Catenary Action Research Articles

Component-based joint model for RC frames with conventional and special detailing against progressive collapse

For reinforced concrete (RC) frames subjected to progressive collapse, severe damage is typically concentrated at beam-column connections or even joint panels. To accurately represent this type of local failure and the corresponding effect on progressive collapse resistance, component-based model (CBM) is employed. This paper proposes a refined configuration and calibration on compression bar force-slip springs of CBM for RC frames with conventional detailing at the joints and members. The compression springs account for varying neutral axis depth with applied loads to solve the issue of inaccurate prediction for progressive collapse resistance of RC structures due to over-/under-estimation of compression spring capacities by simplified methods in previous CBMs. In addition, the influence of joint location and tension–compression beam bar ratio on compression spring properties is also explored. Thereafter, the calibrated CBM is developed to simulate the RC frames with special detailing, including plastic hinge relocation (PHR) technique and additional bar layer (ABL) in the beam. It is shown that relocating plastic hinges further away from the joint interface results in a shorter effective beam length and a smaller ultimate deflection. The former increases flexural-compressive arch action and catenary action capacities, and the latter derives the recommendation of limiting the PHR to 1.5 times the effective beam depth. Providing an ABL significantly improves progressive collapse performance of RC frames. Finally, an alternative approach is proposed, in which an ABL is placed only at the support region, located at either the middle or the bottom quarter of the beam depth.

Numerical Investigation on Progressive Collapse Mitigation of Steel Beam-Column Joint Using Steel Plates.

This research employed extensive numerical analyses to locate the weak areas and determine the structural issues critical to preventing the spread of collapse. As a result, three specimens were tested using scaled models of strengthened and unstrengthened steel beam-column joint assemblies. The data were utilized to verify numerical models. One simple shear joint from the three experimental assemblies was used as the control specimen (unstrengthened joint). The second was a bolted steel beam-column joint utilized as a reference specimen to reflect the ideal beam-column joint generally employed in intermediate moment-resisting frames in seismic zones worldwide. Similar to the control, the third specimen (strengthened joint) had two side plates welded together to strengthen the connection site. Numerical finite element models were developed using ABAQUS (2020) software to extensively investigate the behavior of steel frame assemblies before and after upgrading. The FEM matrix comprised 17 specimens with varying parameters, including plate thickness, steel grade, a joint between the beam flange-strengthening plates, and a column that was either welded or not welded. The effectiveness of the strengthening techniques was established by comparing the mode of failure and load-displacement characteristics of the investigated specimens. The results indicate that the average increase in peak load due to a change in plate thickness for grades A36 and A572 is approximately 22% and 8%, respectively. Plates made of A572 steel increase peak load by 30%. All strengthened specimens attained catenary action, mitigating the possibility of progressive collapse.

Investigation on progressive collapse resistance of prestressed concrete frames with infilled walls

The finite element numerical model of prestressed concrete frames with infilled walls (IW-PC frame) is established to investigate the progressive collapse resistance under the failure of middle column. The simulation of infilled walls adopts the equivalent three-strut diagonal bracing model. The method of node coupling is adopted in the simulation of prestressed tendons Then the pushdown analysis of the structure is carried out. Meanwhile, the effects of the parameters of infilled walls and prestressed tendons on the compression arch action and catenary action are studied. The results show that the infilled walls greatly improve the anti-collapse bearing capacity of IW-PC frame. They enable beams to reach peak loads in advance, and transform the axial force of beam and prestressed tendons in advance. It is more conducive to make the prestressed tendons fully stressed, so that the structure can enter the catenary action in advance, thereby improving the bearing capacity.

Experimental Study on Composite Beam with Various Connections under Midspan Impact Scenarios

The performance of an immediate frame span subjected to a falling-debris-impact scenario is of great importance to investigate the spread of initial damage during a progressive collapse event. If the frame span was not able to resist the impact load from possible falling debris, it would become new debris and impact the lower floor with previous debris. After that, progressive collapse might probably happen. However, the impact mechanism considering the contribution of composite slabs of an immediate span remained unclear currently. Therefore, experimental studies on a series of three half-scale composite beam subassemblies were conducted under impact scenarios. The impact loads were applied via a high-performance drop-weight test machine. Various beam–column connections, including welded unreinforced flange-bolted web, fin plate, and reverse channel connections, which represented rigid connection, hinge connection, and semirigid connection, respectively, were investigated. The structural resistances of the specimens were provided by flexural action, compression arch action, and catenary action. Among them, the contribution of compressive arch action was limited and can be ignored. Flexural action always provided the most structural resistance contribution in the whole impact process, and catenary action provided structural resistance under large deformation. The specimen structural resistance increased with increasing deformation before the failure of connections. Good flexural resistance and deformation capacity were the key factors for structures to resist impact load. The reversed channel specimen had the best impact resistance.

