations set out in the CMPCF
and DOMS approaches demand innovative tools for assessing the risks to customer service
presented by assets. There is an increasing need for these tools to act at individual asset level to
assess both the general (or ambient) risks, and risks that arise when specific events occur.
This paper provides an account of the application of one such approach, Discolouration Risk
Modelling (DRM), to water infrastructure assets. The technique generates a discolouration
performance score for each pipe in a system, and may be applied in different ways to inform both
operations managers and asset planners. A review of these approaches is presented together with
an indication of how the technique might evolve in the future.

The discolouration risk modelling (DRM) approach

Discolouration Risk Modelling (DRM) is an approach originally conceived and developed by Yorkshire
Water and Ewan Group.



The principal objective of DRM is to calculate a risk score for each pipe in a system that reflects its
propensity to contribute to a discolouration event. The key strength of the approach is that it links GIS
data (pipe characteristics, environmental parameters and operational history) with hydraulic models to
simulate the consequences to the flow regime of events occurring in the system. The risk
calculations thus determine realistically, whether a suitable hydraulic trigger such as velocity increase
has occurred and will change that potential into an incident, regardless of the potential of a pipe to
give rise to discolouration because of settled material within it,.
DRM can be used to prioritise mains for replacement, relining or cleaning as part of a proactive
maintenance programme. It then allows a range of rehabilitation scenarios to be modelled, allowing
comparison of their respective risk reduction benefits to identify the most cost effective.
DRM also includes an operational analysis mode that allows water companies to assess the risks
associated with planned valve operations in part of a network. Both of these modes are explained in
more detail in subsequent sections.

Discolouration events require a number of conditions to be present.
First, there must be some change to the system configuration, resulting either from abnormal
demands, or revisions to the normal system zoning arrangements. These changes may be intentional
or accidental; for example a pipe may be closed proactively as part of an operation to isolate a part of
the system for planned works, or it may be reactive where it bursts and needs to be closed for repair.
The likelihood of bursts that tend to result in both accidental demands (burst discharge) and reactive
closure (repair) will be a function of various attributes relating either to the pipe itself or the
environment and conditions in which it operates.
Second, there must be material present within the system that potentially could be re-suspended.
This material may have been generated within the system itself through the corrosion of unprotected
ferrous mains, but may also have originated from pipes upstream of the system under analysis, or
from the water treatment works. The likelihood of a given pipe to contain deposited material will be
influenced both by its normal flow characteristics (i.e. is it working hard enough to be self-cleansing?)
and any recent history of rehabilitation or cleaning.
Finally, there must be a significant hydraulic event caused by the first condition that results in a
sufficient trigger being created to re-suspend material present through the second. It is this aspect
which makes it difficult to model discolouration generically, as the sensitivities of a system to changes
in configuration are a highly local phenomenon. Networks that otherwise may belong to the same
cohort based on age, material and diameter profiles may exhibit widely varying hydraulic responses to
pipe closures, depending on configuration. Even rezoning of a given system may result in a hugely
modified discolouration risk profile.
The principle of hydraulic triggers is illustrated by a simple 8-pipe network shown in Figure 1. When
the system is fully open (Figure 1a), water will flow from the reservoir at the top of the system around
both sides of the loop, to meet the main demand at the bottom of the system, as well as the incidental
demands at the corners. However when pipe 5 is closed (Figure 1b), there are significant increases
in velocity in the pipes on the left hand side of the loop, and flow reversal in pipe 7. Both of these
conditions create a potential hydraulic trigger that is capable of lifting sediment in the pipes that
experience these flow conditions.



Figure 1b

Figure 1a

Figure 1: The hydraulic trigger principle

DRM Software
The DRM software comprises the following components:
Risk algorithms to calculate the likelihood of pipe burst failures and the presence of sediment
based on a range of pipe, environmental and hydraulic attributes.
A control process programmed to simulate the failure of each pipe in the system in turn.
An embedded hydraulic solver, to calculate pipe flows and velocities associated with different
system configurations (i.e. those created by the sequential pipe failure simulations).
A series of searchable databases, providing reference information such as pipe types and risk
scores against attribute values.
A user interface comprising a navigational tree structure and grids within which pipe data,
from which hydraulic results and scores may be sorted and filtered.



A thematic display facility to view discolouration scores on a map.

An export facility allowing csv files to be created from most grids for more extensive
processing outside of DRM.
A SQL server database within which network details and simulation results are stored.
A standard interface that allows import of required data from varied sources via 'xml file'
The architecture is depicted in Figure 2.

Figure 2: A simple representation of the DRM software architecture

Contribution of DRM to system operations

DOMS includes an obligation on the part of water companies to understand and better manage the
risks of unsatisfactory customer service that could arise from operational activities carried out within a
water system.
DRM may be used to simulate the closure of one or more specific pipes within a system, and assess
the impact of these closures on the risk of a resultant discolouration event. The changes to the
hydraulics created by the closures are fed into the scoring algorithms in accordance with the
principles outlined above, so that the pipes that are most at risk of producing coloured water are
identified, allowing for appropriate mitigation measures to be taken. This is illustrated by the example
in Figure 3.



Figure 3 Use of DRM to assess the impact of operational activities

The ability to assess these potential impacts, which are often far from intuitive, presents a valuable
tool to assist operators in meeting their DOMS responsibilities, particularly in networks where there is
a history of discolouration problems, or a perception of high risk. Besides the more obvious technical
advantages, the investigations that have been carried out demonstrate that a diligent assessment of
risks has been undertaken when some adverse effects are experienced.

Application to assist with capital planning

The application of DRM to assess the risks of operational activities is useful for assessing specific
risks, but the approach may also be applied to assess the ambient discolouration risk of each pipe
within a system. This is achieved by simulating the failure of each pipe in the system in turn, in two
phases. Firstly, an extra demand is applied to simulate a burst discharge. Secondly, simulation of pipe
closure for repair is carried out. The hydraulic impact on every other pipe in the system is assessed
with each phase of each pipe failure, and this is fed into the risk algorithms, generating a matrix of
discolouration scores (i.e. pipe failed vs. affected pipe).
A wealth of useful information to help determine discolouration-driven intervention priorities is created
by taking the average impact values across the rows and down the columns of this matrix. The
approach helps to identify pipes that pose the greatest risk in terms of their own likelihood and
consequence of failure, as well as those which are most sensitive to other pipes within the system
during failure.



Figure 4: Use of risk profiles to prioritise interventions

Figure 4 shows the Discolouration Performance Score (DPS) profile for pipes within a DMA, plotted in
descending order. The scores shown are the sum of the scores derived for each pipe by taking
averages across rows and down columns, as mentioned above, labelled by the green and orange
score components respectively. Assuming that a notional acceptable risk threshold may be defined
as illustrated by the red line; it is then possible to select those pipes on which interventions should be
carried out (those between the vertical blue lines). However, by examining the score components for
each pipe more closely, as shown in Figure 5, it is possible to assess the most appropriate type of

Figure 5: Use of score components to guide interventions



The logic behind the conclusions drawn from Figure 5 is that should a pipe have a high DPS at a
given time when other pipes fail, but does involve its own failure, there is little benefit to be gained by
renewing it. From a discolouration viewpoint alone, there may well be other drivers for investment that
would indicate renewal as the appropriate intervention in an integrated assessment. This is because
the biggest additional impact on overall discolouration risk arising from renewal rather than relining, is
to reduce the likelihood of the affected pipe failing. But if the impact of this pipe failure on
discolouration will be minimal, there will be little justification for replacement. Cleaning these pipes in
some network configurations may be sufficient to also substantially reduce the risk in such pipes..
The converse is also true. Pipes that have the biggest impact in the event of their own failure are the
ones which should be given the highest priority for renewal, so that the likelihood of this event is
The DRM approach helps in this way to inform capital maintenance planning through identification of
the mains that should be given the highest priority, as well as indicating the most appropriate type of
intervention to best mitigate discolouration risks.
Validation and enhancement
Discolouration is a complex phenomenon, and the industry is still at an early stage of a
comprehensive understanding thereof. The DRM approach developed to date provides a framework
which currently accommodates a logical set of risk algorithms, based on expert judgement, but all
those involved with its development accept that these approaches will evolve over time as more
informed techniques emerge from science-based research. For example, the approach currently
uses velocity surrogates for the shear stress systems, which in reality will determine when sediment
within a pipe is re-suspended. At present there is insufficient available information to use shear stress,
but this one of several aspects of discolouration being investigated by Sheffield University through
field trial that is being carried out under the PODDS (Prediction and control of Discolouration in
Distribution Systems) programme. A second area of investigation is the rate at which sediment
regenerates following cleaning. The approach currently makes some coarse assumptions about this,
but there is little evidence to support them.
The risk trees themselves are expected to evolve, both as a result of the pooling of expertise which
the Club forum promotes, but more significantly as a result of case-based learning. An ongoing
process of validation over time, will establish where the DRM models have (or would have) predicted
the impact of system events accurately, and where it has not. Investigation of the latter should yield
information that helps understand how the risk algorithms should be improved. It is fully expected that
weighting factors and risk band values will be adjusted, and that the attributes considered by the risk
models will vary over time.
There is also the possibility of accelerating this validation process through purpose-designed trials.
Creating system events, while carefully monitoring the effects that ensue, will provide invaluable
information about how accurately the risk models reflect reality. Such trials are rendered feasible,
thanks to the availability of on-line monitors such as HYDRACLAM', (Salamander: Continuous water
quality monitoring via hydrants, 2005. www.intelisys.co.uk/hydraclam.asp). This system monitors a
range of parameters, including turbidity, on a close-interval basis, and Salamander, its developers,
are in the process of setting up a trial that involves a significant number of units located within a
system that allows for gathering of such information.
Other available analytical techniques will also contribute significantly to the future evolution of the
DRM approach. Currently the risk algorithms, including their structure, their included attributes, the
banded risk scores applied to attribute values, and the weighting factors, have been derived from a
pair-wise comparison process by a panel of experts. Whilst the process is systematic, it depends on
expert judgement. The validity of any modelling approach is enhanced when such judgement is
replaced by data-driven approaches. A range of technologies is now available to help analyse



relationships from data, including Evolutionary Polynomial Regression (EPR) that is being applied by
Ewan to model various service performance risks relating to freshwater and wastewater networks.
Such technologies are likely to play a major part in identifying improved discolouration risk algorithms,
as better datasets become more readily available, .allowing for the linking of causes and effects.
Looking to the future, techniques for identification and selection of suitable interventions for managing
the discolouration risks that are described in this paper are likely to become highly effective, but there
is scope for significant additional improvements through optimisation of interventions. This is because
the relationship between a failing pipe and the others on which it has impact is a highly connected
one. Examining the graphs in Figures 4 and 5, electing to reline a pipe with a high contribution DPS
(the orange part) will reduce the interruption DPS of the predominantly green pipes on the graph,
because there is less material in the system to disturb. The converse is also true. Replace a high
interruption pipe, and the contribution scores of other pipes will diminish because there is a lower
likelihood of a burst that would disturb the sediment therein. The question arises as to which has the
greatest overall beneficial impact? A significant optimisation problem is created if one adds in the unit
costs of the various intervention options available. This is best addressed using an evolutionary
learning technique such as the Genetic Algorithm (GA).
Currently, the number of calculations
carried out by the DRM software demands that the run-times for GA optimisation application involve
many tens of thousands of iterations, each requiring a complete DRM simulation, this being cost-time
prohibitive. A future refinement to the DRM approach is expected to dramatically reduce the number
of calculations needed, without compromising the quality of risk determinations. Optimisation is
regarded as a real possibility that Ewan will be actively exploring in the future.
Other associated approaches
DRM is one example of pipe-level risk models that is being developed and applied by water
companies in their water system management and planning activities. Similar service models are
also being developed for other areas of service including DG2 and DG3. Assigning risks to a pipe
level for these service measures creates challenges because, as with discolouration, the impacts
depend on highly localised factors, particularly system hydraulics. The technology applied for DRM
may however be adapted to provide valuable information to show how individual pipes can contribute
to a risk of other aspects of network performance.

The water industry is moving towards risk modelling of its water infrastructure assets at individual pipe
resolution. This paper has aimed to demonstrate how modelling techniques may be developed to
enable this objective to be met, using the modelling of discolouration risk as a key example.
Infrastructure assets, being underground, are invisible, and it is therefore difficult to understand the
relationships between attributes, environment, operation and performance event with clarity. All riskbased modelling techniques will need to undergo a long-term validation and evolution process to
improve the relationships, as better information becomes available. The collection of good data,
taking advantage of modern monitoring approaches, is a critical part of this process.
It is essential that water companies embrace infrastructure risk models fully to get them to work
effectively, so that service-providers have a chance of ensuring business as usual operations as the
norm. The DRM software outlined in this paper has been built with a view to allowing eventual
incorporation into a companys corporate systems. It is hoped that, in due course, it will, for example,
be a common occurrence for the DOMS team to run a DRM model for a DMA where it is perceived to
have high discolouration risks, before any system operational changes are made.
There is future scope for exploiting significantly greater value from risk models such as DRM by
applying optimisation technology. Ewan sees this as an inevitable development that is becoming ever
more feasible as both computing technology and system understanding continue to improve.



Anglian Waters strategic investment approach - A case study on its initial

application to wastewater infrastructure
G Kemp BSc; CEng; MICE (Member) Strategic Investment Manager, Anglian Water
R Long BSc; CEng; MICE (Member) Mouchel Parkman (Formally Ewan Goup)
(Acknowledgement First published and presented at the 4 CIWEM Annual Conference, Newcastle
Upon Tyne, UK, September 2006)

Anglian Water undertook a fundamental review of how it should approach the development of a longterm forward-looking investment plan following the submission of the company Strategic Business
Plan in 2004 and the associated Final Determination from Ofwat. The increased emphasis on asset
management as a business process, coupled with a more developed understanding of the application
of the planning principles promoted in the 'Capital Maintenance Planning, a Common Framework'
(CMPCF)(UKWIR 2002) has provided a clear focus for enhancing and evolving existing processes
and methodologies.
The review was initiated by a desire to build an integrated process that was an essential part of the
company planning activities, and was not primarily driven by regulatory requirements or directives.
There was a strong feeling that if the processes for developing the forward-looking investment plan
were structured correctly, the regulatory aspects to permit price setting would fall naturally into place,
and thus any periodic reviews would merely punctuate the process, rather than be the introspective
and soul-searching exercises that they seem to have become.
Delivering the right process will lead to significant benefits throughout the company. There will be
increased certainty over customer performance standards and the risk parameters that are used to
develop the plan. Customers will have more transparency regarding the service standards they may
expect from their bill payments, the stakeholders will have more certainty over investment and longterm returns, and our economic regulator, Ofwat, will have more certainty that the business is being
managed in a sustainable manner, with an eye to long-term delivery of excellent customer service
Successful delivery of such an integrated investment approach will provide significant primary and
secondary benefits. Anglian Waters Strategic Asset Management (SAM) team, part of the asset
management directorate, was tasked with developing such an approach, and following approval of the
concept, implementing it across the company. A key objective has been that the approach could be
used to define the investment proposals for the strategic business plan for the periodic review in
2009, as well as forming an integrated business process. It was recognised that involvement of .the
rest of the business was absolutely key to successful delivery of the approach.
Detailed consideration was given to the Ofwat assessments of our capital maintenance submission
following an internal review of the PR04 submission and definition of those areas that were robust,
together with those where improvement was required,
Later stages of this paper focus on the application of the approach to wastewater infrastructure, and
the Ofwat assessment of the industry-wide position on these capital maintenance submissions. Figure
1 depicts this below.



Weighted Mean = 53.0%

Upper Quartile = 61.7%
Lower Quartile = 43.3%

WaSC Sewerage Infra

well-structured case?
efficiency integration
(set to 5)
offset uplifts?
WLC approach
links to co policies
corporate systems based
or standalone


data acquisition
availability / formats

confidence grades
degree of reliance
on expert judgements
risk-based or
degree of risk
top down / bottom up?
Reporter involvement?
evidence of R&D/best practice

Figure 1: Wastewater infrastructure: Ofwat capital maintenance assessment

Note that relatively poor scores are apparent across the industry in a number of areas. Our internal
assessment of those areas where we felt our submission to be weak were:
An integrated business process (corporate systems or stand-alone).
A whole-life costing approach.
Top down / Bottom up process.
Fortunately there was a general alignment between those areas perceived to be weak and those
areas where we believed there were significant gains to be made in an overall asset management
and business efficacy response to counteract these weaknesses.
Integrated business processes and whole-life costing fit very well with our business philosophy, and
were clear areas for improved process and integration.
Past investment in software processes and corporate systems (such as SAP and GIS) had led to a
recording of the asset base at a disaggregated level. Job scheduling, alarm messaging and asset
attributes are, or will be held at this disaggregated level. (To put this into context for noninfrastructure, this equipment level is at the pump or motor level, and for infrastructure items, it
corresponds to a pipe-level analysis).
To fully utilise this past investment and deliver effective asset management strategies there seemed
to be clear alignment between those areas where we perceived the need for improvement and the
regulatory perspective of areas needing improvement. This has lead to the development of the
conceptual modelling approach, as is shown below.



Set St rat egic Object ives The translat ion of

strat egic objectives into t angible measures

St rat egic
Object ives
M anage

Int ervent ion

M anagem ent

Int ervent ion


Int ervent ion

Invest ment
Opt im isat ion

M anage Perf orm ance Risk The monitoring and

managing of asset performance against set
measures t o assess t he future risks of performance
Cost Int ervent ion The costing of solution
interventions f or both capital and operat ional
Int ervent ion Opt im isat ion The optimisat ion of
solutions against strat egic objectives
Int ervent ion Planning The grouping and
planning of solutions in an optimal w ay
Int ervent ion M anagem ent The monitoring of
the delivery of solutions t o achieve strat egic

Figure 2: Strategic investment planning

The key focus at this stage in the development process has been on the manage performance risk
element, where the in-depth analysis of asset deterioration, asset risk parameters, linkage to
customer service standards and definition of appropriate interventions will take place. This process is
where the bulk of the analytical effort will take place in the regulatory and business-planning cycles,
and thus requires one of the longer lead times for effective implementation. Whilst the present focus is
on developing the analytical processes for the capital maintenance elements of our investment
planning process, the fully developed process will be designed to accommodate the investment
analysis to meet all regulatory and customer drivers.
The approach, as shown, is intended to be iterative. However, during the plan development and
delivery phases it is likely that subgroups will be formed which will create an iterative process each in
their own right. For instance, during the development of the periodic review submission it is likely that
the strategic objective setting, manage performance risk, costing interventions and optimisation will
form such a group where revisions to strategic objectives may need to take place when investment
levels are known (perhaps aided by customer willingness-to-pay surveys) and conversely investment
levels will need to be adjusted to meet fixed performance-related strategic objectives.
This extensive workload to develop the manage performance risk element, to provide effective
modelling approaches, is being supported by external consultant expertise for both infrastructure and
non-infrastructure investment areas, and builds upon the extensive risk and value processes created
to support the delivery of our AMP4 programme. These provide valuable consideration of probability
and consequence, which will be fed into our strategic work.
The case study in this paper focuses on the development and application of a modelling approach for
wastewater infrastructure through the approaches proposed by Ewan Group, one of the consultants
appointed to provide support in the development of our strategic approach. Wastewater infrastructure
has been identified by Ofwat as one of the weaker areas of analysis and CMPCF application across
the industry and it was therefore seen to be appropriate to prioritise this area for early consideration.

Overview of technical approach

The method is an application of a failure modes and effects analysis using Ewan Groups IPaD
(Investment Planning and Delivery) approach, to develop a proper understanding of the links
between asset attributes, performance and service. It is a data-driven method that minimises reliance



on expert judgement. The foundation of the project teams approach to investment planning and
delivery is conceptualised in the IPaD matrix illustrated in Figure 3.

Figure 3: Example IPAD matrix for sewers

(The given relationships are for illustration purposes only)

This IPaD structure demonstrates the linkages between asset attributes and failure mode (asset
performance). It also enables demonstration of the linkage between interventions and their impact on
attributes, so that the benefit of interventions may be properly assessed in terms of the improvement
in prediction of a probability of failure that they deliver. A prediction of the impact of an intervention will
therefore be made by considering the intervention impact on attributes, then the effect of the change
in attributes on asset performance, then the impact of the change in asset performance on service
Performance deterioration models will be generated at a pipe or individual asset level for sewers
(critical, non-critical, exS24), tunnels, pumping mains, vacuum sewers and their ancillary manholes,
CSOs and tanks.
Models of deterioration probability or likelihood will be generated from data by an innovative
regression analysis approach used successfully by the project team in the recent UKWIR
deterioration rate of sewers project. This method produces relationships between asset performance
and asset attributes in the form of simple equations that can be tested for engineering reality. The
equations can then be used to test the impact of interventions, so that a programme of measures
providing the required performance at least whole-life cost can be drawn up. The method is ideally
suited to progressive updating, as more and better data are gathered in the future.
Service failure consequence models will be developed for each failure mode, in addition to the
likelihood models. Ewan Groups FastNett sewer network simulator will be modified to do this, to test
the impact of progressive and total failure of each pipe in the system, both in terms of sewage
volumes leaving the sewer system through flooding or CSO spill, and the pathways and receptors that
are affected. These consequences may be flooding of roads and houses, and pollution of
The method is risk-based, forward-looking and service-led, and is wholly compliant with the principles
of the CMPCF.



Risk number
An approach that can assign a risk number to individual pipes is required. The concept of the risk
number is a combined score of the probabilities of each of the service failure modes that can be
predicted for that pipe, and the consequence attached to each. The processes described above will
give all the information needed to determine the risk number for every critical and non-critical sewer
pipe. It is intended in future that the consequences will be expressed in monetary terms, so that
flooding and pollution impacts can be compared on an equal basis.
Anglian Water has continually improved its data collection and processing over recent years, and the
current regime is now providing high-quality data in a form needed for calculating the risk number.
The methodology is based on data streams that will be continually improved over the next few years,
thus also continually improving the accuracy of the results.
The first stage is to develop predictive models for collapse and blockage likelihood at a pipe level.
This requires analysis of a dataset containing sufficient collapse and blockage incidents for which the
identity of the failed pipe is recorded. Critical and non-critical sewers can be taken together for this
analysis, but because limited location or attribute data are available for exS24 sewers, these are
treated separately.
The second stage is to relate performance to service. This requires an understanding of what the
hydraulic effect of a collapse, blockage or change in condition will be in terms of how much sewage
will leave the system, where the flooded or spilt volume of sewage will go, and what consequential
impact it will have. Sewer criticality has less relevance in this context and so critical and non-critical
sewers are considered together. Because exS24 sewers are rarely if ever represented in hydraulic
models, they will again be treated as a separate group.

Data assembly
There are five main groupings of data that need consideration. These are:
Attributes relating to the sewer pipe as it was built, such as age, size, material, etc. These
data will generally be found in the GIS sewer record database to varying standards of
completeness and accuracy.
Attributes relating to the environment in which the pipe exists, such as soil type, traffic
loading, frequency of surcharge etc. These data will be compiled from a variety of sources
such as soil maps (LEACS, HOST etc), traffic data from OSCAR and surcharge frequency
from drainage area plan models.
Attributes relating to the present condition of the pipe, such as intruding connections,
displaced joints, sediment etc. These data will be compiled from the results of CCTV surveys.
Intervention data such as root-cutting, repairs and renovation. These data will be compiled
from corporate systems. It may also be necessary to extract intervention data from previous
systems in cases where earlier intervention could be affecting more recent performance.
Performance failure and service incident data. Collapse and blockage data will be compiled
from incident data held in Anglian Waters corporate system.

Likelihood of performance failure

It is important to recognise that deterioration of sewer systems is a complex process that is influenced
by many variables. Generally there is not a simple relationship between use and condition; in other



words, sewers do not simply wear out with use in the way that for example a pump or a vehicle would.
This raises important issues:
Sewer deterioration can often be triggered by unpredictable external events.
In many cases there is a dearth of knowledge about the circumstances that will tend towards
faster or slower sewer deterioration.
As a result, it is not possible to predict when any particular pipe will collapse.
The large number of potentially significant factors affecting deterioration, and the paucity of repeat
CCTV over a sufficiently long time-span creates uncertainty to methods that rely on observations of
changes in specific pipes over time. Currently, an extrapolation to the general population is liable to
error. This is particularly the case when the method uses changes in condition grade, being a crude
instrument for describing condition.
What is possible, however, is to predict the likelihood of collapse of pipes within a particular
population. This has been demonstrated by studies that Ewan Group and Exeter University carried
out in the 2005/6 UKWIR Deterioration rate of sewers project. In this, the population was defined as
the sewers in a particular geographic area, defined by having a similar age of development. The
method was successful in development of equations that predicted the likelihood of blockage and
collapses in two pilot sewer networks with a confidence as high as 90%.
For Anglian Water, the approach that has been developed for UKWIR is being extended from an
area-based approach to a pipe-based approach, such that populations are defined as cohorts of pipes
of similar characteristics, taking advantage of the improved standards of data collection and storage in
which Anglian Water is investing. These characteristics will include indicators of condition at a pipe
level. It will be possible in this way to develop a linkage between deteriorating condition and
deteriorating performance, to predict the performance of the populations of pipes into the future. This
is important, because although investment strategies under the common framework are driven
principally by service, which derives from performance, the optimal timing of interventions may be
controlled more strictly by condition, as condition could have deteriorated too far for relining to be
feasible before any significant deterioration of service is noticed.
The performance equations developed will be applied to the full asset stock of critical and non-critical
sewers. The output from this process will be predictions of the risk of blockage, collapse and
deterioration of condition at selected future dates, such as 2020, 2030, and later.

Regression modelling
Predictive performance models are being developed using a sophisticated but pragmatic approach to
regression modelling known as Evolutionary Polynomial Regression (EPR).
The EPR software has advantages over other methods such as neural networks and more traditional
regression tools in that expert judgement is minimised, and the analysis can be quickly repeated in
the future as further and improved data become available.

Consequence of performance failure

A failure modes and effects analysis will be specified for each of the asset types defined in the brief.
Service failures resulting from collapses, blockages and deteriorating condition will be flooding other
causes and environmental pollution. There are two stages in predicting service risk from performance
Firstly, when a collapse, blockage or change in condition occurs in a particular pipe, the question
asked is how much sewage exits in the sewer system and where does it leave? Information for both
dry and wet weather conditions is required. This can only be determined quantitatively using hydraulic



models. Initially the process will be carried out using Anglian Waters existing network models, so that
typical profiles can be established and then applied across the region. In time, as more models are
updated and completed, it will be possible to apply the process to each pipe. This work will be carried
out using a development of the FastNett sewer network simulator. This can be adapted to run a
sequence of simulations automatically where the effect of progressive loss of capacity of each pipe in
the system can be tested, pipe by pipe. The output from this process will be a table of flood and spill
locations and volumes for each pipe in the model, as it progressively loses capacity and fails under
both dry weather and wet weather conditions. FastNett interfaces with Wallingford Softwares
InfoWorks and can therefore make use of any models from Anglian Waters model library.
Having determined quantities of sewage leaving the system, it is then necessary to determine the
origin and destiny of this escaped sewage, and what impacts it will have. The impacts may be
flooding, pollution of the environment, or both. Using surface routing techniques based on ground
slope from a digital terrain model, coupled with GIS data defining the location of buildings and the
nature of the surfaces on the flood path, it is possible to estimate the extent of inundation that would
be caused by various volumes of flooding from each potential source. The impact caused by the
inundation can then be monetised using standard impact costs developed by Anglian Water.
Likewise, pollution may be caused by spill from a CSO, or by overland flood flow being intercepted by
a water course. In either case, the location of entry to the watercourse and the volume of sewage is
known, so that the impact can then be estimated on the basis of readily available data such as
available dilution and high-consequence locations downstream, such as SSSIs and abstraction
points. The pollution events can then be categorised and costed, as has been indicated.
Finally, the risk associated with failure of each pipe can be determined by combining the likelihood
and consequence predictions for each pipe.

