Numerical investigation of a novel steel connection for panelized modular houses

Indigenous communities in Canada, particularly those in remote areas, face a persistent shortage of adequate housing. Modular construction offers a potential solution, yet challenges related to transportation and lifting limit its widespread adoption. This study proposes an innovative steel bolted connection using hollow structural sections (HSS) to improve constructability and performance in modular housing. The connection was experimentally tested and validated using three-dimensional finite element models. A parametric study on one- and two-dimensional prototypes examined the influence of stiffeners, bolt arrangement, bolt number, and plate thickness on the connection’s structural performance. The results showed that the AR1.5 bolt arrangement increased capacity through early bolt bearing but reduced ultimate rotation by 50%, whereas the AR0.6 arrangement shifted failure to the column due to local buckling. Increasing plate thickness from 10 mm to 15 mm increased capacity by up to 7% and ductility by 11%, while increasing the number of bolts from six to ten improved capacity by up to 22%, depending on the arrangement. The addition of a 10-mm stiffener reduced ultimate rotation by approximately 60% due to local buckling. These findings highlight the critical role of bolt configuration and reinforcement techniques in optimizing both strength and deformation capacity, providing guidance for the design of efficient and durable modular housing connections.

1 Introduction

Canada is currently suffering a severe housing crisis due to a persistent shortage and increasing homelessness levels, particularly within Indigenous communities. A 2019 report from the United Nations General Assembly on adequate housing emphasized the urgent need for governments to address the poor housing conditions of Indigenous peoples, regardless of whether they reside on reserves or in remote regions (United Nations, 2019). Those reports called for immediate and comprehensive action to improve living conditions and ensure that all Canadians can access safe housing (Schneider, 1992; Canada Office of the Auditor General, 2003; Indigenous Services Canada, 2018). The housing challenges faced by remote Indigenous communities in Canada can be categorized into three main groups related to (i) material, (ii) climate, and (iii) construction challenges (Elhadary et al., 2024b; 2024a). Materials-related challenges include issues such as mold and fire hazards (von Stackelberg, 2019; Garis, 2021). The problem of mold growth often begins before construction even starts, as building materials like wood can become damp during transportation on seasonal ice roads or due to improper storage under harsh weather conditions (Patterson and Dyck, 2015; von Stackelberg, 2019). These conditions are exacerbated by reliance on traditional, labor-intensive building methods that extend construction timelines into the winter months, further exposing materials to moisture and degradation. In addition, fire safety represents another major concern. Many remote and northern communities rely heavily on wood for heating yet lack sufficient firefighting services or emergency response infrastructure, leading to devastating house fires that have resulted in loss of homes and lives (Garis, 2021; National Indigenous Fire Safety Council, 2025). The vulnerability of housing structures to wildfire embers also depends on the type of construction materials used, emphasizing the need for more fire-resistant building systems in future housing designs. Climate-related challenges are intensifying as Canada experiences climate change at roughly twice the global average rate (Canada.ca, 2023). Indigenous communities located in northern regions are particularly exposed to the effects of extreme weather events, including floods, wildfires, heavy snow, and storms (Environment and Climate Change Canada, 2011; Airships, 2020). These climatic impacts not only threaten infrastructure but also shorten construction seasons (Garneau, 2022), making it difficult to transport materials and complete projects within limited time windows. As a result, the winter roads essential for transporting materials to remote sites have become increasingly unreliable and, in some cases, impassable, creating critical delays and cost overruns (Times, 2024). According to the 2016 Census (Statistics Canada, 2017), approximately 63% of Indigenous people resided in mainly remote and rural regions, while only 27% live in urban regions (Figure 1). The population distribution between Indigenous and non-Indigenous peoples in intermediate regions remains relatively comparable. However, Indigenous communities face distinct construction-related challenges, including high transportation costs, limited access to essential services, and lower labor-market participation (OECD, 2016; Transport Canada, 2016; Bleakney et al., 2021). These factors collectively highlight the urgent need for tailored solutions that effectively address unique construction and climate conditions while remaining affordable for these communities. Cold-formed steel modular structures are considered a viable solution to these complex issues, particularly in addressing housing shortages. Steel is a non-flammable and inorganic material that offers significant technical advantages compared to traditional timber structures. Additionally, during transportation, steel structures provide lightweight and space-efficient designs, unlike concrete structures. These modular systems facilitate easier shipping of construction materials to remote and rural areas (Lawson et al., 1999) and expedite on-site construction, resulting in reduced costs compared to conventional construction methods (Kamali and Hewage, 2016; Lopez and Froese, 2016). Steel modular construction is categorized into three types—elemental, panelized, and volumetric—with each contributing to the structural framework of a building (Corfar and Tsavdaridis, 2022). Volumetric modular construction, characterized by box-like units, allows for the efficient manufacture of building components in factory settings, which are then transported to construction sites for assembly. The panelized approach involves fabricating structural elements such as wall panels, roof, and floor systems in a factory, which are subsequently delivered for assembly. This method offers logistical advantages by minimizing transportation complexities and potential damage, as panels can be stacked, enhancing flexibility and adaptability to diverse project requirements (Kilander, 2024). In response to the COVID-19 pandemic, modular structures were utilized to build a post-disaster hospital in Wuhan, China, accommodating 1,000 patients (Suleiman et al., 2021). Frame-supported modular structures transfer loads via edge beams to corner columns, making the connections between module units crucial for maintaining structural integrity and withstanding lateral forces such as seismic and wind forces. Various studies have examined panelized steel modular structure joints, particularly column–column–beam joint assembled on-site with bolts to avoid transport damage (Liu et al., 2017; Zhan et al., 2021). Liu et al. (2017), Liu et al. (2019) developed a H-shaped beam using extended cover plates and an endplate. They conducted cyclic tests to evaluate the influence of bolt number and trapezoidal stiffener on the joint’s stiffness and energy dissipation capacity to enhance its seismic performance. Numerous studies have proposed joints for HSS columns with wide flange beams, often neglecting other effective systems like HSS-to-HSS moment-resisting frames, which offer superior bending, torsion, and compression resistance and thus represent a potentially high-performance system (Korol et al., 1993; Wang et al., 2022a; Wang et al., 2022b). Fadden and colleagues studied the performance of welded HSS-to-HSS connections, both unreinforced and reinforced, under earthquake conditions. They found that reinforcing these connections improves their strength and flexibility during seismic events, which is important for building in earthquake-prone areas (Fadden et al., 2012; Fadden et al., 2015; Fadden and McCormick, 2013). Zhao et al (2024) numerically investigated welded SHS T-joints in steel frame elevator structures, developed a semi-rigid connection model, and showed that semi-rigid behavior can notably increase lateral drift under wind and seismic loading compared to rigid joints. Tsalkatidis et al. (2018) numerically examined bolted hybrid SHS–glulam beam connections using angles and preloaded bolts, evaluating the influence of bolt size, angle thickness, and stiffeners on moment–rotation behavior and proposing an optimal configuration. Numerous studies have proposed different configurations for welding HSS sections to the column using either a central gusset plate or a Tee plate (Kosteski and Packer, 2003a; Kosteski and Packer, 2003b; Park et al., 2021). Elhadary et al. (2024b) introduced a bolted HSS-to-HSS connection with extended plates and long bolts, suitable for transportation and installation in remote Indigenous communities, as shown in Figure 2. This solution reduces damage by enabling modular units to be shipped as panelized sections, improving logistics with standard truck transport. It addresses spatial limitations, offers flexibility for various joint configurations, and maintains architectural integrity. Additionally, the connection can be disassembled for modifications to the panelized floor system, making it beneficial for Indigenous housing by simplifying maintenance and relocation while enhancing structural efficiency and adaptability in modular construction. Elhadary et al. (2024b) tested six bolted HSS-to-HSS moment connections under monotonic load using a DIC camera to analyze the mechanical behavior of the connection for application in moment-resisting frames.

