Not registered?

Register today

Hello. Sign in.

My Account

You are here : At Home > Corporate > News

Latest news

12 October 20

Affinity Water pioneers new AI driven data sharing to allow water companies to better monitor the impact of the weather and Covid19 on water demand in the future

Affinity Water has used AI driven technology to manage the impact of water use in relation to weather patterns and Covid19 which it plans to share as an open data model with other interested water companies.

In the past PCC (Per Capita Consumption) measurements have only been available for analysis every three months, which means understanding incidents such as significant weather changes, water usage campaigns or an outbreak of an infectious disease with subsequent water use messaging such as the Covid19 lockdown can be measured and assessed far sooner. This in turn will allow water companies to meet water use performance targets in the longer term.

Using an AI driven system that measures in real time the water balance at a DMA level (District Metered Area) Affinity Water’s data scientists have been able to accurately predict PCC use across its network and three regions in real-time while also taking into account seasonal demand and weather patterns.

The new AI systems use a collection of analytical tools that belong to the Distributed Machine Learning Community with contributions from many developers. More specifically the Affinity scientists used “xgboost” a gradient boosting machine created by Tianqi Chen.

Using these new AI systems Affinity Water data scientists were able to distinguish between previous periods of hot weather and high demand and this year’s increased high demand for water taking into account Covid19. This represents a major breakthrough for the industry more widely.

Andrew Morris, Affinity Water’s Chief Information Officer said:

“What is new is the idea to use fast logging to infer domestic consumption as well as adopting the powerful machine learning technique known as extreme gradient boosting that has recently been dominating applied machine learning and Kaggle competitions (the World’s largest data science community) for structured or tabular data.

“Our model was completed in four weeks, validated against existing data and then we developed a PCC data dashboard to allow our PCC team access to the real-time data on daily PCC use.

“We can now model scenarios, to see the expected PCC based on a normal year and the expected PCC based on changes such as the Covid19 lockdown or other major events. This is a major breakthrough and as we believe open data is so important to the water industry and to our customers’ we intend to share our new systems openly with the industry.

“In the future this will allow us to plan better for other abnormal events including weather patterns and to be able to monitor the impact much more closely of our work to reduce PCC through our other campaigns as well as it will give us more real-time insight. “

Mr Morris went on to explain the importance of sharing this breakthrough with others in the industry. He explained:

“As a next step we have developed a customer level model to predict the water usage movement for each customer at any time. We are releasing the model openly for all water companies to use freely to benefit the industry. It’s a first for a water company to openly share data science models like this. In the future we hope there will be more open data, and more sharing of proven data models which can ultimately help us all deliver a better service and better value to our customers more quickly. We already use similar data science methods to identify interruptions to supply and to identify leakage and these are already proving highly valuable.

“The next few steps of the PCC modelling journey involve modelling leakage-free DMA consumption on case by case basis leading to an accurate unmeasured consumption model, finally using all of the models to accurately predict the total demand and then using the unmeasured and customer level models to distribute the totals appropriately across every household automatically, every day.

“There is so much value we have already gained from data science and as we further roll out our digital twin capability Affinity Water will change the way we operate in many areas of the business and give a whole different level of insight into how we can be more effective across the water sector. “

News archive

Media enquiries

Our Communications team handles all media enquiries on behalf of Affinity Water and its employees.

Contact us on 01707 277 110 or email news@affinitywater.co.uk.

Our office hours are 8am to 5pm, Monday to Friday.

We are available for emergencies:
5pm to 7am, Monday to Friday
7am to 7pm, Weekends and public holidays
We aim to respond as soon as possible and within two hours.

Enable Recite