Numerical and simplified analytical investigation on RC frame behaviours under progressive collapse scenarios

Although progressive collapse research has been extensively carried out for more than two decades after from the tragic events of 9/11, not many actual dynamic experiments have been conducted due to the constraint in laboratory facilities and expense. Instead, numerical and simplified analytical studies were preferred to achieve a comprehensive understanding of the structural behaviour against these rare dynamic events. In this research, a numerical and analytical investigation programme is carried out to study the mobilisation of alternate-load-path mechanisms, such as compressive arch or catenary action, in RC beam-column substructures undergoing missing column scenarios. First, a physics-based finite element modelling is developed and validated with two published test series conducted under quasi-static and free-fall dynamic conditions. These tests concentrated on 2D reinforced concrete sub-frame structures, investigating two column-loss scenarios, i.e. internal and penultimate columns. After being validated, the model is then employed to further investigate important parameters which can significantly affect the behaviour of a 2D frame structure under both static and dynamic environments, such as concrete grade, release time, column pre-loaded force, etc. Thereafter, a simplified analysis using an equivalent SDOF system is developed to provide a convenient tool for assessing dynamic capacities of RC frame structures. The simplified method produces conservative predictions compared to actual tests and detailed numerical analyses, however it requires very little computing time.

Resilience of prefabricated concrete frames using hybrid steel-concrete composite connections

The progressive collapse resistance of prefabricated concrete frames with hybrid steel-concrete composite connections (HSCC) is rarely studied. To fill the gap, five (one reinforced concrete (RC) and four prefabricated concrete) 1/2 scaled beam-column sub-assemblages were designed and tested. In four prefabricated concrete specimens, different HSCC types, top and seat with web angle (TSWA), end plate (EP), top and seat angle (TSA), and web cleat (WC), were employed. Experimental results demonstrated that the failure patterns of prefabricated concrete frames with HSCC are different from that of the RC frame counterpart. The failure of the prefabricated concrete frames is governed by the shear fracture or thread stripping of the bolts in the connections while that of the RC frame is governed by the fracture of beam longitudinal rebar. Due to brittle failure of the connections in prefabricated concrete frames, both ultimate bearing capacity and deformation capacity of prefabricated concrete frames are smaller than the corresponding RC frame. Among them, the prefabricated concrete frame with the EP connection achieves the greatest first peak load or compressive arch action capacity. However, no catenary action could be developed in this specimen.

Behaviour of vertically loaded steel beams under a travelling fire

As the structural response to a time and space dependent non-uniform temperature field is challenging to manage by code-based design, the behaviour of steel beams exposed to a travelling fire was simulated numerically in this paper. The finite element (FE) models for both thermal and mechanical analyses were validated by two benchmark tests and used to study the effect of uniform and non-uniform thermal exposures and different steel grades on the response of steel beams. The developed five-stage mechanism well characterized the response of the beams exposed to the travelling fire. For the beam in the elastic range, the heating-cooling cycles of gas temperatures induced cyclic axial forces, but the cyclic nature of structural responses was cushioned by material plasticity. For the beams stressed into the plastic range, the repetitions of cycles in axial response created residual deformations that induced catenary action in the beams at lower temperatures. The local heating of the critical sections made the beams more prone to a runaway failure than the uniform heating of the whole beam. Compared to the mild steel beams, the axial force response of the HSS beams showed higher axial compression forces and fluctuated more. The critical temperatures based on the deformation or strength criterion were consistent for the beams with the load ratio of 0.3. However, for designing the beam with the load ratio of 0.5, critical temperature of 350 °C or the strength criteria based on the proportional limit are recommended as the catenary action was activated at low temperatures. These recommendations can be used as a guide for practical design.