Figure 1

Figure 1. Distribution of non-Indigenous and Indigenous populations in different types of regions (Statistics Canada, 2016 Census of Population, 2017).

Figure 2

Figure 2. Installation sequence and connection configurations of panelized modular units using the proposed HSS-to-HSS bolted connection: (a) column module and panelized beam; (b) field assembly (Elhadary et al., 2024a).

The upper and lower limits of each parameter were not randomized to identify the best configuration for the bolt arrangement, the number of bolts, the presence of a stiffener, and the thickness of the extended plate in the connection proposed by Elhadary et al. (2024b). This oversight reveals a significant gap that requires comprehensive investigation. This study performed a comprehensive parametric study involving 81 finite element models (FEMs) to investigate the influence of different parameters on the structural performance of the proposed steel modular connection, such as bolt arrangement, number of bolts, thickness of the extended plate, and strengthening using a stiffener. Additionally, we examined the behavior of a two-dimensional prototype subjected to biaxial bending moments in an L-shape, considering different geometric parameters and applied loading ratios. This is compared to a one-dimensional prototype, which experiences a bending moment from only one side.

2 Summary of tested experimental specimens

Six specimens were tested under monotonic loading at Lakehead University’s structural laboratory to assess the structural performance of the bolted HSS-to-HSS connection. The column base plate was fastened to the W-section using 25.4-mm-diameter high-strength bolts, simulating a fixed-end condition. The beam was supported by a steel roller bracket which maintained the loading point position while allowing rotational movement. Lateral frame supports restricted out-of-plane displacement of the beams. Testing was performed using a hydraulic universal testing machine (UTM) with a maximum capacity of 1,300 kN, applying a displacement rate of 2 mm/min (Elhadary et al., 2024b). The beam and column cross sections were HSS 152 × 152 × 6.4 mm and HSS 152 × 152 × 9.5 mm, respectively, and were fabricated from steel 350W. The joint was assembled using grade A325 high-strength long bolts with a diameter of 5/8 inches (15.875 mm). Table 1 summarizes the geometric parameters of the tested bolted connection, where each specimen is coded based on key design variables: extended plate thickness (EP), number of bolts (B), and bolt arrangement ratio, defined as the ratio of the perpendicular pitch distance to the parallel pitch distance between bolts $\left({\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_{\mathrm{p}}\right)$, categorized as types I, II, and III. “Parallel pitch distance” refers to the spacing between bolts along the longitudinal axis of the cross section, while “perpendicular pitch distance” refers to the spacing normal to the longitudinal axis. The parallel and perpendicular pitch distances for bolts are determined according to the upper and lower pitch distance limits specified in CSA S16-19 (CSA, 2019). Three measurement systems were employed to capture the response of the tested joint. Linear variable differential transducers (LVDTs) measured displacements at key locations, including the beam end, extended plate, and column cap. Strain gauges were installed on the column, beams, extended plates, and, in the case of stiffened specimens, on the stiffeners to record strain distributions. A stereoscopic digital image correlation (DIC) system with two 2.3 Mpx USB cameras and a 1.0 mm speckle pattern was used to acquire full-field displacement and strain data within a 600 mm × 600 mm field of view. The DIC setup included a 1,200 mm object distance, 450 mm baseline, and 22° stereo angle, with calibration performed using a 500 mm target. All displacement sensors, strain gauges, and DIC cameras were calibrated prior to testing to ensure accurate and reliable measurements.

