Anglian Water Services is the largest water company in England and Wales. It covers an area of 27,000km2 in the east of England, and daily serves more than 4.2M customers with drinking water. The distribution input is typically 1.1Mm3 and this is delivered through more than 36,000km of pipelines.

At privatisation, it was realised that significant improvements were required to WTWs and distribution systems to consistently achieve the high standards required by EU regulations and levels of customer service expected today.

But how do we manage today’s expectations with historic assets? Within what was then called the Mid-Northants Water Board, there were more than 200 local water sources taking water from springs, boreholes and river gravel beds. Starting in the 1960s, these were rationalised to three large surface WTWs and associated distribution links established.

Yet the pipeline systems within each settlement were still as originally constructed over the past 150 years. There is no information on the water quality produced, treatment system employed, nor on the specification of the mostly iron pipeline systems. The area was a significant producer of cast iron from local foundries, resulting in a possible plethora of pipe manufacturers, and hence pipe material qualities.

Initially, the treatment works were improved in order to ensure full compliance with the Water Quality Regulations. These regulations have subsequently been revised and tightened, and the works still comply in full. However the regulations also apply to water distribution and this is where the challenge is for maintaining quality.

Throughout the past ten years, attention has been placed on the safe and secure delivery of the high-quality treatment works output from source to tap through the distribution system. However, the materials used in distribution have changed over the past 30 years. (See Fig 1 on facing page.)

During 1992 to 2002, Anglian Water carried out rehabilitation on water-distribution pipes to address water quality in distribution.

These were commonly known as Section 19 obligations and have been completed. The decision was made to use a combination of replacement with polyethylene pipe and rehabilitation using epoxy linings. However, there is still the legacy of some 11,000km of iron pipe, among others, to be managed.

In October 2002, the Drinking Water Inspectorate (DWI) announced its expectations for Distribution Operation and Maintenance Strategies (DOMS). These strategies are in use and being continually developed in conjunction with the DWI to ensure that water quality within distribution is maintained since the completion of Anglian’s Section 19 programme.

Water-quality maintenance requirements

Water quality regulations cover not just chemical but also biological, aesthetic and acceptability parameters. All have to be followed but, with a historical and dynamically changing distribution system, it is not possible or realistic to solely apply the legal maxima of each parameter during maintenance planning.

There needs to be enough difference between legal limits and levels of leading parameters to allow for intervention maintenance to ensure compliance and also affordability. Although the regulations include many specific limits, Anglian Water finds that a selection of parameters are sufficiently representative of the ability of distribution systems to deliver water that complies with regulations and meets customer expectations.

Often, it is found that iron and turbidity are leading measures of water quality deterioration in distribution, usually covered by the description “discolouration”.

Within Anglian’s Policies and Standards for Water Supply Hygiene, there are internal water-quality standards (a selection are in Table 1) and procedures documented to ensure regularity and robustness in the management of its distribution system. The document provides for continual monitoring and regular distribution performance appraisals, and forms the basis of Anglian’s distribution maintenance planning.

Distribution maintenance planning

Anglian’s Section 19 water-distribution rehabilitation programme was completed by September 2002. Since that time, the company has been developing and implementing water quality and pipeline asset management and performance tools to ensure that both short-term and long-term compliance with quality standards are upheld through maintenance planning. These are applied to water distribution pipelines and are described below.

Anglian Water also carries out planned maintenance on fittings installed within water distribution. These include air valves for water-quality risk assessments and customer service, and turn-over flushing at closed valves to maintain disinfectant levels. Here we are looking only at pipeline maintenance.

District meter area risk score

– short-term

Analysis is carried out annually to determine the discolouration incident risk score for every District Meter Area (DMA) throughout Anglian’s region. The tool annually applies the following data as appropriate to each DMA. (See Table 2 on facing page.)

The standard methodology governs how this data is processed and scored. The combination of totalling some scores or taking the higher score from specific selections of scores produces a final score per DMA.

This is used to rank each DMA and, at a predetermined minimum score, those DMAs exceeding this minimum are promoted for planned preventative maintenance (PPM) during the ensuing year.

This analysis is carried out using Anglian’s Geographical Information System (GIS). This can identify locations in detail within DMAs where water quality is known to be at greatest risk of failure. Therefore, not only are the DMAs ranked for PPM, but the locations of distribution where maintenance efforts are best placed are also highlighted. It is usually found that about 100 DMAs in almost 2,000 are identified for annual maintenance. See Figure 2.

Quality maintenance prioritisation score

– long-term

These analyses are carried out at the same time as the DMA risk scoring, and use some of the same data sets, but require a more substantial analysis of water quality and distribution performance within DMAs performance over five years. Any DMA may be advanced into this detailed appraisal whenever Anglian Water has cause to do so. This tool applies the additional data collected during the past five years. (See Table 3.)

Additional data is used in this analysis that provides an indication of future changes in water quality. The effectiveness of previous PPM is included so it is possible to appraise improvements to maintenance and selection of other options. Output from these analyses places each DMA into one of three risk bands. Bands are shown in Table 4.

Any DMA that is categorised within band C is subjected to further detailed appraisal. This includes other investment drivers such as constancy, pressure and growth. The outputs include investment proposals for water distribution that ensure the right pipes are improved at the right time and by the right solution – developing optimum and synergistic maintenance for least whole-life cost of water distribution.