Evaluation on the progressive collapse resistance of infilled reinforced concrete frames based on numerical and semi-analytical methods

The effect of infill walls on the progressive collapse behaviour of reinforced concrete (RC) frames should be fully considered due to the formation of alternative load paths. To explore the configuration that could affect the progressive collapse resistance of infilled RC frames, parametric analyses are performed based on the validated simulation method. In the study, the impact introduced by the extra tie and integrity rebars in walls is emphasised. It is indicated that the variation in the reinforcement ratio of tie rebars leads to a limited change in structural progressive collapse resistance. In contrast, with the arrangement of the integrity reinforcement, the progressive collapse resistance of the infilled frames is enhanced and the propagation of the diagonal cracks in partial infill walls is suppressed. The notable re-ascending trend in the resistance curve for models with integrity rebars reveals the significant development of the catenary action in the integrity reinforcement. Then, with the help of linear regression and gene expression programming (GEP) techniques, a semi-analytical model is proposed for the prediction of the load-bearing capacity of infilled frames after column loss. In the prediction model, the linear effect of geometrical parameters and the non-linear effect of openings on the progressive collapse resistance are included by introducing three modification coefficients. Based on the statistical assessment, the capability of this explainable model is verified, and the appropriate amount of integrity reinforcement to mitigate the progressive collapse is discussed.

Performance of precast concrete beam-column joint with a hidden corbel under progressive collapse scenarios

Collapse incidents of structures caused by extreme loads have attracted the attention of academia and building regulatory bodies worldwide. Due to inherent structural discontinuity, precast concrete (PC) structures are more vulnerable to progressive collapse than cast-in-situ reinforced concrete (RC) buildings. Therefore, studies on resisting collapse performance of PC beam-column joints for PC buildings are needed. In this paper, four 1/2-scaled PC beam-column joints with an in-built L-shaped corbel at each beam end were tested for progressive collapse under column removal scenario. These four joint specimens with identical non-seismic PC details were extracted from adjacent bays of the removed column, and the main parameters were the variations in the boundary and loading conditions. The deformation capacity, load-carrying capacity, catenary action (CA) and failure mode were evaluated and discussed. Parametric studies were also conducted by a proposed component-based mechanical joint model, to investigate the effects of PC interface location, mid-height reinforcement ratio and span-to-height ratio on the collapse resistance of the 2-span substructures. The experimental results show that this type of PC joint could achieve excellent hogging moment resistance but relatively lower sagging resistance. However, both PC joints under downward and upward applied loads could achieve a subsequent increase in resistance due to catenary action at large deformation stage. Numerical parametric studies indicated that shifting the precast interface inwards towards 1/3 point of single beam span and increasing mid-height reinforcement ratio could remarkably improve load-carrying capacity at the compressive arch action (CAA) stage. These detailing features could offset the negative effect caused by the discontinuous bottom longitudinal reinforcement, which is typical for gravity-dominant design in non-seismic countries, such as Singapore.

Anti-collapse performance of concrete-filled steel tubular composite frame with assembled tensile steel brace under middle column removal

To study the effect of assembled tensile steel brace on the progressive collapse resistance of concrete-filled steel tubular (CFST) structures, two 1/4-scaled two-bay two-story composite frames with CFST columns and profiled steel–concrete composite beams were tested pseudo-statically under a middle column removal scenario. The vertical load–displacement responses and failure modes of two specimens were analyzed. Especially, the effects of assembled tensile steel braces on the anti-collapse performance of the frames were discussed. Experimental results show that the assembled tensile steel braces delayed the failure of the upper beam and significantly improved the structural bearing capacity and ductility of the frame. The horizontal reaction forces of the specimens are compressive forces due to the compressive arch action in the initial loading stage, and they change from compression to tension due to the catenary action. Numerical parametric analysis indicates that the splint bolt connection configuration can effectively improve the bearing capacity of the frame models by increasing the bolt preload and friction coefficient. Furthermore, the collapse resistance of the structure is calculated based on the simplified elastic theory method, considering the effect of the steel braces. It is believed that the assembled tensile brace between CFST columns could be used as a reinforcement method to improve the collapse resistance of the framed structure.

A novel principle for improving collapse resistance of steel frame structures

Although the reduced beam section (RBS) connection is an excellent seismic design for transferring plastic hinge, it is prone to fracture under the condition of large deformation. In addition to good seismic response, the performance of RBS connections needs to be improved to ensure good anti-progressive collapse response. In this study, a novel principle is proposed, in which the RBS connection is used to meet the plastic hinge transmission, and kinked reinforced bars are installed in the beam–column connection to realize a second path. Experimental and numerical investigations were conducted on a frame substructure with kinked reinforced bars, considering their diameter d, kinked height a and length l of the reinforced bars. The deformation capacity, vertical force vs. deformation response, failure mode, and contributions of the flexural and catenary actions on the reinforced bars were evaluated. The test results demonstrate that the kinked reinforced bars started to work when the reduced-beam-section connection with reinforced bars (RRBS) is in a plastic state. When the reinforced bars with a > 1.5d were installed at the beam end, the bearing capacity of the RRBS specimens increased with an increase in the diameter d and decreased with an increase in the length l of the reinforced bars. In addition, When the cracking first occurred at the reduced section, the bearing capacity decreased with an increase in the kinked height a. It is suggested to select the kinked height of the reinforced bar according to the deformation capacity of the beam.