The good progress being made on the wastewater infrastructure pilot study has provided reassurance that the overall direction of the investment approach is correct.
Full application of asset analysis at the equipment/pipe level is feasible and will provide
significant benefits in the medium and long term.
Regulatory submissions will become more robust and the delivery of a business wide
integrated investment approach will ultimately deliver customer benefits through the
development of the most effective and efficient approaches.

Other relevant literature

Sewerage Rehabilitation Manual (4


edn.) WRc 2001.

Savic, D, Giustolisi, O., Berardi, L., Shepherd, W., Djordjevic, S., Saul, A., (2005) Modelling Sewer
Failure by Evolutionary Computing. Proceedings of the Institution of Civil Engineers Water
Management. pp 159.



System planning and the asset management process adopted within

Northumbrian Water Limited.
R Wooldridge and S Walker
Northumbrian Water, Abbey Road, Pity Me, Durham, DH1 5FJ, UK.
(Acknowledgement First published and presented at the 4th CIWEM Annual Conference, Newcastle
Upon Tyne, UK, September 2006)

Northumbrian Water Limited (NWL) provides water and wastewater services to a resident population
of around two and a half million in the North East of England. 424 sewage treatment works collect
wastewater via 689 sewage pumping stations, 1471 CSOs and 15711km of sewers.
NWL has adopted a system planning approach to review the performance of assets and link it to the
management of issues in wastewater assets (network and treatment). There are 11 wastewater
systems covering the north-east region. Before commencing the production of system plans for
wastewater, it was necessary to decide what a manageable area for performance analysis was, both
in terms of the sewerage network, including CSOs, SPS and storage tanks / chambers, and the
sewage treatment facilities provided. The result was as follows:
A system is defined as one or more catchments contributing to a river basin or a particular
reach of river, the split of the region into systems can be seen in figure 1.
A catchment is defined as one or more drainage areas, all leading to a treatment works.
A drainage area is defined as a geographical area containing sewers which all lead to a
common point, such as a CSO or SPS.

Figure 1: NWL region showing the 11 Wastewater systems



Figure 2: System 2 showing river basins and drainage areas

Figure 2 shows an example of how a system is then split into river basins and then drainage areas.
System 2 (South Northumberland) is used for illustrative purposes. There are two levels of system
planning meetings, each operating at a different level of detail grouped detailed System Planning
meetings and strategic Regional System Planning Meetings.
In the local detailed system planning meetings the systems are approached in groups which are
based around the rural northern end of the region (Systems 1 North Northumberland and System 2
South Northumberland ) and the 3 major rivers in the region the Tyne (Systems 3 Tynedale,
System 4- Derwentside and System 5 Tyneside), the Wear (System 6 Weardale, System 7
Central Wear and System 8 Wearside) and the Tees (System 9 Teesdale, System 10 Skern and
System 11 Teesside). This can be clearly seen in figure 1.

The concept of system planning has been evolving in NWL for around 5 years, it is now an integrated
part of business as usual with the local grouped system planning meetings held every 6 months and
an annual strategic review which covers the entire region. System planning addresses all issues, and
is not limited to infrastructure, thus ensuring that full interactions are covered for all asset types.
Representatives from a variety of areas of the business attend the system planning meetings
including those involved in asset management and technical strategy, operational activities including
performance and compliance, investment delivery and management of new development.
The early system planning meetings were based on sparser information and tended to focus on
issues within catchments rather than looking for connectivities and integrated solutions. However,
there were some notable examples of successful identification of issues that had wider implications
than realised by their asset owners.



This was exemplified in the Consett Catchment where a new pumping station had been built by a
developer and transferred flows from a small area that previously had been served by a different
treatment works. The increased flows had been noticed by the works manager who was then made
aware of the reason in the meeting.
Previously the main output from the meeting would be a system status report. This would traffic
light the status of a system based upon the perceived performance of the sum of catchments
(drainage areas). This was done by a simplified scoring of all of the specific performance indicators
discussed in the meeting e.g. number of pollution incidents. It crudely identified the performance of
the system by adding up all the following which could contribute to poor serviceability : identified
issues at location level, known flooding issues, compliance trends, health and safety issues,
development potential, customer contacts.

Developing a Risk Based Approach

In May 2006 we embarked on a new round of system planning meetings. These meetings have taken
on a new approach which is the first step towards moving away from issue identification to analysing
a systems performance and starting to look at identifying solutions. It also involves the refining of the
outputs to produce more meaningful system status indicators, which in turn will also prioritisation of
work in drainage areas showing greater risk.
The new system will formalise the scoring of drainage areas up to system level with currently
available information, by use of scoring and weighting methods. Aggregation of drainage areas risk
scores will allow prioritisation to be carried out across systems. Drainage areas with the greatest risk
of failure will be subject to further detailed analysis via the catchment planning process.

Catchment Planning Process

The system planning process is promoting the use of a catchment planning philosophy to support the
development of bottom up analysis to complement the existing top down approach that has been
used for previous Periodic Reviews. This will help address the regulatory pressures to be more
proactive and should provide efficiencies both in capex and opex. Catchment planning also allows us
to embed the common framework principles into business as usual, whilst supporting established
processes such as AMPS. The process consists of a three phased approach, dealing initially with
planning, followed by a diagnostic study with finally the development of solutions. These solutions will
be subjected to the full risk based prioritisation methodology currently in use to determine the future
capital programme of work, and upon resolution the paperwork to enable resolution will be entered
into the asset management system.

Asset Management Process System (AMPS)

AMPS is a new workflow system that was introduced to NWL in 2005. It is an oracle based system
which allows users to log issues, which are then tracked through the process from identification
through to a solution being implemented.
In order for an issue which requires resolution to enter the NWL asset management process , a CP0
form must be raised electronically in the AMP System (AMPS).
An issue in the context of the AMP can be defined as:
a problem that has or will cause the Company:
o To fail to meet target levels of service on a sustained basis.

To fail to meet target levels of performance on a sustained basis.

To make a sustained operational change.



an opportunity that will allow the Company:

o To improve performance, effectiveness or efficiency on a sustained basis.
The solution to an issue may require an intervention and change to the way the asset is operated or
maintained. The intervention can be operational or investment. The process diagram can be seen
below in figure 3.

Issue Originator

Asset Failure, Performance Review, Optimisation, System Planning, Regulatory

Issue Identification


approval by
line manager

More info

Asset Manager

Issue Review

System Review

Asset Needs

Opex solution

Operational Change


Defined solution issues not

exceeding 2000

Investment Project


(Ops & ID)


More info

Figure 3: AMP Process Overview



Asset Indicators used is System Planning

There are a number of sources from which indicators can be drawn of the risk of failure of any of the
assets contributing to the serviceability of a drainage area.
The following table (figure 4) identifies the information that is used, and shows which asset type is

Asset Groups














DG5 hydraulic
incapacity confirmed

Floodlog, DG5







DG5 hydraulic
incapacity under

Floodlog, DG5







Property flooding
due to Blockages

June returns







Percentile sample

Corp docs







UCSOs & SPS eos








Pollution incidents








Development sites








H&S dept







Plus 2















Health and safety
Customer contacts

Figure 4: Asset indicators used for system planning

Each contributing indicator will be given a traffic light of red, amber or green depending upon the
result of the analysis undertaken from the data available as shown in the table. It should be noted
that this approach mimics the current and recent performance of the drainage area, some of the
indicators are resolved by the time of the meeting and are indicative of where issues have been
A map or schematic diagram of the river basin is presented to the stakeholders showing the current
status of each drainage area for each asset indicator, an example can be seen in figure 5a and 5b.



Figure 5a: River Wansbeck River Basin map (part of System 2) showing traffic light indicators for
the number of outstanding CP0s

These maps help to give a visual representation of the special spread of drainage areas and help to
give an indication of the interconnectivity of different asset indicators/asset groups.









Swilder Burn





No indicated risk

STW - compliance traffic light

STW - compliance traffic light or

SPS, CSO - EA unsatisfactory intermittent discharge









Bothal Burn



How Burn





Shieldhill Burn








r Fo








South Northumberland
System 2


May 2006

Risk of Pollution

Water FrameworksDirective: At Risk

Water FrameworksDirective: Probably At Risk:

This schematic is not to scale

Figure 5b: Pollution Risk schematic of the River Wansbeck River Basin (part of System 2)



The schematic shows all the assets (STW, SPS and CSO) that are located within that river basin.
They also show the drainage areas and the point of discharge into the river. The river is colour coded
to show how the EA classify it in terms of perceived risk for the Water Framework Directive. This is
intended, in time, to help us determine if we could be contributing to the risk of failure of a stretch of
river under the Water Framework Directive.
The summation of all the indicators will give the members of the System Planning meeting a visual
picture of how to score the drainage area itself. This will be done by expert opinion, evaluating the
combined traffic lights to a simple score of 1-10 for both the likelihood of failure and consequence of
failure relating to the drainage area. The following tables (figures 6a & 6b) give guidance for scoring,
it should be noted however that in terms of the consequence score the score given can be increased
based on the knowledge of the experts at the meeting. For example, if we know there are pressure
groups in a particular area then we know that there is potentially going to be a higher interest in any
issues within that area, and therefore potentially a higher consequence.

Consequence of Failure

Likelihood of Failure



(Based on D.A. Population)



201 500


501 2,000


2,001 4,000


4,001 7,000


7,001 10,000


10,001 20,000


20,001 40,000


40,001 60,000


Regional / Company Wide






Figure 6a: Likelihood of failure scoring system

Figure 6b: Consequence of failure scoring

To assist in this process, a summary table of the traffic light for each indicator has been produced, for
each river basin. An example is given below (figure 7):



River Wansbeck
02-D13 (p) Morpeth
Miners Homes, Longhirst
Rose Cottage, Longhirst
Scots Gap & Cambo
02-D45 (p) Ashington
02-D46 (p) Bedlington & Cambois




due to


SPS eos


ent sites

and safety

contacts total

contacts odour

Figure 7: Summary table showing traffic lights for asset indicators for The River Wansbeck River Basin

Risk Analysis
The purpose of carrying out risk analysis at System Planning meetings is to understand where there
may be a particular concentration of risk which may need to be addressed in the wider context of
Catchment Planning, for integrated solutions.
However, there are too many assets to consider on a unique or individual basis and too little failure
data linked to assets, so to assess the current risks within each drainage area at the river basin level.
The hierarchy used is:
Assets Drainage Area Catchment River Basin System
Example scoring table for the Wansbeck river basin (part of System 2) is shown in figure 8 below. The
figures for likelihood and consequence scores are illustrative only. Some drainage areas will be all
green and these will be routinely scored prior to the meeting but verified by a quick round the table

Likelihood Conseque Total risk

nce score score
River Wansbeck
02-D13 (partial)
Miners Homes, Longhirst
Rose Cottage, Longhirst
Scots Gap & Cambo
02-D45 (partial)
02-D46 (partial)
Bedlington & Cambois
Figure 8: Risk Scores for River Basin



Each contributing drainage area in each of the river basins will be scored at the local system planning
meeting, and the following chart (figure 9) produced.

System 2 Drainage areas

Wansbeck river basin




and Cambois


Longhirst, Rose
Cottages, Scot's Gap






Figure 9: Drainage Area Risk Score Graph

In each hierarchical level, there has to be assumptions about the aggregation of risk scores, and this
is done simply by adding the scores and then normalising them by applying a weighting factor. Each
process will then have a probability score up to 10, and a consequence score of up to 10.

System 2 river basins













Figure 10: River Basins Risk Score Graph

A plot of the normalised likelihood score against the normalised consequence score for each System
can then be shown, as below in figure 11.



Wastewater System analysis



Increased tendency to failure






Increased potential impact

Figure 11: Wastewater systems Risk Score Graph

Strategic System Planning

The next regional strategic system planning meeting will be held using the results of the new risk
based approach, whereupon it is envisaged that the graphs showing the system risk scores will be
shown, as well as graphs showing the highest risk scoring river basins and drainage areas. There will
also be additional data available for analysis such as customer survey data which will not be neatly
categorized into systems but may help to provide an overview as what is important to our
customers/what perceived problems we have. It is envisaged that the strategic system planning
meeting will concentrate on the systems, river basins and catchments that have the highest risk
scores. The ultimate aim is to be able to identify any common causes in systems/drainage
areas/catchments that are presenting the greatest risk to the business and produce a action plan to
solve these issues.

Future Developments
In the future as part of the PR09 process, the current risk scoring approach will be superseded by
more refined techniques, based on data, including the use of tendency tree analysis by combining
the tendency to fail with the consequence of that failure, thus effectively given a risk score. These
measures will be compared with Company targets, and will generate more sophisticated System
Status reports.
In the future, the data which is being collated into the corporate systems will become the source of
information, based upon comparative performance measures and ratios, and cost performance ratios,
using a new IT package called Amulet. This will allow us to carry out the appropriate level of analysis
to predict performance of drainage areas and their contributing assets against targets, and thus build
in the requirement for using Common framework principles into business as usual



1. The use of System Planning in the business has a proven track record, even though it is still
essentially in its infancy.
2. Planned improvements will drive us to target investment into the right (most risky) areas.
3. Further development will ensure the process is fully compliant with the Capital Maintenance
Planning Common Framework as required by the Industry Regulators, Ofwat.

All data entries are fictitious and do not necessarily represent the state of the asset base.
Any views expressed in the paper are those of the Authors and do not imply that they have
been accepted by NWL.



A case study on water mains rehabilitation planning and implementation,

United Utilities, UK
United Utilities serves a population of 7 million people in the north west of England, extending 100
miles north to south and 50 miles east to west. Water is distributed to customers from 150 water
treatment works in the region though a network comprising 40 000 km of mains. The mains have
been installed over the last 130 years and comprise a variety of materials and sizes. As the region
experienced its greatest growth during the 19 and 20 centuries, many of the mains are as much as
100 years old and the predominant mains material of construction is cast iron.

Figure 1: Service area of united utilities in the UK

Mains were also largely unlined at the time of their installation, although some had bitumen coatings.
Water supply is predominantly upland impoundment and this provides waters that are soft in nature.
In 1985 the European Union set standards for drinking water quality that were incorporated into UK
regulation requiring that an upper limit of 25 mg/l total iron content be applied.
United Utilities broke its water distribution network into 180 supply zones that are routinely sampled at
customer's taps and the results reported to the Drinking Water Inspectorate. It became apparent from
historical records that many supply zones would not comply with the upper limit for iron content.



This case study sets out how United Utilities set about understanding the cause and scale of the
problem, and agreed on a programme of work with the regulators to rectify the situation, then
implementing the improvement programme. A number of innovations arose as a result of the work.

Understanding the problem

United Utilities had a patchwork of water supply zones across its region. Some zones had been
supplied consistently from the same source, but others were in areas where a number of sources
would be used to balance the supply. A series of planning zones were created. These were areas
that not only had a discrete water supply, but areas that also had a similar history of supply. The
simplest example would be an area that had been supplied from one source throughout its life.
Another example might be a zone fed from one source until 1950 and then from another source
thereafter. The region was divided into 180 water supply zones. These water supply planning zones
were mapped onto a geographic information system (GIS) background, on which performance data
could be displayed. This included mains data, although reliability thereof has been variable.
Innovation 1: Historical water supply quality data were used for the purpose of investigations to
divide the zones into characteristic areas; long-term soft supplies, long-term hard supplies, and those
with a mixed supply. Historical records of all water sampling and analysis results were available, and
these could be used to show the historical levels of iron in water, as measured at customers' taps.
Innovation 2: Rigs were built in order to evaluate the corrosivity of different waters. Rigs comprised a
series of pipe-loops into which unprotected specimens of iron were immersed. There were two loops
of different mains diameter to provide results for high and low velocity flow. The rigs were set up at a
number of supply works and the relative corrosivity of the waters was measured. Results from the
tests showed that soft waters are more corrosive than the hard waters.
Innovation 3: A sampling and analysis regime was established, named downstream series
sampling. Samples were taken along the route of supply, starting at the source of a zone and then
following the route to the extremity of the zone. Calculations were used to structure a sampling
regime that drew samples from a broadly-consistent hypothetical slug of water, as it passed through
the zone. Samples were taken at customers taps and these were analysed for soluble and
particulate iron. Routes were selected along mains that from records were known to be
predominantly unlined cast iron, and as a control, a route along non-ferrous mains was also included.
Innovation 4: Remote automatic sampling devices (ALFIs) were developed for use on the network.
These allowed downstream time-series sampling over a 24-hour period to ensure that any diurnal
variations of data were captured.
Results from ALFI sampling were then used to establish the amount of iron accumulation taking place
in water from mains piping within a zone, and this could then be used to establish the need for mains
rehabilitation in that zone.

Agreements with the regulators as to needs

By 1985 United Utilities had undertaken sufficient sampling and analysis to demonstrate that iron
accumulation was evident from significant part of its network. Discussions with the Drinking Water
Inspectorate led to an agreement that the company would continue investigations throughout the
remainder of its region and initiate a comprehensive programme to rehabilitate their mains.
United Utilities, in terms of this agreement, proposed to the economic regulator OFWAT in its Asset
Management Business plan that it would implement this programme. A schedule of mains
rehabilitation actions, with demonstrable water quality improvement was agreed upon, and that this
would be funded from charges to customers.



Developing the solution

United Utilities had information from historical water quality analyses at customer taps to be able to
plot the planning zones for the region, and to show their relative performance. Zones would be
addressed on the worst first basis.
The GIS was used to map the mains in an area and this could show the general pattern of the
network, as well as give information on mains materials and sizes. Historical knowledge of mains
materials could be used to provide informed predictions about material with poor records in some of
the areas, and this would be crucial information for site investigation.
Innovation 5: Network models were created for the region, and these incorporated mains information
from GIS, as well as physical data on lengths, levels and customer locations. Data collation and
model creation were initially and largely manual and time-consuming, prone to inaccuracy, and
resource-consuming. Once a protocol had been established, the collation of data from customer
records, water quality archives and from GIS could be automated, and a package using Map-Info
and Stru-map software could be developed, not only enabling models to be created, but also
incorporating quality control checks and data verification. Field measurements were also taken using
data loggers to record levels in service reservoirs, as well as flows and pressures in the network, as
well as on supplies to large industrial customers, as this could influence the network performance.
Again, the data from these investigations were combined to provide a necessary database from which
the rehabilitation planning would structured. United Utilities undertook an integrated approach to
planning, and whilst the main driver for the system rehabilitation was water quality, it was also
necessary to consider the hydraulic adequacy of the network and its security of supply, as
represented by structural performance. The Map-Info system was used with water quality and
hydraulic performance data to plot mains bursts, as well as include customer complaints (water
quality, no supply etc) derived from the companys archives.
Innovation 6: United Utilities developed Poseidon, a network programme that could be used to
design the optimal rehabilitated network. Any network design has to address conflicting requirements
in the process of an optimisation. Contradictions can include issues such as requisite hydraulic
capacity for fire-fighting, whereas a system just big enough for water quality purposes would
otherwise be optimal. A high level of cross-connection in a network would be desirable in terms of
security of supply, whereas a trunk-and-branch system to customers would be best for water quality.
United Utilities introduced a series of about 2000 discrete leakage-monitoring zones and district
metering areas (DMA) in the network, and flow-modulated valves would manage pressure. . All the
conflicting needs could be incorporated into the Poseidon programme to achieve an optimised design,
but this also needed an intelligence on the suitability of alternative rehabilitation options. Figure 2
shows a screen display of a Poseidon output, with rehabilitation techniques of slip lining, die-draw,
upsizing, and pipeline insertion, all being shown by colour-coding in the example zone.



Figure 2: Poseidon output

Innovation 7: United Utilities has been undertaking water mains rehabilitation over a period of 20
years and has experience of the constraints, benefits and performance of the various technologies
available for replacing or relining pipes. These also take into account the performance on leakagereduction and also how the technologies can work in conjunction with replacement or rehabilitation of
customers' supply pipes. Contract prices for mains laying and associated fittings are available for all
rehabilitation options, and these have been incorporated into the Poseidon programme, enabling a
fully cost-estimated design for the rehabilitation to be generated, and this would be for all works in any
supply zone.
Innovation 8: Once the optimal designs for mains rehabilitation become available, these are then
passed electronically to the contract team who then determine the best approach to the project, taking
into account customer water supply disruption. Historically, United Utilities has taken a conservative
approach to project planning and has built in adequate allowance for site investigation. Programmes
can be developed using experienced contractors, and allowance made for contingencies including
unscheduled complications. This enables significant savings with respect to site investigations
Innovation 9: Project execution involves a comprehensive planning approach, to ensure effective
liaison with all stakeholders, including customers and highway authorities. Customer liaison begins
typically six months prior to work commencement, often with community council meetings, where
plans are set out and feedback recorded on matters under consideration. Liaison with highway
authorities can also influence the rehabilitation technique to be used, with preference for no-dig
technologies. 80% of the mains in an area are typically rehabilitated by inserting a polyethylene pipe
inside an existing main.

The main driver for the programme has been water quality improvement. A follow-up DSS after
project completion in a zone, with results, will show whether an improvement in water quality had



been achieved. The rehabilitated zone can also undergo other changes such as pressure
management and measurement, to ensure that the system performs to expectation. A revised model
of the rehabilitated network is generated, and is used for future management of the system, including
leakage management. Post project rehabilitation assessments for each water supply zone are
formally submitted to the quality and financial regulators, in order to sign off their acceptance
Figure 3 shows a typical result of mains rehabilitation in a water supply zone. The change in iron
levels from routine water quality sampling shows a marked improvement.

Figure 3: Water quality iron results for the Huyton East water supply zone

Achievements, and lessons learnt

Water distribution mains rehabilitation has been carried out by United Utilities over the last 20 years,
and during this time some 14 000 km of the total 40 000 km of mains have been rehabilitated. This
has removed much of the unlined iron piping from the network. Figure 4 shows the improvement
across the region between 2001 and 2006.




Figure 4: Mean iron values in water quality zones across the region for 2001 and 2006



Besides achieving the required improvement in water quality through the region, there have been
significant spin-off benefits in terms of leakage reduction from the network. This has ensured long
term security for water supply resources. There have also been significant savings in operating costs
as a result of leakage reduction, including energy and treatment chemical consumption. The system
now operates optimally with leakage zones, but can be revised should emergencies or other
requirements dictate. Burst rates from mains have been reduced, as a new network effectively
replaces the previous system. Implementation of the software programmes for rehabilitation design
has led to savings of 80 million compared with conventional technology. Effective collaboration with
contractors has brought essential skills to the planning process, bringing minimal adverse impact on
Originally emphasis was placed on planning and design, with on-site investigation and pipe sampling
to determine the structural condition of mains. The volume of work using slip-lining has increased as
the programme has developed, and this has provided a structurally secure pipe, removing any need
to assess the condition of the original pipe. Slip-lining costs are competitive with other reconstructive
techniques. Pipes up to 140mm diameter are delivered as coils, reducing the amount of on-site
jointing required. Approach to site investigation has changed, now being minimal prior to undertaking
the work, the old pipe condition being of no consequence to restorative operations.
GIS maps show that planning zones can be displayed with thematic colours that represent iron levels
from water analyses. These clearly show customers the improvements achieved by the programme.
Customer willingness to pay is part of the business plan development required by the regulator
OFWAT, and the thematic maps are evidence of the companys ability to address its brief.



Change and asset management in Thames Water: building the (im)perfect

A Rees, SR Roberts, BM Bottomley and DM Upton

Some of the main points we aim to demonstrate in this paper are that:
We are building on legacy, where possible, through incremental development of systems,
processes and peoples capabilities. Asset management is not a completely new process, it is
rather an overall framework in which many legacy/current processes can be aligned to give
significantly enhanced outputs.
The universal issue of imperfect data is being addressed in our approach via explicit
uncertainty analyses.
Peoples engagement is crucial.

History - how did we get to here?

Most papers relating the tale of the rise of risk-based investment planning and asset management
point back to the acrimonious debate around Capital Maintenance at the 1999 periodic review (PR99),
with accusations of intellectual neglect prompting the economic regulator Ofwat to issue a letter to
water companies managing directors (Office of Water Services: Maintaining Serviceability to
Customers, letter to Managing Directors MD161, April 2000) (Office of Water Services: Asset
Management Planning to Maintain Serviceability, letter to Managing Directors MD212, February
2006), outlining the steps it saw as essential in developing an economic case for investment.
This led directly to the industrys collective formulation of using risk to service (to customers and the
environment), not age and condition per se, as the basis for investment planning and asset
management, contained within the UK Water Industry Research Common Framework (Capital
Maintenance Planning: A Common Framework, UKWIR, Report ref 02/05/3, May 2002). Although
some companies were already using risk to guide their day-to-day business, many saw the 2004
periodic review (PR04) as providing the first opportunity for both regulated and regulator to apply a
risk-based process to determine investment levels for whole programmes of work over a medium-term
In truth, various forms of risk-based approaches had been used in the UK water industry over many
years, some pre-dating privatisation. Previous steps in recognising risks included:
Wastewater Services
Final effluent consenting on a Monte Carlo basis (recognising that the probability of a given
final effluent flow/concentration combining with a given river flow/concentration, was best
expressed as a distribution, along with the distribution of the consequence).
Urban Pollution Management (UPM) standards the return period/duration/magnitude criteria
in UPM standards basically map onto probability x duration x consequence, being derived
from toxicological studies on impacts on fish life (with safety factors).
Sludge quality control moved to a HACCP basis along the same lines as risk controls in the
food industry.



Water Services
Definition of the economic level of leakage (ELL) presaged most of the principles which
feature in the Common Framework, including assessment of less tangible impacts such as
socio-environmental factors, within a least-cost framework.
In addition, the increasing use (from AMP2 (1995) onwards) of optioneering as a pre-design phase in
project delivery ensured that construction projects were focused on delivering the required outputs at
minimum cost. In practice this led to focus on the specific needs raised, rather than increasing
longevity of service for the asset base as a whole. This cost effectiveness resulted in targeted
expenditure, but did not necessarily give optimal whole life costs for maintaining service.
While some of the wastewater methods were largely applied to assessing investment on individual
projects to meet individual needs in an optioneering context, there were some examples at PR99 of
companies applying UPM to successfully justify investment for non-statutory Quality and
Supply/Demand programmes.
More significantly, the ELL methodology had already been applied within PR99 by most companies,
to provide support for investment in their water Supply/Demand programmes.
The ELL method was influential in the thinking behind the Common Framework, particularly the
elements of least whole life cost (including timing of investment and discounting) and cost-benefit
assessment. At PR04, most companies focussed on capital costs, with little demonstration of
operational cost effects. Similarly, nearly all companies opted for the cost-effective (least cost)
objective of the Common Framework rather than the cost-benefit (best value) objective. As well as
these two near universal gaps across the industry, there was an absence of uncertainty analysis to
demonstrate an understanding of how data and information quality might affect the outcomes of
investment and the likelihood that particular objectives would be achieved through that investment.

