Towards Circular Construction: Material and Component Stock Assessment in Montréal’s Residential Buildings

Abstract

The construction industry is a major consumer of raw materials and a significant contributor to global waste. In Canada, the construction, renovation, and demolition (CRD) sector diverts only 16% of its waste from landfills, underscoring the urgent need for circular economy (CE) practices. This study develops a generalizable and reproducible framework for archetype identification to support CE strategies, with a focus on Montréal, Canada’s second-largest city. We define a new set of exterior shell archetypes for low-rise residential buildings and demonstrate their application in a neighborhood-scale case study. These archetypes enable systematic estimation of material and component stocks, as well as end-of-life recovery flows, across a representative sample of buildings in the Mercier–Hochelaga–Maisonneuve district. Results show that prioritizing reuse can nearly double material recovery compared to conventional sorting and recycling. More broadly, this framework advances engineering design for circular systems by integrating component-level data into reuse strategy assessment and providing a scalable approach for urban circularity.

1. Introduction

The construction industry plays a vital role in shaping our cities but also leaves a heavy environmental footprint—Canada’s construction, renovation, and demolition (CRD) sector alone produces over 4 million tons of waste each year, with minimal diversion from landfills [1]. Addressing this challenge requires rigorous methods to quantify and optimize material flows at scale so that circular economy (CE) systems can be effectively designed and implemented.

The CE, a key application of industrial ecology, encourages smarter resource use through reuse, recycling, and recovery. Yet, applying CE principles in construction remains complex, particularly in the existing built environment where material data is often fragmented or unavailable. In Canada, 89% of CRD waste comes from renovation and demolition, yet 88% could be diverted with better systems [1,2]. Unlocking this potential depends on designing recovery and reuse systems, such as waste management infrastructure and material exchanges, which are informed by accurate building stock data.

Low-rise residential buildings are especially critical in this context. They dominate demolition activity in urban centers, as 97% of demolition permits issued in Montréal between 2022 and 2023 were for low-rise buildings [3]. Yet detailed data on these structures is scarce. Most were built before the digital era, and information is rarely standardized or machine-readable [4]. This lack of data complicates stock assessment but also limits the ability to plan and design future recovery systems.

This study addresses these challenges by developing a methodological framework that supports reuse-oriented strategies through component-level recovery assessments of low-rise residential buildings. Specifically, we (i) propose a generalizable approach for identifying building archetypes, (ii) define representative shell assembly archetypes for Montréal’s post-1940 low-rise residential stock, and (iii) apply them in a neighborhood-scale case study to estimate material and component stocks. By generating the data needed for scenario planning, this framework enables the design of more effective circular systems and provides a scalable basis for advancing urban circularity.

2. Literature Review

Integrating circularity into the existing built environment is hindered by limited building data, especially for pre-digital structures. Quantity takeoff methods help estimate building composition and typically follow bottom-up, top-down, or hybrid approaches. The bottom-up approach starts with individual buildings and scales up, while the top-down method uses aggregated data (e.g., population, building counts) to estimate stock at a broader scale, though with less detail [5,6,7]. The hybrid approach combines both to address data gaps [8]. Within these approaches, three key methods are commonly applied: parametric analysis, machine learning, and material flow analysis (MFA), each offering distinct advantages depending on data availability and scale [9].

Building composition can be assessed through material or component stock, but most studies focus on materials due to broader generalizability. Component stock, however, is crucial for reuse planning and requires more detailed data, making bottom-up and hybrid approaches–often using parametric analysis alone or with MFA–more suitable [10]. Estimating component stock begins with defining archetypes that represent typical building configurations. For instance, Arora et al., 2019 [10], identified public housing archetypes in Singapore to estimate typical component quantities like windows, doors, and fixtures per unit, underscoring the value of archetype-based analysis.

Archetype identification is essential for bottom-up stock and flow studies. This process depends on two main factors: contextual influences, such as local climate, building traditions, and policies, and data availability [9]. Useful data sources include demolition reports, design documents, architectural drawings, and on-site measurements. However, access to such data varies by study location and scale.

Table 1. Overview of the analyzed literature focusing on archetyping for material stock analysis purposes.

Context		Input	Output Type		Reference
Continent	Country	Source of Data	Material Stock	Component Stock
Europe	Netherlands	Demolition reports	✓		[11]
	France	Literature	✓		[12]
	Germany	Design documents	✓		[13]
	Sweden	Design documents	✓		[14]
	Denmark	City data and on-site survey	✓		[15]
	Austria	On-site survey, design documents, and literature	✓		[16]
	Italy	Construction manuals, architectural books, and experts’ opinions	✓		[6]
Asia	China	Literature	✓		[17,18]
	Japan	Building code	✓		[19]
	Singapore	Building code	✓	✓	[10]
	Indonesia	On-site survey	✓		[20]
Oceania	Australia	Land-use and building footprint databases, and experts’ opinions	✓		[21]
South America	Peru	Design documents, on-site survey, experts’ opinions, and the literature	✓		[8]
North America	Canada	Design documents	✓		[22]

summarizes relevant studies, their contexts, and data sources.

As CE practices increasingly emphasize reuse, repair, refurbishment, remanufacturing, and repurposing over recycling, it becomes crucial to reassess the outputs of existing studies. Most focus on material weight, useful for recycling but insufficient for reuse, which requires detailed component stock data.