Table 1

Table 1. Key design parameters of the test specimens (Elhadary et al., 2024b).

3 Finite element model

3.1 Model validation

Three-dimensional finite element models (FEMs) of a modular beam-to-column connection were developed utilizing LS-DYNA software (Livermore Software Technology Corporation LSTC, 2007) which were validated against experimental results obtained by (Elhadary et al., 2024b). The FEM simulations were conducted on the Digital Research Alliance of Canada’s SHARCNET system. Each simulation required an average duration of 48 h to reach the specified termination point, including 760 computational time steps. The boundary conditions of the FEM were set to replicate the experimental test conditions. A prescribed displacement in the Y-direction was applied to a bracket plate with dimensions of 340 × 170 mm and a moment lever arm of 1,080 mm to mimic the relative displacement imposed during the experimental test. Fixed supports were used to restrain the column, while the beam had lateral restraints applied in the X-direction, as illustrated in Figure 3. All model components were represented by eight-node 3D solid elements with reduced integration and hourglass control to prevent shear locking. The beam was discretized using a fine 4-mm mesh, with seven layers of finer mesh used to model the profile thickness because buckling is highly sensitive to local stress concentrations and imperfections, and a finer mesh helps capture these details more precisely. Stress distribution around the bolt holes was captured using inflation layers. This approach was based on a mesh sensitivity analysis conducted using various mesh sizes, including 20, 10, 6, 4, and 3 mm (Figure 3). A 4 mm mesh size was selected as it provided the highest accuracy for both moment and displacement values without the need to increase the mesh size to 3 mm, which would require excessive computational power. All components of the specimens were defined as steel 350W, and grade A325 high-strength bolts were used for the long bolts, with a diameter of 5/8 inches (15.875 mm). Tensile tests were conducted to characterize the steel sections and generate the true stress–strain curve required for accurate FEM modeling (Figure 4). Coupons, CNC-machined from the joint components in accordance with ASTM (2014), were tested using a universal testing machine equipped with two digital image correlation (DIC) cameras and a strain gauge. Figure 4 presents the test setup and the von Mises effective strain distribution for the tensile coupon. The material property values used for the steel coupons from different cross sections and the bolts are summarized in Table 2. The frictional components (i.e., the bolts and their contact surfaces) were modeled using surface-to-surface contact, with a friction coefficient of 0.3 and a penalty formulation for all tangential responses. The friction coefficient was selected to reflect the sandblasted surface finish of the experimental specimens. Welded components, such as the extended plates and columns, were modeled with surface-to-surface tied contact. A 2-mm bolt clearance was used in FEM to accurately simulate the deformation behavior in the bolted connection, ensuring a realistic representation of the connection’s performance. The long bolts used in the connection were 8.5 inches (215.9 mm) long and were installed using snug tightening (i.e., without pretension). The joint was designed as a bearing-type connection. To ensure that the connection could develop the full plastic moment of the beam without significant bolt deformation, 5/8-inch-diameter bolts (15.875 mm) were selected to provide sufficient bending stiffness. All models used nonlinear material properties, implicit analysis with nonlinear time steps, and large deformation; the convergence tolerance was set to 1.0*10⁻²⁰ to ensure high precision in the solution process and a full Newton–Raphson method with a modified arc length approach. FEM was validated by comparing its moment–rotation curves and failure modes with experimental data (Figures 5, 6). Table 3 presents the average error in predicting the connection’s ultimate capacity, which was only 5%, demonstrating the model’s high accuracy. The lower initial stiffness and minor fluctuations observed in the FEM moment–rotation curves can be attributed to the modeled 2-mm bolt clearance (Figure 5). This clearance allowed an initial slip between the bolt shanks and the hole surfaces, with load transfer occurring primarily through plate–beam bearing. Once the bolts fully engaged with the hole surfaces, the connection began to resist loads through bolt bearing, resulting in an increase in stiffness.

Figure 3

Figure 3. (a–b) Boundary conditions of FEM and experimental tests; (c) FEM mesh configuration; (d) mesh sensitivity analysis for moment and displacement accuracy (Elhadary et al., 2024b).

Figure 4

Figure 4. (a) Uniaxial test associated with a DIC camera; (b) graphic representation of the effective strain distribution (von Mises) for the tensile coupon; (c) true stress-true equivalent plastic strain curve for steel 350W and grade A325 high-strength bolts.