Application and development of risk-based asset management in Thames Water

Past use of risk approaches in Thames Water
In AMP2, Thames, developed a prioritisation tool for capital investment delivery which, had it been
more directly engaged with a wider set of business users, would have been markedly ahead of its
time. This included delivery against pre-defined business targets in the following categories:
Legal & regulation.
Customer impact.
Company image.
Group commercial potential.
Value for money.
Investment payback.
Source of funding.
Likelihood of failure.
Consequences of failure.



Figure 1: StepUK Asset Lifecycle vision for Strategy & Planning

A Leader in Strategy & Planning

an Integrated Approach

Similarly, at PR99 Thames Water (TW) undertook customer research which was as good as any in
the industry in terms of establishing customers willingness to pay a range of bills associated with a
range of levels of service.
In 2001, TWs work on network asset performance modelling received the Operational Research
Society Presidents Medal in 2001, a lead which others followed in obtaining peer review.
In PR04, TW made reference to existing operational risk registers, which were business-as-usual
tools used routinely to prioritise reactive maintenance against achieving defined business objectives.
Further, the use of risk modelling for trunk mains in the PR04 business plan was independently
reviewed as being leading edge.
In developing TWs investment planning and asset management capabilities, it has always been our
intent to build on existing practice (be that systems or processes) where at all possible.
In 2004, TW began a process of business change, under the banner of StepUK, to understand and
drive the changes in processes, systems and people required to transform the business in several
streams of activity. One of these was the asset lifecycle (AL) stream. An example of one aspect of the
AL Vision is shown in Figure 1.



Table 1: Components of risk-based asset management

Risk Framework
Risk Analysis

Agreed 5x5 severity &

consequence scales;
Still within an Ops heatmap
method for risk escalation

Ops heatmap categories

Water / Wastewater
mostly qualitative
trunk mains modelling of
severity and quantity

Common methods across

Water / Wastewater:
infrastructure modelling tool
(probability & severity)
Relex non-infra FMEA &
reliability tool
trunk sewers as well as mains

Risk Register
Costed Interventions Register
Portfolios via Optimisation
inc. constraints & weightings

System links to live programmes

of work (capex, opex)


Ops heatmap categories

Water / Wastewater not
largely problems not risks

Ops risk register

engineering estimating system
(EES) primarily for Programme
Delivery optioneering
existed briefly, more
prioritisation than optimisation;
not used for developing an AMP
customer valuation not
formally applied

Not present

Asset Planning System:

subsumes Ops risk register
holds problems, risks, needs,
solutions, costs, post-solution risks
Investment Management System:
uses APS outputs with customer
&/or business weightings &
produces investment scenarios
to meet defined scenarios
Not dynamic links, more via
business process

In late 2005, a team began to put a range of projects together to execute and deliver the Asset
Lifecycle Vision. Throughout several iterations of business challenge, including dealing with drought
and a sales process, this programme of work has been given the highest priority of business critical
and continues to receive a very high degree of scrutiny from TWs senior management team.
Delivering the Asset Lifecycle Vision
Components & precursors
Table 1 shows the components of the TW approach, and what the precursors of each of these were, if
appropriate. Figure 2 shows the overall cycle which has been, or is being further, developed in
Thames Water.
(a) Risk Framework: our risk measures align to both business drivers and service to customers
and the environment. These have their origins in the criteria contained in the Operational Risk
Registers (ORR).
The ORR had 5 categories of likelihood and 8 of consequence:
Likelihood categories:
Improbable (within 25 years).
Remote (within 10 years).
Possible (within 5 years).
Probable (within 2 years).
Almost certain (within 1 year).
Consequence categories (of varying magnitude):
Customer impact (number of properties, customers).
Regulatory/governmental action.
Media attention/PR.
Compliance failure.



Figure 3: Operational Risk Register likelihood/consequence categories



This has been modified for the forward-looking assessment of risk of asset and service failure in the
Asset Planning System (APS). We define risk as:
severity x quantity x probability of asset failure leading to impact consequence attributes
x frequency of asset failure/yr probability attribute

This ensures that we represent the reality that

Not every failure of an asset results in service failure, and
An asset failure can result in several types of impact, each with a different chance of
The consequence categories in APS are:



Unplanned interruptions

Internal flooding from sewers

Low pressure

External flooding from sewers

Water quality (microbiological/chemical)

Intermittent discharge compliance

Water quality (aesthetic)

Effluent quality compliance


Treated flow compliance

Security of supply

Pollution incidents

Internal flooding from mains

Sludge quality / disposal route

External flooding from mains

Incinerator performance

Pollution incidents
Common across services:
nuisance (noise, odour, traffic disruption); Health & Safety; statutory compliance

These consequence categories have been deliberately designed to allow us to assign weights which
either the business has derived for itself, or which have come from customer valuations through
choice experiments, both of which are held in the Investment Management System (IMS see
Forecasting over a time period of 25 years of frequency of asset failure can be either by human
judgement, or by means of forecasting tools embedded in APS (see below). Similarly, estimating the
probability of asset failure resulting in a service consequence can be derived from an experts
judgement or via the Relex reliability model (see below).
(b) Asset Planning System (APS): this is the core of our approach, and is a web-based system
for use across the company.
It holds and reports assessments of current levels of risk against various levels in the asset hierarchy
as appropriate to the risk being considered (e.g. pipe or pump, through to supply zone, catchment, or
geographic area, or up to company level). Examples of some of the information items being held in
APS for use by the operational business or for IMS optimisation are:



Failure information e.g. reference numbers; asset reference numbers and locations; probability,
severity, quantity and frequency of failure now and over a 25 yr period pre-solution;
interdependencies with other risks and solutions; organisational and geo-political tags
Solution information e.g. failure and asset reference numbers; solution references; investment
categories; post-solution risk attributes (P, S, Q, F); interdependencies with other solutions and risks;
must do attribute; start/end dates; outputs and activities (e.g. properties and pipe lengths)
Financial information (each solution) e.g. CAPEX/OPEX split; investment allocations; discount rates;
inflation indices
In addition to holding information items such as those above, which allow IMS to run cost-effective or
cost-benefit scenarios, APS has these key attributes:
Confidence categories: to allow uncertainty modelling.
Links to Infrastructure Tool: this allows quantitative modelling of below ground asset
likelihoods of failure.
Links to Relex: a reliability software tool which captures Failure Mode Effect Analysis survey
information and quantifies probability of asset failure resulting in service impact.
Links to the Engineering Estimation System (EES): this provides transparency between the
basis for the costs used in both the investment planning and capital delivery processes, along
with the ability to track efficiency and outperformance.
As an example, the application of confidence categories in our planning approach breaks information
Modelling of probability and severity (inc. base data).
Solution cost estimation.
Solution effectiveness.
Solution deliverability.
Graphically, uncertainty will be handled as noise around estimates of risk positions (see Figure 4).
This can be used by IMS as both a filter on types of risks and solutions (e.g. only include in a
snapshot solutions of a particular level of confidence), or as a data item to feature in the optimisation
(i.e. use expected value of benefit rather than a fixed value).



Figure 4: representing uncertainty in investment planning

-P and S



uncertainty (solution
effectiveness): mostly P



intervention cost

uncertainty (solution
mostly S

(c) Investment Management System (IMS)

To some extent, the functionality of IMS is the most straightforward, albeit it is highly dependent on
information received from sources upstream in the process. The purpose of IMS is to either deliver a
specified reward for the least cost, or to deliver the maximum reward within a given budget.
Chief among the upstream sources are:
Risk and cost information APS.
Business weightings and constraints Executive Management team.
Customers valuations and preferences stated preference choice experiments (willingness
to pay).
IMS calls a snapshot of information through from APS, with the ability to apply filters to what comes
in. One example was given above of using information confidence as a filter criterion; other examples
might be to use risk thresholds, time horizons or investment types. Each snapshot consists of the
relevant risks, associated solutions and costs, and the associated EES cost models.
Once a snapshot has been received, IMS can then apply business weights or customer valuations to
use as the basis for the reward provided by each and every solution within the snapshot. It is fairly
simple to see how this would then form the basis for a simple prioritisation based on bang per buck
of each solution.
However, the complexity of optimisation (c.f. prioritisation) lies in the application of constraints, or
targets, to a given snapshot and set of weights. In some sense, once any constraints are applied, the
resultant selection of solutions or actions in a portfolio is sub-optimal compared to an entirely
unconstrained selection.



Consequently, changing constraints (e.g. levels of service or spend profile) will not simply result in the
cut-off line being drawn at a different place in an list of ranked solutions, it can result in significantly
different selections of the solutions contained in the portfolio. Consequently, the business investment
planners have to learn a new discipline of understanding the effects of what may seem small changes
to constraints, weights or preferences, ad communicate this to the wider business.
The ability to apply constraints within investment scenarios to generate potential portfolios is only
limited by the information upstream of IMS. We intend, for instance, to understand what a portfolio
would look like which met compound constraints of <level CAPEX profile> by <operational area> to
achieve <minimum flooding performance level of service X> and <minimum security of supply service
level Y>.
In theory, we could use customer valuations segmented from socially disadvantaged groups as the
basis for customer valuations if we wanted to explore affordability constraints. As meter penetration
increases, it would be feasible to start to understand the implications of differential levels of service
down to high levels of spatial resolution. These are things for the future which we should start to
consider now, to inform the industry as a whole.
Organisational capability People & business Processes
In addition to developing the above systems, we have from the outset understood that achieving
business change is a tripod approach requiring People, Systems and Processes if it is to be
successful. To some degree, the Systems are the least difficult aspects of change.
As part of StepUK, a significant amount of time was spent in defining business processes, both as is
and to be. This has been built upon in developing the risk-based asset management approach in
Thames. Examples of altered or new processes resulting from this are:
Risk identification and approval.
Uncertainty banding.
Solution costing and approval.
Portfolio selection and agreement.
Interaction with live programme.
Each of these could be the subject of a paper in its own right. If we take item iii) above as an example,
it clearly cuts across functions within Thames Water and other companies that historically have been
separated in time i.e. there is a planning function which estimates costs for planning purposes and a
capital delivery function which develops costs for solution optioneering and delivery.
The People aspects are significant. Our approach requires an even closer working relationship, with a
need to move from understanding costing of options and cost-effectiveness to address a particular
need at a project level, to seeing how a need may be interdependent with others and their associated
solutions, in the context of programmes of work trading off different levels of service benefit,
investment and operational risks.
In the past, much of this has been possible for experienced individuals to carry out on an ad hoc
basis, because they have been working on priorities within separate programmes such as flooding or
environmental schemes. Now, in a period of an aging workforce with experience leaving the industry,
companies are looking to maximise value for money for customers by making decisions across all



types of investment simultaneously. To do so means moving to more structured and formal

assessments of risk and associated investment requirements. The approach has led to a number of
challenging issues arising, such as listed below:
Is day to day running of deterioration models an operational function, as operations provide
the data to update the models, or a strategic planning function?
A business as usual approach requires data, systems, people and processes to be aligned.
Simultaneous change in all four areas is a tremendous challenge. Incremental change
reduces risk, but significantly reduces the benefit of the changes until all aspects are in place.
It is important that the benefits of change are seen early, to encourage the business on its
Piloting change is difficult when organisations have strict expenditure controls, as the benefits
are hard to define at the outset. Hard specifications are difficult to establish in a vacuum when
dealing with complex interfaces between business process and data. A compromise has to be
drawn between financial efficiency and project effectiveness.
Paradoxically, while we are attempting to provide improved systems to enable this to happen, these
are after all just decision support systems, not decision making systems. These are an attempt to free
peoples time from the minutiae of analysis to allow them to focus on making better, more transparent
decisions. At the end of all this, we do not have a computer says no situation: accountability will
always remain with a companys people.

Relevant information
Foundation for Water Research (1998) Urban Pollution Management Version 2, FWR CD-ROM,
October 1998 [Version 1 (1994) Report No. FR/CL 0002].



Capital maintenance Good practice guide Leading Edge Asset Decisions

Assessment (LEADA) - Yorkshire Water
This document sets out the principles of a Leading Edge Asset Decisions Assessment (LEADA)
initiative driven by the Water Service Provider (WSP). LEADA was conceptualised during the postPR99 review of capital maintenance planning, triggered by Ofwats letter MD161 (Office of Water
Services: Maintaining Serviceability to Customers, letter to Managing Directors MD161, April 2000),
and forms the basis of Part B of the WSPs approach to the Common Framework. Development and
ongoing data collection commenced in the summer of 2001.
It is important to recognise that LEADA has built on the work undertaken by the WSP at the PR99
review in the areas of capital maintenance/serviceability and service enhancement. The concepts
underpinning LEADA are well embedded. The early versions of these systems were initiated in 1996,
and used at and after PR99.
For PR99, a number of risk-based serviceability concepts and tools were developed in the area of
capital maintenance by the WSP to determine which elements of the treatment and network assets
presented greatest risk in terms of potential service loss. The document looks at how the WSP has
identified risk on a consistent basis using a unified risk process which allows the comparison of risk
across the entire asset base.
Also for PR99, the WSP utilised benefit-cost analysis in the area of sewer flooding,, based upon being
able to meet customer aspirations for service improvement within a related Willingness-To-Pay (WTP)
for that same service improvement. This was used to justify the 1999 Draft Business Plan (DBP)
submission to Ofwat. LEADA has widened and enhanced this benefit-cost approach to cover the
majority of services provided by the WSP
These building blocks form the basis for LEADA and the WSPs decision to follow the benefit-cost
route within the UKWIR Common Framework to provide an economic approach to investment

The objective at a technical level has been to develop the benefit-cost methodology outlined in the
UKWIR Common Framework across the widest possible range of services and to use this to identify
where greatest service value is delivered to customers. The desired outcome from this has been to
assist the WSP in identifying where and when to invest in the assets to maximise the service gain to
A further and significant objective for LEADA has been to create a planning methodology, supported
by IT systems, to ensure that the data and processes used for selecting which asset interventions are
included within the DBP are those to be used in determining which receive investment as part of the
live capital programme.

Data requirements
What drives the data?
The WSP considered the problem from an economic perspective, and particularly the economics of
supply and demand, in order to run a benefit-cost approach to capital maintenance and investment
The demand element relates to the customers of the WSP and their priorities and WTP for service



The supply element relates to the ability of the WSP to reliably utilise and improve its assets to meet
customer demand for service quality at least cost.
Demand Element Data
The WSP needed to identify the services deemed important by customers, and estimates of the
annual monetary benefit associated with changes in service quality for each service, to estimate the
demand from customers for the services. Figure 1 demonstrates total annual benefit delivered to
customers from alternative levels of service.






Figure 1: Customer demand for service

Water Service
Inadequate mains pressure
Interruption to supply
Security of supply
Drinking water quality (Biological/chemical)
Drinking water quality (discolouration)
Personal injury

Wastewater Service
Sewer flooding of property
Area flooding
Ecological quality of rivers
Bathing water quality
Nuisance (odour and flies)
Personal injury



The service areas had to be sufficiently broad to cover maintaining current service via capital
maintenance, and service enhancement via enhancement programmes.
The resulting service areas used by the WSP for data collection are indicated in the table below.

Supply element data

To estimate the supply element data, it was necessary to understand the capability of the WSPs
assets and operation to deliver the service at acceptable cost.
To do this the WSP has identified the risk of not delivering service across the asset base. Risk was
expressed as a function of probability (p), severity (s) and quantity (q), defined as follows:


probability of at least one service failure occurring

severity of the impact on the customer or environment
scale of the impact

To determine the risks to address, it was necessary to identify:

Solutions to risks.
Estimates of the benefits from undertaking a solution.
Assessments of the costs involved in undertaking a solution.
Consequently the information required for a solution is:



probability of at least one service failure occurring (post solution)

severity of the impact on the customer or environment (post solution)
scale of the impact (post solution)
any necessary capital expenditure
any operating cost effects






Figure 2: Comparison of customer demand
with WSP cost

The OPEX and CAPEX have been used in the calculation of an annualised net present cost, for
direct comparison with the annual monetary benefit. Consequently there was essential physical asset
data required: asset reference, asset type and asset class.



CAPEX and risk are included in the WSP to ensure that the most appropriate balance of OPEX. It is
necessary to consider cost build-up, where appropriate, more than one solution to a risk, each one
having different cost make-up and different risk improvement.

Determining probability, severity and quantity

The methodology of the WSP recognises that service is placed at risk via assets failing in different
ways (failure modes). These can essentially be split in to three categories operational, asset
capability and asset-death related.
Where asset death was involved in the consideration of a specific failure mode, additional data were
required for the purposes of asset planning over 5, 10 and 15 years, in order to facilitate the
estimation of probability. In this instance a series of probability distributions based upon Weibull were
created to cover many of the asset groups operated by the WSP.
The data collected to drive the Weibull distribution were:
Commissioning data or age.
Cumulative Probability of Failure

Condition grade.


Asset loading (working effort).


Duty arrangement.






Age (years)

Standby backup.

Figure 3. Weibull distribution for probability

This data then allowed the WSP to use generic asset group distributions and make them asset
specific by reflecting the physical attributes of a given individual asset and its working environment.
The WSP, in terms of determining the collection of severity data, has created a series of standard
definitions used in the assessment of risk impact. This also defines the measurement of quantity
placed on each risk.




Internal Flooding



Damp patch - unused cellar

Damp - used cellar

Foul water in cellar


Dampness in living

Foul water in living



Significant area &/or

Area Flooding

Damp Patch / seepage

Sewage escape

Localised flooding

volume - Access/Traffic

sensitive customers

disruption sensitive


Cat 4

Cat 3

Cat 2

(including Bathing Beach

A reported polluting discharge

but not severe enough to
damage the environment

Noticible but minor impact on

habitats & amentity value (no
fish kills)

Significant but localised effect

on habitats (e.g a few fish kill)
+/or minor health risks to man


Flow Compliance
Final Effluent Compliance


Bathing Water Compliance

No. of properties

Severe & significant

depth. Major road


disruption / closure
Cat 1 or other EA
Major impact or contamination
with habitat devastation (e.g.
many fish kills) +/or serious
health risks to man

Consent Failure
Sample Failure(s) but not


Consent Failure

Individual failures but not

Mandatory or Guideline

beach fail


No. of Incidents



Local action or pressure

Env. Health active



Loss of normal disposal

Loss of normal disposal

route - alternative or

route - alternative

similar process


Injury with no Time Off

Non reportable accident (<

Reportable accident


3 days off work)

(> 3 days off work)

Customers Aware

e.g sewage smells, noxious

gases, swarms of flies, traffic or
mechanical noise

Sludge Quality /

Personal Injury

Incident but no injury

Statutory Nuisance Order

Loss of normal disposal

route - alternative landfill

No. of Properties

Quantity in p.e.

Fatality or permanent
disability resulting in


incapacity to work

Figure 4: An example of a risk impact table for wastewater (severity and quantity)

The important element to these scales is that the definitions of impact severity are aligned with the
service areas included in the demand-side data, so that identified risks can be mapped and compared
directly with customer benefit.
These scales are used in the assessment of risk at defined studies, or during dedicated data
collection exercises, and are coded in to the WSPs software associated with the prediction of asset

Data for network assets

The prediction of risk was less obvious where network assets were involved, due to the hidden
nature of the asset, and the transferable nature of the consequence in relation to the causative asset.
In this case the collected data has involved the analysis of Geographical Information System (GIS)
and other spatially-related data, to estimate the probability and consequence of failure. Before this
could occur, the WSP reviewed its failure mode trees created in the run up to PR99, to account for
new and better data sources for the assessment of risk.
The tree detailed below for each catchment is populated as fully as possible at the bottom level, to
inform the assessment of risk at the upper level. As in all other cases the risk is presented in the form
of probability, severity and quantity.







Reduction in
carrying capacity

Dead ends

Unlined CI

Changes in

Generation of
corrosion products

Number of
dead ends

Lining material

No - 0

Unlined/Bitumen- 10
Insitu Cement Mortar - 8
Factory Cement Mortar - 7
Epoxy Lined-3
PU Lined/Thin walled-1
Lining Material

>60 years - 10
50-60 yrs - 7
25-50 yrs - 5
10-25 yrs - 3
<10 yrs - 1



<3" -10
3-5" - 8
5-7" - 4
7-9" - 2


High Difference
In Elevation


Student areas

No - 0

in relation
to inlet

15 to 25m above - 10
5 to 15m above - 6
-5 to +5m - 4
-5 to -15m below - 2
>-15m below - 0

No - 0


Reduction in
carrying capacity


in zone

Changes in

High Difference
In Elevation

Caravan Parks

>=4 per zone - 10

1- 3 per zone - 5
0 pe rzone - 0

<45 per zone -10

20 - 45 per zone - 8
10 - 20 per zone - 6
1 - 10 per zone - 2
0 per zone - 0

Proportion of pipes lined

(Cement Mortar/Bitumen
/Unknown ONLY)

Figure 5: Network asset risk tree for water mains pressure



Asset Surveyor
Below Ground

Above Ground




Stage 1
Asset Failure Predictor

Source to Tap &

Sink to River

Quality and
Supply &

Business Risk Model (BRM)

Stage 2


Risk Matrix



Inventory AI


Update actual delivery data






Accepted or Rejected flag

Stage 6

to Pay

Business Plan Displayed on BRM Strategies

Stage 3

Stage 4


Stage 5

Figure 6: LEADA systems for asset management and investment planning

The LEADA approach to capital maintenance and investment planning
Figure 6 illustrates the LEADA planning process and its key components. It is broken down into the
key stages the WSP considered necessary for efficient asset management and investment planning.
Stage 1 - Identification of risks to service
The WSP has assessed the risks of non-delivery of service via the collection and analysis of asset
related data on its above and below ground assets. This has been referenced back to asset
inventories to allow for detailed programming as part of its capital programme following the final
determination. The collected data allows the WSP to identify the probability of asset failure at given
times, using the mathematical distributions already referred to, thus allowing it to identify current and
future risks to service delivery. The use of these distributions to determine deterioration of asset
performance is a key assumption in the WSP's methodology, and it is critically important that the
output is validated and/or calibrated with historical observations, where it is possible.
Also critical at this stage is data quality and quantity. In terms of data quality, the WSP ensures
consistency by the use of technical approach manuals. These set out the detailed requirements of
the data, formats for collection, and definitions. The process has been managed and audited by
centrally-based teams to ensure consistency and completeness. The appropriate degree of
resolution is of significance in terms of quantity of data, and much consideration has been placed on



what asset level gives sufficient resolution for consistency with project delivery, while at the same time
not overloading the asset management processes.
In this case the WSP has selected element component level as the general level, but in some cases
goes lower into the asset hierarchy to assembly level, where there are assets of a significantly
different type or age performing one joint task. This is of particular significance to wastewater
treatment works where similar assets of a similar type (e.g. filters) have been built at different periods
in time, as demand on the works has grown.
An example of a process element would be:
















Tank 1



Tank 2

Figure 7: Illustration of data resolution used for identifying risk



Asset death related risk

The WSP has completed site surveys to collect data for the purposes of assessing asset death
related risk, and has developed software to ensure consistency of collection. This is vital for the data
that will be used in failure mode and consequence modelling. Below is an example of the collected
data, as stored within the WSPs IT application, referred to as its asset surveyor.
The assets have been classified in terms of their failure modes and how they would contribute to an
impact on service provision. This work was undertaken utilising the expert opinion of the asset
management and operation teams, who deal with these assets on a day-to-day basis.
The data from the survey were passed in to a risk model, identified as asset failure predictor in Figure
6, where probability is calculated utilising the Weibull distribution and this is then combined with failure
mode definitions to identify risks of service failure at status quo, and for 20 years later. The risks are
expressed in terms of the probability, severity and quantity parameters identified earlier.

Operational Risks
For risks that occur for reasons other than asset death, but lead to service failures against current
objectives from issues such as reliability, or catchment deterioration, then the WSP captures these
risks via Source to Tap and Sink to River studies. These are facilitated events where catchments
are studied in detail by operational and asset management teams, covering above and below ground
assets. The teams reviewed historical data and knowledge to identify and challenge risks, and to
propose outline solutions to risks.
In dealing with these types of risks, it was necessary to distinguish between those already occurring
and those which are forecast to occur for the first time at some given future instant, because correct
treatment of this information is critical for the correct estimation of probability. To this end, the WSP
devised a methodology for assisting the study teams in capturing the relevant information, to allow the
derivation of probability.

Asset referencing

Failure Modes

Asset data
collected in system

Figure 8: Tool for the collection of data for calculating probability












Route A


(on a scale of 0 to 100%)




Route B


(e.g. 1 in 3 years)

Route C





A) Between 2 and 5 years
B) Between 6 and 10 years
C) Between 11 and 15 years
D) Between 15 and 20 years
E) Beyond 20 years

Figure 9: Decision tree guidance for the calculation of probability

Further quality assurance and consistency work has been undertaken by the WSP by holding a series
of regional challenges, where company experts have reviewed the assessment of risks and solutions
to ensure that processes and procedures have been followed in accordance with technical approach
guidance issued to the asset management teams. Stage 2 outlines how the WSP has used IT to
assist with consistency.
This same methodology has been equally applicable to asset capability risks where a new obligation
would render an asset or group of assets incapable of meeting the future need. It was important to
capture this information if the objective of undertaking the benefit-cost analysis was to be fulfilled, and
an assessment made of the enhanced service levels.
Stage 2 - Consistent risks and solutions
All risks, regardless of their origin, will have been stored within a system referred to as the Business
Risk Model (BRM). The BRM formalises the Company risk methodology and ensures that all risks are
scored and stored in a consistent manner. It is also used to capture relevant output and activity
information needed for effective asset management and completing the relevant DBP tables.

The BRM system allows the WSP to cost solutions using unit cost models contained within its unit
cost database. Again this is about achieving consistency. The process allows for alternative
solutions to the same risk, each requiring different levels of CAPEX, and/or OPEX, with each having
potentially different levels of risk improvement to be captured, and hence made available for the
economic modelling (benefit-cost) assessment, in Stage 4. The costs used in the economic
optimisation are the CAPEX and OPEX, expressed as an annualised NPC, to make them directly
comparable with data from willingness-to-pay (WTP) described in Stage 3.



Statement of
risk prior to
solutions to
a risk

of risk after
the solution

Figure 10: Illustration of system for scoring risk, before and after solution

Stage 3 - Market research willingness-to-pay

As a result of the WSP adopting the cost-benefit assessment approach of the common framework
across a wide range of services, it was essential to generate estimates of customer benefit for as
many services as possible. These have been listed earlier.
This was achieved by following a three-step process:
A qualitative phase to determine general priorities.
A quantitative phase to capture willingness-to-pay (WTP) information.
Benefit estimation associated with service change.
The qualitative phase consisted of 8 focus groups and a number of business depth interviews to
discuss service issues generally, before focussing on the water and waste water services
experienced. The areas covered during this phase were:
The definition of what good service was, and the perception of an ideal service provider.



The WSPs service performance.

Clean water and waste water areas for improvement and investment (spontaneous views, no
Clean water and wastewater areas for improvement and investment in the light of prompting
o actual and alternative targets for service
o alternative measures and attributes
Ranking or prioritisation of service areas in terms of importance to the customer.
The quantitative phase was a much larger exercise, where stated preference choice experiments
were undertaken to generate WTP for the priority services identified during the qualitative exercise.
To undertake this in a robust manner, 1 500 face-to-face interviews were completed, covering
representative samples of domestic and business customers. Each customer was offered a series of
service packages, each with combinations of decreasing, improving, or stable service levels. These
service packages were traded-off against potential impact upon the individual customers account. In
the example shown the customer would be trading-off changes from the current situation (now) with
two alternative service packages (B and C), both including a mixture of improving, and deteriorating
service, together with changes in billing levels. It has been possible in this way to elicit service
valuation, and see how this value changed with service level change.