As shown in Table 1, most studies emphasize material recovery. A notable exception is Arora et al., 2019 [10], which quantifies specific components, such as windows, doors, and fixtures, offering valuable insights for reuse planning. However, this study also presents some limitations. It focuses on only two primary materials–steel and concrete–and a limited set of components. Moreover, archetypes are defined solely based on bedroom count, neglecting architectural diversity within Singapore.

A key factor in component-level studies is access to detailed data. Arora et al., 2019 [10] used standardized drawings from Singapore’s Housing Development Board (HDB), which represents 75% of the country’s residential stock. This uniformity enabled precise component quantification. However, replicating this approach in more diverse, less centralized contexts is difficult due to a lack of standardized design data.

Few studies have expanded component-level analysis beyond limited building elements or applied it in diverse urban contexts–an essential gap for advancing reuse planning in CE. Most archetypes rely on broad attributes like building use or construction year [11,13], which, while useful, do not capture key details. Recent studies highlight the value of more granular predictors, such as roof and exterior wall types, for improving stock estimates [23,24]. Incorporating these details into archetypes is essential for better assessing reuse potential and guiding circular practices.

Finally, the separation of component and material stock analyses further limits the ability to plan realistic, circular interventions. While reuse is prioritized in CE, it is not always feasible due to degradation or regulatory limits. Including both components and materials in stock assessments would support more realistic and comprehensive planning for material recovery.

3. Materials and Methods

This study estimates the component stock of low-rise residential buildings and evaluates the impact of various end-of-life scenarios using a four-phase methodology: Context Analysis, Criteria Definition, Composition Identification, and Recovery Assessment (Figure 1).

3.1. Context Analysis

The first crucial step in estimating building composition on a large scale is performing a context analysis, which provides a foundational understanding of the geographic area of interest. This phase informs all subsequent steps by gathering comprehensive information on two key aspects: the building stock and the existing material and component recovery infrastructure status quo.

To analyze the building stock, the process begins with collecting information from various sources, including construction documents, architectural drawings, specifications, building codes, building images and on-site measurements and inspections. Additionally, existing databases and previous research studies are consulted to gather relevant data on low-rise residential buildings. Following data collection, a thorough analysis of the building stock context is conducted, focusing on historical and current construction practices. This analysis emphasizes commonly used materials and components in key elements such as exterior walls, windows, doors, and roofs. By doing so, the study identifies important factors that have influenced variations in building compositions, as well as typical components and materials used throughout the building stock.

Another critical aspect of the context analysis is understanding the existing process for handling the outflow of building materials and components from the building stock. The objective is to identify how the local recovery process works, including the available types of facilities (e.g., recycling centers, reuse centers), their locations, and the applicable regulations governing material recovery. The data collection for this begins by gathering government reports and policy documents to understand the overall regulatory framework and procedural guidelines for material recovery. Next, an extensive web search is performed to identify both public and private facilities within the study area. This provides an understanding of where these facilities are located and the services they offer. Finally, an in-depth analysis of each material is performed to obtain specific information on its recovery processes. This is achieved through the analysis of governmental reports.

The insights gained from this context analysis directly inform the next phase, where key building criteria are defined to characterize typical groups of low-rise residential buildings within the study area.

3.2. Criteria Definition

The second phase of the methodology aims to identify typical groups of buildings within the selected context, with an emphasis on building criteria related to material composition. This process follows a structured, two-step decision approach to ensure that only relevant criteria are included in the analysis. The first step involves determining the relationship between each criterion and the composition of building components and materials. This begins by compiling a comprehensive list of typical building criteria, followed by a thorough literature review of existing conceptual models for building composition estimation. The review identifies the most commonly used criteria in these models. Each criterion is then evaluated based on its relevance to building composition, with those not directly linked to material composition excluded from further consideration.

In the second step, the remaining criteria are further assessed for their relevance to the specific characteristics and conditions of the selected context. Insights gathered during the Context Analysis phase, particularly from the Building Stock Analysis, are used to evaluate the applicability of each criterion. Criteria that do not accurately reflect the local context are excluded, while those that capture the unique features of buildings in the area are retained. The final selection of criteria ensures they accurately represent the distinct characteristics of buildings within the selected geographic area, providing a solid foundation for the subsequent phases of the study.

These defined criteria are then applied in the next phase to systematically identify and analyze the typical exterior shell compositions that characterize each building group.

3.3. Composition Identification

The third phase aims to determine the exterior shell composition of the typical buildings identified in earlier phases. Based on the ‘Context Analysis’ findings, this phase involves a detailed examination of each building archetype to identify common assembly types. These assembly types are crucial as they represent the structural framework of the typical buildings, serving as a foundation for the detailed analysis of component and material compositions. The first step is to categorize the common assemblies, which represent the typical building elements used in the construction of these buildings. Each identified assembly type then undergoes a thorough analysis to determine the common components and materials used in its construction. This detailed examination provides a comprehensive understanding of the composition of building assemblies and supports the subsequent evaluation of their potential for reuse.

The composition analysis defines both the types and quantities of the components that constitute the building assemblies (bill of components) and the materials that form each component (bill of materials). This output provides the necessary data for the Recovery Assessment phase, where the recovery scenarios are evaluated. By offering a detailed breakdown of both component and material compositions, this phase enables a more detailed understanding of what can happen to these components and materials at the end of life, supporting more effective circularity planning strategies.