Forward-looking and risk-managed maintenance

Great reliance has been placed on using data already collected, as described above, and extrapolating this into the future. So only what we know now can be used. But what about what we don’t know? If a pipe has never burst, or never failed a water-quality sample, or flows never changed, then it is likely little will be known about the water-quality risk potential in that pipeline. Sometime, something will change. And when this happens, to be in the area supplied by this pipe, a water-quality incident may occur with a severity that was not predicted.

In order to include these latent risks to water quality in distribution in maintenance plans, Anglian Water is working with UKWIR, consultants (Ewan Group), and the University of Sheffield to develop new thinking and understanding about maintenance of water distribution through research and development.


UKWIR has national projects research ongoing into water mains condition grading and serviceability. The National Failure Database compiled and maintained by Bodycote Pipeline Developments is the largest national water distribution failure database in the world. From this, companies can develop best practice to understand their mains failure rates better. Several companies’ data, including Anglian Water’s, has sufficient quality and quantity for researching the correlation of explanatory factors.

Water mains condition grading and serviceability indicators are current projects to examine how the water companies have interpreted existing Ofwat definitions. It is expected that proposals will be published for improvements to definitions and methodologies as appropriate that will assist Ofwat and water companies to better

monitor and manage water distribution in the future.

Prediction of discolouration in distribution systems (PODDS)

The University of Sheffield (Pennine Water Group) has two research projects under the title of Prediction of Discolouration in Distributions Systems (PODDS). The first project, funded through EPSRC, researched the characteristics of water-pipeline sediment and developed a mathematical expression to predict how discolouration, once in suspension, responded within distribution networks. Details are available at Anglian Water, along with Yorkshire Water, supported this project. PODDS has led to a shift in understanding of the causes and processes involved in discolouration, with the potential for increased justification of best operational and management practices.

The PODDS model, developed by the Pennine Water Group is capable of predicting the discolouration response in distribution networks to changes in hydraulic conditions. Results highlighting the capabilities of the full model are presented in Figure 3. The model is based on the novel application of cohesive transport theory and its development has led to a step change in philosophy and perception relating to discolouration. The initial PODDS research project, however, did not include development of the model into a tool suitable to be readily utilised by the water industry.

A follow-up project is therefore under way for the project collaborators to lead the industry in developing and producing a practical tool for planning discolouration related operational and maintenance strategies. Further research and field trials are being currently carried out at more sites within England and Wales. At least six major water companies, including Anglian Water, support this project, where one of the major objectives is to understand discolouration regeneration (Figure 4).

Ewan Group has established a club of water companies, including Anglian Water, to develop and implement Discolouration Risk Modelling (DRM). Details are available at DRM will enable Anglian Water to gain benefits through application for qualitative risk prioritisation using hydraulic models, water distribution attributes, and water-quality data (Figure 5).

It applies an algorithm which simulates the bursting and then closure of each pipe in the system in turn and the impact of each of these events on the flow condition in every other pipe in the system. A series of risk trees then allows the calculation of a hierarchy of factors, the top level of which is a Discolouration Performance Score (DPS). Each pipe’s DPS is assessed from the product of likelihood and consequence. In DRM terms, likelihood is assessed from each pipe’s tendency to burst, its potential to cause discolouration and its sensitivity to flow changes as a result of events in the system which could trigger discolouration. Consequence is based simply on the flow through the pipe as a surrogate for the scale of customers who could potentially be affected.

Analysis of these results by area, maybe by DMA or combination of DMAs, will enable discolouration risk, pipe by pipe, to become visible and allow Anglian Water to prioritise maintenance to those pipelines deemed to present an unacceptable discolouration risk. After maintenance, the discolouration risk borne by the business will be reduced (Figure 6), the ability to target pipelines using DRM outputs ensuring that this will have been achieved cost-effectively.

The model provides a framework for further development either of discolouration risk itself as scientific understanding improves, or of other serviceability risk aspects.

A recent pilot trial of DRM within a supply zone, consisting of four DMAs, has shown very good alignment to the existing scoring system, but DRM is more detailed by being down to pipe level.


  • Maintaining water quality in distribution systems by planning maintenance based on analysis of historic data is no longer sufficient.
  • Current research (PODDS) has improved our understanding of the characteristics of discolouration.
  • A new risk-based model (DRM) that works at pipe level has been developed and successfully trialled.
  • Better visibility of water quality risk in distribution systems enables proactive planning for optimised and targeted maintenance.
  • Water quality will be maintained by intervention being scheduled through a more cost-effective and risk-managed approach.
  • Discolouration in water distribution was previously managed through the Section 19 mains-rehabilitation programme, Anglian Water is moving forward with new knowledge and models to manage water distribution maintenance with greater foresight (Figure 7).
  • Implementation of forward-looking maintenance planning is a major move forwards for Anglian Water to achieve its goal of least whole-life cost for water distribution.

The author wishes to express his gratitude to Anglian Water for its support in producing this article. Opinions presented are those of the author not of Anglian Water. Support from University of Sheffield (Pennine Water Group) and Ewan Group is also gratefully acknowledged.

Action inspires action. Stay ahead of the curve with sustainability and energy newsletters from edie