The key output from this stage was the ability to place a monetary benefit on the service provided by
the WSP, and indeed every solution the WSP could consider.
The WSP followed the draft version of the DEFRA stated preference non-market valuation guidelines
[Bateman et al (2002) Economic Valuation with Stated Preference Techniques, A Manual, published
by Edward Elgar, in association with DTLR
and DEFRA].
The WTP functions provide monetary
values of the changes in the quantity of
1 year
1 year
1 year
service provided. Changes in the severity
in 500
in 750
in 250
and probability of service impacts are
incorporated into WTP by weighting the
WTP for quantity service changes.
The benefits of a programme of
investment to the average residential
customer are described by the equation
below, taking the example of a quadratic
benefit function. Clearly a similar function
is required for business customers, and
the two aggregated to estimate the total
benefit across the WSPs region from a
programme of investment.

j 1



Water Bill




Figure 11: Pictorial representation of choice





wij Qij
i 1



i 1

wij Qij
i 1


Figure 12: Form of benefit equation for residential customers derived from WTP


service areas
weight, composed of change in severity and probability arising from solution
quantity of service

The and coefficients have been derived from estimated utility through the WTP study. The first
term is the current bill, and the second is a service-area-specific constant. The change in the weight
term ( w) relates to the change in severity and probability.
The weights were required, as all solutions that could potentially enter the programme are quantified
during the risk process by their pre-solution risk (i.e. probability, severity and quantity) and their postsolution risk. Severity is classified into one of five classes from very low to very high. The weight is
calculated as the pre-solution severity weight multiplied by the pre-solution probability, minus the
post-solution severity weight, multiplied by the post-solution probability.
Stage 4 - Economic optimisation
The WSP has identified individual asset and company level risk at this point in the process and
expressed these in terms of service risk. The WSP has also generated one or more possible
solutions to the risk (each of which has a specific improvement in risk following implementation)
together with an associated cost. The WSP has also estimated the monetary valuation of service
benefit from the market research WTP.
At this point the two sets of data are brought together and put through an economic optimisation
engine, developed by the WSP. This is identified as the Economic Level of Service Assessment
(ELSA) optimisation tool. Within the optimisation engine is the WTP benefit equation which is used to
assess risks in terms of their contribution to improvement in risk and allocates a monetary benefit to
each solution. The optimisation routine looks at benefit at a programme level, given the non-linear
nature of the benefit function. This is achieved by accounting for when individual solutions enter the
programme, where they are on the benefit curve, what scale of benefit the solution should
consequently attract, and ultimately determine whether solutions are cost-beneficial, i.e. are they
making a positive contribution to the net benefit to customers? This is illustrated below.





Net Benefit


Net Benefit




Optimal Level of Investment

Figure 14: Illustration of how the benefit-cost assessment is

carried out in ELSA

Clearly the above process would identify an economic level of service based on the information
gathered on risk, cost and benefit. However, to meet the needs of MD161 and testing the balance of
service, risk, cost and economics, the WASP has then built the ELSA optimisation system with the
facility to constrain the model in different ways to build programme scenarios. The model can be
used to apply:
Risk profile constraints to the risk data.
Efficiency assumptions to CAPEX and OPEX.
Maximum and minimum service objectives.
Maximum and minimum CAPEX and OPEX objectives.
o At a programme level.

At a functional unit level.

At an investment category level.

Some examples of how this is achieved are illustrated below:



Low Risk Scenario

High Risk Scenario

Figure 15: Application of risk profiles for modelling in ELSA

In the examples illustrated above, the risks contained in the red areas of the risk matrix would be
constrained into a programme. Those in the green area would not be considered for inclusion in the
programme. Any in the white would be subjected to the economic optimisation (benefit-cost
assessment). Testing these risk levels allows the WSP to understand the service and cost
implications of operating at different levels of risk.



Select constraint
to apply

Set specific
(absolute or range)

Figure 16: Application of service and budget objectives in optimisation

The objective of being able to undertake these constraining actions within the context of the economic
optimisation is that ELSA enables the WSP to answer different questions and develop different
scenarios, balancing for example, capital maintenance with quality enhancement programmes.
Examples of the scenarios the system is designed to model are to:
Identify the economic level of service, given no cash or service constraints.
Identify the required level of risk the WSP would have to accept, to achieve given service
levels within a given budget.
Identify the economic service levels possible for different levels of efficiency.
Identify the impact of increased quality obligations on capital maintenance for an assumed
The output from the process is a view of service levels, expenditure requirements and risk profiles.
The WSP has adopted a matrix approach for presenting risk, as illustrated on the ELSA system
screens, to help understand the risk benefits being delivered. Examples of how risk might change
following a programme of investment are illustrated below.
ELSA Programme ID 150 - All Investment Areas
Risk position 2010 pre investment































Total red risk

Total Amber risk
Total green risk


Figure 17: Risk movement from programme over 5 years, including




The figures in the matrix cells relate to the number of risks identified across the company asset-base
that have been considered as part of a specific scenario. In the case of the post investment matrix
the small numbers in the top right hand corner of the cell represent the net change following
Stage 5 - Financial implications
The final stage of the process is to assess the impact of the investment programme on customer bills
and company financing implications. To do this the WSP takes a programme output and passes it
through a scheduling tool to apply spend-profiles to each solution, with the objective of fitting all the
programmed solutions within a target spend profile.

Figure 18: Output from scheduling, to illustrate year-on-year breakdown for 5

year programme

Following this, the profiled CAPEX and OPEX is modelled through the WSP's financial modelling
package to determine the price and financing implications of running with a given programme.
Stage 6 - Investment authorisation
The final stage of the LEADA process is an ongoing review of existing business processes associated
with the WSP's Investment Authorisation Procedures (IAPs). These IAPs are part of the WSP's
internal project investment authorisation processes. The WSP recognises the importance of
integrating the LEADA processes and systems, with its existing ways of working to ensure that the
implementation of the common framework continues to deliver benefit to the business throughout the
delivery of the resulting capital programme.




Active, Beautiful, Clean (ABC) Waters
sustainable assets management




Singapore, the smallest city-country in South East Asia, is faced with a water scarcity. With an area of
700 km and a population of 4.7 million, Singapore tries to conserve every single drop of rainwater by
utilising as much of the land available as catchment. Marina Reservoir, a dam that has just been built
across the Marina Channel with the most urbanised catchment at 10000 ha or one-sixth the size of
Singapore, is Singapores 15 reservoir, and together with two reservoirs under construction, it will
convert two thirds of Singapore's area into water catchment. However, the maintenance of 15
reservoirs, 2 reservoirs under construction, 32 rivers and a 7 000 km network of drains and canals is
not an easy task. PUB encourages citizens to enjoy and value its water resources, and not see
themselves as trespassers thereon. This call will create a better sense of stewardship of Singapores
water assets. It is a strategic, wide-ranging and long-term programme that has been named the
Active, Beautiful, and Clean (ABC) Water Programme, which aims to open up the reservoirs and
waterways for public recreational use. These projects will bring people closer to the water and foster a
greater sense of ownership thereof. Thus, in the long term, the people will see themselves as the
custodians and maintainers of Singapores water assets, improving the overall quality of Singapores
living environment.


Singapore is a small island city-country with a population of 4.7 million people in an area of 700 km .
Located in near the equator, Singapore enjoys an equatorial climate with average annual rainfall of 2
400 mm. However, being a small island-nation, collection of rain which falls throughout Singapore
must be balanced with land use for socio-economic growth. Water is nevertheless a scarce resource
in Singapore, despite the fact that half of Singapore land area is already utilised for water catchment.
Singapore has been categorised in the lowest band of water scarcest countries in the world in the
United Nations 2003 report, Water for People, Water for Life.
Population increase, coupled with rapid industrial, economic, and social development, has resulted in
massive increases in water demand. The consequence is that Singapore has to continually innovate
with new technologies and approaches to be able to be sustainable. At the beginning of the 21
century, institutional restructuring made Singapore able to close of the water loop. Closing the Water
Loop has become a phrase which PUB uses to manage entire water cycle from the sourcing of water
by rainwater collection in reservoirs, desalination, water importation, to treatment and distribution of
drinking water; and the collection of waste water, which then undergoes treatment to make available
NEWater, a term PUB uses for recycled water (Figure 1). Institutionally, this closing of the water loop
has been accelerated by the integration of the Sewerage and Drainage Departments in the Ministry of
the Environment (ENV) with PUB.
This paper focuses on the collection and transport of rainwater by drains and reservoirs, especially
with respect to how Singapore maintains these water assets through its Active, Beautiful, Clean
(ABC) Waters Programme.



Figure 1: Singapores integrated water loop

Singapores rain water collection and transport

In the past, canals and drains were built primarily to prevent flooding. Flooding has been a regular
phenomenon in Singapore, caused by the seasonal monsoon rain. About 30% of Singapore is less
than 5 m above mean sea level, with some older sectors of the city below high tide level. This
renders Singapore more vulnerable to flooding. These factors have been exacerbated by inadequate
drainage in the early years, as the city became more urbanised. In fact, in the 1970s, about 3 200 ha
of Singapore was at risk to regular flooding.
The obvious engineering solution was to create large stormwater drains or canals to direct stormwater
quickly to the sea. But in land- and water-scarce Singapore, this approach has conflicted with other
demands for land use (giving up precious land for wide canals) and water supply (losing the precious
water to the sea, instead of collection and storage in reservoirs for use).
Thus, the drainage and reservoir construction needed to be planned in conjunction with urban
development. One of the recent examples is Marina Barrage, a dam that has just been built across
the Marina Channel to form Singapores first reservoir in the city. The Marina Reservoir, Singapores
15 reservoir, has the largest and most urbanised catchment of 10 000 ha, or one-sixth the size of
Singapore. Marina Barrage, together with two reservoirs under construction, namely Punggol and
Serangoon, will turn two thirds of Singapore's area into water catchment zones.
To date, 32 rivers, 15 reservoirs, and the other two reservoirs under construction, as well as a 7 000
km network of drains and canals, are a visible part of the Singapore physical landscape, devised to
collect and harvest rain to meet Singapores water demand. However, the management of those
water assets is not an easy task. PUB believe that for sustainable asset management, there should
be a shared responsibility in management of these assets, including keeping the water clean. This
holistic approach is known as the ABC (Active, Beautiful, Clean) Waters Programme, where public
and private sectors, as well as the citizens of Singapore are united in purpose.



Singapore assets management through the ABC Waters Programme

The ABC Waters Programme focuses on a community outreach. However, it is not just a simple
community outreach programme, but a far more strategic, wide-ranging and a long-term approach.
Through the ABC Waters program, the citizens are encouraged to enjoy and value its water
resources, and not to regard themselves as trespassers on property that is in their ownership. This
will create a better sense of stewardship of Singapores water assets, which in a long term, will lead to
a comprehensive asset management a structured program to minimise the costs of asset
ownership, while maintaining required service levels and sustaining infrastructure, words quoted from
N. S. Grigg, in Water, Wastewater and Stormwater Infrastructure Management (first edition, CRC
Press LLC, 2003).
A gradual and systematic programme to open up the reservoirs and waterways for public recreational
use has been started. This has been a radical move away from the old keep-away policy,
characterised by ever prevalent and prohibitive NO signs, with threats of heavy fines.
Reservoirs are open for water sport, including activities, such as kayaking, sailing, dragon boating,
and wakeboarding. Sailing has been introduced in Lower Seletar Reservoir. Walk-in centres where
the public can rent a kayak have been set-up at MacRitchie, then at Bedok, Jurong Lake and Lower
Seletar. Events such as the World Wakeboarding Championships and Dragon boat, kayaking and
sculling races, as well as carnivals have been held with very strong public support, especially from the
schools and community organisations.
The utilitarian, concrete and granite-lined drains, canals and reservoirs are transformed into active,
beautiful and clean streams, rivers and lakes. The drainage reserve on either side of canals or
monsoon drains has since become useful as a park connector, amenable to jogging or cycling, in
addition to the creative ABC Waters projects (Figure 2).

Figure 2: An example of ABC transformation SengKang Floating Islands

enhances the areas liveability

All the ABC Waters Programme have been placed under a strategic ABC Waters Master Plan in
2008, based on the Blue water assets map of Singapore (Figure 3).



Figure 3: Singapores Blue Water Asset Map

The Master Plan identifies at least 150 opportunities to create various projects. The first ABC Waters
demonstration project, Kolam Ayer ABC Waterfront, was officially declared open to the public on 5
April 2008 (Figure 4). During construction phase, a workshop and series of brainstorming sessions
were held, to gather ideas from affected communities as to what kind of activities would support public
education and promote clean waterways, besides promoting care of these facilities. PUB will be
setting up this workshop as a model for all subsequent ABC Waters projects.

Figure 4: Residents enjoying themselves playing with the water wheel, and dragon boating at Kolam Ayer
ABC Waterfront

In addition to Kolam Ayer ABC Waterfront, two other projects are currently being implemented, and
will be completed within the next two years. Another 25 will be implemented over the island within the
next five years. These projects will bring people closer to the water, and encourage the people
themselves to assist in maintaining their own environmental assets. The value of water will go beyond
that of just being a resource, but will be an environmental asset that will help to improve the quality of
Singapores lifestyle environment.



ABC projects maintenance is achieved through collaboration with commercial operators as well as
community engagements. PUB is collaborating with the National Park Board (NParks), Singapore's
science-based authority on nature conservation, and this will involve commercial operators in
maintenance and management of landscapes and parks around reservoirs and waterways. Sengkang
Floating Island and Fruit Park at Anchorvale Street is currently under construction (Figure 5).

Figure 5: Sengkang Floating Islands and Fruit


Other than commercial operators and communities are also being engaged as water adopters (Figure
6). Currently there are 53 adopters that give school presentations, involve themselves in clean-up
projects and community outreach programmes to raise awareness of keeping the water clean and
taking ownership thereof. One example is the Waterways Water Society who have adopted the
Kallang Basin and Singapore River. Their programmes include river patrol on weekends, and a
outreach programme to schools to educate scholars on their role in keeping our waters clean. St
Andrews Secondary School has adopted Kallang River and has been conducting clean-up projects
along the river near their school. They have also embarked with a study on the impact on aquatic life
when saline marine water is ultimately displaced by fresh water, which will occur when the Marina
Barrage is fully operational. Temasek Polytechnic, Eunos Primary School, and Tampiness North
Primary Schools have united to adopt Bedok Reservoir.



Figure 6: Water adopters in action

As a water-scarce city-state, Singapores approach to asset management is somewhat unique. Water
scarcity itself has forced Singapore to close its water-loop with a consequence of needing to maintain
all its water assets throughout this water-loop, which is not an easy task. Faith in good management
starts from a sense of ownership. Singapore, through PUB, has embarked on the ABC Water
program, a far-reaching yet flexibly and innovatively tailored program to solve immediate problems, as
well as putting in place the underpinnings for the future. Instead of a traditional hands-off approach,
PUB has adopted a 3P strategy that involves the Public, Private and People Sectors in the ABC
Water program, to encourage environmental stewardship. The ABC Waters Programme aims to
transform utilitarian drains into active, beautiful and clean streams, rivers and lakes, for community
recreational enjoyment. Through enjoyment and public participation, we hope that the citizens of
Singapore will take greater ownership of their sustainable water assets management, both in the short
as well as I the long term.

Relevant Reading
Public Utilities Board (2008), Annual Report, Singapore.
Public Utilities Board (2008), Active, Beautiful, Clean Waters Programme, Master Plan, Singapore.
Public Utilities Board (2008), SQA Application Report, Singapore.




European Union/DPLG Pilot Project Asset Planning for Four Municipalities
The South African National Government Department of Provincial and Local Government (DPLG) and
the European Union (EU) appointed a consortium of consulting firms in 2004 to assist them with a
programme for the improvement of management of infrastructure in four municipalities as part of a
wider EU-assisted Programme to strengthen local government in the Mpumalanga and Limpopo
provinces. While the initiative comprised a number of interventions, one key activity was to prepare
asset management plans for infrastructure networks in these municipalities.
The project was the first in South Africa where infrastructure asset management plans were prepared
in line with the concepts set out in the "International Infrastructure Management Manual", New
Zealand National Asset Management Steering Group and Institute of Public Works Engineering of
Australia, April 2000
Four pilot municipalities were identified as incubators for methodologies, tools and other systems that
could be applicable elsewhere to strengthen municipalities ability to deliver infrastructure services to
communities sustainably.
At the end of the project, the municipalities were provided with documentation as to the nature, extent,
condition, performance and value of the various infrastructure networks; existing and target service
levels; life-cycle management needs; long-term funding implications, and the actions required to
improve asset management practice. In addition, a number of site-specific interventions were
undertaken at each municipality.
The greatest value of the project was the crafting of an appropriate South African approach for lessresourced municipalities.
The pilot municipalities have proved to be interesting in the context of application of the IIMM
principles, because of issues such as their backlog in basic services, non-payment for services and
capacity limitations. Challenges included officials who are unfamiliar with the terms of asset
management (let alone the process), and large gaps in data availability.
The principal best practice/innovation point arising from this project is that:
substantial insight has been gained into key challenges and benefits of infrastructure asset
management planning in the lesser-resourced municipalities in South Africa, and no doubt also in
lesser-resourced municipalities elsewhere.

The replacement cost of infrastructure, even in a small South African municipality, can amount to
several hundred million Rands (1 US$~ 7 ZAR). Water services invariably comprise a significant
proportion. Not surprisingly, the Local Government: Municipal Finance Management Act 2003,
introduced in 2004, specifically highlights the responsibility of all municipalities to safeguard and
maintain assets in their custodianship. Indeed, it aims to raise the bar in asset management practice
in that it sets out the need for municipalities to establish appropriate systems and controls, and to
institute specific accountability and transparency provisions. The high value and long serviceable life
of infrastructure (such as streets, water reticulation and treatment works) points to a need for this
particular group of assets to receive specific management attention.



Fortunately, an understanding of best practice in infrastructure asset management has been evolving
in the international arena over the past decade, albeit predominantly based on first-world experience.
An initiative by the national Department of Provincial and Local Government (DPLG) and the
European Union (EU), began in 2004 under the banner of the Programme to Strengthen Local
Government in the Mpumalanga and Limpopo Provinces, for the Improvement of Management of
Infrastructure in four municipalities. While the initiative comprised a number of interventions, one key
activity was to prepare asset management plans for infrastructure networks in these municipalities.
The project, undertaken between 2004 and 2005 with the assistance of a consortium of consultants,
was the first where the development of infrastructure asset management plans for municipalities in
South Africa has been based on the concepts proposed in the "IIMM 2000", viz "International
Infrastructure Management Manual", New Zealand National Asset Management Steering Group and
Institute of Public Works Engineering of Australia, April 2000.
Substantial insight has been gained as to the key challenges and benefits of infrastructure asset
management planning in the lesser-resourced municipalities in South Africa, and no doubt also in
lesser-resourced municipalities elsewhere.

The four local municipalities selected for the project were Mkhondo and Albert Luthuli in Mpumalanga
province, and Lephalale and Mookgophong in Limpopo province. The total current replacement cost
of their civil engineering infrastructure (water, sanitation, roads, and solid waste) in 2005 was of the
order of R 500 million for each municipality.
The following is an overview of the infrastructure status of this group of municipalities as it was in
All had localities where the services provided were less than basic, with the majority of the
backlog being in rural areas.
There was a variability in financial management performance. Payment for services ranged
from 50 to 90%. One municipality had no audited statements for the preceding 3 years).
All had a vision in terms of providing basic infrastructure, but no detailed policy on technical
and functional service levels.
None had formal maintenance strategies. Existing practice and maintenance history was
poorly documented.
Formal condition-based programmes of replacement/renewal of infrastructure were
The compilation of an asset management plan promised to provide a significant step forward in
understanding the big picture challenges of each of the municipalities, and appreciating the
important issues to be addressed in the pursuit of sustainable service delivery.

Points of departure
The IIMM gives guidance on how to prepare an infrastructure asset management plan, following the
logical progression that is summarised in Figure 1.



Figure 1: Asset management planning flowchart

This can be enhanced for larger municipalities to include deterioration modelling, optimised decisionmaking (such as is employed in some pavement management systems) and other advanced
techniques, but this needs to be supported by a robust policy environment and accurate and detailed
asset data, none of which was available in these municipalities prior to the project. Thus, the asset
management planning started at a basic level.
The methodology used in the project had two key points of departure.
One needed to relate to defining the target level of service associated with a particular network. While
regulations and national planning frameworks provide a framework for provision of basic
infrastructure, the specific needs of the local community, and the ability of the municipality to deliver,
need to be taken into account when determining a municipal policy with regard to level of service.
Furthermore, other technical factors such as minimum condition and quality should apply, and
functional issues such as response times and methodology of informing the community of
closures/outage, could also be part of the definition. A sustainable policy on levels of service would
provide a framework in which appropriate technical and financial responses for the life-cycle
management of the infrastructure could be planned.
The second departure point of any asset management planning must be numerical data and other
information -- the more focused, complete, reliable and recent, the better. The objective is to establish
the existing extent, nature, condition, capacity, utilisation, and performance of the assets. A realistic
prediction of future changes in demand is also needed.

Asset management planning approach adopted

The task is multi-faceted, covering key financial, technical, legal and social aspects with regard to a
diverse range of services to be delivered by the municipality. The approach taken in compiling these
plans has been to prioritise to cover all the main aspects of the process of asset management
planning, but to a limited level of detail. In other words, to get a fix on the big picture, to focus on the
status quo, and to decide on the top priority, follow-up actions are required.
Accordingly, the following approach was adopted:
Focus on the high value assets using the 80/20 principle.



Make optimal use of existing information (e.g. asset registers, as-built drawings, budget
Make best use of the judgement of experienced staff or consultants, especially where there is
no/little/unreliable information.
The approach is summarised in Figure 2.

Figure 2: Overview of approach

It was essential to work closely with the municipal officials, particularly because these were the first
infrastructure asset management plans ever compiled for these four municipalities. This was also
because they had valuable insight into the history, nature and challenges of the municipality and its
infrastructure assets, and there was a need to foster their understanding and ownership of the
process and its outputs, as this is fundamental to the longer term objective of improved infrastructure
The methodology included a broad-based awareness-raising programme covering the following:
For councillors:
o Infrastructure components, processes and the professionals involved.
For senior officials:
o Life-cycle management of infrastructure assets.
o Levels of service.




Asset management systems and data.

Risk management.
Asset accounting.
Infrastructure asset management planning.

A document was prepared that set out the following points, to ensure consistency of approach within
each municipality, and from municipality to municipality:
Grouping of infrastructure types (see Table 1 for an example) to enable a similar level of
detail for each infrastructure type and to group infrastructure assets that would be expected to
have similar deterioration patterns and risk profile.
The type of information to be collected.
Unit costs for asset components, based on current replacement cost.
Nominal asset lives.
The approach to condition grading of each asset type (typically a simple 5-point assessment
The approach to assessing asset management practice (a checklist with a scoring system).
The approach to deriving operations, routine maintenance and periodic renewal budget
requirements, based on industry norms.
Asset management











Table 1: Example of infrastructure groupings for water



Asset management plan deliverables

Scope, value and condition of infrastructure
The asset management plan of each municipality has summarised the extent and replacement cost of
the infrastructure by sector (for water sector example, see Figure 3). Depreciated Replacement Cost
was also calculated, based on the assessors view of asset lives, which for these municipalities,
ranged from 25 to 45% of the current replacement cost. This gave a broad indication of the overall
consumption of each of the asset groups.

Figure 3: Scope and cost of water assets at Mookgophong Municipality

The condition of infrastructure components was assessed on a simple 5-point scale that ranged from
Excellent [1] to Very Poor [5]. Figure 4 shows the overall condition of water infrastructure at
Mookgophong Municipality.

Figure 4: Condition of water assets at Mookgophong Municipality



The Figure flags assets in poor condition (demonstrating significant deterioration), and a very large
proportion (40%) of all assets in the fair category, i.e. marginal with deterioration clearly evident.
Even though this is a somewhat rudimentary approach, it clearly points out that the water services
infrastructure assets with a replacement value of around R30 million are at risk.
One consequent recommendation has been that a more detailed survey of the current condition of
these assets (e.g. excavate and expose representative sections of piping) was needed, that the likely
rate and nature of future deterioration and the risk to service delivery performance should be
considered, and a that programme of replacement/rehabilitation should be prepared.
The current and target availability of infrastructure within each municipality has been summarised on
another 5-point scale, and depicted on maps in the asset management plan.
It will be useful with future iterations of the asset management planning to augment this with a risk
analysis, focusing on the consequences of inaction.

Existing asset management practice in each municipality has been reviewed, and improvements
recommended with regard to systems, procedures, information and organisational issues that should
be implemented.
In addition, preliminary performance measures have been established in consultation with the
municipal officials. Target performance has been defined, and current performance assessed. The
measures include accessibility, affordability, reliability, health and safety, sustainability, quality,
payment collection, and water losses. The intention is that these measures and targets will be
reviewed and refined by each municipality before the next asset management plan is prepared, and
can be formalised by the municipal council.

Budget needs
The life-cycle planning part of the DPLG/EU project documented for each municipality the capital
works has been planned and committed over the next 3 to 5 years, with a first order vision of the
longer term capital needs. Multi-year budget needs for operation and maintenance have been
estimated, based on industry norms. In addition, the assessments of age or remaining useful life of
the assets, together with the condition data, has enabled an assessment to be made of the multi-year
budget required for periodic renewal of infrastructure. Additionally, identification of areas that needed
more detailed condition/functionality assessments before budgets could be drawn up.
The life-cycle budget requirements have been aggregated to give a financial vision of the needs of
infrastructure over the next 10 years. The ability of each municipality to meet these financial needs
has been assessed, and options for accessing external grant funds and loans, optimising financial
performance, and note taken on any implications regarding tariff structures.

At local level
The value of this DPLG/EU-sponsored project lies in the way in which the exercise of asset
management planning in each municipality has brought the message firmly home to that municipality.
It has motivated officials from different sections of the municipality to discuss and debate issues on
common ground, and take stock of their inter-dependency at strategic and planning level. It has given
a clear picture of the main challenges facing the municipality, now in both the short- and long-term,



and has drawn attention to the fundamental responsibility of a municipality to decide what services it
should deliver, and to plan how to do it. It has positioned sustainability alongside development on
the municipal agenda.
In addition, a number of site-specific interventions have been undertaken at each municipality.

In a wider context
The value of the project lies in the insight that it has given with respect to the preparation of asset
management plans for lesser-resourced municipalities in South Africa (and no doubt in other
countries as well). The four municipalities were used as incubators for methodologies, tools and other
products that can be replicated elsewhere to strengthen municipalities ability to sustainably deliver
infrastructure services to communities. The project represented a significant step forward in efforts to
improve the ability of less-resourced South African municipalities to manage their infrastructure.
The pilot municipalities have proved to be an interesting context for the application of the IIMM
principles, because of issues such as a backlog in basic services, non-payment for services, and
capacity limitations. Challenges included officials who are unfamiliar with the terms of asset
management (let alone the process), and large gaps in available data.
The preparation of the plans highlights the need for other guiding policies to be established by the
municipalities. A particular need is an asset management policy, firstly as a broad statement of its
intentions in this regard, and then as supplementary policies dealing with issues such as target levelof-service, service delivery performance monitoring and reporting, asset management planning, and
resourcing. Whilst it would have been logical to have such policies as a starting point, the prior
preparation of the asset management plans, as was done on this project, provides a sound backdrop
that demonstrates to participants the reasons why such policies are required, and gives pointers on
the main issues that need to be included.