3.4. Recovery Assessment

The fourth and final phase, Recovery Assessment, focuses on mapping the potential recovery pathways for building components and materials based on local infrastructure and practices. This phase begins with a hierarchical assessment of the typical building components to determine which ones can be recovered in whole or in part. Recovery strategies considered include reuse, refurbishment, remanufacturing, and repurposing, but for the purpose of this study, all these strategies are categorized under the broader term “reuse”. Components that cannot be reused are further evaluated based on their material composition to determine their eligibility for recycling or energy recovery, using available local technologies and waste management infrastructure. Materials that are unsuitable for either recycling or energy recovery are directed to landfills for disposal.

To quantify these flows at scale, a Material Flow Analysis is conducted using component stock data derived from the building archetypes analyzed in earlier phases of the study. Then, relying on the local knowledge obtained regarding the possible routes for different components and materials, material quantities are allocated to one of four end-of-life pathways: reuse, recycling, energy recovery, or landfill. This analysis is performed across multiple end-of-life scenarios, each representing a different level of material recovery effort, from full disposal to progressively more circular approaches involving recycling and reuse. This flow-based assessment enables a spatially and contextually grounded understanding of the extent to which a city’s building material stock is recoverable within existing local infrastructure. It also helps identify material-specific bottlenecks in the system and supports the development of targeted CE strategies for the built environment.

For the scenario analysis, the term “diverted” refers to all materials that do not enter landfill or direct reuse streams, including recycling, energy recovery, storage, and export. Materials recovered for direct reuse are reported separately under “reuse.”

4. Results

This section presents the key output from the archetyping analysis of low-rise residential buildings in Montréal. It begins by outlining the composition-relevant criteria used to define representative buildings, followed by a characterization of the most common building types within the Montréal context. Finally, the typical composition of exterior shell assemblies is detailed to support downstream assessment of material recovery potential.

4.1. Archetyping Criteria

A key gap in the literature is the overly broad categorization of buildings using limited criteria. This section identifies the most relevant criteria for analyzing building composition and proposes additions based on observed shortcomings. According to the Canadian guide for residential building energy archetyping, buildings are typically characterized by four main categories: general building information (e.g., year of construction, building type) [25,26,27]; geometric characteristics (e.g., footprint, height) [25,27]; thermal characteristics of the envelope (e.g., heat retention performance of the envelope) [28]; and systems and equipment (e.g., heating and hot water systems) [29]. These categories form the foundation of this study, as energy archetypes offer a well-established and structured framework that can be adapted for material and component-level analyses.

However, not all energy modeling criteria are directly applicable when estimating building composition, especially for exterior shell components. For example, thermal characteristics are reframed in this study to reflect the envelope’s material composition rather than its insulation performance. Similarly, systems and equipment are excluded, as they do not contribute to the shell’s physical makeup.

To identify composition-specific criteria, we reviewed conceptual models from bottom-up studies that examine building stock at the neighborhood or city scale [9]. These studies typically rely on data models or expert-informed categories to identify compositional traits. This review yielded 18 relevant criteria spanning general building attributes, geometry, and envelope characteristics. We then cross-analyzed these with energy archetyping criteria to identify overlaps and gaps.

Table 2. Criteria matrix for defining energy archetypes and conceptual models in stock composition estimation (‘→’ = direct match; ‘↝’ = indirect alignment).

Energy → Building Criteria Composition ↓	Building Use	Construction Year	Location	Household Size	Household Type	Building Height	Surface-Volume Ratio	Building Footprint	Number of Stories	Roof Type	Thermal Insulation	Window-to-Wall Ratio
Building use	→
Construction year		→
Location			→
Building structure
Region
Number of dwellings				↝	↝
Building configuration					↝
Number of bedrooms				↝	↝
Building height						→
Gross floor area								↝	↝
Building footprint								→
Number of stories above ground									↝
Number of stories below ground									↝
Volume							↝
Exterior wall material											↝
Roof material											↝

presents an interactive matrix showing the synergies between both sets of criteria, highlighting both direct matches and indirect alignments. Detailed sources for the criteria are provided in Appendix A.

Several foundational criteria–such as building use, year of construction, location, building footprint, and building height–appear in both domains (direct matches). In energy modeling, these factors influence thermal performance expectations; in composition studies, they suggest likely construction techniques and material choices, though more detailed information is needed to capture the full range of building assemblies.

Other criteria show indirect alignment, reflecting the different objectives of each field. For instance, thermal insulation–typically expressed through R-values–is central in energy analysis, while composition estimation focuses on the actual materials used–hence, criteria such as exterior wall material and roof material are prioritized. Finally, the comparison also revealed underutilized criteria that could strengthen composition models. Geometric attributes like roof type (e.g., gable, flat) signal structural differences, while the window-to-wall ratio offers insight into the proportion and type of window assemblies–components often overlooked.

This synthesis validates existing archetyping methods while introducing new composition-relevant attributes. The result is a consolidated list of 20 criteria that combines established practices with enhancements tailored to material estimation. This list supports the methodological framework presented in the next sections and offers a transferable reference for studies beyond Montréal.

