Thames Water is the UK’s largest water and wastewater service company. It supplies 9 million water and 14 million wastewater customers across London and the Thames Valley. It manages 32,000 km of water mains, 12,000 km of Trunk Mains, 108,000 km of sewers, 100 water treatment works and 350 sewage treatment works.
Thames Water have used AIM to optimize over £30 billion capital value of assets to determine their 40-year investment plan submitted to the UK regulator OFWAT. AIM gives Thames Water the control and ability to run unlimited scenarios in-house to determine optimal investment plans. This has resulted in running hundreds of scenarios with multiple customer service and budget constraints in a short period of time. AIM is now fully embedded within asset and investment planning team and is being used as business as usual to not only set the strategic plan but also inform the programme of delivery in a cyclic manner.
Thames Water is Probit’s first client and is using AIM for water and sewer pipe assets to quantify the investment required to achieve levels of service for customers.This feeds into their five-year planning cycle, in which proposals are submitted to their regulator, Ofwat, to determine the investment companies can make and future bill limits for customers.
Asset Investment Manager has also been used for water resource management planning to identify how Thames Water will maintain water supplies over a 25 year period.The complexity and size of Thames Water greatly benefit the development of AIM, which all our clients ultimately benefit from as well.
Adopting AIM within a business as an investment planning tool is a journey. AIM allows a transparent and step by step progression from simple to complex asset management in a time frame that reflects the maturity required.
For Thames Water, the journey started by replicating their existing investment planning approach within AIM. This quickly demonstrated the speed and value of AIM and allowed for staff to quickly learn how to set up and run their own investment scenarios by themselves. Training sessions were also provided to get new users up to speed.
The next step in the journey involved enhancing existing and developing new approaches to determining asset risk and investment requirements.
The final step in the journey involved Thames embedding AIM within their asset planning process and building an in-house team around it.
Throughout the journey, we worked with Thames Water to help them get the most value out of using AIM at each step.
Asset Level Modelling
Robust statistical models were developed to predict asset performance and consequence of failure at asset level. A step by step approach was adopted that involved updating any existing models with the latest data and expanding the models to include additional predictor variables.
A variety of standard statistical techniques based on an understanding of the assets, their underlying failure process and the data that was available, to produce burst rates, leakage and deterioration models.
Consequence models covering interruptions to supply, low pressure and customer complaints were developed based on
consequence data and pipe characteristics.
All analysis and resulting models were undertaken at pipe level and incorporated within an AIM Risk Map. This framework links pipe level performance and deterioration, with downstream consequence and service drivers. This creates a cause and effect and effect relationship between asset performance and service delivered to the customers.
Interventions and Costs
The Risk Map was populated with several interventions and costs including asset replacement and repeat interventions, such as cleaning.
AIM has allowed Thames Water to ask and answer investment decisions in hours that previously took months. Using AIM’s powerful and fast optimization engine Thames have been able to run 100’s of scenarios at asset level with varying objectives and multiple constraints.
A key scenario for Thames is to hold all network performance and customer service drivers’ constant over a fixed time interval. This is then used to compare other potential scenarios.
Each scenario generates a plan of investment at asset level and the final chosen scenario is used as a basis for their 5-year delivery programme.