A final caution
The big-picture-focused" approach that has been adopted would not have worked without the insight
of well-experienced senior practitioners in the respective technical and financial fields, and without
affording them regular and extensive access to the municipal officials.
The principal best practice/innovation point has been that:
Substantial insight has been gained into key challenges and benefits of infrastructure asset
management planning in the lesser-resourced municipalities in South Africa, and no doubt
also in lesser-resourced municipalities elsewhere.



Free State Province and South African national monitoring of water and
wastewater quality and improvement of asset management
The statutory water services authorities in South Africa (generally municipalities or combinations
thereof) are responsible for quality of drinking water and wastewater effluent. Many of them,
particularly those away from the metropolitan areas, are commonly unable to comply with the
prescribed standards.
The Department of Local Government and Housing of the Free State province has used an external
agency for two decades to monitor drinking water quality and wastewater discharge quality from
treatment works. It was up to the municipalities as to what they did with this information. However in
2002, in response to increasing non-compliance with standards, the decision was taken firstly to
expand and improve the data capture and feedback system, and secondly to use this information to
manage an improvement in water quality more actively., An intervention programme for quality
monitoring and asset management improvement was implemented, using this information to harness
effort to best effect. Elements of this programme, led by the Provincial Government, included on-site
technical assistance and training of officials and municipal councillors.
Simultaneously, an electronic data-capture, information dissemination, and management system has
been developed nationally with the use of open source software. This guides municipal officials
towards good water and wastewater quality management practices, as well as providing real-time
quality data for the information of, and action by the Provincial Government and the National
Department of Water Affairs and Forestry (DWAF). Piloted from 2004 at a limited number of sites, the
so-called eWQMS" was found to be effective, easy-to-use, robust, practical and inexpensive.
Implementing it in all municipalities in the Free State province is now well advanced, and it will follow
on to other provinces. Associated with it is a risk-based methodology to summarise the current status
of key water and wastewater quality management sustainability indicators for each municipality.
According to a self-survey by municipalities, 15% of municipalities in the Free State province supplied
acceptable drinking water quality in 2003.. Compliance has now risen to 80% in the most recent
The principal best practice/innovation point has been:
An appropriate and accessible information that is essential to improvement of water services
infrastructure asset management.
An electronic data capture system, that provides real-time information, is promising to be of
valuable assistance, but external intervention might be necessary if any improvement is to be

The South African National Government Department of Water Affairs and Forestry (DWAF) is
responsible for leading and regulating the water sector, developing policy and strategy, and providing
support to water services institutions. The statutory water services authorities (generally municipalities
or combinations thereof) are responsible for planning, ensuring access to, and provisioning of water
services within their area of jurisdiction. They may provide services themselves and/or contract
external water services providers to undertake such function on their behalf. They are responsible for
securing licences from DWAF to abstract water from, and to discharge wastewater to any water



Many of them, particularly those away from the metropolitan areas, are unable to comply with the
prescribed standards. A key contribution to this undesirable water services gap has been a lack of
requisite capacity at municipal engineering level, compounded by poor awareness as to the
importance of water quality management by municipalities, and a consequent lack of prioritisation

Potable water quality

All households should be supplied with drinking water that consistently meets the minimum standards
for potable water set out in the relevant SANS document (SANS: South African National Standards.
SANS 241-2005: Drinking-Water (6 edn.) 2005).
A survey of drinking water quality is annually undertaken by DWAF in order to gauge compliance.
The methodology used is one of self-assessment. Findings from the survey are depicted on the
map in Figure 1.

Figure 1: Municipal drinking water quality compliance (DWAF Strategic Overview of the Water
Sector in South Africa 2006)



Wastewater quality
All wastewater treatment works should achieve the prescribed receiving water quality objectives and
comply with the licensing criteria, in order to minimise environmental impact and human health risk.
As can be seen in Figure 2, a high proportion of the approximately 1000 municipal-owned wastewater
treatment works currently fail to operate satisfactorily in terms of the key criteria for both
environmental and health impact and authorisation and compliance.

Figure 2: Current status of municipal wastewater treatment works

(From DWAF, Strategic overview of the water sector in South Africa 2006)

Key challenges to improving water and wastewater quality management by

These key challenges to improving water and wastewater quality management (WQM) by
municipalities are:
To improve and implement the governance and regulatory framework.
To ensure common understanding of quality criteria.
To improve monitoring and sharing of information.
To establish municipal and sector appropriate management and reporting systems.
To provide institutional support and develop sustainable capacity.
To promote ownership and accountability (culture of performance).
To improve asset management (specifically treatment works).
To provide adequate funding for water and wastewater quality monitoring and management.
To improve public perception through continuous and consistent information-sharing on water
and wastewater quality compliance.



The long-standing Free State water quality management programme, as spelt out in WIN-SA, 2006
Free State water quality management programme.
The Department of Local Government and Housing (DLGH) of the Free State Provincial Government
has for two decades used an external agency to sample drinking water quality and wastewater
discharge. It was up to the municipalities as to whether they took any action on the basis of this
information. Nonetheless the mere possession of the information was a significant improver in terms
of the drinking water quality and wastewater quality achievements of municipalities in this province,
performance being consistently better than the national average for non-metropolitan municipalities
for many years.
In 2002, however in response to escalating non-compliance with standards in the Free State
Province, the decision was taken to firstly expand and improve the data-capture and feedback
system, and secondly to use the information to manage improvement in water and wastewater asset
management more effectively, and thus influence water quality. The DLGH led the intervention
project, using available WQM information to direct effort to best effect. Elements included on-site
technical assistance and training of officials and municipal councillors.
Significant improvements have been brought about in most of the municipalities. Some municipalities
needed to address problems which would take longer to resolve. Examples are finding alternative
water resources, significantly improving existing infrastructure (e.g. equipment repairs, treatment
works over-capacity requiring expansion), and improving staff capacity (e.g. new appointments, and
improved training of operators and supervisors).
The achievements of municipalities in the Free State, and not least the culture of monitoring and
improvement which had been engendered, have made the province very amenable for piloting of the
first provincial-wide implementation of the electronic water quality tool described below.

Electronic water quality data management system

Arguing on evidence, including the long-standing Free State experience that water and wastewater
quality monitoring and management of the resultant information was essential for improved
performance, DWAF initiated a process to identify a suitable electronic data capture, information
dissemination, and management systeml. Ideally, this would provide real-time quality management
information, and would provide it to both the municipality and DWAF, as well as to other partners,
such as the Department of Health. In this task, DWAF found a ready ally in the Institution of Municipal
Engineering of Southern Africa (IMESA), a voluntary professional association that is very familiar with
the challenges concerned with delivering effective, sustainable municipal services.
We developed the so-named eWQMS" a package using Open Source software. Piloted from 2004 at
a limited number of sites it has been found to be effective, easy-to-use, robust, practical and
inexpensive. Extending use to all municipalities in the Free State province is now well advanced, and
it is being introduced in other provinces. Associated with it is a risk-based methodology to summarise
the current status of key sustainability indicators for each municipality.
We have observed that eWQMS enables municipalities to work smarter. It also results in both
increased WQM awareness and effectiveness within the municipality, and awareness of the need for
improved asset management. Where municipalities have an existing WQM system, eWQMS is linked
via a computer patch, whereby WQM data can be seamlessly captured and passed on to DWAF
national office for regulatory purposes. The web-based eWQMS system is provided where
municipalities have no existing WQM system.
IMESAs specific role is as an honest broker, ensuring that the eWQMS solution is appropriate and
supportive of municipal requirements. This is achieved by IMESA's participation in both initial



municipal interaction fora, and subsequently by IMESA-led iterative feedback sessions with municipal
users, determining incremental improvements and enhancements in the software system.

The Free State recent developments

Thus, the DLGH of the Free State provincial government now has an additional means of assisting
municipalities with the consultative audit of drinking-water provision and wastewater treatment. For
example, at wastewater works, consultative audit technical staff can review the status of operations
and maintenance, provide on-site technical assistance, and collect samples of the final discharge
effluent. Data on treated wastewater are loaded onto eWQMS, thereby enabling auto-reporting of
legislative compliance and treatment plant efficiency to all stakeholders.
Figures 3 and 4 below give examples of eWQMS outputs. Figure 3 is the management dashboard
screen. Figure 4 highlights compliance at particular sites on an overview screen that uses a map of
the Free State Province.

Figure 3: Example 1 of eWQMS outputs



Figure 4: Example 2 of eWQMS outputs

WQM sustainability analysis

DWAF has over the course of the last year analysed each municipality at strategic level with respect
to its WQM sustainability. Indicators of WQM sustainability include:
Status of raw water sources, water treatment and other infrastructure considerations.
Availability of laboratory facilities, sample points monitored, water quality parameters
monitored, frequency of monitoring.

Staff skills, staff numbers and capacity, communication protocols.

Ability to interpret drinking water quality results, availability of emergency procedures,
logistical issues hampering effective WQM (e.g. transport, distance, condition of roads).
Management awareness and engagement; and financial status.
Using a risk-based methodology, weighted scores are used to summarise the current status of the
particular category and highlight where capacity building efforts should be focused. These scorings
are then graphically represented on a so-called spider diagram which readily communicates to all
parties the current standing and risk profile of the municipality as regards WQM. The outputs of the
strategic level WQM sustainability analysis are intended to be used both internally (in the municipality)
and externally (at provincial and national level), in order to guide the necessary capacity development
and funding. The performance categorisation used is shown in Table 1 below.



Category Score



70% 100%


Usually requires minor interventions to

ensure sustainability

45 69%


Sustainability is in question if interventions

are not implemented



Effective and sustainable WQM is not

possible without external support and

Table 1: WQM performance categorisation

We expect that effective use of the eWQMS tool by municipalities, together with broader proactive
and supportive intervention to ensure that WQM is adequately resourced and capacitated, should
lead to further considerable improvement in WQM.
Due to these various measures, and to increased awareness of the need for water quality monitoring
and improvement, performance has improved as asset management has improved. Evidence,
according to regular DWAF-initiated self-surveys by municipalities, is that 28% of municipalities
throughout South Africa in March 2003 supplied drinking water that was fully compliant with the basic
drinking water quality criteria.. Compliance by November 2006 had risen to 62% (DWAF November
2006. "Drinking water quality assessment in municipalities in South Africa"). (Note that not fully
compliant does not necessarily imply that the drinking water is unfit for potable consumption.)
The principal best practice/innovation point is that:
Appropriate and accessible information remains essential to improving water services
infrastructure asset management. An electronic data capture system, providing real-time
data, is promising to be of valuable assistance. External intervention may nevertheless be
necessary to effect improvement.



Johannesburg Water and its "Operation Gcin'amanzi" service upgrading, asset

management and demand management programme
Johannesburg Water was established in 2001 as a business entity of the Johannesburg City Council.
It inherited the water supply and distribution system of a municipality of 3.5 million people, created by
amalgamating a number of local governments that had over the years practised very different
infrastructure asset management regimes.
In 2003, Johannesburg Water launched "Operation Gcinamanzi" ("Save Water") for water services
upgrading in Soweto. The objective of this programme, with the co-operation of communities and
each householder, has been to put a management plan in place, and to repair all pipes and fittings
leading to and on all properties. Thereafter, all properties will be metered, (allowing for the first 6
kilolitres of free water supplied to households each month, this allowance being in terms of the
National Governments programme on free minimum basic services).
Due to leakages in the municipal infrastructure or on individual properties, or to wasteful practices,
approximately 70 000 megalitres have been unaccounted for in Soweto each year, costing about R
160 million per year (US$ 23 million). Given this situation, the expected total cost (at 2003 prices) of
R 450 million (US$ 65 million) could be recovered in as little as three years.
Reduced water leakages to sewer will also reduce sewer flow,, directly resulting in environmental
benefits and reduced cost of sewage treatment. And although the population of the area served by
the nearby treatment works is increasing steadily, and will continue to do so, the capacity recovered
by Operation Gcin'amanzi will allow a 10-year postponement of already-planned increase in capacity.
The work has been completed in respect of about 40% of the total of 170 000 properties. The work
has for the most part been carried out by Soweto-based small contractors, the product of a contractor
development programme.
The principal best practice/innovation point is:
Substantial long-term operational and capital savings, as well as environmental benefits and
improved service to customers, can accrue from targeted short-duration repair and
rehabilitation programmes and long-term infrastructure asset management.

Johannesburg Water was established in 2001 as a business entity of the Johannesburg City Council.
The municipality owns 100% of the shareholding of Johannesburg Water (Pty) Ltd. The utility has a
25-year contract with the municipality. It inherited the water supply and distribution system of a
municipality of 3.5 million people, created by amalgamating a number of local governments that had
over the years practised very different infrastructure asset management regimes.
Addressing the myriad water services problems of Soweto has been one of its highest priorities, but
by no means its only priority.
This case study describes as briefly as possible the legacy inherited by Johannesburg Water, and its
process and purpose. This then serves as background to the description of water services in Soweto
and to Operation Gcin'amanzi, in particular.

Johannesburg Water



The Johannesburg City Council was created in 1995 by the amalgamation of 13 autonomous local
authorities. The Municipality was preoccupied for the next two years with rapid and large-scale
development spending on the areas underserved during the apartheid era, and it did not pay
adequate attention to financial discipline. Thus, for example, revenue collection fell while capital
expenditure escalated. Municipal debt grew rapidly in this period of growth without sustainability.
A new administration In 1997 committed itself to building a sustainable institution. A key element of
its programme was to ring-fence a number of municipal services, and to create utilities wholly owned
by the municipality. One of these was a water services (water and sanitation) utility.
Transforming the financial situation of the municipality required substantial economising.
Programmes of new capital works were cut back, but so, too, were maintenance and refurbishment
programmes. Thus, by the time the utilities were established, infrastructure services had suffered
from nearly a decade of gross under-investment in infrastructure replacement of items that had
reached the end of their service life, and in planned maintenance and refurbishment.
Johannesburg Water was set up with the assistance of an established international water supply
management company, which then managed it for the first five years, progressively building the
capacity of utility staff in the process. An average 1 250 M/day of water was distributed in 2007 (30%
of it supplied to Soweto); 900 M/day of wastewater treated; involving a R 2.5 billion annual turnover.
In 2003, asset replacement cost R 11.6 billion (USD 2 000 million). Johannesburg Water, when
established in 2001, thus faced a number of issues that had been accumulating over the years,
Amalgamating the water services of 13 local authorities.
Addressing the backlog of lower levels of service, particularly in Soweto, the largest of the
poorer residential areas. In 2001, as a whole, 3.5 million people in the municipality were
linked to the water and sewer reticulation, while more than half a million people lived in
informal settlements, dependent on alternative forms of sanitation and on standpipes.
Addressing the massive unaccounted-for water (in 1999, this constituted 42% of the water
purchased from the bulk supplier, Rand Water).
From inception, Johannesburg Water gave high priority to preventative maintenance and monitoring
of key maintenance indicators. The focus of attention in the first two years was on the bulk
infrastructure, and results soon became apparent. For example:
Pump-station breakdowns halved within two years.
Spills and overflows at wastewater treatment works fell in two years from 750 per annum to
150 per annum.
Final effluent compliance of the wastewater treatment works rose in two years from 85% to
Thereafter, Johannesburg Water turned its focus to other matters, such as taking over billing from the
municipality, and improving billing efficiency and revenue collection. It also began to give a great deal
of attention to the problems of Soweto and other poorer areas.

At that time, even in the reticulated areas (not forgetting that a large number of Sowetans have lived
in informal settlements, and many still do) the level of service was not the same as in the more
affluent areas of Johannesburg. Both water and sewerage mains had been laid midblock, with toilets



in separate structures close to the back boundary. Often, the midblock water mains were 50 mm
steel pipes.
Very little maintenance had been done to the system over the years - either to the municipal system,
or to the pipes and fittings on each property. Population densities had grown in some suburbs of
Soweto over the years, without concomitant increase in bulk supply to these areas. While the quality
of water remained high, the levels of serviceability had in some suburbs dropped to the extent that the
residents could only draw water during the night, when the draw-down was low, and the network had
sufficient pressure.
The steel pipes were leaking profusely for many reasons, including corrosion caused by stray
electrical currents
Pipes and fittings on many individual properties were also in a bad state, and deteriorating by the
year. In many cases, taps could no longer be turned off, and likewise, toilets would flush
continuously. Because households were charged a flat rate for the water they consumed, they had
no incentive to repair these pipes and fittings. There was no "ownership" by households of their pipes
and fittings, nor of their water consumption.
The estimated water loss in 1987 from the system was somewhere between 50% and 60%. At that
stage, and for the next 15 years, the authorities did not have the funds to repair the leaks or replace
the systems, and the water losses steadily increased. By 2002, the rate of growth of water supply to
Soweto was nearly double the rate of growth of household formation in Soweto, reflecting the rate of
network deterioration and the escalating wastage. The average supply to each household with a
water connection was nearly three times the supply to equivalent but well-maintained areas
elsewhere in Johannesburg.
It was calculated that the total unaccounted-for volume was over 70 000 M/annum, equivalent to an
average continuous flow of 200 M/d.
Highly relevant was that few households paid the flat rate charge, anyway. This arose from a number
of factors, including the aforesaid lack of ownership, politically motivated resistance to paying,
unaffordability (even if willing to pay), and a widespread opinion among householders that payment
should be contingent on improvement of the service.

Operation Gcin'amanzi
Johannesburg Water and the municipality recognised that "the current situation represents a vastly
unsustainable service delivery option that requires immediate, intensive and comprehensive
intervention on a number of fronts". The following "interrelated and fairly independent" dimensions
thus all needed to be tackled by "a holistic and integrated approach that proactively and
simultaneously addressed all of the facets". The quotation is from a Johannesburg Water proposal
call in 2002. The points are:
A shortage of technical and engineering management, including management of bulk
purchases; storage, pressure and zone management; distribution and supply.
Insufficient systems, procedures, programmes etc and readily available plant and material to
perform operation and maintenance functions.
Neglect of metering, billing, customer management, and customer service issues.
Non-payment by customers.
A programme was needed that would:



Address the water supply problem in Soweto.

Additionally, address the issues surrounding affordability, wastage and loss, and by this
means create an environment conducive to payment for water services.
Interventions, both technical and social, would therefore be needed as follows:
Technical interventions:
Reduce water that is unaccounted-for in Soweto.
Rehabilitate the water network in Soweto.
Allow customers full access to the Government's free basic service programme of a free
water allowance of 6 k/month, billing them for quantities that exceeded this basic water
Provide a better level of service in the form of larger diameter pipes laid in road reserves, as
opposed to midblock location.
Locate leaks and perform once-off repair of plumbing fixtures.
Improve the level of integrity of technical information systems and customer records.
Social interventions:
Educate the community on water as a scarce and precious resource.
Create an awareness on water conservation.
Educate the community on the water cycle.
Secure a buy-in from residents/consumers regarding their water consumption.
Provide education on tariffs and free basic water.
Provide customer education on rights and obligations.
Thus, in 2003 Johannesburg Water launched "Operation Gcinamanzi", a Zulu phrase that, means
"save water". The objective of this programme has been to put a management plan in place with the
co-operation of communities and each householder, and to repair all pipes and fittings leading to and
on all properties.
Due to leakages in the municipal infrastructure or on individual properties, or to wasteful practices,
approximately 200 M was unaccounted for in Soweto every day, as has already been noted. This
was costing Johannesburg Water about R 160 million per year (US$ 23 million). The expected total
programme cost (at 2003 prices) of R 450 million (US$ 65 million) could be recovered in respect of
savings in water purchases by Johannesburg Water over as short a period as three years.
The work to be done would consist of:
Laying of the new water mains in the road reserves.
Decommissioning of midblock mains.
Leak detection and repair of all existing water
renewed/upgraded/replaced as part of the programme.









Renewing/replacement of all individual house connections in areas where secondary water

mains were to be replaced.
Once-off leak repair retrofitting exercise of private plumbing fixtures located on the formal
Provision of a metering infrastructure on each formal property. The meters are generally of
the prepayment type.
Customer Care, awareness programmes, social aspects, education, training, facilitation, etc.
for all above aspects, including a house-to-house survey.
This broad range of interventions would be phased in over a period of six years. 170 000 formal
properties would be affected, including schools, health facilities and commercial properties.
Attention is drawn to the point that all properties were to be metered because, stated in terms of the
proposal call:
The metering of all customers is required in terms of national legislation.
It is widely accepted that the metering of individual consumers leads to a stronger sense of
ownership of consumption, and it can therefore be expected that metered consumers will be
more likely to maintain pipes and fittings, and also cut down on needless usage of water, with
beneficial effect to their own pockets.
Individual metering and billing will empower Johannesburg Water to manage customers,
which will inevitably lead to an increase in payment percentages.

Programme progress
The programme has been presented and discussed extensively with the community, and continues to
be presented, in stages, as it reaches successive suburbs in Soweto. Johannesburg Water is
advocating the programme on the basis of improvement in levels of service and of serviceability.
A significant number of residents have objected on various grounds. Most, if not all, refuse to make
any payment for water, regardless of the quantity consumed. In addition to disruptions at some public
meetings, site workers have on occasion been threatened, and operations vandalised.
However, Johannesburg Water is confident that by far the majority of the residents in Soweto support
the efforts to improve the levels of service and serviceability of the water delivery system. The
programme, although delayed, has consequently gone ahead.
The work was completed by April 2007 to the extent of about 40% of the eventual 170 000 properties.
The work has for the most part been carried out by Soweto-based small contractors, the product of a
contractor development programme.
Arrangements for the receipt of customer complaints and processing thereof, to resolution of
complaints, vending of payment meter tokens, and for the inspection of meters, are keeping pace with
the technical interventions. This is essential.
By April 2007, reduction in bulk purchase of water was of the order of R 120 million.



Consequent effects
The financial savings on the cost of water hitherto purchased from the bulk supplier have been
considerable. The benefits to householders in terms of improved service, namely access to service,
reliability, and improved pressures, are also considerable. However other peripheral benefits have
Deferred upgrading of water reservoir capacity.
Deferred upgrading of wastewater conveyance and treatment capacity.
Environmental benefits, such as less nuisance with ponded water and wastewater, and
reduced overflows from sewer mains and at wastewater treatment works.
Improved financial status of Johannesburg Water and of its sole shareholder, the municipality.
Improved circumstances for other water demand management measures.
The significant benefit of a deferred wastewater treatment plant upgrade arises from the following
reason. Water that is wasted through leaks in reticulation systems percolates into the soil. Water that
is passed by defective fittings (such as continuously flushing toilet cisterns, and dripping taps) enters
the sewerage system, and eventually contributes to the hydraulic volume that enters the treatment
works. Savings through repair of defective fittings should not only be measured by the cost of water
saved, but also by the saving arising from treatment of a diminished wastewater hydraulic volume.
Similarly, judicious household water use diminishes hydraulic volume inflows at the treatment works.
While Operation Gcin'amanzi is in progress, reticulated water services and waterborne sanitation are
being extended to other households in Soweto. The total wastewater flow from Soweto is not
projected to increase, despite an increase in the number of households that will be reticulated.
Johannesburg Water has gained a reprieve of at least 10 years on the upgrading of bulk outfalls and
wastewater treatment capacity as a consequence of the programme.
Bushkoppie and Olifantsvlei Treatment Works serve those areas of the City of Johannesburg lying
south of the ridge which runs east/west and forms the Witwatersrand watershed. Soweto constitutes a
large portion of the catchment area of both sides. We anticipate that the reduction in effluent flows to
these works as a direct result of Operation Gcin'amanzi will by the end of the programme result in a
combined saving of about 120 M/day. Instead of additional treatment capacity having been required
in 2005, it should only be required in about 2015.
The proposal call stated that it was "extremely important" to note that the programme "can be
financially motivated purely on savings in water purchases by Johannesburg Water", and is thus not
dependent on improvement (if any) in payment percentages, or any other factor.








OGA reduction
B.K.+OFV less OGA (22kl/stand)





















Figure 1: Historical and projected average annual daily flows to Bushkoppie (BK) and Olifantsvlei

Many householders in Soweto now enjoy more reliable and otherwise improved water services, and
others can look forward to the same improvement, as the programme proceeds.
As a direct result of progress with the programme, Johannesburg Water has been able to motivate for
and gain approval for measures such as:
Consumer metering and billing.
Community awareness of water wastage, and households consequent reduction of wastage.
Development of an asset inventory as a first step to full asset management.
The principles of Operation Gcin'amanzi appear sound, based on the evidence, so far. Using
improvements and local modifications, these principles are already being applied in other, smaller,
low-income areas in Johannesburg where similar water services situations exist.
The principal best practice/innovation point is that:
Substantial long-term operational and capital savings, as well as environmental benefits and
improved service to customers, all accruable from targeted short duration repair and
rehabilitation programmes and long-term infrastructure asset management.



DWAF National Water Services Infrastructure Asset Management Strategy

Insufficient attention is being paid by the majority of South African water services institutions to the
management of their infrastructure. In addition, many have, due to years of neglect, built up a
backlog of need in respect of maintenance and also refurbishment, renewal and replacement. The
competing demands made on limited operational budgets, staff and other resources, severely
constrain proper management of infrastructure.
Thus the national Department of Water Affairs and Forestry (DWAF) decided in 2005 that, as sector
leader, it needed to investigate the infrastructure asset management situation and thereafter to
provide guidance to water services institutions. Accordingly it commenced the formulation of a
national "water services infrastructure asset management strategy".
The case study describes the background, and then the work completed thus far, viz.:
A desktop strategic study of the state of water services infrastructure and of the state of water
services infrastructure asset management in South Africa.
Analysis of this information, identifying the key factors that drive the current state, and
identifying elements needed for an enabling environment to ensure sound asset
Identifying priority strategic action areas.
The priority strategic actions take cognizance of DWAFs mandated responsibility, and what DWAF
needs to do within its own sphere as well as in conjunction with others, particularly with other national
government departments.
DWAF is currently in the next stage, one of planning and programming for implementation of the
The principal best practice/innovation point is that:
A national government department has taken the lead in developing a strategy for water
services infrastructure asset management in all water services institutions.

Introduction and problem statement

The water services sector in South Africa is responsible for infrastructure assets with a replacement
cost of several hundred billion Rands (calculated at 1 US$ ~ 7 ZAR). During the next decade much
more infrastructure will be provided, yet, as noted in documentation by DWAF, May 2005, "terms of
reference for WS infrastructure asset management strategy: " many water services authorities do
very little infrastructure asset management, and do not budget sufficiently for this."
Insufficient attention is being paid by the majority of South African municipalities and the regional
water boards owning water services infrastructure to the ongoing commitments that they have
incurred to manage their infrastructure. Public sector institutions owning water services infrastructure
comprise principally the municipalities and the regional water boards. In addition, many have, due to
years of neglect, built up a backlog of need in respect of maintenance and also refurbishment,
renewal and replacement. The competing demands made on limited operational budgets, staff and
other resources, severely constrain the proper management of their infrastructure. Thus DWAF
concluded in its May 2005 strategy document that "The Department, as sector leader, needed to



investigate the infrastructure asset management situation and provide guidance to the sector; thus the
need for this project"
Accordingly, during 2005, DWAF, assisted by consultants, commenced the first stage of formulation
of a national "water services infrastructure asset management strategy".