4.2. Typical Low-Rise Residential Buildings

With the comprehensive list of composition-relevant criteria obtained in the previous section, the next step is to identify which criteria are pertinent to our scope and exclude the irrelevant ones. Focusing on low-rise residential buildings constructed after 1940 in Montréal naturally narrows the range of suitable criteria and influences their interpretation. Several key criteria, including building use, location, and construction year, are effectively fixed by this scope.

Building use is limited to residential buildings, removing variability associated with commercial or industrial structures and allowing for a more in-depth understanding of possible variability in composition. Similarly, location is fixed to Montréal, which implies region-specific construction practices and material choices, both of which are highly relevant to compositional modeling [24,30]. The year of construction is also restricted to buildings built after 1940, a period characterized by more standardized building methods. This cutoff is particularly relevant in the context of Montréal, where buildings constructed prior to 1940 are often classified as historical heritage and are therefore subject to strict preservation regulations that limit demolition or substantial alteration. As such, they are excluded from this study [31].

In addition to these, other attributes are indirectly fixed. Most notably, building structure is considered to be predominantly wood-frame, which is the standard construction method for post-1940 low-rise residential buildings in Montréal [31,32]. The number of floors below ground is similarly constrained; most buildings in this category feature no more than one level below ground, reducing variability in subsurface materials and structural design.

Certain criteria are excluded entirely due to their limited utility in this context. For instance, the number of bedrooms, while sometimes used as a proxy for building size [22], offers a less precise measure than geometric or dwelling-based criteria. Moreover, considering that nearly half of Montréal’s low-rise residential stock consists of non-single-family dwellings [31], bedroom count does not reliably correlate with envelope composition and is therefore omitted from the framework.

Following these exclusions and considerations, eleven criteria remain central to our analysis. These can be grouped into either qualitative or quantitative types. On the qualitative side, relevant criteria include building configuration, roof material, façade material, and roof type. Building configuration is especially important when firewalls are present, as their construction differs significantly from that of exterior walls [22]. Common configurations in Montréal include detached, semi-detached, and row houses, each of which influences envelope composition in different ways [33]. Roof type, which varies from flat to sloped designs, affects framing requirements and overall structural makeup. Façade and roof materials are directly linked to the mass and type of construction materials present in the envelope and are thus essential to any compositional assessment [23,24].

Quantitative criteria include the number of dwellings, number of windows, building volume, number of stories, building height, gross floor area, and building footprint. The number of dwellings distinguishes between single-family homes, duplexes, triplexes, and small multiplexes, all of which affect the expected size and layout of the building envelope [31,34]. The number of windows informs the number of window systems present in the building envelope. Building volume provides a general indication of the structure’s size by integrating both its height and footprint. Vertical dimensions are captured through building height and number of stories, while horizontal extent is described through gross floor area and footprint.

Together, these criteria form the basis of the archetyping framework applied in this study, which is specifically tailored to the context of Montréal’s post-1940 low-rise residential buildings. While the selection process draws from a broader set of composition-relevant criteria (as outlined in Section 4.1, Archetyping Criteria), the final framework reflects local construction practices, regulatory constraints, and building characteristics. This is not intended as a universally scalable model, but rather as a methodological tool suited to estimating material composition in this specific urban and historical setting. The complete framework is summarized in Figure 2.

4.3. Shell Composition

This study focuses on low-rise residential buildings in Montréal constructed after 1940 that share several key characteristics: they feature one basement floor, rise no more than four stories above ground, and use wood-frame construction. Classification of these buildings is based on the number of dwellings, building configuration, and exterior shell materials, specifically those of the façade and roof. Although building size indicators contribute to classification, they are primarily used during the calculation phase.

To define representative shell archetypes, we drew on multiple data sources, including visual surveys of the existing building stock (Figure 3), the National Building Code of Canada (1941–2020), historical architectural records, RS Means construction cost data, relevant literature, and over 200 Central Mortgage and Housing Corporation (CMHC) house blueprints dating from 1951 to 1979 [32,35,36,37,38,39]. These sources informed the selection of typical façades, roofs, and opening assemblies, ensuring the resulting archetypes reflect both the historical evolution and prevailing construction practices of Montréal’s post-1940 low-rise housing stock.

Given the dominant use of wood framing in these buildings, particularly in the exterior shell, two essential factors in estimating material composition are the assembly size and the exterior materials. Due to the overall homogeneity in construction methods within this typology, a standard set of shell assemblies was identified, encompassing three main groups: façades, roofs, and openings (i.e., windows and doors). Drawing on national building codes, cost data, and architectural records, five typical façade types and eight roof types were defined. These assemblies, some of which are shown in Figure 3, include structural components such as wood framing and sheathing, where applicable, along with a range of exterior covering materials.

The façade assemblies are classified into five types, each defined primarily by its exterior cladding. These include: (1) brick veneer walls, consisting of a wood-stud frame and a brick exterior; (2) stone veneer walls, similarly structured but clad in stone; (3) siding façades, with wood studs, plywood sheathing, and a siding finish; (4) stucco façades, comprising wood studs, sheathing, and a stucco exterior; and (5) firewalls, constructed from wood studs with gypsum board on both sides but no exterior finish, serving primarily as fire barriers. All wood-framed walls are assumed to use 16-inch stud spacing, a regional standard [32,35]. Additional assembly details are provided in

Table A2. Details of archetypal façade assemblies.