Table 2

Table 2. Material properties used in the FEM for the steel cross-section.

Figure 5

Figure 5. Comparison between the moment–rotation curve of the proposed FEM and the test specimen.

Figure 6

Figure 6. Comparison of the failure modes between the experimental specimens and the proposed FEM: (a) specimen EP13-B6-I; (b) specimen EP13-B8-I; (c) specimen EP10-B6-I.

Table 3

Table 3. Comparison between experimental test and FEM results.

3.2 Parametric study

A parametric investigation encompassing 81 FEM simulations was performed to explore the influence of various geometric parameters on the mechanical behavior of the proposed innovative modular connection for the one- and two-dimensional prototype connection. These parameters included the extended plate thickness (i.e., EP = 8, 10, 13, and 15 mm), the number of bolts (i.e., B = 4, 6, 8, and 10), bolt arrangement ratio (i.e., ${\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_{\mathrm{p}}$ = 1.5, 1, 0.6), and the existence of stiffener of 10 mm below the lower extended plate (i.e., ST) (Table 4). The extended plate dimensions depended on the bolt arrangement in the HSS sections and the number of bolts. Throughout the parametric study, several parameters were constant, including the specimen length of 1,250 mm for the beam, beam cross section of 152 × 152 × 6.4 mm, column cross section of 152 × 152 × 9.5 mm, bolt diameter of 5/8 inches, and pitch distance between bolts perpendicular to the member axis, consistent with the experimental specimens. The perpendicular pitch distance was defined as a factor of the section depth (H), expressed as $P_n=0.5\mathrm{H}$ (Figure 7). Additionally, the normal edge distance was set equal to the parallel edge distance. The coding for the tested specimen included the following elements: type of analysis (either EXP for experimental test or FEM for finite element model), thickness of the extended plate, bolt number, and the ratio of the perpendicular to parallel pitch between the bolts $\left({\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_{\mathrm{p}}\right)$, which were implemented using three symbols (I, II, and III), where ${\mathrm{I}=\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_{\mathrm{p}}=1.5,{\text{II}=\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_{\mathrm{p}}=1$, and ${\text{III}=\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_{\mathrm{p}}=0.6$ (Figure 7). For example, in the designation FEM-EP15-B8-II-ST, “FEM” stands for finite element model, “EP15” indicates an extended plate with a thickness of 15 mm, “B8” signifies that there are eight bolts, the “II” represents the ratio ${\mathrm{P}}_{\mathrm{n}}/{\mathrm{P}}_{\mathrm{p}}$ with a value of 1.0, and “ST” denotes a stiffener located beneath the lower extended plate. Six models were analyzed for the two-dimensional prototype (i.e., L-shape) under monotonic load, with each beam subjected to a different displacement rate to assess its impact on the general performance of the proposed connection. The L-shape connection had a fixed bolt configuration of EP15-B6-I-ST for both sides. For the L-shape prototype connection, the displacement load for both beams 1 and 2 was applied simultaneously. The displacement percentages applied to beam 2 were 20%, 30%, 40%, 50%, 60%, and 70% of the failure displacement of beam 1 (Table 5). The value of 210 mm displacement was determined when the load dropped to 80% of the ultimate load based on the one-dimensional connection studied. The boundary conditions and the used mesh configuration for the L-shape are illustrated in Figure 8.

Table 4

Table 4. Main design parameters of the one-dimensional prototype connection.

Figure 7

Figure 7. Main studied parameters within the parametric study.

Table 5

Table 5. Studied parameters of a two-dimensional L-shape prototype connection.

Figure 8

Figure 8. Boundary conditions for the two-dimensional prototype of L-shape HSS connection.

4 FEM results and discussion

Table 6 summarizes the numerical results of the 75 models of one-dimensional prototypes. The ductility index is defined as the rotation ratio corresponding to the ultimate moment (φ_ult) to the rotation at yielding resistance (φ_y). The ductility coefficient of the specimens ranged 2.28–7.77, exceeding the values specified in the Canadian code CSA S16-19 (CSA, 2019) for moderately ductile and ductile moment-resisting frames of 3.0 and 5.0, respectively. This indicates that the specimens demonstrated exceptional ductility, enhanced structural resilience, and improved safety performance. Seven failure modes were identified based on the geometric parameters studied, including rupture failure of the upper extended plate, bolt shear failure, local buckling of the lower extended plate, bolt-bearing failure on the beam, local buckling of the beam, local buckling of the column, and yielding of the column due to stiffener bearing (Figure 9). The rupture of the upper extended plate and bolt shear failure is classified as premature failures occurring at lower moment capacity. In contrast, local buckling of the beam is considered a favorable failure since it enables the hinge to form at the beam. However, local buckling of the column is an unfavorable failure as it violates the “strong column, weak beam” concept.

Table 6

Table 6. Summary of FEM results.

Figure 9

Figure 9. Dominant failure modes of different specimens: (a) EP15-B10-II-ST; (b) EP8-B10-II-ST; (c) EP13-B8-I-ST; (d) EP8-B10-III; (e) EP13-B4-I; (f) EP15-B10-III.

