Abstract :
[en] Accelerating the energy renovation of existing residential buildings is essential to achieving Europe's long-term climate goals. This renovation is even more important to ensure a certain degree of energy independence for the continent. As a result, several actions have been taken by EU public authorities, but without significantly accelerating the rate of renovation. This doctoral thesis fits into this context. It examines the dynamics of renovation in Wallonia's residential building stock through a multidimensional approach that integrates the technical characteristics of buildings and the socio-economic profiles of occupants to recommend more targeted and effective actions in favor of energy renovation.
Chapter 3.1 characterizes Wallonia's residential housing stock through an analysis of the correlations between the technical attributes of dwellings and the socio-economic data of their occupants. Using comprehensive databases on buildings and inhabitants, statistical analyses identify the main determinants of energy performance and highlight the importance of considering socio-economic categories such as low-income households and property ownership. The study focuses on targeted renovation strategies for specific types of buildings, particularly semi-detached houses, where unfavorable technical and socio-economic factors converge.
Chapter 3.2 builds on these foundations by grouping together archetypal "building-occupant" pairs to shed light on energy renovation strategies. Using ordinal logistic analysis and beta regression followed by clustering using the K-means method, the study identifies representative typologies that reflect the diversity of Wallonia's housing stock. Among the main conclusions are the presence of low-energy-performance houses occupied by low-income households (more than 17% of the housing stock) and, unexpectedly, low-performance houses occupied by high-income households (more than 11%). These typologies provide actionable information for policymakers to implement targeted and scaled interventions and demonstrate the value of integrating heterogeneous datasets.
Chapter 3.3 is based on a review of the literature on InfraRed Thermography (IRT) analyzed considering the general objectives of the thesis. This literature review highlights the advantages and disadvantages of IRT for analyzing the energy performance of buildings, and for quickly and inexpensively estimating the energy losses of a large group of buildings. The study highlights the potential of this technique, not only to improve knowledge of the energy performance of homes, but also to contextualize this performance in relation to the behavior of the occupants.
Chapter 3.4 presents a case study using mixed methods in a historic neighborhood of Liège, combining household surveys, IRT, and participatory focus groups. This chapter examines how the technical conditions of buildings and the socioeconomic characteristics of occupants jointly influence renovation decisions. The results reveal that socio-economic factors, including income, occupancy status, and energy culture, often outweigh technical considerations. Thermography is effective in identifying hidden heat losses and raising occupant awareness, but information alone is not enough; ongoing education and personalized engagement are essential to influence renovation choices. The study also highlights the specific challenges associated with renting and co-ownership, underscoring the importance of policies tailored to the local context.
In general, this thesis provides details on the statistical relationship between technical data on housing and socio-economic data on occupants. Two sets of housing/inhabitant clusters representative of the Walloon region have been constructed. The distribution of these clusters across the Walloon territory also makes it possible to provide recommendations tailored to specific areas. IRT also contributes to improving knowledge of building stock. All this data provides a better understanding of renovation behavior and market inertia. Based on these results, several recommendations can be made to the various levels of government. Overall, this thesis provides a set of recommendations to guide energy renovation policies in Wallonia. By combining statistical modeling and participatory field research, it offers both scientific knowledge and practical recommendations for improving the efficiency, fairness, and context sensitivity of residential energy renovation initiatives. The combined study of the technical characteristics of housing and the socioeconomic levels of residents demonstrates its value, even if a third, more specifically behavioral axis is gradually emerging.
