Social Connect developed an innovative method for identifying customers who are in, or at risk of, fuel poverty with greater granularity than ever before, by leveraging newly available datasets.
What is the project about?
Tackling fuel poverty is a key focus of our Consumer Vulnerability programme. One of our main challenges is accurately identifying those who need additional support.
Social Connect has developed an innovative method for identifying customers who are in, or at risk of, fuel poverty. By applying machine learning to both new and existing datasets, the tool provides insights with greater accuracy and presents the information in a clear, user-friendly format.
How we’re doing it
Social Connect was a collaborative project between Energy Systems Catapult (ESC) and UrbanTide. Collectively, the project team merged a deep understanding of vulnerability and fuel poverty with expertise in artificial intelligence (AI) and data science. This combination was essential for identifying and analysing relevant data, and for developing a predictive tool capable of estimating, at a street-by-street level, the prevalence of customers who are in or at risk of fuel poverty. An existing smart data platform served as the foundation for this development.
The project was delivered in several phases. The first was a data discovery phase, aimed at identifying key behaviours and indicators associated with people experiencing, or at risk of entering, fuel poverty. These indicators were used to assess the availability of relevant datasets and informed the direction of future tool development.
The selected datasets were integrated into a digital tool designed to analyse and present insights on customers estimated to be at risk. A user-friendly interface was also developed, enabling users to explore and interrogate the data with ease.
What makes it innovative
Recent advancements in Artificial Intelligence and Machine Learning mean that tools such as uSmart have recently come to market that can be in this type of project. The techniques applied in this project have not been applied before as they are emerging in the marketplace, enabling the best possible detection and prediction.
The key data source that will be analysed in this project is from the Data Communications Company (DCC) system metadata. This contains information such as when emergency credit is activated, or a prepayment meter is topped up which will be the best quality and most complete secure sharing of the data to date. The release of such data is emerging along with Government guidelines for data protection, data sharing and sector-specific initiatives such as the Energy Data Taskforce.
Social Connect is a truly unique project which we believe is a first of its kind globally, not just in the UK.
What we’re learning
Through this project, we developed a better understanding of the diversity of situations that can cause a household to be at risk of fuel poverty, and how those patterns can be spotted using data. We also learnt more about how households themselves perceive their needs and the use of their data to support them.
Social Connect developed a first-of-its-kind consumer-led method to detect the risk of people falling into fuel poverty and experiencing fuel poverty. The project harnessed the value of smart meter system data to accurately identify the incidence and risk of fuel poverty as part of a wider data integration solution.
Ultimately, the project will help us better target our support services so that more people can receive personalised support.
Share your ideas
If you have an idea that could reshape or revolutionise the way we work, we want to hear from you.
We use cookies to give you the best possible experience when using our website. Many of these cookies are essential to the efficient operation of this site.