National government has made substantial progress on its promise to improve the lives of previously
disadvantaged citizens. The expenditure on infrastructure has been considerable and there has
An increase of 80% (1994-2004) in the number of people with access to a basic level of water
An increase of 56% (1994-2004) in the number of people with access to a basic level of
sanitation service.
Furthermore, the basic services programme is nowhere near complete, and it is clear that the
government is intent on continuing to fund infrastructure until the backlog is eradicated. Public sector
and parastatal infrastructure owners can thus expect that their portfolios of infrastructure
responsibilities will continue to expand, as the effort continues to provide services to all, and
especially to provide free basic services to the poor. In official terms of the nationwide "free basic
services" policy, indigent households are not charged for services such as water, sanitation and
electricity, provided that they stay within stated limits. In respect of water, for example, they are not
charged for the first 6 kilolitres they use each month.
The definitive "Strategic Framework for Water Services" in the document, "Strategic framework for
water services: water is life, sanitation is dignity." DWAF, Department of Provincial and Local
Government, National Treasury, South African Local Government Association, and South African
Association of Water Utilities, September 2003, clearly sets out the sectors national goals with
respect to access to basic water services, healthy living practices, accountability of municipalities, and
regulation of services that are provided equitably, affordably, effectively, efficiently and sustainably.
The key challenges listed in the "Strategic Framework" deal with the need to extend coverage of
water services, for water services to support economic development, and for the institutional reform of
water services provision. Most pertinently, however, "Services and the use of the water resource
must be sustainable, to ensure that we continue to make progress, and to ensure that future
generations benefit from this progress. Under-expenditure in maintenance and under-investment in
rehabilitation is a significant challenge to overcome."
The "Strategic Framework" states that it is incumbent on owners of water services infrastructure to
maintain a register of water services infrastructure assets and put a system in place to manage this
infrastructure in terms of a maintenance and rehabilitation plan. This plan must be based on the
principle of preventative maintenance and must be part of the water services development plan.
Infrastructure assets must be rehabilitated and/or replaced before the end of their reliable and
economic life and the necessary capital funds must be allocated for this purpose.

Scope of the first stage

The first stage has consisted of three phases:
A desktop strategic study, a "scan", of the state of water services infrastructure, and of the
state of its management in South Africa.



A process of analysing this information, identifying the key factors that drive the current state,
discovering underlying problems and opportunities, and identifying elements needed for an
enabling environment, to ensure sound asset management.
Identification of priority strategic action areas.

Phase: Scan
The desktop strategic study (scan) of the state of water services infrastructure and its management
found that there are many different reasons for the great differences found in infrastructure
management, ranging from best practice through to unacceptable management. Nevertheless there
are common patterns, and it is clear that all or nearly all water services institutions have to contend
with many of the same issues, chiefly:
Inadequate budgets.
Inadequate skills (especially technical skills) and experience.
A dearth of guidelines, norms and standards.
This is despite there being a very broad range in the capacity of water services institutions, the state
of their infrastructure, and the state of the management of their infrastructure. Consequently, there
could definitely not be a "one-size-fits-all" set of measures to improve the management of their
infrastructure assets.
Budgets and staffing policies are often severely inhibitive of sound infrastructure management,
thereby placing much infrastructure at risk. The great majority of municipalities are not making
adequate provision for long-term preventative maintenance, refurbishment and eventual replacement
of their infrastructure. (This is not just inadequate provision for water services infrastructure, it should
be noted, but also for other infrastructure services for which municipalities are responsible ,particularly
roads and stormwater.)

Phase: Proceeding from fact-finding to solution-identification

The scan phase findings have been the foundation upon which the work of the next phase has been
built. This next phase commenced with a process of identifying the key factors that drive the existing
state of water services infrastructure and the state of its management. This phase involved not just
problem identification, but also discovery of underlying opportunities, and analysis and classification
of problems.
More than 400 generic challenges have been identified, rigorously analysed, and classified into
"challenge areas". This analytical approach has facilitated better understanding of individual
challenges, as well as seeing the bigger picture in terms of priority needs. For the record, the
challenge areas are:
Planning problems.
Technical/design problems.
Construction/installation problems.
Infrastructure operation problems.
Repair/maintenance/refurbishment problems.
Inadequate skills for infrastructure asset management.



Statistical/management problems.
Financial problems.
Social/cultural problems.
Economic/poverty problems.
Natural environment problems.
Political/tactical problems.
The following priority challenges have been identified within these areas:
Life-cycle management (service delivery does not end with infrastructure projects).
Knowing your infrastructure (including asset register).
Implementing infrastructure asset management processes and procedures.
Clear responsibility and accountability for infrastructure asset management.
Hands-on approach (and also that one size does not fit all).
Water services infrastructure asset management is a part of management of the total assets
of the owning institution.
Funding requirements and processes for infrastructure asset management.
Infrastructure asset management staffing requirements (number and skills).
The analysis has then proceeded from challenges to the preliminary identification of a solution for
each of the 400-plus generic challenges. Evaluation (including noting some repetition of solutions)
and identifying commonality of solutions has enabled classification of solutions into one or other of
nine solution types, viz.:
Human resources (HR) (i.e. including skills and appointments).
Legal and procurement.
Monitoring and evaluation.
Management and leadership.
Operation and maintenance.



These identified solutions have been categorised "Priority 1", "Priority 2", or "Priority 3".
Only the "Priority 1" solutions are depicted in pie diagram form below:

This figure makes it clear that much needs to be done with respect to the aspects of human
resources, skills development and capacity building. While the focus of capacity building is on
municipal capacity building, capacity empowerment must also include DWAF and other national and
provincial role players that have to manage the process and regulate effective service delivery.
Management and leadership is another important area. Specific actions need to be taken by DWAF
as sector leader, and by municipal managers and politicians, in general. To make a strategic
intervention of this kind, it is essential that politicians and senior managers fully understand,
appreciate and support infrastructure asset management.
Financial solutions have come up in third position in the order of frequency. This implies that finance,
also, is a very important intervention area, and a key success factor for sustainable infrastructure
asset management. The solutions include, amongst others, improved budgeting and allocations for
infrastructure asset management, financial incentives for effective infrastructure asset management
performance, cost recovery and various other planning, regulation and administration issues.
Given the way in which the solution types have been defined, and that operation and maintenance
problems are the direct result of skills or leadership problems, flagging was under "human resources"
or "leadership" and not under "operation and maintenance". "Operation and maintenance", as a class
of solutions, is ranked only fifth in frequency, and this is not surprising. Other key operations and
maintenance solutions can be found under finance, management and technical.



The way in which the analysis has been done also enables identification of the parties:
with responsibility to lead the way forward for each solution
that should be involved, or merely informed.

Phase: Priority action areas

This phase, completed early in 2007, comprised the identification of priority strategic actions.
Identification of these actions took cognizance of DWAFs mandated responsibility and what DWAF
needs to do within its own sphere and also in conjunction with others, particularly with other national
government departments. The priority actions identify elements needed for an enabling environment
to ensure that infrastructure assets are properly managed, and that they take into account the need to
synergise with asset management initiatives external to DWAF. They express what needs to be
done, irrespective of whether some of the elements of an action are or are not already under way.
The immediate basis for the recommendations is the 150 Priority 1 solutions, which have been
packaged into the form of multi-faceted priority actions.
The priority strategic actions, amongst other recommended measures, include recommendations with
respect to awareness raising, funding, legislative review, performance management review, improved
incentives to councillors and officials to responsibly manage the infrastructure in their care, improved
infrastructure asset management guidelines, and a skills plan.

Way forward: The second stage

The set of recommended priority actions does not constitute an implementation plan. DWAF, in
coordination with other major stakeholders, and with the assistance of a new external team, is
currently busy with the second stage of formulation of the water services infrastructure asset
management strategy, and is designing the implementation plan and programme. The plan will
identify the who, what and when to be considered by DWAF, and will indicate prioritisation in terms
of both urgency and importance. It will also determine the how, including tactics, culture and
incentives, and it will identify key performance areas and will set key performance indicators.
The details of selected aspects of the plan and programme are being formulated with the assistance
of an external team, and with the involvement of key sector partners such as National Treasury, other
national government departments responsible for public works and for provincial and local
government supervision, the South African Local Government Association (representing elected
municipal councillors) and the Water Research Commission, and taking into account the roles of the
various water services institutions. The result will be an integrated and co-owned implementation
Cognizance is being taken of the main other national infrastructure asset management initiatives, and
how they are complementing an achievement of the objectives of DWAF. Among these are the
National Infrastructure Maintenance Strategy (approved by Cabinet in 2006), the national Guidelines
for Infrastructure Asset Management in Local Government, and National Treasurys measures to
increase provincial and local government accountability for assets.

It is timely that increasing attention is being paid to water services infrastructure asset management.
The recent work by DWAF and others in discovering and documenting the poor state of so much of
the water services infrastructure is serving to underline the importance of the DWAF water services
infrastructure asset management strategy, and the need to complete the strategy formulation and to
start implementing the strategy. The principal best practice/innovation point is that: a national
government department has taken the lead in developing a strategy for water services infrastructure
asset management in all water services institutions.



Sebokeng/Evaton water saving through pressure management

The minimum night flows in many South African residential suburbs are unacceptably high. An
example is the minimum night flow of 3000 cubic metres per hour, recorded in Sebokeng/Evaton,
south of Johannesburg, in July 2003. This represented 75% of an average daily flow.
With leakage at this magnitude, the potential for water saving is significant. The purpose of a project
reviewed in this paper was, to reduce this leakage in one of the worst areas through pressure
management. The high losses were resulting in the municipality paying R110 million (US$ 16 million)
per annum to the bulk water supplier, whereas it was estimated that this cost could be reduced by of
the order of two-thirds through pressure management. The high levels of wastage were also creating
excessive sewer flow, leading to a demand for upgrading wastewater treatment works capacity.
The situation is similar to that in Khayelitsha, in Cape Town, where the first large pressure
management installation in South Africa was commissioned, in 2001. This is still operating at full
potential, and continues to give the savings that were estimated during design, thus demonstrating
that such measures are effective and sustainable.
The project undertaken in Sebokeng/Evaton is the first of its type in South Africa where a publicprivate partnership, a partnership between a firm of consulting engineers and the municipal water
entity was formed to undertake the work. The firm and its private sector funding partners financed the
full cost of the installation to regulate water pressures. On installation the equipment immediately
became the property of the municipality.
A proportion of the savings (savings in terms of reduced purchases from the bulk supplier) accrues
to the firm and its funding partners for a period of five years after commissioning of the installation in
2005, with the remaining savings going to the municipality. If there are no savings, then the firm and
its partners receive no return on their investment. Thus far, the savings (which are independently
audited) have been substantial The savings are so large that, as predicted, the installation has paid
for itself in the first four months and there is little reason why this trend should not continue.
We consider the installation to be one of the largest pressure management installations in the world.
This arises from the magnitude of the problem of excessive minimum night flows.
The principal best practice/innovation point is that:
Under appropriate circumstances, public-private partnerships can be profitable for the private
sector, even though income may not come from direct sales of a product or service, but may
come from savings, on a risk-sharing basis, with a water services authority.

Sebokeng and Evaton form part of the Emfuleni Local Municipality to the south of Johannesburg, the
main commercial and industrial centre of South Africa. The areas are supplied with potable water
from a 240 M concrete reservoir through two large water mains (1000mm and 675 mm diameter
respectively) which run parallel to each other before diverging into the two respective localities. The
new pressure management installation is located just before the divergence.
Sebokeng and Evaton (population around 500 000) are predominantly low-income residential
suburbs, with more than 65 000 connections and an equivalent number of sewer connections. At the
start of the project, the residents experienced average water pressures of between 30m to 60 m water



The combination of low income, coupled with high unemployment, has resulted in a general
deterioration of the internal plumbing fittings over a period of many years. The poor quality fittings
cause excessive levels of leakage to sewer, as is clearly evident from the unusually high levels of
sewer flow during the late evenings and early mornings.
The Minimum Night Flow (MNF) of 3 000 m3/h in July of 2003 (see Figure 1) for the
Sebokeng/Evaton area represented 75% of the Average Daily Flow (ADD) which was measured to be
4 020 m /h. In a typical well-managed system, with no leakage problems, the MNF to ADD ratio is
usually in the order of 10% to 15%, implying that a MNF of less than 400 m /h would be the norm.
The figures clearly highlighted the scale of the leakage problem in the area and that the potential for
saving water was significant.

Figure 1: Flow entering the Sebokeng/Evaton areas in July 2003

Because leakage is a function of water pressure, any reduction in pressure, even if only for a short
period each day, will result in lower leakage, as well as fewer pipe bursts. If water pressures can be
lowered significantly during the off-peak periods (especially at night) then substantial savings can
usually be achieved.
While it is true that pressure management will not stop leaks, and that all existing leaks will remain
after pressure management has been implemented, the key issue is the amount by which leakage will
be reduced, if the pressure is reduced. Pressure management is not appropriate for certain systems.
In particular cases, however, such as the Sebokeng/Evaton area, pressure management is by far the
most appropriately cost-effective means of intervention, and should therefore be implemented as the
first main intervention. This does not imply that other forms of water demand management should not
at some other stage be implemented. It simply highlights the fact that the greatest savings can be
achieved for the least cost through pressure management, after which the other forms of intervention
can be considered.



Why install pressure management

Most water reticulation systems are designed to provide a minimum pressure at all points in the
system throughout the day. This means that the minimum pressure (normally specified in the local
by-laws) occurs at some critical point in the system which is often either the highest point in the
system or the point furthest from the supply.
All water distribution systems have significant fluctuations in demand throughout the day, with
morning and evening peaks coupled with periods of low demand during the night, and sometimes
also during the early afternoons.
Many systems also have seasonal fluctuations caused by
climatic factors that influence irrigation requirements or by holiday migration that can significantly
influence the demand for periods of days or weeks at a time.
Because the water supply systems are designed to provide a set minimum pressure throughout the
day, they are generally designed to meet this pressure requirement during periods of peak
demand, when the friction losses are at their highest, and inlet pressures at their lowest. As a
result, higher pressures than necessary only occur close to the inlet point of the reticulation during
the peak-demand periods.
This concept is shown in Figure 5, a typical pressure situation for a zone at peak demand periods
where the minimum pressure required is 20 m. The same zone is shown again in Figure 6 for the
periods of low demand. Note the unnecessarily high pressure throughout the entire system.
It can be understood from these two figures that unless some form of active pressure management
has already been implemented, the pressure in a water distribution system is likely, for most of the
time, to be unnecessarily higher than required. Moreover, the higher the pressure the greater the
leakage, thus, leakage levels in most systems are higher than they need to be.
It is often possible to substantially reduce water losses by adjustment of pressures during off-peak
periods without locating or repairing any leak. After addressing the excessive pressures, other
measures, such as repairing leaking pipes and/or retrofits can then be addressed.

The purpose of the project was to reduce the unacceptably high levels of leakage and wastage of
water in the Sebokeng and Evaton areas, as this was resulting in a water account to the municipality
from Rand Water, the parastatal bulk water supplier, of R110 million ($US 16 million) per annum at
2003 prices. Based on normal acceptable levels of service, the annual water bill for the area should
have been closer to R30 million ($US 5 million).
The high leakage levels within the properties resulted in water bills that few residents could afford, or
were willing to pay. This led to very low levels of payment. Householders simply ignored leakage loss
on their properties, and refused to settle the large water accounts arising for these leakages. As with
a typical circular problem, one issue leads to the next, and the only solution is to break the cycle of
high leakage, after which diminished water accounts could possibly lead to resumed payments by

The sewer flow originating from Sebokeng/Evaton at night were measured at 2 500 m /h, representing
about 80% of the water flowing into the area at night, and indicating that most of the leakage was
occurring within the properties, i.e. household plumbing leaks that were draining directly to sewer.



The high level of water wastage to sewer, increasing wastewater works inlet volumes, was also
creating a need for upgrading the wastewater treatment works capacity.
The situation was very similar to that experienced in Khayelitsha where the first large advanced
pressure management installation was commissioned in 2001 for the City of Cape Town. This project
is still operating at its full potential and continues to generate the predicted savings, verifying that such
measures are effective.
The Sebokeng/Evaton installation is regarded as one of the largest pressure management
installations in the world, a status arising from the magnitude to which the minimum night flows had
The pressure management project represents the first phase of a long-term initiative to reduce water
consumption in the area to normally acceptable levels, leading, in turn, to realistic levels of payment
by consumers, and finally to a financially solvent and sustainable water utility.
Annexure A presents more information on the principles of pressure management.

The installation
A box-shaped concrete structure 10m long by 10m wide and approximately 5m deep, houses the
pipes and valves that manage the water pressures in Sebokeng and Evaton (Figures 2 and 3).
The installation has involved cutting into the two existing water mains and replacing a short section
with a series of smaller pipes and associated valves, meters, and strainers. The new pipe-work and
fittings enable the pressures into the two areas to be controlled in such a manner that the water
pressures can be reduced during off-peak periods, and are then restored to the original standard
pressures during periods of high demand. The leakage from the system, as well as the incidence of
new burst pipes, has been greatly reduced in this manner.

Figure 3: Elevation layout of the Sebokeng/Evaton installation



Figure 2: Plan layout of the Sebokeng/Evaton installation

Project team
The project undertaken in Sebokeng/Evaton is the first of its type in South Africa of a public-private
partnership, where a firm of consulting engineers and the municipal water entity were linked to
undertake this kind of work. The firm and its private sector funding partners funded the full cost of the
installation to manage water pressures, after which the installation immediately became the property
of the municipality.
Twelve partners were involved in a variety of roles. The smallest firm was (and still is) an
independent entity appointed as external auditor of the savings. Auditing of the savings is key to
success, because it is from a proportion of these savings that the payments are made to the
More than 50 public meetings were held in the local communities to inform the residents of the
project, and to address any concerns they had with regard to the project.
The construction used labour-based practices, where possible, in order to maximise the employment
opportunities for the local communities.

Savings achieved through the project exceed 8 million m3/annum, representing 20% of the total water
supplied to the area.
The savings are clearly shown as the red area in Figure 4, which is based on the initial logging results
from July 2003 compared to the corresponding period in July 2005. Note that the graphs do not take
the escalation in water demand from 2003 to 2005 into account. This would result in a larger red
area. But this has been accounted for when calculating the annual savings.



Figure 4: Savings, being the difference in water use in July 2003 compared with July 2005

Further savings will be achieved through in-service adjustments to the installation.

The project cost approximately R5 million (US$ 0.7 million) to complete. The cost of water saved
(savings in terms of reduced purchases from Rand Water) in the first full year of operation was in the
order of R 20 million (US$ 3 million). The savings are thus so large that, as predicted, the installation
paid for itself in the first four months.
A proportion of the savings (capped at 20% of the total savings) accrues to the firm and its funding
partners for a period of five years after commissioning of the installation, which took place in 2005.
Remaining savings go to the municipality. If there are no savings, then the firm and its partners will
receive no return on their investment. We see no reason that this substantial level of savings will not
The principal best practice/innovation point is that:
Under certain circumstances public-private partnerships can be profitable for the private
sector, even though income may not come from sales of a product or service, but may come
from savings, on a risk-sharing basis with the water services authority.



Figure 5: Typical zone pressure distribution during peak-demand periods

Figure 6: Typical zone pressure distribution during low-demand periods



Rand Water major pipeline asset management

The parastatal Rand Water is the bulk water supplier to the industrial and commercial heartland of
South Africa, including the metropolitan areas of Johannesburg and Pretoria. It abstracts an average
of 3 500 M/d from the Vaal River, treating it to potable standard, and pumping it nearly 400 m
upwards to the service reservoirs, 65 km and further away at the local municipalities. The reliability of
its major pipelines is of obvious great importance to the national economy, and the asset
management regime of these pipelines therefore needs to be of a very high order.
Management of the major pipelines is dependent upon:
Reliable and comprehensive information.
A disciplined approach to operational and investment decisions.
That Rand Water is financially robust, giving it the freedom to treat infrastructure as a valuable longterm investment, and to make decisions based on optimising the service life-cycle of that
In respect of reliable and comprehensive information are, inter alia:
Good records of infrastructure performance that have been kept over the years.
Environmental factors that are recorded.
Coupled comprehensive condition assessment procedures, where necessary, using the most
sophisticated appropriate technology available.
Linkage to a common GIS database.
Being backed by a record of experience from elsewhere, with similar infrastructure in usefully
comparable conditions.
Concerning a disciplined approach to asset management decisions, for example, we advocate that
Comprehensive information is utilised to identify the problem probability profiles of all
significant infrastructure elements.
Strategic importance of these infrastructure elements is also identified, and the risks (i.e.
consequences of infrastructure failure) are matched with the problem probability profiles.
Type of attention be identified, in order to specify priority, namely, to consider the alternatives
and options to cover the identified risks).
Operational and investment decisions can then be based with confidence on this reliable and
comprehensive information, and a disciplined approach.
The principal best practice/innovation point is that:
That Rand Water pipeline asset management represents best practice water services
infrastructure asset management in South Africa, and that there is no substitute for the
underlying factors that enable Rand Water to manage its assets in this way, these factors
o Financial stability.



Top managements willingness to budget adequately for infrastructure asset


Sufficient technical capability to advise, to plan, and to implement, and on brief,

procure and supervise outsourced skills.

Rand Water background and importance to the economy

The National Water Services Act 108 of 1997 allows for the establishment of water boards to assist
the local government sphere to meet its constitutional obligation of effective, efficient, sustainable and
affordable services delivery. Rand Water, established in 1903, is by far the largest of these water
boards, and is the bulk water supplier to the industrial and commercial heartland of South Africa,
including the metropolitan areas of Johannesburg and Pretoria. Sixty percent of South Africa's GDP
is generated in this area.
It is of obvious and utmost importance to the economy of South Africa (and to the people living in its
service area) that Rand Water provides a service of the highest reliability and quality. It is essential
that provision of this service be without interruption.
Rand Water turnover is in excess of R 3 billion (equivalent to US $ 400 million) per annum, and the
substantial surplus which is generated, year after year, is either reinvested or committed to
accumulated reserves. As a result, Rand Water is financially secure and continuously enjoys the
highest credit rating in South Africa. Because of the way in which its infrastructure investment is (and
has in the past been) planned and funded, and because of increasing efficiencies in operation, tariff
increases (the water being sold in bulk to the municipalities that retail it) have, for some years, been
less than the escalation in the national consumer price index.
That Rand Water, being financially robust, gives it the capacity to treat infrastructure as a valuable
long-term investment, and is able to make decisions based on optimisation the life cycle of that
infrastructure. There is no substitute in respect of what it needs to do to ensure a service of the
highest reliability and quality, for Rand Water being financially stable, and for its top management to
be willing and able to budget adequately for infrastructure asset management. Nor is there a
substitute for Rand Water having access to sufficient technical capability to advise, to plan, and to
implement, and where appropriate, to brief, procure and supervise outsourced skills.

Supply area

The supply area covers 18 000 km , accommodating about 10 million people (Figure 1).
Raw water that is purchased from the national Department of Water Affairs and Forestry (DAWF) is
treated to potable standard, and is then supplied in bulk to 16 municipalities who distribute to their
consumers. There is also a direct supply to a number of other users, principally mines.
Rand Water abstracts an average of 3 500 M/d from the Vaal Dam, on the Vaal River to the south. It
is then processed by one of the largest water treatment works in the Southern Hemisphere. The
potable-grade water is then pumped nearly 400 m upwards to the service reservoirs of Rand Water
and other local municipalities, 65 km and further away. Water is treated at Vereeniging and
Zuikerbosch primary stations, and pumped at a head of approximately 190 m into the network.
Secondary booster stations then elevate about 90 % of this water a further 200 m to 53 strategicallylocated service reservoirs on the high ground of the Witwatersrand. Water is then gravity fed from
these reservoirs to the boundaries of the area of supply, with some additional localised booster
pumping. Extrapolation of the average rate of growth of demand of recent years (2.76% per annum)
indicates that average consumption in 2020 will be 5 200 M/d. However, because of the anticipated



effects of water demand management and of HIV/AIDS, an alternative prediction is that the average
in 2020 will have risen to only 4 100 M/d.

While the reliability of overall key infrastructure is clearly extremely important, the reliability of major
pipelines is crucial. The asset management regime of these pipelines therefore needs to be of a very
high order.

South Africa


Figure 1: Rand Water area of supply

Pipeline asset management regime: overview

Rand Water had the foresight nearly 20 years ago to commence a process that has led to a Pipeline
Asset Management Plan, the primary elements of which are described in the following case study.
For a brief overview of the early history of this planning process, see Annexure A. This case study
describes the asset management regime of the pipelines, which, as noted above, needs to be of a
very high order. Water that has been unaccounted for (leakage losses) is currently 3.5%. The target
is 2.0%.
The major transportation and distribution system comprises:
2 700 km of steel pipelines, with diameters ranging from 1 000 mm to 3 500mm.
180 km (about 6% of the total length) of pre-stressed concrete pipelines, with diameters from
1 000 mm to 2 100 mm.
Management of these major pipelines is characterised by:
Reliable and comprehensive information.
A disciplined approach to operational and investment decisions.



In respect of reliable and comprehensive information, principally:

Good records of infrastructure performance have been kept over the years (including when
built, repair records, refurbishment records, leakage records).
Environmental factors are recorded (including dolomitic areas, water and groundwater
chemistry, surface encroachments).
Coupling to comprehensive condition assessment procedures, where necessary, using the
most sophisticated appropriate technology available.
Linking all to a common GIS database.
Back-up by a database of experience elsewhere, with comparable infrastructure in usefully
similar conditions.
In respect to a disciplined approach to, for example, asset management decisions:
The comprehensive information is utilised to identify the problem probability profiles of all
significant infrastructure elements.
The strategic importance of these infrastructure elements is also identified, and the risks (i.e.
consequences of infrastructure failure) are matched with the problem probability profiles, In
order to identify priorities for attention, and what that kind of attention should be (e.g.
alternative ways to cover the risks identified).
In terms of this approach, Rand Water completed a priority ranking in 2005 of the pipelines that
needed renovation over the following next five years. This ranking is reviewed each year, in the light
of the work that has been completed, and in the light of accumulating knowledge of pipeline condition.
The purpose of this priority ranking is in order that timely financial provision can be made for whatever
needs to be done. Operation and investment decisions can be based with confidence on this reliable
and comprehensive information, and a disciplined approach.
The priority ranking is primarily based on:
The importance of the pipe in terms of Rand Water commitments to provide reliable supply.
This can be viewed as Rand Water's awareness of its commitment to absolute reliability in
terms of service delivery. But, reinforcing that, it can also be viewed as awareness of
business risk. Non-delivery has revenue implications, and would erode the financial stability of
Rand Water.
The type of the pipe.
The age of the pipe (see Figure 2) and the expected lifespan.
The performance record of the pipe.
Water quality criteria (e.g. if problems are being identified in the quality of the water being
carried in the pipe).
The window of opportunity (if any).
An example of how this influences implementation, Rand Water is currently augmenting a particular
pipeline with a new parallel pipeline, a few years in advance of what would otherwise be warranted,
so that the existing pipeline can be taken off service in order to renovate it. The older pipeline could
not have been shut down until the parallel pipeline was in operation, as it is critical for Rand Water to
maintain an uninterrupted water supply.



Figure 2: Age profile of Rand Water pipelines

Steel pipes have a life of up to 70 years. Pre-stressed concrete pipes have a life of 30-35 years, and
many are reaching this end of life. Fitting internal steel lining to pre-stressed pipes will extend their
lifespan by another 30-35 years.
Rand Water is currently well advanced with a survey of all of pre-stressed concrete pipes in the
network. Step 1 has been to aerial scan the pipelines, and take thermal images. This shows the
position of significant leaks. Step 2 has been to prioritise as to which pipelines should be temporarily
taken out of service for eddy current scanning, and to undertake sonar scanning to locate, inter alia,
where pre-stressed wires have broken. The scanning will be redone at regular intervals, to assess
the rate of degradation. Step 3 has been to prioritise the lining of pre-stressed pipelines. Step four
involves renovation work.
It will cost R 35 million for Steps 1 and 2 on pre-stressed concrete lines only, and will take 5 years.
Rand Water will later assess the steel pipelines, using the same techniques.
Rand Water's approach to asset management of its pre-stressed concrete pipelines is described in
more detail below.

