Assessment of real losses in potable water distribution systems: some recent developments

Considerable progress has been made over the past 10 years in the assessment and benchmarking of real losses in potable water distribution systems. Most of the advances have been based on the burst and background estimate (BABE) methodology, which was first developed in the mid-1990s by the UK water industry and has since been widely accepted and used in many parts of the world. Since the original BABE methodology was developed, several other key concepts have been added to the ever-growing list of water demand management tools. In particular, the infrastructure leakage index (ILI) and unavoidable annual real losses (UARL) introduced by A. Lambert, and the fixed area variable area discharge (FAVAD) theory by J. May, are now recognised as key "tools of the trade" in any water demand management assessment. One of the first main developments where the above-mentioned concepts were applied in practice to benchmark leakage was in South Africa, where the local Water Research Commission supported the production of the BENCHLEAK Model. This was basically the first comprehensive model to assess real losses in potable water distribution systems using the UARL and ILI concepts. The model was developed by one of the authors together with A. Lambert, and was soon followed by similar developments in Australia (BENCHLOSS) and New Zealand (BENCHLOSSNZ). Both models incorporated additions and enhancements to the original South African model, and were tailored to suit the local conditions in line with the clients' requirements. Similar developments took place in parallel by various leakage specialists, most notably in Brazil, Malaysia and Cyprus, to mention just a few of the similar initiatives. Each time a new model was developed, certain improvements were made and the "science" of leakage management and benchmarking was enhanced. Through the use of the different models and from discussions with various researchers from around the world, it has become clear that there is a genuine need for such models, and they are being readily accepted by clients in most areas. The discussions have also raised many questions concerning the derivation of the terms used to calculate the UARL and the ILI, and, to address these concerns a specialist group was created through the IWA to investigate the various issues. This paper will highlight the progress that has been made to date with regard to the key issues that have been raised by the task-team members, and recommendations based on the feedback that has been received from around the world. The paper will also present some of the results that have been obtained from different parts of the world to highlight both the progress and the problems associated with the assessment of real losses. The paper will conclude with a short description of several new models that have been developed and are in use, which demonstrate the latest improvements to an ongoing process to assess and benchmark real losses in water distribution systems.