Articles & Case Studies

How Thames Water uses Data Science in its Water Operations

Posted: Thursday 15th July 2021

Here at Water Active we recently discussed the potential that artificial intelligence can bring to detecting network blockages and now we’re going to take a closer look at how Thames Water, the UK’s largest water provider, uses data science to do just that, and more. We spoke with Shaun Dippnall, CEO of EXPLORE Group, Thames Water’s data science partner.

“Data science combines multiple fields, including statistics, scientific methods, artificial intelligence (AI), and data analysis, to extract value from data,” begins Dippnall. “In this world of digital transformation, it’s increasingly being talked about as something that every business needs to get better at, but few actually manage to do so. While some industries struggle to keep pace, the utilities sector appears to be leading the charge and Thames Water is one of the leading examples.”

Supplying customers with 2.6 billion litres of drinking water every day, Dippnall says Thames Water is using data science and analytics to deliver products that reduce leakages, respond quickly to supply interruptions, prevent blockages and flooding, reduce wastage and eliminate environmental pollution, all while saving vast amounts of money. He adds: “We currently have 70 data scientists involved in this project, helping to build out Thames’ Smart Water and Smart Waste platform.”

Before Dippnall and his team got involved, it’s interesting to note that London’s sewer system was built for a population of four million back in 1865, when only two and a half million people lived there. Fast forward to today, and the capital is home to more than nine million people, but it was only in recent years that major changes began to take shape.

Dippnall explains, “Up until two years ago, more than 500 factories assisted Thames Water in the treatment of its water or waste, but blockages resulting from flooding or pollution in this vast delivery network were a continual problem and they needed to be able to pre-empt and prevent these.”

So how did Thames Water and EXPLORE meet? Dippnall tell us: “When the water crisis hit Cape Town in 2018, we tasked our students at EXPLORE Data Science Academy (EDSA) with creating a database using historical water consumption data by suburb, available on Cape Town’s open data portal.”

He continues: “The goal was to develop insights into the city’s water shortages and provide recommendations to alleviate the crisis. It was here that specific water supply-related analytical tools were developed.”

Then in 2019, after becoming aware of EDSA’s work in Cape Town, EXPLORE’s consulting arm was invited to deliver the analytics resulting from the monitoring of water levels from Thames Water’s vast network of water collection and distribution piping.

At the time, Dippnall recalls that Thames Water’s Chief Digital Officer, John Beaumont, said, “The decision to appoint Explore AI was based on its in-house data science capability and its proven expertise in mapping the various data sources that contribute to understanding the factors affecting Cape Town’s water supply. By working with Explore AI, Thames Water has entered an unprecedented new era in disruptive technology that enables us to digitise our business in ways previously not possible.”

You don’t need a team of 70 to start harnessing the power of data science and following in Thames Water’s footsteps, Dippnall reassures. “In fact, in the current situation, it would be quite difficult to find so many skilled candidates,” he adds. “According to IBM, last year saw demand for data scientists and data-savvy candidates surge by almost 40%. That’s because, in a world where businesses are leaning increasingly on digital ecosystems to engage with potential customers, they need to be able to put all of the data they’re harbouring into action.”

He concludes: “If data science is to be embraced as a philosophy and an approach to overcoming challenges and finding opportunities, businesses instead need to imbue their already competent workforce with skills in analytics, programming, business intelligence and data engineering.

“And the magic you get when you upskill your existing team, is that they know your business, your customers, and love what you do. You can nurture a truly data-driven culture instead of trying to retrofit one artificially by slowly bringing in outside talent. Data science skills should be viewed as team pursuit, blending an affordable recruitment pipeline with the upskilling of in-house staff to gradually uncover the myriad benefits of a more scientific approach to problem-solving.

Read the magazine online

August 2021

About the magazine »
Magazine archive »

Pulsar New Banner