Façade System Type	Component (Unit of Size/10 ft2)	Material (lb/10 ft2)	Assumptions
Brick veneer wall	Framing—2″ × 4″ studs (75 linear ft)	Wood (97. 5 lb)	Studs have 16-inch spacing, which is standard in the Montréal area [32,35]. Quantity of studs estimation does not consider corners of the wall and is calculated by dividing the length of the wall by the stud spacing. A linear foot of a 2″ × 4″ stud weighs 1.3 lb [62]. 6.5 bricks are needed to cover 1 ft2 and each brick weighs 3.88 lb [63]. 17 lb of stone cladding covers 1 ft2 of wall [64]. A ft2 of a 7/8″ plywood board weighs 2.6 lb [65]. A ft2 of a vinyl siding weighs 0.44 lb [66]. Stucco mix weighs about 3.28 lb per ft2 [67]. A ft2 of a ½″ gypsum board weighs 1.6 lb [68].
	Sheathing—7/8″ plywood board (10 ft2)	Engineered wood (26 lb)
	Wall cover—Bricks (65 units)	Bricks (252.2 lb)
Stone veneer wall	Framing—2″ × 4″ studs (75 linear ft)	Wood (97. 5 lb)
	Sheathing—7/8″ plywood board (10 ft2)	Engineered wood (26 lb)
	Wall cover—Stone (10 ft2)	Stone (170 lb)
Wall siding wall	Framing—2″ × 4″ studs (75 linear ft)	Wood (97.5 lb)
	Sheathing—7/8″ plywood board (10 ft2)	Engineered wood (26 lb)
	Wall cover—Vinyl siding panel (10 ft2)	Vinyl (4.44 lb)
Stucco wall	Framing—2″ × 4″ studs (75 linear ft)	Wood (97.5 lb)
	Sheathing—7/8″ plywood board (10 ft2)	Engineered wood (26 lb)
	Wall cover—Stucco (10 ft2)	Stucco mix (32.8 lb)
Firewall	Framing—2″ × 4″ studs (75 linear ft)	Wood (97. 5 lb)
	Sheathing—2 × 1/2″ gypsum boards (20 ft2)	Gypsum (32 lb)

.

Roof assemblies are grouped into eight types, differentiated by their exterior covering. Regardless of pitch, all roofs are composed of wood rafters with plywood sheathing. Covering materials include asphalt, slate, or wood shingles; metal or clay tiles; and built-up roofing systems, with or without gravel. These roof types reflect the materials and construction techniques commonly used in Montréal’s post-1940 residential buildings and serve as the basis for estimating recoverable material quantities. Further technical descriptions are available in

Table A3. Details of archetypal roof assemblies.

Roof System Type	Component (Unit of Size/10 ft2)	Material (lb/10 ft2)	Assumptions
Sloped roof (asphalt)	Framing—2″ × 10″ rafters (81.225 linear ft)	Wood (268 lb)	Rafters have 16-inch spacing, which is standard in the Montréal area [32,35]. Roof is assumed to be at a 22.6° angle (conversion factor is 1.083 [69]). A linear foot of a 2″ × 10″ rafter weighs 3.3 lb [62]. A ft2 for 5/16″ plywood board weighs 1 lb [65]. Shingles are counted in standard bundles that cover 33 ft2 of roof area. 1 bundle of asphalt shingles weighs 60 lb [70]. A ft2 of slate tiles weighs 10 lb [71]. A ft2 of wood shingles/shakes weighs 3 lb [71]. A ft2 of metal roofing weighs 1.4 lb [72]. 2 clay tiles are needed to cover 1 ft2 and they weigh 9 lb [71,73].
	Sheathing—5/16″ Plywood board (10.83 ft2)	Engineered wood (10.83 lb)
	Roof cover—Asphalt shingles (1/3 of a bundle)	Asphalt (20 lb)
Sloped roof (slate)	Framing—2″ × 10″ rafters (81.225 linear ft)	Wood (268 lb)
	Sheathing—5/16″ Plywood board (10.83 ft2)	Engineered wood (10.83 lb)
	Roof cover—Slate shingles (10.83 ft2)	Slate (108.3 lb)
Sloped roof (wood)	Framing—2″ × 10″ rafters (81.225 linear ft)	Wood (268 lb)
	Sheathing—5/16″ Plywood board (10.83 ft2)	Engineered wood (43.32 lb)
	Roof cover—Wood shingles (10.83 ft2)
Sloped roof (metal)	Framing—2″ × 10″ rafters (81.225 linear ft)	Wood (268 lb)
	Sheathing—5/16″ Plywood board (10.83 ft2)	Engineered wood (10.83 lb)
	Roof cover—Metal tiles (10.83 ft2)	Metal (15.162 lb)
Sloped roof (clay)	Framing—2″ × 10″ rafters (81.225 linear ft)	Wood (268 lb)
	Sheathing—5/16″ Plywood board (10.83 ft2)	Engineered wood (10.83 lb)
	Roof cover—Clay tiles (53 tiles)	Clay (97.47 lb)
Flat roof (built-up)	Framing—2″ × 10″ rafters (75 linear ft)	Wood (247.5 lb)	Rafters have 16-inch spacing, which is standard in the Montréal area [32,35]. A linear foot of a 2″ × 10″ rafter weighs 3.3 lb [62]. A ft2 for 5/16″ plywood board weighs 1 lb [65]. Ethylene propylene diene monomer (EPDM) weighs 0.32 lb per ft2 [74]. Built-up roofs with gravel should have 50 mm of gravel ballast and it weighs 16.38 lb/ft2) [75].
	Sheathing—5/16″ Plywood board (10 ft2)	Engineered wood (10 lb)
	Roof cover—EPDM (10 ft2)	EPDM (14.75 lb)
Flat roof (built-up with gravel)	Framing—2″ × 10″ rafters (75 linear ft)	Wood (247.5 lb)
	Sheathing—5/16″ Plywood board (10 ft2)	Engineered wood (10 lb)
	Roof cover—EPDM (10 ft2)	EPDM (14.75 lb)
	Roof cover—Gravel (10 ft2)	Gravel (163.8 lb)

.

