• J5-60093 - Advances in modelling to bridge theory and practice gaps in energy poverty forecasting
The Client : ( J5-60093 )
Project type: Research projects ARRS
Project duration: 2025 - 2027
  • Description

Energy poverty, defines as a situation in which households cannot access essential energy services and products (European Commission, 2023), is taking an increasingly central position in the European Union (EU) energy policy. Citizen-focused energy systems in the Clean Energy for all Europeans package required Member States to tackle energy poverty in their National Energy and Climate Plans (NECPs). Moreover, the European Green Deal has also required Member States (MS) to reduce energy poverty in the framework of the sustainable energy transition, including considering energy poverty in their Long-Term Renovation Strategies under the Renovation Wave. To date, a rapidly growing body of literature on the topic of energy poverty has made significant progress in developing a robust decision-making framework to mitigate energy poverty per the EU requirements, particularly in formulating the definitions and setting the monitoring indicators, even if the general picture at the EU scale remains highly uneven. Now, further steps are needed, which will contribute to solve issues of the most vulnerable households and individuals who are experiencing a reduction in essential energy services (Brosemer et al., 2020; Carfora et al., 2021; Hoang et al., 2021). As a response, this project aims to support academic and policy efforts to better target interventions by focusing on energy poverty as one of the critical areas in the NECP. Within the context of energy poverty research, it targets forecasting, an area that has remained largely underexplored. Energy poverty forecasting is an integral part of the NECP; therefore, developing algorithms based on the existing theoretical knowledge and policy requirements is imperative. The EU MS are expected to provide quantified targets for each energy poverty objective and introduce a comprehensive timeline to monitor the implementation of energy poverty measures within their NECPs. However, the rapidly growing body of literature on the measurement and monitoring of energy poverty has so far not produced any pan-European model to reliably forecast energy poverty that incorporates its most relevant micro and macro determinants. The goal of this project is to build a model in the form of algorithms, involving machine learning components, that will enable forecasting in EU MS. Its contribution to the field in this context is twofold. First, the project will compile, organise and critically assess available secondary data on socio-demographic, economic, political, geographical and environmental factors that can be used to identify and forecast energy poverty. In this context, the projects second contribution is to develop novel identification and energy poverty forecasting methodologies. Currently, no forecasting methodology would allow a consistent quantification of the NECP targets across EU MS (López-Vargas, 2022). Therefore, the use of machine learning techniques will be paramount to this task. Both contributions ultimate aim is the more efficient implementation of policy measures that shelter consumers from the energy crisis. In addition to the requirement for forecasting in the NECPs, two other facts underscore the importance of this project. The first is our recent publication in Nature Energy (González Garibay, Primc & Slabe-Erker, 2023), where we highlighted the significance of forecasting EP and the urgent need for its development for formulating effective policies in alleviating EP. The publication is an introduction to this project, where we aim to implement what we have written in theory. Second, the importance of our work is also evidenced by the attention of Eurostat, which has been awarding prizes for the most accurate forecasting and transparent methods for various energy-related indicators (such as electricity prices) for the past few years.