Tynemarch Publication

Capital Maintenance Planning in the Water Industry:
Practical Applications of the UK Common Framework

Dr Jeremy Lumbers, Tynemarch Systems Engineering Ltd
George Heywood, Tynemarch Systems Engineering Ltd

Introduction

The Capital Maintenance Planning Common Framework (CMPCF) was developed as a joint project between the water companys’ research company UKWIR and the economic regulator, Ofwat to provide guidance for the estimation of the economic level of capital maintenance. The steering group for the project included representatives from the Environment Agency, the Drinking Water Inspectorate and DEFRA, as well as the water companies and Ofwat.

A number of key concepts form the basis of the CMPCF as follows:

• capital maintenance should normally be justified on the basis of current and forecast probability and consequence of asset failure, with and without investment
• ‘consequences’ are expressed as direct or indirect impact on service and company costs
• ‘service’ is defined as service to customers and the environment (including all relevant third parties and regulatory requirements)
• service is assessed using suitable indicators, such as interruptions to supply, effluent quality, etc
• opportunities for trade-offs between operating costs and capital costs should be evaluated
• the analysis should recognise the integrated nature of both the water and wastewater systems in order to assess the direct and indirect impacts on customers and the environment.

The CMPCF is illustrated in the following diagram:

The Common Framework was launched in May 2002, although the main principles were presented at a conference in October 2001 to provide early guidance for the preparation of the draft asset management plans submitted in August 2003.

The Common Framework was received well by both the Regulators and the Companies. A pilot application by a selection of volunteer companies resulted in responses including:

• Company A - "The Common Framework for CMP provides a logical and well-structured methodology...which complies with the needs of MD161¹."
• Company B - "...was able to fulfill all of the steps and produce an asset plan for the asset set studied."
• Company C - "...believed that the framework in essence is what all good companies should be undertaking..."
• Company D - "...the general principles of the model are aligned with the Asset Management approach that is being implemented within the company."

• "We are pleased to agree and confirm that the framework when implemented fully and effectively by a water company will provide a robust basis for assessing capital maintenance needs."
• "We encourage its application for the 2003-04 draft and final business plans with an expectation that each company will use its expert judgements where information gaps exist".
• "The Common Framework is a large step in the right direction."

Ofwat AMP4 guidance and the Common Framework

The Common Framework is referenced widely in the Ofwat guidance for PR04 in the sections related to capital maintenance planning².

The introduction of 'service risk' to replace the historical performance grading system was another aspect of the integration of the Common Framework into the price review process. For example Ofwat stated in connection with the asset inventory:

"The key change from the PR99 review is to incorporate aspects of service risk into performance assessment, in line with the principles of the UKWIR 'Common Framework for Capital Maintenance Planning'."

The high level guidance provided regarding service risk was as follows:

"The concept behind performance categories that reflect service risk is:

Grade 1	No concerns
Grade 2	No threat to service, but operating costs increasing
Grade 3	Capital maintenance on economic or health & safety grounds or by third party initiation
Grade 4	Preventative/proactive capital maintenance
Grade 5	Reactive capital maintenance (including reactive response policy)"

It was further stated that:

"Capital expenditure to maintain the asset base would normally be allocated across performance (i.e. service risk) grades 3, 4 and 5."

The descriptions of the service risk grades were expanded in more detail as an example in relation to the water distribution system.

The implication of the above is that companies could justify capital maintenance without necessarily having observed a deterioration in service measures (serviceability indicators) which was a requirement in previous price reviews. Similarly the explicit recognition of Heath & Safety as a driver for capital maintenance is a new development bringing the water industry into line with other utilities such as gas where H&S issues are primary factors in justifying investment.

There were a number of issues arising from the first AMP prepared using the Common Framework, including:

• The availability of asset data
• Level of detail of cost data, both refurbishment and replacement
• The full cost of asset failures
• Integrated system models for consequence prediction
• Cost and benefits of Opex alternatives to Capex
• Balancing local versus company-wide service standard objectives

Case studies

Tynemarch assisted seven water/water and sewerage companies in the estimation of their capital maintenance requirements covering both underground (infrastructure) assets and above ground (non-infrastructure) assets.

For example all the non-infrastructure assets of United Utilities were 'modelled' using Tynemarch forward-looking risk-based analysis tools. The asset groups included in the assessment were:

• Water treatment works
• Water pumping stations
• Service reservoirs
• Sewage pumping stations
• Wastewater treatment works
• Sludge treatment works

In addition large diameter trunk mains/aqueducts were included in view of their role in the strategic water supply system.