Opening assemblies encompass both windows and doors, each varying in size, material, and configuration. Window frames are typically made from aluminum, steel, fiberglass, PVC (vinyl), or wood [36,37]. Due to poor thermal performance, aluminum is generally avoided in Montréal’s cold climate, while fiberglass–despite its thermal advantages–has not seen widespread adoption because of earlier manufacturing inconsistencies. PVC is now dominant for its cost-effectiveness, low maintenance, and insulation properties, while wood remains common in heritage buildings and among owners who prefer traditional aesthetics. This study focuses on wooden and PVC-framed windows, as they are most representative. Each window system includes the frame, casing, and glazing, with full component details provided in

Table A4. Details of archetypal openings.

Opening System Type	Component (Unit of Size/Unit)	Material (lb/Unit)	Assumptions
PVC window	Vinyl frame (47.2 inches × 47.2 inches)	Vinyl (23.13 lb)	Standard window unit has the following dimensions: 47.2 inches × 47.2 inches [76]. The glass is double-glazed and weighs 3.2 lb/ft2 [77]. The vinyl window frame + casing weigh 1.495 lb/ft2 [77]. Vinyl casing is assumed to be ~10% of frame + casing weight: 2.4 lb. The wooden window frame + casing weigh 0.7209 BDFT/ft2 [77] and 1 BDFT weighs 1.95 lb [62]. Wood casing is assumed to be ~10% of frame + casing weight: 2.2 lb.
	Vinyl casing (47.2 inches × 47.2 inches)
	Double glazing (47.2 inches × 47.2 inches)	Glass (49.5 lb)
Wood window	Wood frame (47.2 inches × 47.2 inches)	Wood (21.75 lb)
	Wood casing (47.2 inches × 47.2 inches)
	Double glazing (47.2 inches × 47.2 inches)	Glass (15.87 lb)
Wood door	Wood frame (36 inches × 84 inches)	Wood (98.78 lb)	Steel door without polyurethene insulation weighs 92 lb and their size is: 80 inches of height; 32 inches of width; and 1.75 inches of depth [78]. Steel frame weighs 45 lb and measures: 80 inches of height; 32 inches of width [79]. Wood door and frame weigh 98.78 lb and their size is: 80 inches of height; 32 inches of width; and 1.75 inches of depth [80]. Wood frame is assumed to be ~10% of total door weight: 10 lb. Rigid polyurethane density is 2 lb/ft3 [81].
	Wood door (32 inches × 80 inches)
Steel door	Steel frame (36 inches × 84 inches)	Steel (92 lb)
	Steel door (32 inches × 80 inches)
		Rigid polyurethane foam (5.18 lb)

.

Exterior doors are either solid wood, which is prevalent in older buildings, or have insulated cores encased in metal, fiberglass, or wood shells. Insulated steel doors are commonly used for their enhanced thermal performance [38]. In this study, solid wood and insulated steel are considered representative of exterior door types in Montréal’s low-rise housing stock. Each door system includes both the door leaf and its frame. A full breakdown of their materials and components can be found in Table A4.

Together, these shell assembly archetypes–covering façades, roofs, and openings–capture the typical construction features of Montréal’s post-1940 low-rise residential buildings. All archetype–component–material relationships, including unit weights, dimensional assumptions, and per-unit material intensities, are provided in the Appendix A with both imperial and SI units to enable replication of city-scale stock and flow calculations.

These assembly archetypes were defined based on typical construction practices for post-1940 low-rise residential buildings in Montréal, reflecting regional building codes, material availability, and climatic conditions. The construction assumptions used for each assembly are summarized in Table A2, Table A3 and Table A4. While these parameters capture Montréal’s building stock, their applicability to other regions or construction periods would require verification and possible recalibration. In the following section, the archetypes are used in a case study to estimate material composition and evaluate the potential for component recovery in the local context.

5. Practical Application

With the building assembly archetypes defined, this section examines Montréal’s current recovery infrastructure and its capacity for reuse, recycling, and disposal. It concludes with a neighborhood-scale case study illustrating the practical benefits of obtaining composition archetypes.

5.1. Recovery Pathways

CRD waste management in Montréal primarily involves two types of waste handlers and sorting facilities. The first are eco-centers–municipally operated sites designed to increase landfill diversion by sorting CRD waste and directing recyclable or reusable materials to appropriate streams. Montréal residents, small businesses, and non-profit organizations can access any of the city’s seven eco-centers [40]. The second type includes private companies that offer waste collection, container rental, sorting, and transportation services. These companies dominate the sector, handling approximately 71% of the CRD waste sorted in Quebec [41]. In addition to these primary pathways, CRD waste may also enter the industrial, commercial, and institutional waste stream or the municipal solid waste system, although these flows are less systematically tracked [41]. While precise quantities remain unknown, second-hand building components are available through at least three reuse centers in Montréal (Figure 4), offering additional informal recovery routes.