4.1 Effect of bolt arrangement

Three different bolt arrangements were studied (i.e., AR1.0, AR1.5, and AR0.6) to investigate their impact on the proposed connection’s failure mode and ultimate capacity. It was observed that the arrangement of the bolts plays a crucial role in the connection mechanism, as it could shift the failure model from the beam to the column. It was observed that the use of the AR1.5 bolt configuration caused bolt bearing to occur early on the beam, enabling it to achieve greater capacity. This, in turn, resulted in local buckling at the compression flange of the beam and reduced the ultimate rotation of the extended plate by 50% compared to the other bolt arrangements—AR0.6 and AR1.0. The AR0.6 bolt configuration influenced the connection’s loading mechanism by shifting the failure mode toward the column’s compression flange. This effect became more noticeable in the eight- and ten-bolt arrangements, as the extended plate length increased with the bolt pitch distance. Consequently, increasing the distance between bolt rows delayed the initiation of bolt bearing without eliminating it, leading to the applied load being transferred primarily through bearing on the two extended plates rather than direct bolt bearing on the beam flanges. The extensive rotation of the extended plates induced greater bending in the column, which ultimately failed by local buckling of the compression flange. Bolt configuration AR1.0 is considered a balancing effect between AR1.5 and AR0.6 as it effectively postpones the occurrence of bolt-bearing failure, so bearing forces were distributed along the bolt rows in almost equal values compared to bolt arrangement AR1.5. This led to a focus almost at the bearing force on the outer bolt rows and without inducing significant rotation in the extended plate compared to the AR0.6 arrangement. To maintain the “strong column, weak beam” principle, it is advisable to use bolt arrangements AR1.5 and AR1.0. These configurations enable the connection to reach its ultimate capacity through bolt-bearing failure.

4.2 Influence of extended plate thickness

Four different extended plate thicknesses (8 mm, 10 mm, 13 mm, and 15 mm) were analyzed to evaluate their impact on failure modes and their ability to transfer moment capacity to the column. The study revealed that an 8 mm extended plate thickness led to rupture failure of the upper plate at the net area, particularly with bolt arrangements AR1.0 and AR0.6. In these cases, the delay in bolt-bearing, caused by the bolt arrangement, affected how the forces were distributed. The shear forces arose from the extensive rotation of the extended plate, which was a result of the beam transferring load primarily through bearing on the extended plates, rather than through immediate bolt-bearing. In contrast, no premature failure was observed with the AR1.5 arrangement, as this bolt configuration relied on bolt-bearing mechanics without causing extensive rotation to the extended plate. Furthermore, using a stiffener below the lower extended plate did not prevent premature failure for bolt arrangements AR1.0 and AR0.6, as the column flange began to buckle due to the stiffener bearing, altering the load path through the extended plate and resulting in rupture failure of the upper extended plate. It was noted that increasing the extended plate thickness from 10 mm to 13 mm improved the connection’s ultimate capacity and ductility by approximately 4% and 9%, respectively. Meanwhile, increasing the thickness from 10 mm to 15 mm resulted in enhancements of 7% in ultimate capacity and 11% in ductility across various bolt arrangements. This enhancement is due to the larger cross section of the plates, which helped prevent bolt-bearing failure on the extended plates and facilitated greater force transfer through them. This led to delayed plate yielding during the loading phases, which increased the connection’s stiffness by relying on the stiffness of the two extended plates during the initial rotation.

4.3 Influence of the number of bolts

The number of bolts was crucial to achieving higher moment capacity across different bolt arrangements (i.e., AR1.0, AR1.5, and AR0.6). Using four bolts with a diameter of 16 mm led to premature shear failure within the connection, as the bolt-bearing failure was unevenly distributed among the bolt rows, causing the outer bolt row to fail first. For the AR1.5 bolt configuration, increasing the number of bolts from six to eight and from six to ten resulted in an 11% and 17% improvement in connection capacity, respectively. This improvement was achieved without inducing significant rotation in the column, as the length of the extended plate was increased based on the number of bolts (Figure 10). For the AR1.0 bolt configuration, increasing the number of bolts from six to ten changed the failure mode and increased capacity by 22%. This increase was due to a change in the loading path within the connection, postponing bolt-bearing failure and allowing the two extended plates to rotate until local buckling of the column occurred. However, increasing the number of bolts from six to eight enhanced ultimate capacity by 10% without causing unfavorable failure to the connection, leading to local buckling of the beam. For the AR0.6 bolt arrangement, increasing the number of bolts from six to either eight or ten caused local buckling of the column. In contrast, using six bolts for AR0.6 resulted in local buckling of the beam (Figure 10). Therefore, using six bolts for different bolt arrangements is recommended to avoid any unfavorable failure modes within the three-bolt configurations studied (i.e., AR1.5, AR1.0, and AR0.6).

Figure 10

Figure 10. Relationship between number of bolts vs. extended plate thickness vs. ultimate moment for different bolt arrangements: (a) AR1.5; (b) AR1.0; (c) AR0.6; (d) AR1.5 with stiffener; (e) AR1.0 with stiffener; (f) AR0.6 with stiffener.

