Supporting fuel-poor households through data integration and AI
This project developed uZero, a tool combining UK-wide smart meter data with other data and artificial intelligence to identify households that are at high risk of fuel poverty.
£599,483 (across two phases)
Summary: impacts and findings
The project created an application to better identify households in or at risk of fuel poverty, by using smart meter data and integrating it with other key data sources in a detailed mapping tool.
All of the organisations involved in the project shared that they continue to face serious strategic and operational barriers that this tool would help them overcome.
Project aims and approach
Homes account for 20% of UK emissions. There are 29 million homes in need of retrofitting to improve energy performance, and an estimated 2.4m households in fuel poverty.
For the first time, this project aimed to combine UK-wide anonymised smart meter system metadata with a wide range of other data sources and artificial intelligence, to create a new way to identify households in fuel poverty and those that would most benefit from energy efficiency programmes such as retrofitting.
The aim is to support planning that can maximise progress toward Net Zero whilst reducing fuel poverty, and support better use of data, advanced statistics and machine learning in delivering benefits to the fuel-poor.
In Phase 1, the project:
- Accessed anonymised smart meter metadata from 11 million meters through the Data Communications Company, which is responsible for Britain’s smart meter communications network
- Showed that combining this with other data can uncover patterns of behaviour and provide insights on fuel poverty current approaches, including helping to identify fuel poverty risk
- Moved forward thinking on ethical use of data, working with Ofgem and UCL
- Worked with the Connected Places Catapult to engage with key stakeholders and prospective customers such as local authorities, managing agents, energy companies and charities
- Built an alpha prototype application, uZero, based on UrbanTide’s uSmart data integration platform.
In Phase 2, the team developed the uZero tool further, including incorporating data from the Smart Energy Research Laboratory at UCL which improved accuracy in identifying fuel poverty risk. The tool progressed through alpha and beta testing, including trials with local authorities using the tool to help inform planning of large-scale retrofit programmes. UrbanTide worked with the Greater South East Net Zero Hub to facilitate this.
You can watch a detailed ‘show and tell’ presentation and demonstration of the project, recorded in June 2022.
The show and tell from Phase 1 of the project, recorded in July 2021, is also available here.
April 2021 to August 2022
Achievements and barriers
UrbanTide has developed its application, uZero, to the stage of market-readiness, and it was publicly launched in November 2022.
While uZero use grows, the insights gained in the project are also informing a new UrbanTide project, uEco. This builds upon the foundations of uZero and will seek to develop new identification models and test new prototypes that can pinpoint households that would most benefit from energy efficiency programmes in UK and internationally.