Asset management of pre-stressed concrete pipelines

The majority of the pre-stressed concrete pipes were laid during the 1960s, and are reaching the end
of their design lives. Investigation, monitoring and renovation of the pre-stressed concrete pipes are of
a high priority.
A two-prong strategy of proactive and reactive asset management has been adopted. The aim is that
reactive (i.e. emergency repair and renovation) work done, as and when required, will be replaced
over a five year period by a proactive strategy, in terms of which planned shutdowns will be done,
based on asset management principles, with condition assessment being a key component.
Secondary objectives are:
Developing guidelines.
Determining roles and responsibilities.



Giving consultants and other stakeholders the opportunity to assist and learn from Rand
Water to obtain solutions, and vice versa.
Preparation and approval of a budget to suit the action plan covering an initial five-year
window period.
Implementation of a long-term monitoring procedure to determine trends.
Focus groups or fora be instituted with external stakeholders as a basis for sharing
information and experiences.
In brief, the concurrent steps in formulation of the strategy are:
Research, including field investigations, into in particular environmental factors contributing to
pipeline problems, methods of pipeline inspection and problem detection, and remedial
Raising awareness (principally at Board level).
Pilot implementation (which is the B4 pipeline -- see below).
Accumulating and analysing asset condition information, for example, seeking correlation with
pipeline serviceability record, and determining whether problems are once-off or recurrent.
The objective of the strategy is to improve risk management. Much of the investigation is therefore
into failure prevention. Many factors contribute to the development of failures in pre-stressed concrete
pipes. Failures range from small to catastrophic, from continuous leaks to sudden bursts. The
consequences may typically include interruption of service, repair costs, insurance claims, property
damage, confidence-loss in terms of serviceability, and queries as to fit for purpose.
Some elements of the research and field investigations include:
Condition assessment comparatives by external remote aerial thermal imaging, remote field
eddy current, acoustic emission, and other methods.
Destructive and non-destructive materials testing.
Investigation of correlation between soil resistivity and stray currents on the one hand, and
occurrence of broken pre-stressed wires and external corrosion to steel specials on the other.
Discussions with suppliers of alternative repair methods.
Hydraulic modelling, and water-hammer modelling.
Derivation of failure predictability.
Gathering and sharing of information, worldwide.
A pipeline that has been a key to assured supply, and having been laid in 1965, is at the end of its
design life. This is the Zuikerbosch to Palmiet pipeline, known as the B4 pipeline (Figure 3). A 10 km
section is a steel pipeline with a diameter of 1 670 mm, and another 35 km is in pre-stressed
concrete, with the diameter ranging from 1 600 mm to 2 100 mm.



Figure 3: Location of the B4 pipeline

Numerous leaking joints have been reported at various localities on this pipeline during its operating
lifetime. A burst occurred in 1975, but there were no further bursts until 1998. Further problems have
since occurred, in particular, a burst in 2002.
The ranking process gave renovation or replacement of this pipeline the highest priority.
The most expensive option was to replace this pipeline at a cost of R500 to R600 million. However,
the Board approved the recommendation from the officials that the pre-stressed concrete portion of
the pipeline be renovated at a cost of R130 to R150 million (20 million US $).
Renovation of the B4 pipeline was accordingly selected as a pilot project for renovation,, gathering
information, and furthering experience of the strategy through investigations during restoration work.
The first step in planning renovation was to review the available alternatives to maintain supply when
the pipeline was out of service for the renovation work. The importance of this pipeline was most
apparent at those times when it had to be shut down in order to allow emergency repair work to take
Renovation work on the B4 pipeline started in 2005 and is proceeding in phases, as the opportunity to
shut down portions of the pipeline permit.
The renovation work is now nearly complete. The option has been taken to test different alternatives,
for example, various methods of analysis, or methods of repair. Two different techniques have been
used to install the steel slip lining. Carbon fibre composite has been used for repair. Where broken
pre-stressed wires have been located, but pipe-lining not planned.
Combinations of factors have contributed to bursts and leaks. Further research and development
work is therefore still being conducted (or is proposed) in order to understand failure mechanisms
more comprehensively. This should make it possible to reduce the incidence of bursts and leaks, and
to address the consequent problems with better competence. Research is thus covering, inter alia:
External environmental factors, such as land use, including farming practices and the
(improper) use of fertilisers; also the effect of dolomitic strata, water and groundwater
chemistry, surface encroachments.



Latest technology in acoustic sounding, without having to drain a pipeline for installation and
operation of test equipment.
Alternative operating scenarios that can be implemented to reduce operating pressures and
water hammer pressures.
Finite element modelling, developing graphs to assist with risk assessment and proposed
remedial action, and in so doing, combining certain crucial parameters to determine
Soil and ground water testing, also linking the potential of cathodic protection measures at
isolated sections, or full sections along the pipeline.
Various renovation methods, such as internal steel slip lining, internal strengthening in
isolated places by means of carbon fibre or other material, external strengthening by means
of various methods.
Tests on material types currently used to perform internal and external sealing of joints,
assessment in terms of service life and effect of exposure to a variety of conditions.

It is clearly of the greatest importance to know the condition of each section of pipe, so that it is
possible to categorise the sections to line, to replace, or leave, as is, with confidence.
As Rand Water progressively records the cost of the remedial work that will be required, it will be able
each year to make tariff recommendations more accurately, ensuring that it will have the financial
resources at hand when it becomes necessary to do the work.
The principal best practice/innovation point is that:
Rand Water pipeline asset management represents the best practice water services
infrastructure asset management in South Africa, and that there is no substitute for the
underlying factors that enable Rand Water to manage its assets in this way, these factors
o Financial stability.

Top managements willingness to budget adequately for infrastructure asset


Sufficient technical capability to advise, to plan, to implement, and to brief, procure

and supervise outsourced skills.



Case Study: Pipeline asset management regime: history

It had become apparent as early as in 1990 to senior officials of Rand Water that the extent and
strategic importance of their major pipeline network required that a properly structured, long term,
planned pipeline renovation plan be drawn up and a programme for its implementation
commenced. An internal working group was set up the following year to this end, with the approval
of the Board. This group was representative at senior level of the disciplines involved in operating
and maintaining the pipeline network, controlling water quality in the system, and the design and
construction of new pipelines.
Work progressed steadily, and in due course, reports detailing the need to develop a Management
Information System (MIS) were submitted to the Board. Further reports detailed a longer-term need
for the development of two important basic elements of the Pipelines Renovation Programme
(PRP), namely: a Pipelines Asset Management Plan (PAMP) and a Strategic Renovation Plan
(SRP). These were followed by other reports setting out implementation proposals and cost
Major renovation work commenced in 1996, with an amount of R16 million spent during that
financial year. This has subsequently grown into an annual budget in excess of R150 million (US$
20 million).
A 5-year SRP was published during 1998. A 15-year SRP has subsequently been developed.
The PRP depends on accurate and up-to-date information for effective decision-making, greater
operational efficiencies then being made available. This it receives from the in-house maintained
MIS. The MIS, which is under constant development and improvement, comprises of a wellpopulated Geographic Information System (GIS), Access and/or Oracle databases, multiple links to
different Rand Water-wide databases, links to outside databases, and links to equipment such as
cellular (mobile) phones.



Rustenburg public-private partnership for upgrading, operation

maintenance of water and wastewater treatment infrastructure


The town of Rustenburg, in the North West Province, South Africa, obtained most of its water supply
from the municipality's water treatment works at Bospoort Dam, downstream from the town since the
1950s. As the town grew, it found additional sources of water. Two municipal wastewater treatment
works were in the intervening years built downstream from the town but upstream from the dam.
These installations were more than adequate in terms of treated effluent discharge of an acceptable
standard into the dam. Some of this treated effluent was sold to nearby mines, as they found it
adequate for purposes that did not require potable water.
In recent years, due to rapid growth of the area and of its water consumption, these wastewater
treatment works have become overloaded, leading to deterioration in the quality of the treated
effluent. When quality fell to unacceptable limits, the mines refused to receive any more of the poorlytreated effluent. As the quality deteriorated further, the Bospoort Dam became seriously polluted, and
highly eutrophic. The consequence, in turn, was that the potable water treatment works at the dam
was no longer able to cope with the high pollution load of its influent water, and the plant was closed.
As a result of all of this, Rustenburg Municipality was faced with:
Reduced revenue (due to the mines no longer purchasing treated wastewater).
Three unusable treatment works.
Increasingly polluted waters and dams downstream.
Forced increase in reliance on (more expensive) water supplies from sources elsewhere.
In 2003, a consortium consisting of Magalies Water (the parastatal water authority for the region north
of the town), an engineering consultancy and a bank won a bid to refurbish and upgrade the two
wastewater treatment works, refurbish and upgrade the potable water treatment works, and refurbish
a bulk water pipeline. The offer included funding (total project cost R 280 million (US$ 40 million)
including capitalised interest) and operation of the infrastructure.
The first of these treatment works was returned to service in 2006, and the second during 2007. The
municipality owns the assets, and as is usual, charges users of the water and sanitation services, but
the income, less the municipalitys administration costs, is paid to the Rustenburg Water Services
Trust, a municipal entity. This Trust, which was formed to develop the project, pays the operator
(Magalies Water), pays for the professional services, and repays the loan that was made to the Trust
by the bank. Thus, the financial benefits are shared by the municipality and the consortium.
The principal best practice/innovation point is that:
The private sector and parastatals can offer a municipality the resources (principally,
expertise and funding) that are otherwise unavailable to them, address infrastructure
problems that have financial, health and social or environmental consequences, and can do
this to the benefit of all parties.

Rustenburg, in the North West Province, is one of the fastest growing municipalities in South Africa.
Rapid population and economic growth experienced in the vicinity has been driven by the expansion



of platinum, chrome and allied mining activities, as well as the incorporation of a considerable portion
of the surrounding rural land into the municipal area during the 1990s.
The town obtained most of its water supply since the 1950s from the municipality's potable water
treatment works at Bospoort Dam, downstream from the town. Two municipal wastewater treatment
works were subsequently built downstream from the town but upstream from the dam. These were
more than adequate to discharge treated effluent of an acceptable standard into the dam. Some of
the effluent was sold to nearby mines, as they found the treated effluent to be adequate for purposes
that did not require potable water.
As the town grew, it received additional water supplies from further afield. Magalies Water, a
parastatal water authority for the region north of the town, is one supplier, and Rand Water, a
parastatal water authority for the Johannesburg and Pretoria area (to the south-east of Rustenburg,
and on a completely different catchment system, is the other.
The municipality did not increase its wastewater treatment capacity, after the second of the
wastewater treatment works was commissioned in the early 1990s.
By 1997, the increased influent load had overwhelmed the treatment capacity of the two wastewater
treatment works. The result was an increasingly unacceptable effluent being discharged into the
rivers which fed the Bospoort Dam, the water impoundment for the supply to the Bospoort potable
water treatment works. The works failed to treat the highly eutrophic water to potable standards, and
was consequently mothballed. The prevailing shortfall of about 20% of the municipalitys needs was
made up by Rand Water, but at a substantially escalated cost to the local municipality.
Rustenburg faced a four-fold set-back that was largely of its own making, because it had neglected to
address its key infrastructure needs. It faced:
Reduced revenue (due to the mines no longer purchasing treated effluent).
Three unusable treatment works.
Increasingly polluted waters and dams downstream.
Forced increased reliance on (more expensive) water supplies from external sources.
Troubles mounted as the national Department of Water Affairs and Forestry (DWAF) and the
provincial government of the North West province both threatened Rustenburg with penalties, should
it fail to upgrade its wastewater treatment works, and a restraint was placed by DWAF on further
residential development in the town. Additionally, DWAF warned Rustenburg to either use its
allocation from the Bospoort Dam, or forfeit it in future.

Addressing the issues

The poor state of the Rustenburg municipal balance sheet was not conducive to commercial banks
lending the necessary funds required for the capital expenditure that would address the water
services infrastructure problems. Moreover, in terms of South African legislation, municipal debt may
not be underwritten by central government.
Following extensive consultations, Rustenburg issued a call for proposals to address the pollution of
the Bospoort Dam and the river upstream of it through a municipal service partnership mechanism,
namely, a public-private partnership (PPP). The scope would include financing, design, construction,
operation and maintenance of refurbishments, augmentation and extensions to the three treatment
works, and other additional works.



Its motivation in calling for proposals was based on the premise that the most cost-effective source of
additional water would be from recommissioning the three municipal treatment works, after
refurbishment and upgrading. Upgrading of the two wastewater treatment works would improve the
quality of the treated effluent discharging to the Bospoort Dam. Improvement in the dam water quality
would on its own ease the load on the potable water treatment works. But the eutrophic state of the
dam water would nevertheless demand an upgrading of the potable water treatment works capacity,
to enable it to cope with what would still be the very eutrophic water from the dam, and satisfy the
demand criteria in Rustenburg at a consumption as high as 95 Ml/day.
Restoring these treatment works to service would reduce the probability of the town having a water
shortage during periods of peak demand. It would also enable sales of non-potable grade treated
water to the mines to be resumed, and reduce their need to draw more expensive water from Rand

The consortium
A consortium consisting of Magalies Water (the parastatal water authority for a region north of the
town), an engineering consultancy and a bank won a bid in 2004 to refurbish and upgrade the two
wastewater treatment works, refurbish and upgrade the water treatment works, and refurbish a bulk
water pipeline. The offer included funding (total project cost R 280 million (US$ 40 million) including
capitalised interest) and operating the infrastructure in the short-to-medium term.
In terms of the contract, the municipality owns the assets, and charges users of the water and
sanitation services in the usual manner, but the income, less the municipalitys administration costs, is
paid to the Rustenburg Water Services Trust, a municipal entity. This Trust, which was formed to
develop the project, pays the operator (Magalies Water), pays for the professional services, and
repays the bank the loan that was provided to the Trust by the bank. Thus, the financial benefits are
shared by the municipality and the consortium.

The engineering works

The capital works have consisted of:
Refurbishment and extension of the larger wastewater treatment works. This will increase the
treatment capacity from 22 M/d to 42 M/d. Increasing treatment capacity from 22 M/d to 42
M/d by building a new 15 M/d activated sludge unit designed for full biological nutrient
removal, and by upgrading the existing 10 M/d activated sludge unit to treat 15 M/d, also
with full biological nutrient removal. The existing 12 M/d biological filter works is being
refurbished to full capacity. This works has a tertiary treatment facility that chemically
removes phosphates.
Refurbishment of the smaller (8 M/d) wastewater treatment works, with no increase in
capacity. Operational procedures will also be improved.
The upgrade of the Bospoort potable water treatment works involves complete mechanical
and electrical refurbishment, and the addition of improved unit treatment processes, to assist
with high influent pollution levels.
Refurbishment of the pipeline from the water treatment works to the town.
Construction began in 2005. The water treatment works is fully operational, and all of the previous
capacity of the wastewater treatment systems has been restored. The extended wastewater
treatment capacity will be operational by the end of 2007. The mines are once again accepting treated
effluent from the wastewater works.



Rustenburg Water Services Trust

The Rustenburg municipality is retaining ownership of the infrastructure, and remains responsible for
service tariffs to end-users and for controlling quality.
The Mati ya Vanhu (Water for the people in Venda language) consortium was appointed as the
preferred bidder, to form the municipal service vehicle, and to execute the project.
The implementation approach proposed by the consortium, and accepted by the municipality, is
unique in South Africa. The consortium proposed the formation of a business unit, the Rustenburg
Water Services Trust (the Trust), as the special purpose vehicle to implement the project. The
project is fully ring-fenced through the Trust and the rights and obligations of all parties are clearly
spelt out in a number of agreements. The legal status of the Trust is that of a municipal entity under
the Municipal Systems Act. Due to legal changes made by National Government, existing trusts may
continue, but in future, the special purpose vehicle will need to be a company. The main disadvantage
of this change will be that it will have to pay tax on earnings before dividend distribution to the parent
local authority. The rest of the structuring can remain unchanged whether it be a trust, a CC, a
Section 21 company, or a conventional company
The Trust obtained funding for the project from a member of the consortium, one of the largest South
African banks. Debt will be serviced by:
The sale of treated effluent to the mines for use as process water.
The sale of potable water from Bospoort water treatment works to the municipality at
prevailing Rand Water tariffs.
Charging the municipality for wastewater treatment.
The combined income stream is projected to be more than sufficient to service all of the Trusts debt,
as well as its operational and maintenance requirements. The income stream will, it is projected, be
sufficient to service capital redemption, pay for operation and maintenance by Magalies Water, and
generate a profit with a net current value of approximately R 150 million (US$ 20 million) over a 15year loan period.
The special purpose vehicle, in trust form, was selected for the following reasons:
A trust has no assets that can be attached (ownership of the infrastructure lies elsewhere, in
this case with the municipality).
A trust is able to appoint a turnkey contractor and pay for professional services.
Once the capital works are complete, they are handed over to Magalies Water, to operate on
behalf of the owners.
The municipality charges its customers, and collects monies for the water supplied to it by the
The income that the municipality generates, less the municipal administration charges, is paid
to the Trust, which in turn pays the ongoing operation and ancillary costs while repaying the
capital and interest to the bank.
An aspect of the arrangement is that all parties involved are apportioned a degree of risk,
which they undertake to accept for the period of loan (18 years).
The majority of trustees are nominated by the municipality, and thus the Trust is locally



Being the only beneficiary of the Trust, the municipality receives all profits therefrom, and can
use these to fund further water services development.
The only financial liability of the municipality is to buy the water from the Trust at Rand Water tariffs,
and to pay the agreed tariff for wastewater treatment. No capital comes from the municipality, and it
does not carry any loan against its balance sheet.
The consortium member, Magalies Water, is responsible for all operation and maintenance, inclusive
of taking over all of the municipal staff directly employed on the works, or employed in the existing
municipal water laboratory.
Nearly all of the treated effluent from the refurbished wastewater treatment works has since
December 2006 been supplying the mines needs for industrial quality water.
Demand and pricing for this lower-quality water for industrial purposes was determined through the
tender and negotiation process with the three platinum mining groups in the Rustenburg area. The
mines required a guaranteed quality and quantity. This is not a cross-subsidy, but a benefit to them,
as they obtain this industrial grade water at a significantly lower cost than they would otherwise have
to pay Rand Water for unnecessarily high-quality potable grade water. Negotiations have set the tariff
for industrial grade water at 48% of the domestic tariff charged by the municipality.

The innovative approach that has been described has turned neglected and underperforming facilities
into valuable assets that are already making a significant contribution towards addressing the water
quality and quantity issues experienced by Rustenburg for more than a decade. Responsible use of
both local financial resources and local water resources is being demonstrated.
One of the advantages of this approach is the fact that the bank will insist on best practice operation
and maintenance, to ensure that the project retains its sustainability and their investment is protected.
The community enjoys better service delivery, and the mines benefit from a reliable and cost-effective
supply of industrial water. Waters in the local catchment are also being restored to acceptable quality.
An integrated solution ensures that the water treatment works are not upgraded in isolation from the
issue of the pollution sources in the catchment. This approach would have been futile. The project is
an example of how engineering, progressive municipal management and sound investment policies
can work together to solve water services problems that commonly confront local authorities.
The principal best practice/innovation point is that:
The private sector and parastatals can address infrastructure problems that have financial,
health and social or environmental consequences by bringing to a municipality the resources
(principally, expertise and funding) that municipalities otherwise find impossible to secure.
Moreover they can do this to the benefit of all parties.



Western Cape Provincial Government collaborative programme. - Asset

management planning support for local municipalities
The Western Cape Provincial Government is collaborating with municipalities in order to improve the
management of municipal infrastructure (including water services infrastructure) in the province. The
essence of this programme is the adoption and implementation of asset management principles in a
consistent and consultative manner by all municipalities in the province. The Western Cape
Department of Local Government and Housing has spearheaded the programme through information
sharing, funding pilot programmes, funding the training of municipal officials, and providing a forum for
feedback and interaction between municipal officials.
The case study describes the leadership role of the provincial government, and also two initiatives of
particular value, namely:
Preparation of a standardised framework for compilation of asset registers.
Selection of pilot projects to test asset management initiatives before implementing them in all
Initial pilot projects for the compilation of infrastructure asset registers at Drakenstein and
Stellenbosch municipalities have been followed up with a pilot project, still under way, to prepare an
asset management plan for water services at the Drakenstein Municipality. The intention of the
second pilot project is for the municipality, assisted by the Provincial Government, to develop an
analysis framework and methodology that will address the municipalitys asset management
challenges. This analysis framework can then be reviewed by the other municipalities, and refined to
provide a basic structure of analysis and reporting.
The principal best practice/innovation point is that:
The provincial government of the Western Cape is showing how a provincial or regional
government can considerably assist municipalities in making their infrastructure asset
management processes more effective and efficient.
o For example, the provincial government has brought into being and has tested a
basic analysis framework and reporting structure, thereby reducing the uncertainty of
both the inputs and the outputs of asset management planning studies. In turn this is
clarifying deliverables, and reducing cost and improving the quality of asset
management studies. It is also providing comparable information, municipality with
municipality, thereby assisting municipalities and the provincial government to
determine priorities.

South Africa has three spheres of government, namely national, provincial (nine provinces) and local.
The local authorities, these are the municipalities, are in terms of national legislation the designated
water services authorities. All municipalities are of course responsible for infrastructure service
delivery in respect of a wide range of services, of which water services is a part.
As the experience described in this paper evidences, the provincial government can play a strong
leadership role in persuading municipalities to improve their infrastructure asset management, and
then facilitating their improvement.
The Provincial Department of Local Government and Housing (PDLGH) of the Western Cape
provincial government have over the last few years provided strong infrastructure management



leadership and support to municipalities in the Western Cape. This has been done through means
such as information-sharing, the funding of pilot projects, the development of technical resources, the
training of municipal officials, and the provision of a forum for feedback and interaction between
municipal officials. In the process it has addressed several key issues, including quantification of
infrastructure backlogs, master planning to address future demand, and management of infrastructure
through asset management planning. These areas of infrastructure management do partially overlap
in data needs and analytical methods, but each requires a specialised approach to address the
infrastructure management needs of municipalities adequately.

Initiation of the collaborative approach

The PDLGH had long realised that there was a need for infrastructure asset management to be
practiced more effectively by municipalities in the province. It was understood that methods to
provide support and an enabling environment for municipalities to engage in infrastructure asset
management practices would need to be both collaborative and incremental.
Implementation of the vision of infrastructure asset management for the province began with the
raising of awareness amongst the municipalities. Monthly technical meetings, convened by the
PDLGH, provided the forum for the awareness-raising. Critical at that stage was the development of a
commitment and support for infrastructure asset management as a technical discipline that would
provide value to municipalities and assist the technical personnel in the management of their
The constraint on implementing asset management at the individual municipalities was the lack of
experience and technical guidelines on methods. While it was as a starting point, accepted that the
South African edition of the International Infrastructure Management Manual (IIMM) (South Africa
edition. Association of Local Government Engineering NZ Inc., 2006) is descriptive of best practice,
and outlines the principles of asset management, and is a good reference, it lacks the clear analysis
methodology and structure needed for individual implementation at any one municipality. The PDLGH
funded an initial study for that reason, to clarify the principles contained in the IIMM, and to structure
them into a more rigid analysis framework that was applicable and attainable at a municipal level.
The inputs into the analysis framework were workshopped and discussed at the monthly technical
meetings, to reach consensus, and to develop a common understanding of the guidelines. The result
was a guideline document Monitoring the Condition of Municipal Infrastructure Assets in the Western
Cape Province (Rev4), Provincial Government for the Western Cape, March 2006. This provides the
rationale for monitoring the condition of municipal infrastructure assets.
The guidelines include:
A standardised asset hierarchy. As an example of material in the guidelines, Annexure A
describes the standardized asset hierarchy in the context of water services.
Standard asset register data requirements.
A clearly defined condition grading method and scale.
A risk assessment methodology.
Asset valuation unit rates.
Life-cycle budget estimation models.
These guidelines form the basis of further knowledge development.



The collaborative approach continues, both in learning terms, by applying further asset management
planning developments in pilot areas, and by implementation of the already piloted techniques in all
the other municipalities.
The next two sections of this case study describe the two principal pilot processes thus far
Compilation of asset registers.
Preparation of an asset management plan (for water services).

Data collection pilot studies

A pilot study was initiated by the PDLGH for the compilation of infrastructure asset registers to test the
principles and analysis methods contained in the document. Two municipalities, Drakenstein and
Stellenbosch (both in the wine-lands area of the province) were selected for the pilot study. The pilot
study yielded mixed results, with several lessons learnt to the benefit of other municipalities.
The main lesson was the need for clear data collection guidelines, documented data rules, and
customised field assessment forms for each asset type. These were not fully developed at the time of
the pilot studies, and although the data collectors were briefed in the data requirements, the
complexity of data collection was too great. It adds to the complexity of the data collection that each
municipality owns a diverse range of assets, including housing, community buildings, roads,
reticulated systems and treatment works - each service having its own data requirements. The water
supply service, which is one of seven services, has twenty asset types, each with its own
requirements. The net result was data of poor quality being fed into valuation and budgeting models.
Examples were entering the holding volume of a dam into a valuation model based on the dam wall
volume, or entering a propertys land area instead of the floor area of the building located on it.
The condition of data collected in the pilot studies was found to predominantly group around the fair
rating on a five-point condition scale. This was due to:
Inexperienced data collectors who are reluctant to rate at the extremes of the continuum of
Inadequate data resolution by not breaking down the hierarchy to a great enough level of
This bias of data collectors can be eliminated through clear definitions, training and experience, but
the data resolution problem can only be resolved by striking a good balance between sufficient detail
and the cost of data collection.
Another lesson learnt was the limitations of using existing data for use in compiling asset registers.
The compilation of an asset register requires several typical data types, including:
Identification (GPS position, naming, numbering, location description).
Description (attributes that describe the asset).
Quantification (data that quantify the size of the asset).
Condition (data that describe the condition and performance of the asset).
Valuation (data needed to value the asset).
Risk (data used for prioritisation).



Utilisation (data describing the usage and capacity of the asset).

Survey data (data that give the survey date, assessor, accuracy, etc.).
The data in many cases are purportedly readily available for direct use in compiling an asset register,
only to then find that the data are only partially available, or that data are presented in a format that
cannot be readily used, such as CAD drawings and text-based reports. Existing GIS data typically
only identify location, CAD drawings typically provide description and quantification data, and planning
reports may provide capacity and utilisation data. The existing data need to be supplemented with
condition assessments in the field in all these cases. For invisible assets such as buried pipes, that
cannot practically be inspected, reliance has perforce to be on such data as is available, and on the
experience of maintenance personnel.
The perception is often misguided that the assistance of municipal staff that are familiar with the
assets would reduce the need for condition assessments. Although the officials responsible for
operation may have useful insight into the performance of the assets, they may not have insight into
all the components at a facility, may not know the data requirements, and are unfamiliar with the
condition grading scale. Nonetheless experienced municipal staff can be invaluable for condition
assessments for two reasons. They assist greatly with finding assets in the field, especially when
limited information is available on the location of infrastructure, and their knowledge of the
performance history of many assets assists the assessor to rate the condition of the assets.
Based on the experiences with the pilot studies, in order to collect good quality data that will be
suitable to support a condition analysis, it is important to:
Develop comprehensive field guides with clearly specified data requirements for each asset
Compile data collection forms for each asset type.
Extend the asset hierarchy to a sufficient level of detail.
Make financial provision for the physical inspection of all visible assets.
Use experienced data collectors.
Have municipal staff available to support the condition assessments.
The pilot studies served the purpose of refining the data requirements, improving the data collection
methodology, and developing the understanding of asset registers. Following the development of
guidelines and data specifications for asset registers and the pilot studies, the PDLGH made asset
register software licences available for municipalities who undertook to compile their own asset
registers. The software provides the means to manage and store the data and provides standardised
reports to support asset management planning.