4.4 Impact of stiffener

A 10-mm stiffener was installed under the lower extended plate to examine its impact on connection performance across various parameters. The stiffener reduced the ultimate rotation by an average of 60% for different bolt arrangements because it prevented extensive rotation and yielding of the extended plate (Table 6). It was observed that for bolt arrangements AR0.6 and AR1.0, the presence of the stiffener altered the load transfer mechanism to rely more on the bolt bearing on the beam flange rather than the rotation of the extended plates, leading to the connection failing through local buckling of the beam. However, the column flange could not endure the bearing force from the stiffener, resulting in plastic deformation (Figure 10). For the AR1.5 bolt arrangement, the stiffener had no significant impact on the connection’s failure mode or capacity. Nonetheless, it caused the yielding of the column flange since part of the load was directly transferred through the stiffener to the column flange.

4.5 Performance of the L-shape connection under different applied forces

Six L-shaped specimens were tested, utilizing the bolt configuration EP15-B6-I-ST for two beams. This configuration aimed to induce bolt bearing on the beam at an early stage, achieving a favorable failure mode by forming a hinged support at the beam. The moment–rotation curves for both beams under different displacement ratios are presented in Figure 11. It was observed that the column experienced varying rotations for the two beams. After reaching the target load for beam 2, the load path was primarily transferred by the stiffener from beam 2 to the column, leading to plastic deformation of the column beneath the stiffener (Figure 12c). Beam 2 exhibited a 50% higher rotation rate than beam 1, especially with applied displacements from 40% to 70%, due to the biaxial bending moment applied to the column. The L-shaped connection demonstrated similar behavior to the one-dimensional prototype, particularly for beam 1, where the connection failed due to local buckling of the beam at various displacement ratios (any failure signs in the extended plates, as shown in Figure 12).

Figure 11

Figure 11. Moment–rotation curves for the two beams in the L-shape prototype connection under different applied displacements: (a) 70%; (b) 60%; (c) 50%; (d) 40%; (e) 30%; (f) 20%.

Figure 12

Figure 12. Failure modes of L-shape prototype connection under different applied displacement ratios between beam 2 beam 1: (a) 20%; (b) 50%; (c) 70%.

5 Conclusion

We studied the mechanical performance of the corner HSS-to-HSS connection for panelized modular houses in remote Indigenous regions. A total of 81 finite element models were developed for both one- and two-dimensional prototypes (such as the L-shape) to examine the impact of different geometric parameters on the ultimate moment capacity, ultimate rotation, failure modes, and ductility. The studied parameters showed that the bolt arrangement, stiffener, number of bolts, and extended plate thickness significantly influenced the connection capacity gain and failure modes. The following conclusion can be drawn from the above analysis.

• The FEM results were in good agreement with the experimental results, with an average error of 5%. It was found that the extended plate thickness influenced ductility and failure mode differently, depending on the bolt arrangement used. To prevent premature connection failure due to rupture failure of the extended plate, using a thickness greater than 8 mm is recommended.

• Bolt arrangement plays a crucial role in the connection’s performance, influencing both its capacity and rotation. The AR1.5 arrangement improved the connection’s capacity by enabling early bolt bearing on the beam, allowing for a greater load transfer, but it reduced the ultimate rotation of the extended plate by 50% due to local buckling at the beam’s compression flange. The AR0.6 arrangement shifted the failure mode to the column’s compression flange, effectively transferring the load to the column and increasing the capacity. However, this arrangement also reduced the ultimate rotation as the beam’s participation in the failure mechanism was diminished. The AR1.0 arrangement struck a balance between AR1.5 and AR0.6, postponing bolt-bearing failure and allowing forces to be distributed more evenly along the bolt rows, resulting in a moderate increase in capacity while maintaining better control over the rotation compared to AR1.5.

• Increasing the extended plate thickness improved the connection’s capacity and ductility. A 4% increase in capacity and 9% increase in ductility occurred when the thickness increased from 10 mm to 13 mm, while a 7% increase in capacity and 11% in ductility occurred when it increased from 10 mm to 15 mm. The greater plate thickness helped delay bolt-bearing failure, improved load transfer, and increased stiffness, resulting in more controlled rotation

• Increasing bolts from six to ten for the AR1.5 arrangement improved capacity by 17%, while for AR1.0, it increased capacity by 22% and allowed for greater rotation before failure. However, increasing bolts in the AR0.6 configuration could lead to column buckling, affecting ultimate rotation and causing unfavorable failure. Using six bolts in various arrangements is recommended to maximize performance and avoid premature failure.

• The addition of a 10-mm stiffener reduced the ultimate rotation by 60% across different bolt arrangements, shifting the load transfer to the beam flange and causing local buckling of the beam in AR0.6 and AR1.0 configurations. For AR1.5, the stiffener had little effect on the failure mode but caused column flange yielding due to direct load transfer.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material; further inquiries can be directed to the corresponding author.

Author contributions

ME: Writing – original draft, Methodology, Visualization, Validation, Software. AB: Writing – review and editing, Supervision, Resources, Funding acquisition. AE: Writing – review and editing, Supervision, Funding acquisition, Conceptualization, Project administration.

Funding

The authors declare that financial support was received for the research and/or publication of this article. This research was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) under Grant No. RGPIN-2022-04755.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The authors declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Airships, ISOPolar (2020). Addressing Canada’s Indigenous housing shortage. Available online at: https://isopolar.com/addressing-canadas-indigenous-housing-shortage/ (Accessed July 26, 2022).