The analysis of the impact of failures of water treatment works, pumping stations and large mains on the security of supply was modelled³ systematically assessing the population likely to be affected by an interruption to supply of a given duration. A typical schematic is shown in Figure 1 as used for South East Water.

Figure 1. MISER water supply optimisation model

The overall process involves the determination of the economic capital maintenance required for each asset group, based on whole-life costing principles taking into account the cost of unplanned failures, repairs, refurbishment and replacement. The future service indicators such as pressure, interruptions to supply, water quality, etc, are then forecast for the next 15 to 20 years. Where the predicted service standards were unsatisfactory a formal optimisation method is then used to identify the most cost-effective additional investments required to avoid deterioration in future service.

Primary data were used to estimate such relationships as failure rate changes with time as influenced by such factors as age, type of equipment and operating environment. Other data such as unit capacity, configuration, telemetry coverage, etc are required to undertake conventional reliability analysis.

Hence the risk analysis for non-infrastructure assets delivers estimates of the probabilities of failure for each failure mode of each asset, together with failure durations and estimates of consequence magnitude measured against each level of service criterion. The results are assessed at a variety of functional and spatial levels in support of investment planning. At the works level for example, the customer service risks associated with individual unit processes can be compared for a range of planning horizons and investment scenarios. Equally, comparisons can be made across the set of treatment works to identify areas of unacceptably high risk. The results can be displayed in the form of a Pareto plot as in Figure 2.

Figure 2. Water Quality risks organised by Treatment Works

Analysis of infrastructure assets employs a similar approach but distribution mains are considered at the zonal level. The relationships between modes of failure resulting in reduced customer service and zonal or environmental characteristics are encapsulated within a formal structure. Each zone can then be analysed for risk under a variety of investment and asset deterioration scenarios. In addition to prioritising zones for further analysis and comparing zones for relative Corporate risk contribution, the analysis can be used to assess the risk-reduction impact of investment and to identify the sensitivity of risk to different capital maintenance options. Sensitivity analysis also assists in prioritising the collection of data to support the current and future analysis.

A key component of the analysis of water distribution systems is the modelling of mains bursts which as well as being an Ofwat reportable performance measure also have an impact on other measures of service such as leakage, discolouration and interruptions to supply.

Statistical models are used that take into account influential factors such as those in Table 1.

Mains data	Date of bursts
	Date laid
	Material
	Diameter
	Pressure
Environmental data	Soil corrosivity
	Soil fracture potential
	Surface type
Climatic data Max and min temperature
	Rainfall
	Soil moisture deficit

Table 1. Burst modelling data

Reasonably good predictions of historical bursts confirmed the accepted general belief that hot summers and cold wet winters have a strong influence on the observed burst frequency, as illustrated in Figure 3, for a number of future scenarios.

Future development

The capital maintenance planning assessment process used by Ofwat taking into account the Common Framework was reviewed by an independent assessor. Two reports have been produced which are available in the public domain^4,5.

Some of the conclusions and recommendations related to the CMPCF are as follows:

"the process, being aligned with the CMPCF, addresses the issues and concerns raised at PR09.

The process is conceptually robust, being based on the CMPCF, and has already been accepted in principle by the industry

[We recommend that Ofwat] continues to provide incentives for the industry to improve its asset information and work towards full application of the CMPCF

[We recommend that Ofwat] continues to monitor the companies' progress in the application of the CMPCF in their day-to-day decisions..."

Future developments are likely to focus on the assessment of capital maintenance requirements within the broad guidelines of the CMPCF, but in more detail and as part of routine business rather than on the 5 year AMP cycle. Greater consideration of the integrated nature of water systems will be required together with better data and modelling capabilities in some areas.

Summary

The Common Framework:

• is the agreed way forward for identifying Capital Maintenance requirements in the UK Water Industry
• has sufficient flexibility to encourage individual company innovation and hence competition
• provides a rational basis for gathering asset data regarding failure probability & consequences
• is a platform for on-going collaborative improvement involving both the industry and the Regulators

Acknowledgements

The author wish to acknowledge all who have contributed to the development and practical realisation of the Common Framework, including the sub-contractors, Ofwat, Environment Agency, the Drinking Water Inspectorate, DEFRA and the water/water and sewerage companies, the Steering Group for their constructive guidance and support, and UKWIR for permission to write this paper.

Note that this paper represents the views of the authors only.

The Common Framework report is available from UKWIR at www.ukwir.org.