Welcome to the LEAKS Suite Website

Leakage Evaluation and Assessment Know-how Software

The aim of LEAKS Suite is simple – to assist and encourage Water Utilities anywhere in the World to improve their leakage management performance

LEAKS Suite softwares use the following internationally recognised concepts, which have been developed during the period 1993 to date.

BACKGROUND AND BURSTS ESTIMATES (BABE) (1993)

This internationally applicable overview concept was developed by Allan Lambert for the UK National Leakage Control Initiative, for calculating components of Real Losses based on the various parameters which influence them. The original paper (Lambert, 1994) and a follow-up paper (Lambert & Morrison, 1996) are available for downloading from the Technical Papers page of this website. The concept was applied in several of the ‘Managing Leakage’ Reports of the UK National Leakage Initiative in 1994.

In BABE analyses components of Real Losses on different parts of the infrastructure (service reservoirs, mains, service connections etc.) are considered to consist of:

• Background leakage at joints and fittings, flow rates too low for sonic detection if non-visible
• Reported leaks and bursts – typically high flow rates but short duration
• Unreported leaks and bursts – moderate flow rates, average duration depends on method of active leakage control

BABE analyses can be used for calculating components of Annual Real Losses, or components of night flows. For annual Real losses calculations, by assuming that average run-times of detectable leaks and bursts consist of three time elements – Awareness, Location and Repair time – Utility policies and standards of service can be modelled. Typical burst flow rates are specified at a standard pressure, and are adjusted to actual pressure using appropriate assumptions for FAVAD N1 values (see next item). The concept can be used for component analysis of annual real losses volumes from water balances, economic leakage levels and component analysis of night flows.

FIXED AND VARIABLE AREA DISCHARGES (FAVAD) (1994)

Pressure: Leak Flow Rate Relationships
Japanese and UK research data on pressure/leakage rate relationships have been reconciled with other test data worldwide using the FAVAD concept, first proposed by John May in 1994. The velocity of flow of a leak varies with the square root of pressure and a Coefficient of Discharge (Cd); but FAVAD also recognises that the effective area of some leakage paths – Cd x A – may vary with pressure. The simplest versions of the FAVAD equation are:

Leakage Rate L (Volume/unit time) varies with Pressure N1 or L1/L0 = (P1/P0)N1

N1 values can be calculated from tests on Sectors at night. Values derived for sectors in the UK, Japan, Brazil, Cyprus, USA, Australia and New Zealand have shown that N1 generally varies between 0.50 and 1.50, with occasional values up to 2.5. Small undetectable leaks at joints and fittings (background leakage) typically have N1 values around 1.50, as do larger leaks and bursts on flexible pipes. Detectable leaks and bursts on rigid pipes normally have N1 values close to 0.50.

The FAVAD concept is used in many of the LEAKS Suite softwares for modelling relationships between pressure and leak flow rates, and pressure and components of residential consumption. The N1Test worksheet explains the theory of an N1 night step test, but for most purposes the N1 exponent can be assessed based on the Infrastructure Leakage Index (ILI) and the % of rigid pipes in the system, using an equation shown in the ‘N1Calcs’ Worksheet.

The BABE and FAVAD concepts are used together in many of the papers in the ‘Technical Papers Library’ page. The paper by Thornton and Lambert, 2005, provides a concise summary, but the analysis of laminar/turbulent flow in that paper did not provide a satisfactory explanation of why high values of N1 are associated with background leakage at joints and fittings.

UNAVOIDABLE ANNUAL REAL LOSSES (UARL) (1999)

The first IWA Water Loss Task Force (1996-99), chaired by Allan Lambert, developed a system-specific equation for the lowest technically achievable Real Losses, for well managed infrastructure in good condition (Lambert, Brown, Takizawa and Weimer, 1999). Substituting appropriate parameter values in a simplified ‘BABE’ component analysis of annual Real Losses, an equation was derived which allows the Unavoidable Annual Real Losses (UARL) to be calculated for most systems (subject to some lower limits on size of system and average pressure).

UARL is a useful concept as it can be used to predict, with reasonable reliability, the lowest technically annual real losses for any combination of mains length, number of connections, customer meter location and average operating pressure – assuming that the system is in good condition with high standards for management of Real Losses. UARL is also used in the calculation of the Infrastructure Leakage Index (ILI).

There is also a different equation for calculating Unavoidable Background Leakage in volume/hour, for component analysis of night flows. A ten-year review of the international application of UARL and ILI (Lambert, 2009) can be downloaded from the Technical Papers Section of this Website.

IWA BEST PRACTICE STANDARD WATER BALANCE (1999/2000)

Drawing on the best practice from many countries, the first IWA Water Loss Task Force, and the IWA Performance Indicators Task Force, produced a standard approach for Water Balance calculations with definitions of all terms involved, which was first published in 2000. This was a major step forward, and the IWA Water Balance is now used and recommended in an increasing number of countries, by technical organisations, regulators, and international funding agencies.