Google Scholar

ASTM (2014). A370: standard test methods and definitions for mechanical testing of steel products. West Conshohocken, PA: ASTM Int., 1–50.

Google Scholar

Bleakney, A., Masoud, H., and Robertson, H. (2021). Labour market impacts of COVID-19 on Indigenous people living off reserve in the provinces: march 2020 to August 2021. Canada: Stat. Canada website. Available online at: https://www150.statcan.gc.ca/n1/pub/45-28-0001/2021001/article/00037-eng.htm (Accessed February 25, 2024).

Google Scholar

Canada. Office of the Auditor General (2003). Report of the auditor general of Canada to the house of commons, April 2003. Ottawa, ON, Canada: Office of the Auditor General of Canada. Available online at: www.oag-bvg.gc.ca.

Google Scholar

Canada.ca (2023). Climate change adaptation in Canada. Gov. Can. Available online at: https://natural-resources.canada.ca/climate-change/what-adaptation/10025 (Accessed October 14, 2022).

Google Scholar

Corfar, D.-A., and Tsavdaridis, K. D. (2022). A comprehensive review and classification of inter-module connections for hot-rolled steel modular building systems. J. Build. Eng. 50, 104006. doi:10.1016/j.jobe.2022.104006

CrossRef Full Text | Google Scholar

CSA (2019). CSA-S16-19: design of steel structures. Available online at: https://www.concrete.org/store/productdetail.aspx?ItemID=318U19&Language=English.

Google Scholar

Elhadary, M., Bediwy, A., and Elshaer, A. (2024a). “Experimental investigation on a novel steel connection for modular indigenous houses,” in Canadian society of civil engineering (CSCE) 2024 annual conference (Ontario: Niagara Falls).

Google Scholar

Elhadary, M., Bediwy, A., and Elshaer, A. (2024b). Novel steel connection for modular houses in indigenous communities: an experimental study. J. Constr. Steel Res. 220, 108850. doi:10.1016/J.JCSR.2024.108850

CrossRef Full Text | Google Scholar

Environment and Climate Change Canada (2011). Warm season weather hazards - canada.ca. Goverment Can. Available online at: https://www.canada.ca/en/environment-climate-change/services/seasonal-weather-hazards/warm-season-weather-hazards.html#toc7.

Google Scholar

Fadden, M., and McCormick, J. (2013). Evaluation of HSS-to-HSS moment connections for seismic applications. Struct. Congr. 2013 Bridg. Your Passion Your Prof. Proc. 2013 Struct. Congr. 2334–2345. doi:10.1061/9780784412848.204

CrossRef Full Text | Google Scholar

Fadden, M., Asce, S. M., Mccormick, J., and Asce, A. M. (2012). Cyclic quasi-static testing of hollow structural section beam members, 138, 561–570. doi:10.1061/(ASCE)ST.1943-541X.0000506

CrossRef Full Text | Google Scholar

Fadden, M., Wei, D., and McCormick, J. (2015). Cyclic testing of welded HSS-to-HSS moment connections for seismic applications. J. Struct. Eng. 141, 04014109–04014114. doi:10.1061/(asce)st.1943-541x.0001049

CrossRef Full Text | Google Scholar

Garis, L. (2021). Fire risk for Indigenous people, firefighting in Canada. On-line Annex Bus. media. Available online at: https://www.firefightingincanada.com/fire-risk-for-indigenous-people/ (Accessed November 2, 2022).

Google Scholar

Garneau, M. (2022). The effects of the housing shortage on Indigenous peoples in Canada: report of the standing committee on Indigenous and northern affairs. Available online at: www.ourcommons.ca.

Google Scholar

Indigenous Services Canada (2018). Evaluation of the on-reserve income assistance pogram. Canada: Gov. Canada. Available online at: https://www.sac-isc.gc.ca/eng/1557321693588/1557321741537 (Accessed July 26, 2022).

Google Scholar

Kamali, M., and Hewage, K. (2016). Life cycle performance of modular buildings: a critical review. Renew. Sustain. Energy Rev. 62, 1171–1183. doi:10.1016/j.rser.2016.05.031

CrossRef Full Text | Google Scholar

Kilander, A. (2024). Unraveling modular construction: Volumetric vs panelized approach. Available online at: https://blog.framecad.com/blog/unraveling-modular-construction-volumetric-vs-panelized-approach (Accessed February 25, 2024).

Google Scholar

Korol, R. M., Ghobarah, A., and Mourad, S. (1993). Blind bolting W-shape beams to HSS columns. J. Struct. Eng. 119, 3463–3481. doi:10.1061/(asce)0733-9445(1993)119:12(3463)

CrossRef Full Text | Google Scholar

Kosteski, N., and Packer, J. A. (2003a). Longitudinal plate and through plate-to-hollow structural section welded connections. J. Struct. Eng. 129, 478–486. doi:10.1061/(asce)0733-9445(2003)129:4(478)

CrossRef Full Text | Google Scholar

Kosteski, N., and Packer, J. A. (2003b). Welded Tee-to-HSS connections. J. Struct. Eng. ASCE 129, 151–159. doi:10.1061/(ASCE)0733-9445(2003)129:2(151)

CrossRef Full Text | Google Scholar

Lawson, R. M., Grubb, P. J., Prewer, J., and Trebilcock, P. J. (1999). Modular construction using light steel framing: an architect’s guide. Ascot, Berkshire, United Kingdom: The Steel Construction Institute.