Although eco-centers aim to sort CRD waste for reuse, recycling, and energy recovery, available data on their output destinations typically cover only recycling, waste-to-energy, storage, export, or landfill. To better assess reuse potential, this study also examined second-hand markets. Field visits to two local reuse stores–Habitat for Humanity’s ReUse Store [42] and RÉCO [43]–identified three reusable systems and twelve reusable components from typical assemblies. These include systems such as wood-framed windows and doors, and components like wood studs, rafters, steel doors, and glazing.

Even though many elements may theoretically be reusable, actual reuse depends on their condition and the methods used during disassembly [44,45]. Fully demolished buildings often yield fewer reusable components than those carefully deconstructed. For elements that cannot be reused–either due to poor condition or market limitations–their materials must be evaluated for recycling or energy recovery. The potential for recycling and energy recovery was assessed using data from Québec’s 2021 residual materials management review, as summarized in

Table 3. Status quo of material pathways for Montréal’s CRD waste [41].

Material	Recycling	Energy Recovery	Storage	Export	Landfill
Dimensional wood	12.9%	34.1%	✗	3.5%	49.5%
Engineered wood	✗	1.4%	✗	✗	98.6%
Clay	✗	✗	✗	✗	100%
Stone (aggregate)	39.7%	✗	6.8%	✗	53.5%
Vinyl	✗	✗	✗	✗	100%
Stucco mix (fine residue)	5.3%	✗	10.5%	84.2%	✗
Gypsum	2%	✗	✗	✗	98%
Asphalt	1.3%	35.5%	✗	6.6%	56.6%
Slate	✗	✗	✗	✗	100%
Metal	41.4%	✗	✗	✗	58.6%
EPDM	✗	✗	✗	✗	100%
Gravel (aggregate)	39.7%	✗	6.8%	✗	53.5%
Glass	62.5%	✗	✗	✗	37.5%
Steel	41.4%	✗	✗	✗	58.6%

. This table reports the fate of materials that enter sorting facilities under current flows. Materials directed to reuse, as modeled in Scenario 3, are prioritized before any sorting occurs and therefore fall outside the baseline statistics in Table 3.

Despite the availability of sorting facilities, only 53% of Québec’s CRD waste is processed by them, while approximately 47% is landfilled directly, highlighting a continued reliance on disposal over recovery. Even among sorted materials, diversion remains limited: a significant share still ends up in landfills. From a CE perspective, material reuse is the most desirable outcome, followed by recycling, energy recovery, and finally, landfill. Table 3 outlines the actual destination of typical materials that arrive at Montréal eco-centers based on their mass, showing the challenges of moving up this hierarchy.

5.2. Neighborhood Case Study

With the defined exterior shell archetypes and recovery pathways identified, this section demonstrates their combined application to assess the neighborhood-scale impact of prioritizing reuse at the end-of-life in the Montréal context. The Montréal neighborhood of Mercier–Hochelaga-Maisonneuve was selected as the case study due to its predominantly residential character and the prevalence of low-rise buildings (four stories or fewer) [46]. Moreover, the area has undergone considerable redevelopment in recent years, resulting in a higher number of issued demolition permits compared to other boroughs, particularly in 2023 [3,47].

This case study evaluates the impact of different end-of-life recovery scenarios on buildings demolished in Mercier–Hochelaga-Maisonneuve during 2023 by considering three scenarios. These scenarios aim to represent potential end-of-life outcomes for the generated CRD waste in the analyzed context: Scenario 1—‘no diversion’, in which all materials are sent directly to landfill with no recovery effort; Scenario 2—‘diversion’, where materials are directed to sorting and recycling facilities to extend their life cycle; and Scenario 3—‘reuse + diversion’, in which reusable materials are recovered first, and the remaining materials are sent to sorting facilities. These scenarios reflect increasing levels of material recovery and are used to assess how local infrastructure and practices influence the fate of building materials at end of life.

To support this analysis, several data sources were used to assist the material and component flow estimations resulting from demolition activities. First, local demolition permit records provided information on the number and location of demolished buildings–12 in total within the study area. Second, building footprint data, combined with the number of stories (identified via Google Street View images dated from 2022 or earlier), enabled estimation of the total roof and exterior wall areas. This yielded a total façade area of 28,384 ft² and a roof area of 13,622 ft² [48].

Google Street View images were further used to identify the number of available windows and doors, as well as visible characteristics of the building façade and roof based on the appropriate assembly archetypes defined in this study, allowing for the calculation of both component and material quantities for each demolished building shell. The demolished buildings reflected a variety of exterior assembly types, including brick veneer, stone veneer, wall siding, and firewall facades, as well as asphalt shingle and built-up flat roofs. Across all buildings, 25 exterior doors and 62 exterior windows were observed. In the absence of detailed data on window and door framing materials, it was assumed that doors were evenly split between wood and steel frames, and windows between wood and PVC frames. To evaluate the sensitivity of this assumption, additional tests were performed varying the opening-material distribution between 0/100 and 100/0 for both windows and doors. The resulting variation in total material quantities was below 2%, as indicated by the error bars in Figure 5, confirming that this assumption has a negligible impact on the overall scenario outcomes. The distribution of the demolition sites is shown in Figure 5.

