23 Apr

Transforming EDF's Energy Insights Through Strong Data Foundations to Reach Net Zero

TJ
Tom Jenkin

Key Facts

  • Reducing the risk of human error, mitigating the frequency of incidents and reducing their financial impact.
  • Saving further costs in the millions through improved cost forecasting accuracy.
  • Reducing the validation time of energy trades from 150 minutes to under 30 minutes boosting efficiency.
  • Cut document generation time from months to minutes.

As the biggest generator of zero carbon electricity in the UK, EDF plays a crucial role in the journey to net zero. At the core of this sits the Wholesale Market Services (WMS) team, responsible for accurately forecasting energy usage. 

Amid unprecedented upheaval in the energy industry, the team needs to make smart decisions based on accurate and quality information. 

“The energy landscape is changing really quickly at the moment. We want to prepare for the future and build a strong data foundation that will not only help us make data centric decisions, but will also put ourselves in the strongest position to help Britain achieve Net Zero. Ultimately we want to free up as much of our analysts time as possible, so they're able to invest in innovation and find the best solutions possible.” Cici Whitcomb, Data Engineer, EDF

Reacting to a Shifting Energy Landscape

The WMS team needed to respond faster to price changes caused by increased market volatility in wholesale energy prices. The complex analysis of historical and future-looking data was a manual process lacking in accuracy and scale. The team wanted to improve cost measurement and forecasting capabilities to navigate the changing market landscape. 

The WMS team sought out the expertise of Mesh-AI to assist in building an organisation-wide data strategy, to increase access to crucial data, make WMS analyst’s jobs easier and reduce the risk of inaccurate forecasting. 

“We design and implement hedging strategies to try and reduce the chance of being on the wrong side of the market. To put that into context, we trade over 10 billion pounds worth of energy annually. So the risk of getting a strategy wrong is in the 10s of millions of pounds worth of impact.” Cici Whitcomb, Data Engineer, EDF

Increasing Accessibility to Crucial Data

Mesh-AI performed an initial Proof of Concept and discovery engagement to:

  • Demonstrate the art of the possible with AWS native technologies to the WMS team.
  • Assess the feasibility, viability and value of candidate use cases.
  • Established an EDF-wide data strategy that identified key people and process changes the organisation needs to deliver, in order to become a data-driven business.
  • Create a prioritised roadmap aligned with WMS and the wider EDF business strategy.

Through cataloguing and categorising existing workloads in the initial discovery phase, the Mesh-AI team worked with EDF to agree on an architecture blueprint and migration strategy for the initial candidate use cases. 

A combination of AWS and Snowflake solutions were deployed to extract data from a variety of source systems, migrate, model and transform it, and expose it to users through visualisations and ad-hoc analysis tooling. Leveraging existing tooling, systems and processes across EDF for efficiency, we created new AWS resources where required to bring the pilot data use cases into production. 

With a shared vision, we delivered these new data products with all the underlying supporting infrastructure, drastically increasing the accessibility of information critical to accurate forecasting.

“Historically accessing our data has been a long process… this data capability has made it much more accessible. And the data is now ready to use, which has meant that we've reduced the time that analysts have to spend accessing and transforming data and really freed up a lot of their time.” Cici Whitcomb, Data Engineer, EDF

Smarter Decisions to Invest for Net Zero

By implementing a modernised data-driven approach through cloud-native technology, EDF has significantly improved its operational efficiency. Notably, the time needed to generate Ofgem’s ‘Requests For Information’ documentation was reduced from months to minutes.

Mesh-AI has enabled EDF to reduce the risk of human error in their energy trading, mitigating the frequency and financial impact of incidents. We have also saved further costs in the millions by improving cost forecasting accuracy, and helping EDF reduce the validation time of energy trades from 150 minutes to under 30 minutes. All of this helps EDF make smarter decisions about energy forecasting and understand where to invest for net zero.

“We have also unlocked massive value from our data, as we're able to be much more flexible with the granularity that we're able to perform analysis at, as well as the types of operations that we're able to perform. This means that we're able to get more accurate insights, and perform analysis that we weren't previously able to do.” Cici Whitcomb, Data Engineer, EDF

Latest Stories

See More
This website uses cookies to maximize your experience and help us to understand how we can improve it. By clicking 'Accept', you consent to the use of these cookies. If you would like to manage your cookie settings, you can control this in your internet browser. Find out more in our Privacy Policy