Asset management planning pilot study

The PDLGH supported the pilot study for the compilation of an asset management plan in a similar
approach to that which it adopted for the development of an analysis guideline and the execution of
the asset register pilot studies,. The Drakenstein Municipality was selected for the pilot study, with the
project focused on the water services. This asset management plan will (when it is complete, before
the end of 2007) serve as a template for other municipalities from which they can evaluate the data
inputs required to compile the plan and evaluate how the outputs will be of use to them in the
management of their infrastructure.



An area that needs more clarity is the relationship between the asset management plan and the other
master plans that municipalities are by national statute obliged to prepare, such as a Water Services
Development Plan.
The methodology and lessons learned from the compilation of Drakensteins asset management plan
will be conveyed to the other municipalities, so that they can improve their outputs and simplify their

Benefits of a collaborative approach

The benefits of asset management are widely touted, but the lack of experience in the practice across
South Africa makes the implementation of infrastructure asset management in any single municipality
a daunting task. The IIMM provides the principles, a broad methodology and an implementation
framework, but this is too broad for small municipalities to go directly to implementation, without the
development of the supporting technical resources and a clear analysis methodology. Given that
there are few consultants with asset management experience, and municipal resources are currently
stretched, the effective implementation of infrastructure asset management at medium and small
municipalities is unlikely without support.
Collaborative development of technical resources and shared knowledge reduces the barriers to
infrastructure asset management implementation. The existence of standardised data requirements
and a robust analysis framework reduces the uncertainty of the inputs, analysis method, and the
outputs of infrastructure asset management. This in turn clarifies deliverables, reduces cost, and
improves the quality of infrastructure asset management studies. Knowledge-sharing and
collaboration leads to accelerated incremental learning and the subsequent efficiency gains for
participating municipalities. Technical resources such as field guides, data specifications, valuation
models, and maintenance estimation models are very much in the growth stage, and require further
research to bring them to maturity.
There are substantial benefits of standardisation that extend to the provincial and national spheres of
government. Standardisation firstly enables the scaling-up of infrastructure quantitative and condition
data to form a picture of asset management across the whole province, and secondly, allows direct
comparison between municipalities. The collection of data over time will show trends in infrastructure
condition and responses to investment and management. These facilitate benchmarking and
performance monitoring at the provincial level, enabling the Western Cape Provincial Government to
fulfil its constitutional mandate of monitoring and supporting municipalities more effectively.
Province-wide information, in turn, informs national infrastructure initiatives, support programmes, and
policy. Whereas at national level there has been a strong emphasis on the quantification of the need
across the country for new infrastructure, limited attention has been given to quantification of the
ongoing life-cycle needs. A fundamental principle of infrastructure asset management is that assets
require ongoing reinvestment beyond creation to maintain them in serviceable condition. Failure to
make provision for this reinvestment will lead to premature infrastructure failure and the decline of
services to the communities they support. It is therefore imperative that municipal data be converted
into provincial and national information to support decision-making at these levels.

Municipalities in the Western Cape are being supported by the provincial government in their efforts to
improve their infrastructure asset management practices. The support has been ongoing for the past
three years and is focused on the collaborative development of shared technical knowledge and the
development of a common understanding of infrastructure asset management. The intention of the
provincial governments support programme is to improve the technical resources available to the
municipalities, and to reduce the uncertainly and cost of implementing infrastructure asset



management. The support programme has emphasised the development of a standardised approach,
to enable benchmarking and comparison between municipalities. The intention is to systematically
improve the information on infrastructure condition and life-cycle funding needs across the province,
so that the provincial government can be more responsive and supportive to the needs of the
municipalities. Such information will also provide valuable input into national policy formulation.
The principal best practice/innovation point of this case study is that:
The provincial government of the Western Cape is showing how a provincial or regional
government can considerably assist municipalities to make their infrastructure asset
management processes more effective and efficient.
o An example is that the provincial government has brought into being and has tested a
basic analysis framework and reporting structure, the existence of which is reducing
the uncertainty of both the inputs and the outputs of asset management planning
studies. In turn, this is clarifying deliverables, and reducing cost and improving the
quality of asset management studies. It is also providing comparative information,
municipality to municipality, thereby assisting municipalities and the provincial
government to determine priorities.

Annexure A: Standardised asset hierarchy

Abstracted from Chapter 2 of Monitoring the Condition of Municipal Infrastructure Assets in the
Western Cape Province (Rev4). Provincial Government for the Western Cape. March 2006.
The aim has been to establish a clear, holistic and logical breakdown of the principal infrastructure
asset sectors that may typically be found in South African municipalities. Thus the standardised asset
hierarchy was framed to accommodate the following needs:
To enable simple life-cycle cost modelling.
To recognise the different risk profiles of different assets.
To record a minimum level of detail, that is nevertheless sufficient for PDLGH to monitor
trends in infrastructure condition.
To provide a minimum level of detail that is accurate enough to provide a foundation for
municipalities to make informed strategic and tactical decisions, and to promote the
establishment of maintenance and rehabilitation programmes to control and safeguard their
infrastructure assets (Note: it is not the intention to provide asset information at a level
necessary for day-to-day operations or maintenance).
To complement existing software in general use by some municipalities.
To cater for nationally legislated financial requirements (e.g. isolating key asset groups with
different expected useful lives).
Table 1 provides an overview of the advantages and disadvantages of adopting different levels of
detail of asset data.



Level of



High level

80/20 rule
Quick results
Not resource-intensive
Data collection is needs-driven

Completeness/quality of data?
Risk of some inappropriate decisions
if overdue reliance on inaccurate data
Cannot do advanced analysis


Accurate, comprehensive data

Highly reliable outputs
Can do advanced modelling and
economic evaluation

Resource intensive, expensive

Time consuming
Risk of overdue focus on data, lose
focus on key objectives

Figure 1: Level of detail Assets

International experience in infrastructure management in recent years suggests that municipalities do

not need over-detailed data in order to make the necessary informed decisions. Furthermore, the
benefits of the provincial monitoring and municipal infrastructure asset management planning
exercises are not likely to be measured so much in terms of efficiency gains in day-to-day operations,
but rather in challenging the prevailing attitudes of decision-makers and impacting on strategic and
tactical planning of municipalities. Consequently, while bearing in mind the capacity constraints of
many of the municipalities, the standardised asset hierarchy cuts off at the level of asset indicated in
Figure 1.





Table 1: Level of detail of asset data

Some municipalities may have proprietary asset management systems or maintenance management
systems that adopt more detail, including component (or even sub-component) level. In such cases,
these data can be rolled up to asset level for reporting to the province. Whilst these more
capacitated municipalities may already have processes in place that support strategic and tactical
decision-making, the preparation of data in a consistent fashion, albeit only to the level of data
required by the standardised asset hierarchy, will facilitate benchmarking across all municipalities in
the Province.



Table 2 is an example of the standardised asset hierarchy, the selection being sanitation (abbreviated
from the original).





(for info only)

Sanitation Collection

Sewerage reticulation




pipelines Land (servitude)
(outfall sewer)
Rising mains

Bulk meters
Control valves

Gravity mains (outfall)


Pump station

Civil works and pipe-work

Thrust blocks

Small buildings

including internal and
external, plumbing, etc.

Perimeter security




Mechanical plant

Control valves
Water meters

Electrical plant

Electrical reticulation
Control equipment and


treatment Land
Civil works and pipe-work

Intake works



Sedimentation tanks
Oxidation ponds
Small buildings

including internal and
external, plumbing, etc.

Perimeter security


Site works


Mechanical plant

Control valves
Water meters

Electrical plant

Electrical reticulation
Control equipment and

Table 2: Standardised asset hierarchy: Sanitation

A potential alternative approach to complex facilities, such as treatment works, is to include each
process type (e.g. filtration, stabilisation, etc.), and then break each of these down into civil, electrical,
electronic, and mechanical assets. This clearly represents a substantial increase in the level of detail
of the exercise, and consequently the effort required. Whilst breaking a treatment facility into 30 to 50
assets (as opposed to, say, 7) does provide a greater extent of information, the reliability and benefit
of this in terms of strategic and tactical decision-making needs to be considered. Thus more detail
should only be considered for large works, say over 10M/d.




Guidelines, software, publications and papers

American Public Works Association (APWA)

APWAs publications found in the bookstore webpage, include manuals, books, CD-ROMs,
and others, cover many topics in public works such as Facilities and Grounds, Leadership,
Management, and Administration, Water Resources, and Construction and Municipal
Engineering. Some of selected APWAs publications in Asset Management are:
o Infrastructure Asset Management Manual.
o GASB 34 Statement & Implementation Guide.
o Managing Infrastructure Assets - Knowledge Product.
o The Facility Management Handbook, 2nd Edition.
The National Association of Clean Water Agencies (NACWA)
Managing Public Infrastructure Assets to Minimize Cost and Maximize Performance (Asset
Management Handbook) published by the Association of Metropolitan Sewerage
Agencies (AMSA) is a useful handbook in water and wastewater.
The Water Infrastructure Network (WIN)
Two reports by WIN discussing water and wastewater infrastructure financing programs:
Clean & Safe Water for the 21st Century and Water Infrastructure NOW: Recommendation
for clean and safe Water in the 21st Century For more information please visit WINs Web
site at win-water.org.
International Perspective on Infrastructure Asset Management - American Water
In a number of countries around the world, asset management programs have become a vital
part of utilities; among these Australia, New Zealand, United Kingdom, and South Africa are
the most experienced and successful.
International Infrastructure Management Manual - American Water
The Institute of Public Works Engineering Australia (formerly the Institute of Municipal
Engineers Australia-IMEA) and the New Zealand National Asset Management Steering
(NAMS) Group have jointly developed an International Infrastructure Management Manual [1].
This manual is divided into five sections: introduction, implementing asset management,
implementing techniques, asset management information systems, and country specific
information. The manual defines asset management; introduces the concepts of total asset
management and life cycle asset management; shows the benefits provided by asset
management techniques; outlines best practices; provides principles and processes of asset
management implementation guidelines; shows how to evaluate and implement information
systems to support good asset management planning and decision making; cites case
studies, and includes an implementation plan.



Australian National Audit Office (ANAO)

The ANAOs Asset Management Handbook helps asset managers implement asset
management principles. The Asset Management Handbook provides an overview of asset
management principles, instructions on the application of asset management principles and
concepts, a selection of case studies, and a glossary of terms.
National Guide to Sustainable Municipal Infrastructure
The National Guide to Sustainable Municipal Infrastructure is funded under the Infrastructure
Canada Program (ICP) and managed by the Federation of Canadian Municipalities (FCM) in
partnership with the National Research Council of Canada (NRC). The Guide provides a road
map to the best available solutions (Best Practices) for addressing municipal infrastructure
Municipal Infrastructure Investment Planning (MIIP)
The Municipal Infrastructure Investment Planning [45] project is a research initiative of the
National Research Council Canada and a number of municipalities. Its objectives are:
o Evaluating existing tools and techniques to assist municipal infrastructure investment
planning, and.



Developing prototype tools and techniques for asset managers to better manage their
municipal infrastructure.
Specific deliverables of the MIIP project include: evaluations of Information Technology (IT)
tools, surveys of the use of asset management in practice, case studies of asset
management, and guidelines and manuals on asset management

Water Environment Federation (WEF)

Managing the Water and Wastewater Utility is one of the best books published by WEF in
asset management.
Please visit www.wef.org.
Water Environment Research foundation (WERF)
Sustainable Infrastructure Management Program Learning Environment (SIMPLE):
This web-based knowledge management tool assists wastewater treatment facilities in
developing a proper life-cycle asset management system. By gathering asset management
best practices and processes from around the world, SIMPLE offers users a look at state-ofthe-art asset management programs and provides the tools they need to begin a program of
their own. For more information about WERFs activities and products please visit the Web
site at: www.wef.org.
Asset Management: A Handbook for Small Water Systems (EPA 816-R-03-016,
September 2003)
EPA has developed a Simple Tools for Effective Performance (STEP) Guide that
emphasizes how effective asset management is a key element of small system sustainability.
Various sample worksheets are provided to help small systems organize data and determine
the best approach to maintenance and replacement of major physical assets. An electronic
copy of thie document can be found at
Taking Stock of Your Water System: A Simple Asset Inventory Guide for Very Small
Drinking Water Systems (EPA 816-K-03-002, October 2004)
EPA has developed a STEP Guide to assist very small systems in conducting a simple
inventory of infrastructure for capital planning purposes. This STEP Guide can help these
types of water systems run properly and ensure that the drinking water they produce is
reliable, safe and affordable. An electronic copy of the document can be found at
Strategic Planning: A Handbook for Small Water Systems (EPA 816-R03-015,
September 2003)
EPA has developed a STEP Guide to assist small systems in strategic planning. The guide
provides worksheets and related tools to help systems organize data and systematically
assess their strengths, weaknesses, challenges, and opportunities. This guide is based on
the strategic planning workshops held around the country in 2000. An electronic copy can be
found at www.epa.gov/safewater/smallsys/pdfs/guide_smallsystems_stratplan.pdf.
Sources of Technical and Financial Assistance for Small Drinking Water Systems (EPA
816-K-02-005, July 2002)
EPA has developed a guide that identifies major sources of technical and financial assistance
specifically targeted at small drinking water systems. Each listing describes the sources
mission and types of assistance that can be provided, and lists contact information. An
electronic version available at : www.epa.gov/safewater/smallsys/pdfs/tfa_sdws.pdf.



TEAMS (Total Electronic Asset Management System) Asset Management Software for
Small Utilities
Developed by the Maryland Center for Environmental Training (MCET), this software is
targeted for small wastewater utilities and is accompanied by a training tool kit which includes
training modules on a range of asset management topics. The software can be obtained by
U.S. EPA Advanced Asset Management Training Workshops
The Office of Water is collaborating with partner organizations, hosts, and co-sponsors to
provide training on best practice in Advanced Asset Management. The workshops are
primarily designed to meet the Advanced Asset Management training needs of water and
wastewater utility CEOs, and senior level personnel. For more information and a list of
upcoming sessions, go to www.epa.gov/owm/assetmanage/index.htm%20and click on
Training Workshops.
WERFs Sustainable Infrastructure Management Program Learning Environment
EPA has collaborated on the development of an intuitive and interactive Web-based asset
management strategy tool, SIMPLE, which has been developed under the aegis of a Water
Environment Research Foundation (WERF) research project (03-CTS-14). SIMPLE contains
a set of user-friendly online processes and practice guidelines, templates, and decision
support tools that utilities and wastewater industry professionals can apply to asset
management. For more information, visit www.werf.us/and click on interactive tools.
NACWAs Managing Public Infrastructure Assets to Minimize Cost and Maximize
This handbook, funded by EPA, establishes an understanding of asset management
principles and program benefits and assists public water and wastewater utilities with the
development of asset management programs. To obtain a copy, visit www.amsacleanwater.org/pubs/index.cfm.
IPWEAs International Infrastructure Management Manual (2006 Edition)
Published by the Institute of Public Works Engineering Australia, the 2006 edition of the
International Infrastructure Management Manual is the premier handbook on asset
management practices and provides a detailed road map for preparing an asset management
plan. The manual contains extensive information on benchmarking, condition grading,
valuations, asset hierarchy structures, and information systems. It presents simple economic
evaluation tools and other techniques for project decision-making and prioritization. To obtain
a copy of the manual, visit www.ipwea.org.au/news/169.html.

Other related publications

Institute of Asset Management, Publicly Available Specification PAS 55 Asset Management,
BSI Publications (2004) [P].
Korving H, Sewer System Management: Decision-making under uncertainty, PhD Thesis,TU
Delft Sanitary Engineering (2004) http://www.gezondheidstechniek.tudelft.nl/korving.htm [C].
European 5th Framework funded project Computer Aided Rehabilitation of Sewer Networks,
CARE-S (http://care-s.unife.it/)[O].
UKWIR Deterioration rates of Sewers, completes Dec 2005 [O].



5. Beenen A.S., SPIRIT: Towards and interactive Tool, for Predictive Asset Management of
Sewer Systems - 10th International Conference on Urban Drainage, Copenhagen (2005) [O].
WERF report 97-CTS-7, Development of Predictive Tools for Measuring Sewer Degradation
(2004) [C].
European 5th Framework funded project Computer Aided Rehabilitation of Water Networks,
CARE-W (ref: http://care-w.unife.it/) [O].
UKWIR report 03/RG/05/7 - Nationally Agreed Failure Data and Analysis Methodology for
Water Mains: Volume I - Overview and Findings (2003) [C].
UKWIR project RG05/B Water Mains Serviceability Indicators and Condition Grading
(completes in 2006) [O].
AWWARF Project 3052, Dynamic Influences on the Deterioration Rates of Individual Water
Mains (completes in 2008) [O].
AwwaRF project 2927, Installation, Condition Assessment, and Reliability of Service Lines,
Connections and Fittings (completes in 2006) [O].
AwwaRF project 2883, Risk Management of Large Water Transmission Mains [O].
WERF project 03-cts-20 CO and AWWARF (3048) Protocols for assessing the condition and
performance of water and wastewater assets [C].
WERF project 97-cts-7, Development of a tool to prioritise sewer inspection (2004) [C].
WERF project 01-cts-7, An examination of innovative methods used in the inspection of
wastewater collection systems (2004) [C].
WERF project 99-cts-3ET, Ultrasonic-based defect characterisation in wastewater concrete
pipelines using invariance transformation techniques [C].
AwwaRF project 2612, Techniques for Monitoring Structural Behavior of Pipeline Systems,
(2004) [C].
AwwaRF project 2564, Electromagnetic Inspection of Pre-stressed Concrete Pressure Pipe,
(2001) [C].
UKWIR 22 Ref No. 05/TT/02/1
AwwaRF project 2871, Workshop for Condition Assessment for Water Mains, (2004) [C].
AwwaRF project 2772, Assessment and Renewal of Water Distribution Systems (2004) [C].
AwwaRF project 2953, Implementing Reliability Centered Maintenance (RCM) Programs at
Water Utilities (draft report due 2005) [O].
UKWIR project - Implementing the Common Framework: Meeting Asset Data Requirements
for PR09, 04/RG/05/12 [C] [O].
UKWIR project 02/RG/05/3 Capital Maintenance Planning: A Common Framework (2002) [C].



WERF project 03-cts-14 Asset management strategic planning and implementation guidelines
for wastewater infrastructure [O].
AwwaRF project 2939, Risk Analysis Strategies for More Credible and Defensible Decision
Making (completes 2006) [O].
A Risk Management Standard, Published by IRM, AIRMIC and ALARM (2002) [C]
Risk management Vocabulary Guidelines for use in standards, ISO/IEC Guide 73
WSAA report, Asset Management Performance Benchmarking Programme Project Industry
Report, (2005) [C].
UKWIR project 00/RG/04, International Benchmarking of Water Industry Costs and
Performance (2000) [C].
AwwaRF project 2848, Asset Management Planning and Reporting Options for Water Utilities
AwwaRF project 2745, Development of a Strategic Planning Process [C].
AwwaRF project 3125, A Guide to Triple Bottom Line Decision-Making: Social, Economic,
and Environmental Sustainability [A].
AwwaRF project 2849 Strategic Planning and Organizational Development for Water Utilities
(2004) [C].
UKWIR project Economic Level of Leakage Methodology (1997) [C] restricted circulation.
Report to the Tripartite Group of the Department for Environment, Food and Rural Affairs, the
Environment Agency and Ofwat: Future approaches to leakage target setting for water
companies in England and Wales: Best practice principles in the economic level of leakage
calculation. (2002) [C]
UKWIR 23 Ref No. 05/TT/02/1
WRc report Ref: UC3893, Key principles in the Economic Level of Leakage Calculation,
January 2001 [C].
WRc report Ref: UC3894 The Development of Leakage Key Performance Indicators [C].
AwwaRF Project 2811, Evaluating Water Loss and Planning for Loss Reduction Strategies
(2005) (report in progress) [C].
AwwaRF Project 2928, Leakage Management Technologies (completes 2006) [O].
UKWIR project, An Integrated Network Management Roadmap, starting 2005) [A].
UKWIR project, Capital Maintenance Planning Common Framework Review of Current
Practice (completes Sept 2005) [O].
UKWIR project, Review of Serviceability Indicators (completes Sept 2005) [O].



Boxall, J. B., O'Hagan, A., Pooladsaz, S., Saul, A. J. and Unwin, D. M. (2004). Estimation of
burst rates in water distribution mains. Research Report No. 546/04, Department of
Probability and Statistics, University of Sheffield. Submitted to Journal of Water Supply:
Research and Technology - AQUA. [C].
UKWIR project, The Role of Cost-Benefit Analysis (completes 2006) [O].
AwwaRF project 2870, Customer Acceptance of Infrastructure Reliability (completes 2005)
AwwaRF Project 2774 Investigating International Principles and Customer Views on Utility
Rate Structure (2005) [O].
UKWIR project, DOMS What Can We Learn? (completes 2006) [O].
AwwaRF Project 2762 Creating Effective IT Solutions (2004) [C].
AWWARF project 2933, 2003 Application of Knowledge Management to Utilities 92003) [C].
AwwaRF RFP 3119 Improving Water Utility Capital Efficiency [A].
AwwaRF Project 3133, Ground Penetrating Radar (GPR) Development [A].
AWWARF report 2524, 2001, New Techniques for Precisely Locating Buried Infrastructure
(2001) [C].
UKWIR Report 05/WM/127, Underground Asset Location: Review Of Current Technology
(2005) [C].
AwwaRF project 3126, Life Expectancy of Field and Factory Applied Cement-Mortar Linings
in Ductile-Iron and Cast-Iron Water Mains ( starting in 2005) [A].
Smith, A Capital Maintenance: a Good Practice Guide, Water Asset Management
International, Issue 1, Vol 1 (March 2005).

Training aids and initiatives

Australia and New Zealand
Already back in 1986 and 1987, the South Australia Public Accountants Committee published
a series of eight reports alerting all Australian governments to the need to seriously consider
the management of their infrastructure if deterioration of valuable public services were to be
avoided. Following these reports, Penny Burns, Professor at the University of Adelaide
(Australia) played an essential role in bringing the attention to the importance of the subject
and formalising some key concepts and principles. The Australian and New Zealand AM
school is synthesised in the International Infrastructure Management Manual (IIMM),
which is dedicated to different types of public infrastructures and promotes the Total Asset
Management Process. This publication was jointly elaborated by the Institute of Public Works
Engineering Australia (IPWEA, www.ipea.org.au) and the National Asset Management
Steering (NAMS) Group (www.nams.au.com) and has greatly evolved over time. The
currently available third edition (IIMM, 2006) counted on contributions from other countries,
such as the United States, South Africa and the United Kingdom, and is a true international
reference. With regard to water supply and wastewater AM, Australia and New Zealand



apparently present a bottom up approach, from the practice (utility associations and utilities)
to the theory (academics and researchers). There are rather relevant experiences (e.g.,
Melbourne Water, Hunter Water, Westernport Water, Hobart Water).
National Research Council Canada and the National Guide to Sustainable Municipal
Following the clear identification of the need for higher expenditures in maintenance and
rehabilitation of municipal infrastructures in Canada, the National Research Council (NRC)
Canada has had multiple initiatives directed to creating awareness and establishment
guidelines for the implementation of AM approaches adequate to municipal infrastructures.
(Vanier & Rahman, 2004). The approach recommended by NRC may be synthesised in the
structures reply to the following six what: What do you own? What is it worth? What is the
deferred maintenance? What is the condition? What is the remaining service life? What do
you fix first?. The AM approach recommended by the NRC was incorporated into the
InfraGuide: National Guide to Sustainable Municipal Infrastructure. The guide contains a
good number of independent documents presenting the best practices applicable to the
management of various types of municipal infrastructures. The policy adopted of keeping the
documents produced in the public domain, freely available in electronic version from the web,
has greatly contributed to the dissemination of the knowledge created.
(http://sustainablecommunities.fcm.ca/Infraguide, ref, July 2007)
The European CARE-W and CARE-S systems
Within the 5 Framework Program of the European Union, the twin systems CARE-W
(Computer-Aided Rehabilitation of Water Networks) and CARE-S (Computer-Aided
Rehabilitation of Sewer Networks) are developed. CARE aims at assisting water utilities in
setting up strategic and tactical rehabilitation plans. The projects ended respectively in 2004
and 2005, and since then a number of applications took place in various countries. The CARE
systems comprise an integrated approach, providing tools that assist in the main phases of
decision making process: characterisation, analysis and diagnosis, long term (strategic)
planning and short term (operational) planning. The focus of the CARE systems is on buried
assets. The emphasis is put on the engineering aspects. Figure 5 shows the modules of
CARE-W. CARE-S has a similar structure and architecture, although containing different
modules, adequate to the wastewater systems.


Graphical user interface (GUI)

CARE-W central database
(PI tool)


Water supply



To identify the
performance of a water
supply system for
monitoring, target
setting and

To identify individual
pipes and areas with a
high failure potential

To identify the most

vulnerable points in a
water supply system
and the associated

To explore the future

rehab investment needs
and consequences and
define the optimal
rehab strategy

To select the most costefficient rehab projects

for an annual rehab
programme under
consideration of
various criteria

Modules of CARE-W prototype



The Common Framework for Capital Maintenance in the UK

The Common Framework for Capital Maintenance (United Kingdom) is a project
undertaken with the support and collaboration of the Office of Water Services (OfWat), the
Drinking Water Inspectorate (DWI), the Environment Agency (EA), the Water Industry
Commissioner for Scotland and the Department for Environment, Food and Rural Affairs
(DEFRA) together with UK Water Industry Research (UKWIR). It is based on the analysis of
risk and encompasses an economic approach which allows the trade-off between capital and
operational cost options to be considered. The following key concepts form the basis of the
o capital maintenance should normally be justified on the basis of current and forecast
probability and consequence of asset failure with or without investment;
o consequences are expressed as direct impact on service and company costs;
o service is defined to customers and the environment (including all relevant third
parties and regulatory requirements);
o service is assessed using suitable indicators, building on the approach applied by
OfWat at the 1999 Periodic Review;
o there is a need to demonstrate a least cost approach to Opex versus Capex and
proactive versus reactive maintenance;
o an integrated system approach is required for some systems to assess the direct and
indirect impacts on customers and the environment.
Nowadays, the effect of the Common Framework is felt not only in the UK, but also in
many other countries, particularly in Europe, where the importance of its principles
are being recognised and in some way implemented.



GWRC Compendium.qxd 4/5/10 3:23 PM Page 1 (1,1)

Water Environment Research Foundation

635 Slaters Lane, Suite G-110 Alexandria, VA 22314-1177
Phone: 571-384-2100 Fax: 703-299-0742 Email: werf@werf.org
WERF Stock No. SAM7C07

April 10