Using the estimated material composition of the demolished buildings, the three end-of-life scenarios were assessed to evaluate the status of the current infrastructure for recovering building shell components and identify key opportunities for improvement. The results presented in Figure 5 include Sankey diagrams for each scenario and a comparative summary of the total waste mass (categorized as landfilled, diverted, or reused).

Scenario 1 (no diversion) reflects a common default pathway when no sorting or recovery infrastructure is engaged, highlighting the burden of having all 398 t (877,393 lb) of waste sent to the landfill. Scenario 2 (diversion) introduces material separation via sorting facilities, resulting in a reduction in landfilled waste. This scenario results in 356 t (782,145 lb) sent to landfill and 43 t (95,248 lb) diverted to recycling processes, with no materials recovered for direct reuse. Despite improved diversion, substantial material loss still occurs due to limitations in sorting precision, market demand, and available recycling technologies. Lastly, Scenario 3 (reuse + diversion) demonstrates the greatest potential for waste reduction by prioritizing the recovery of reusable components. In this scenario, 100 t (219,633 lb) are landfilled, 23 t (51,183 lb) are sent to recycling facilities, and 276 t (607,313 lb) of materials are recovered for direct reuse. This represents a nearly six-fold increase in reused material compared to the ‘diversion’ scenario and highlights the potential of reuse strategies to reduce landfill contributions significantly.

As anticipated, the level of diverted waste increases with the intensity of recovery efforts, from minimal intervention in Scenario 1 to significant material reuse in Scenario 3. Importantly, Scenario 3 also highlights the untapped potential of certain components, which, if supported by improved recovery practices and enabling policies, could become integral to a circular construction material stream. These findings help in answering critical questions for policymakers and waste management stakeholders aiming to support a circular construction sector in Montréal: (a) Do we have sufficient infrastructure capacity to support reuse practices at scale? (b) Given the current infrastructure’s recovery capabilities, what targeted incentives should be implemented to promote the recovery of materials that are already feasible to divert or reuse? (c) When planning new facilities or processes, which components and materials should be prioritized for recovery?

It is important to acknowledge that the present analysis does not capture all sources of waste generation. For instance, additional losses may occur during the dismantling, transport, or on-site processing of reusable components, depending on the deconstruction practices adopted. As such, the recovery rates applied here likely represent optimistic upper bounds. Nonetheless, given the magnitude of the potential reductions observed, these findings still demonstrate the significant opportunities that reuse-oriented strategies offer at the neighborhood scale. Future work should further examine how variations in deconstruction techniques and on-site handling practices influence the effective recovery and waste generation of reusable components.

6. Conclusions

This study developed a methodology to estimate the component and material stock of low-rise residential building shells in Montréal, enabling the assessment of three end-of-life recovery scenarios and their impacts on waste generation and circularity outcomes. By defining building archetypes and their typical façade and roof assemblies, the research supported material composition estimates and the evaluation of recovery pathways: direct landfill disposal (Scenario 1), diversion through sorting facilities (Scenario 2), and a combined reuse–diversion approach (Scenario 3).

The Mercier–Hochelaga-Maisonneuve case study demonstrated the practical implications of these scenarios. An analysis of 12 demolished low-rise buildings showed that prioritizing reuse can nearly double recovered material volumes compared to the conventional sorting path alone. Components made of dimensional wood, engineered wood, glass, and vinyl emerged as key reuse opportunities, especially where recovery infrastructure already exists. These findings highlight the value of leveraging current recovery capabilities and implementing targeted policies to support the reuse of materials already feasible to divert.

This study presents three key contributions. First, it introduces a generalizable framework for identifying and categorizing building archetypes using composition-relevant criteria, which can be adapted to other geographic contexts. Second, it defines a new set of exterior shell archetypes specific to Montréal’s low-rise housing, enabling city-scale estimation of component and material stocks. Third, it demonstrates how these archetypes can be applied to scenario planning for end-of-life recovery, providing evidence to support reuse-oriented circular system design.

Several limitations remain. Archetype definitions were based on fragmented historical data, primarily sourced from Montréal’s low-rise housing stock. While this ensures contextual relevance, it constrains the geographic extent and generalizability of results beyond this context. The analysis focused on major exterior components, excluding finer details such as fasteners, air/vapor barriers, and insulation. Assumptions regarding component recoverability were idealized: the reuse scenario considered full recoverability under optimal demolition and handling conditions, without accounting for potential degradation, contamination, or design-related constraints that may reduce real-world yields. Although the approach enables more granular assessments than typically found in the literature, some uncertainty persists due to variability in construction practices and the limited availability of consistent building composition data. A persistent challenge—both in this study and the field—is the lack of accessible, comprehensive data on existing buildings, which limits the validation of model results.

Future research should expand the scope to include other building types and interior or non-structural components. Empirical data from on-site demolition audits and recovery operations would help refine recovery estimates. Future research should also explore component-specific recovery constraints, such as disassembly requirements, material degradation, and market availability, to translate scenario results into actionable guidance for recovery planning. Furthermore, integrating computer vision and machine learning could support automated identification of assembly types and building dimensions at scale, enhancing both accuracy and applicability for urban circularity planning.