Google Scholar

Liu, X. C., Yang, Z. W., Wang, H. X., Zhang, A. L., Pu, S. H., Chai, S. T., et al. (2017). Seismic performance of H-section beam to HSS column connection in prefabricated structures. J. Constr. Steel Res. 138, 1–16. doi:10.1016/j.jcsr.2017.06.029

CrossRef Full Text | Google Scholar

Liu, X. C., Cui, F. Y., Zhan, X. X., Yu, C., and Jiang, Z. Q. (2019). Seismic performance of bolted connection of H-beam to HSS-column with web end-plate. J. Constr. Steel Res. 156, 167–181. doi:10.1016/j.jcsr.2019.01.024

CrossRef Full Text | Google Scholar

Livermore software technology corporation (LSTC) (2007). “Livermore software technology corporation (LSTC),” in LS-DYNA keyword user’s manual, 1. California, USA.

Google Scholar

Lopez, D., and Froese, T. M. (2016). Analysis of costs and benefits of panelized and modular prefabricated homes. Procedia Eng. 145, 1291–1297. doi:10.1016/j.proeng.2016.04.166

CrossRef Full Text | Google Scholar

National Indigenous Fire Safety Council (2025). NIFSC highlights national crisis in Indigenous fire safety in new article - canadian firefighter magazine. Available online at: https://www.cdnfirefighter.com/nifsc-highlights-national-crisis-in-indigenous-fire-safety-in-new-article/ (Accessed October 13, 2025).

Google Scholar

OECD (2016). OECD regional outlook 2016: productive regions for inclusive societies. Paris: OECD Publishing.

CrossRef Full Text | Google Scholar

Park, K., Jiansinlapadamrong, C., and Chao, S.-H. (2021). Double-HSS seismic resistant beam-to-column moment connections. J. Struct. Eng. 147, 04021098. doi:10.1061/(asce)st.1943-541x.0003057

CrossRef Full Text | Google Scholar

Patterson, D., and Dyck, L. (2015). Housing on first nation reserves: challenges and successes. Available online at: https://publications.gc.ca/site/eng/480465/publication.html (Accessed November 9, 2022).

Google Scholar

Schneider, L. (1992). A time for action: aboriginal and northern housing: fourth report of the standing committee on Aboriginal affairs. Statistics Canada (2017), 1–17.

Google Scholar

Suleiman, M., Elshaer, A., Billah, M., and Bassuony, M. (2021). Propagation of mouth-generated aerosols in a modularly constructed hospital room. Sustainability 13, 11968. doi:10.3390/su132111968

CrossRef Full Text | Google Scholar

Times, T. R. (2024). Impassable winter roads create ‘dire’ situation for Ontario first nations. Available online at: https://tworowtimes.com/news/impassable-winter-roads-create-dire-situation-for-ontario-first-nations/(Accessed February 25, 2024).

Google Scholar

Transport Canada (2016). Minister-led Indigenous roundtable on the future of transportation. Available online at: https://tc.canada.ca/en/corporate-services/minister-led-indigenous-roundtable-future-transportation-summary-discussion (Accessed February 25, 2024).

Google Scholar

Tsalkatidis, T., Amara, Y., Embaye, S., and Nathan, E. (2018). Numerical investigation of bolted hybrid steel-timber connections. Front. Built Environ. 4, 48–51. doi:10.3389/fbuil.2018.00048

CrossRef Full Text | Google Scholar

United Nations (2019). “Adequate housing as a component of the right to an adequate standard of living, and the right to non-discrimination in this context.”

Google Scholar

von Stackelberg, M. (2019). Homes on remote first nations are mouldy before they’re even built, experts say. CBC News. Available online at: https://www.cbc.ca/news/canada/manitoba/first-nations-housing-mould-1.5074196 (Accessed November 2, 2022).

Google Scholar

Wang, H., Zhao, X., and Ma, G. (2022a). Experimental study on seismic performance of column-column-beam joint in panelised steel-modular structure. J. Constr. Steel Res. 192, 107240. doi:10.1016/j.jcsr.2022.107240

CrossRef Full Text | Google Scholar

Wang, H., Zhao, X., and Ma, G. (2022b). Novel coupled modular steel structure and seismic tests on high-performance interconnection. J. Constr. Steel Res. 189, 107058. doi:10.1016/j.jcsr.2021.107058

CrossRef Full Text | Google Scholar

Zhan, X. X., Liu, X. C., Feng, S., and Yu, C. (2021). Seismic performance of a square HSS column to H-section beam bolted connection with double cover plate. Eng. Struct. 231, 111729. doi:10.1016/j.engstruct.2020.111729

CrossRef Full Text | Google Scholar

Zhao, B.Da, Xu, C. C., Zhu, H. C., Chen, R. S., Wu, J. L., Qi, Y. S., et al. (2024). Wind-resistance and seismic behavior of a steel frame elevator with square hollow section members and semi-rigid joints. Front. Built Environ. 10, 1463663. doi:10.3389/fbuil.2024.1463663

CrossRef Full Text | Google Scholar