PhD Oral Exam - Rafaela Orenga Panizza, Building Engineering

When studying for a doctoral degree (PhD), candidates submit a thesis that provides a critical review of the current state of knowledge of the thesis subject as well as the student’s own contributions to the subject. The distinguishing criterion of doctoral graduate research is a significant and original contribution to knowledge.

Once accepted, the candidate presents the thesis orally. This oral exam is open to the public.

Abstract

The construction sector is both highly resource-intensive and a major generator of waste. Advancing circular economy practices in the built environment, therefore, requires effective systems for managing construction, renovation, and demolition (CRD) materials. A key prerequisite for such systems is a detailed understanding of the composition and evolution of the existing building stock, which determines the type and quantity of materials that may become available for recovery. While municipal datasets can provide useful insights into building stock dynamics (e.g., demolition activity), they rarely contain standardized, machine-readable information about building composition. This limitation is particularly pronounced for low-rise residential buildings, which dominate urban demolition activity yet often lack accessible digital records.

This research addresses this information gap by developing a framework to enrich urban-scale digital building models with estimates of the materials and components stored in the exterior shells of low-rise residential buildings. The main goal is to enable scalable characterization of building material stocks to support circular economy planning and resource management. The research is structured around four objectives. First, a building shell archetype framework is developed to represent typical exterior envelope assemblies and their component compositions for low-rise residential buildings, enabling component-level circular mining analysis. Second, a computer vision pipeline is created to automatically extract observable building shell attributes, such as roof type, roof material, façade material, and opening counts, from widely available aerial and street-level imagery. Third, an automated integration workflow links geospatial building datasets with imagery-derived attributes and archetype-based assembly definitions, allowing city digital models to be enriched with building-level estimates of component and material quantities. Finally, a scenario-based planning framework is developed to evaluate how demolition activity and building stock characteristics influence the availability of recoverable materials, accounting explicitly for uncertainty in demolition patterns and building selection.

The framework is demonstrated through a neighborhood case study of low-rise residential buildings in Montréal, Canada. By combining archetype-based composition modeling, computer vision-derived building attributes, and geospatial data integration, the research enables scalable estimation of exterior shell material stocks across large urban areas. Scenario analysis further illustrates how this enriched building stock information can support probabilistic planning of circular economy strategies for CRD waste management.

Overall, this work contributes a data-driven methodology for generating composition-level building stock information from widely available data sources, providing a foundation for improved urban-scale circular economy planning in the construction sector.