Allan Lambert has since been involved in introducing this Water Balance to Austria, Australia, Canada, Bahamas, Bulgaria, Croatia, Italy, New Zealand, Serbia, South Africa and the U.S.A, often using ILMSS software. The use of confidence limits in Water Balance calculations, included in the PIFastCalcs software, provides valuable insights into the reliability of the calculated Real Losses volume, and the priorities for action for improving the reliability of the calculations.

IWA BEST PRACTICE PERFORMANCE INDICATORS (1999/2000)

The first IWA Water Loss Task Force identified the best traditional simple performance indicators for Non-Revenue Water and operational management of Real Losses, and also improved performance indicators for these parameters (% NRW by value, and Infrastructure Leakage Index). These were initially published in several papers and reports between 1999 and 2000. Since then, many hundreds of ILI values have been calculated and compared, from many different countries.

PERFORMANCE ASSESSMENT or METRIC benchmarking is used to compare performance indicators from different Utilities having different key characteristics. In 2010 the IWA Specialist Group on Benchmarking recommended replacement of the term ‘Metric benchmarking’ by ‘Performance Assessment.’ The ILI was specifically developed for this purpose by the 1st Water Loss Task Force, and is now widely recognised as the best international performance indicator for comparing performance in Real Losses management between Utilities with diverse system characteristics (mains length, number of service connections, meter location, average pressure).

PERFORMANCE IMPROVEMENT or PROCESS benchmarking is used for measuring improvements in performance. In 2010 the IWA Specialist Group on Benchmarking recommended replacement of the term ‘Process benchmarking’ by ‘Performance Improvement’. If pressure management is part of the Real Loss reduction strategy, the ILI is not the most appropriate performance indicator for assessing reduction of real losses. Instead it is recommended to use:

• litres/service connection/day, if connection density is greater than or equal to 20/km of mains
• m3/km of mains/day, if connection density is less than 20/km of mains

4 COMPONENTS APPROACH TO MANAGING REAL LOSSES (1999)

The first version of this widely used diagram was developed by David Pearson (UK), and it has since been further enhanced to include economic leakage levels. It is now widely used internationally as a simple means of explaining the basic activities that are required for effective operational management of Real Losses.

TWIN TRACK APPROACH TO MANAGING REAL LOSSES (2002)

This first appeared in a paper by Lambert & McKenzie at the IWA Conference ‘Leakage Management – A Practical Approach’ in Cyprus in November 2002. It emphasises that:

• the Infrastructure Leakage Index (ILI) measures how well real losses are being managed at the current pressure
• but the current pressure regime is not necessarily optimal, and there are often pressure management options that will have additional and substantial benefits in terms of reductions in leak flow rates, new burst frequencies, repair costs and other benefits

ECONOMIC INTERVENTION (2005)

Calculation of economic intervention frequencies in the UK, based on continuous night flow measurements in small zones, had been in use since 1994. However, outside the UK, many Utilities did little or no active leakage control, and a simpler method was needed. Lambert & Fantozzi (2005) and Lambert & Lalonde (2005) proposed a simpler method of calculating economic intervention frequency based on regular survey, using only 3 key parameters:

• Variable cost of Real Losses (CV)
• Cost of an Intervention (CI)
• Rate of Rise of Unreported leakage

Using this practical approach, Utilities can quickly calculate annual budget requirements for an economic frequency of intervention in any size of system, together with the consequent annual average volume of unreported leakage. See the Lambert & Lalonde (2005) in the Technical Papers Section of this website for more details.
This approach also greatly simplifies the assessment of Short-Term Economic Leakage Levels. The Economic Intervention methodology forms an integral part of the ALCCalcs and other LEAKS Suite softwares.

PRESSURE: BURST FREQUENCY RELATIONSHIPS (2005 ongoing)

Collection of burst frequency data ‘before’ and ‘after’ pressure management in over 100 individual zones by IWA Task Force members (Lambert & Thornton, 2006) clearly demonstrated that remarkable reductions in new burst frequencies can be achieved in some cases, by quite moderate reductions in maximum pressure (including reduction of surges). Within the Pressure Management Team of the IWA Water Loss Specialist Group, Allan Lambert leads the research on how best to model and predict this effect.

The initial (2005 - 2010) analysis and prediction methods were included in PressCalcs, but ongoing improvements which now include the extension of infrastructure life for AC pipes, are now in new professional software such as LAPMETCalcs (Leakage and Pressure Management Evaluation and Targeting) and P&BACalcs (Pressure and Bursts Analysis). See papers by Lambert & Thornton (2011 and 2012), Lambert Hicks, Kay, May and Waldron (2010), Lambert & Fantozzi (2010), and Thornton & Lambert (2007) in the Technical Papers Section; it is recommended to read the most recent papers first.

ECONOMIC LEAKAGE LEVELS, WITH AND WITHOUT PRESSURE MANAGEMENT (2008)

The development of quick and practical methods for calculating economic leakage levels has been an objective of the Water Loss Task Force since 2001. Methods developed in the United Kingdom in the 1990s were data intensive and did not consider the influence of pressure management on leak flow rates, burst frequency and repair costs, and frequency and costs of economic intervention. Confidence Limits also needed to be included for meaningful calculations.