The EnergyFlex project has been developing microsimulation tools for generating synthetic housing stocks for local authorities which can be used to explore inequalities in energy efficiency and target homes in need of retrofit. As well as the development of the microsimulation tools a workshop with stakeholders was held to identify a clear agenda for the challenges in residential decarbonisation in the UK and the possible avenues for using microsimulation models such as EnergyFlex to overcome these. As part of the project with stakeholders from government and industry alike have been engaged with to put EnergyFlex to use in supporting decision-making and strategy on residential decarbonisation.
Explaining the science
EnergyFlex takes a microsimulation approach to modelling the energy performance of the housing stock for a given local authority. This allows the model to combine data from multiple different public datasets to create a representative synthetic population of homes and the households that live in them, along with estimates for energy performance of the house itself. The model starts by generating a synthetic population which is then enhanced by sampling energy intensity for each home based on the type and age of housing. Finally a Bayesian calibration step is used to infer further energy performance characteristics of the housing stock.
The aims of this project have involved developing a microsimulation model of the housing stock and the household which live in them that can be used to:
- Simulate retrofit solutions – The synthetic population produced by EnergyFlex can be used to assess and compare the impact of retrofit solutions for homes in a given Local Authority by age and property type.
- National trends analysis – Comparing the outputs from EnergyFlex for all Local Authorities across the country allows for national scale analysis of spatial trends and inequalities in housing energy efficiency which can inform policy and target support for retrofits.
- Targeting fuel poverty – EnergyFlex can be used to target fuel poverty interventions by identifying households in more vulnerable socio-economic circumstances who are also living in homes more likely to be energy inefficient and as a result cost more to heat.
At present there is a distinct lack of local, individual data on the energy performance of the housing stock at a local scale that can be used for analysis and to support policy tailoring and targeting – microsimulation approaches such as that proposed by EnergyFlex provide a solution to this with synthetic yet representative populations of homes and households for a local area.
This project has focused from the start on the practical applications of the microsimulation model developed. Active engagement with a range of stakeholder has been key to this project including:
- Local authorities interested in exploring the energy efficiency of their local housing stock, and the impacts of this on their most vulnerable.
- Policy makers focused on effectively targeting support for households at risk of fuel poverty at a national scale.
- Housing associations interested in targeting their decarbonisation strategies to tackle worst performing housing, as well as identify opportunities to roll out standardised improvements across housing in different areas.
Following a workshop at AI UK, as the project completes its initial phase of development, there have been a number of further steps taken ranging from documentation and dissemination to engagement with a range of public and private stakeholders in the energy and housing domains to explore the use of the tool for strategic planning and policy-making.
This project has involved a close collaboration with the Data & Analytics Facility for National Infrastructure (DAFNI), through which modules of code have been uploaded as standalone executable versions of the model which can be run in the cloud without the need for coding know-how or computing facilities. This will facilitate use of EnergyFlex by a wider audience of interested stakeholders and their own analysts and researchers.