Experimental investigation on progressive collapse resistance of precast concrete frame structures with different beam-column connections

In this paper, twelve 1/4-scale precast concrete (PC) substructural specimens with different beam-column connections under progressive collapse scenario were tested and the results were compared with that of the corresponding conventional reinforced concrete (RC) subassemblies. For assembled monolithic concrete (AMC) specimens, the top longitudinal reinforcement bars of beam passed though the beam-column joint continuously. However, the bottom longitudinal reinforcement bars of beam were either anchored as a 90° bend or lap spliced in the cast-in-place joints. Moreover, the flange effects as well as the effects of transverse beam and slab were also investigated according to spatial subassemblies with or without precast composite slabs. For fully assembled concrete (FAC) specimens, dowel bars and corbel connections, as well as steel plates and hidden corbel connections were used in the beam-column joints. Test results showed that the AMC specimens with or without precast composite flanges (slabs) exhibited almost the same progressive collapse resistance as RC specimens in the literatures, except for the peak value of the vertical load at the compressive arch action (CAA) stage for the AMC specimen with lap-splice of bottom rebars in the joints. Significant CAA at small deformation stage and subsequent tensile catenary action (TCA) at large deformation stage developed in the laminated beams under column removal scenarios. Furthermore, the transverse beam and composite slab effects could obviously increase the capacity of an AMC two-span beam. However, unlike RC specimens, CAA and TCA were not completely developed in the FAC specimens due to pull-out failure of the rebars in the beam-column joints or premature failure of the welding points between steel plates and rebars.

Numerical analyses of post-fire beam-column assemblies with WUF-B connections against progressive collapse

This paper presents the progressive collapse experiment efforts in beam-column assemblies with WUF-B connections after fire exposure. A refined numerical simulation approach that considers the damage and failure for ductile metals was introduced to understand the structural response of the post-fire specimens. The established numerical model could effectively simulate the failure patterns, key fracture phenomenon, load–displacement response, as well as strain evolutions of the specimens. Subsequently, an abundance of parametric studies was performed based on the validated numerical approach to investigate the effects of variable configurations of WUF-B connections on the progressive collapse resistant performance of the beam-column assemblies. It was demonstrated that the vertical resistance and deformability of the post-fire assemblies could be obviously improved by enlarging the shear plate thickness. However, the bolt preload exerted little influence on the structural behaviors of the post-fire assemblies under column removal scenarios. For the assemblies with WUF-B connections, increasing the end distance between the inner bolt hole and the middle column flange (namely, parameters a) as well as decreasing the vertical bolt hole spacing (namely, parameters c) could effectively enhance the vertical resistance and deformability of the specimens. On account of the remaining unchanged crack propagation and force transfer path of the specimen, the horizontal bolt hole spacing (namely, parameters b) negligibly influenced the structural response of the specimens. In the anti-collapse design of the assemblies with WUF-B connections, it was suggested to appropriately increase the values of parameters a and b, but properly reduce these of parameter c on the premise of meeting the designed bolt configurations, which was more conducive to the formation and development of the catenary actions. Eventually, two typical reinforcement methods used for the post-fire assemblies were also numerically studied and compared in terms of the structural behaviors of strengthening specimens against progressive collapse.

Disproportionate collapse prevention analysis for post and beam mass timber building

Disproportionate collapse prevention of mass timber buildings is becoming a relevant design consideration due to the increasing height of these buildings. Linear static alternative load path analysis (ALPA) is well-established with extensive details and worked examples for concrete, steel, and light-frame wood buildings. The research presented herein illustrates its application to a mass timber building with floors supported by beams and columns, referred to as ‘post and beam’ system. The performance of a six-storey post and beam mass timber building, using predominantly wood-to-wood bearing-type connections, following three notional element removal scenarios (removals of ground floor internal, edge, and corner columns) is investigated. It is evaluated whether the case study building’s structural system can trigger floor collapse-resistance mechanisms by checking if its primary elements and connections satisfy disproportionate collapse acceptance criteria, defined by the demand-to-capacity ratio after element removals. The results demonstrated that: i) element removal analyses should be conducted in 3D and should include all primary structural elements, with floor carrying its own weight at every level; ii) compressive arching is dominant for members with small span over depth ratios, whereas catenary action governs for larger spans; iii) continuous beams and slabs help the floor system to cantilever over the removed elements; and iv) the considered connection detail may be improved by the addition of axial connectors to trigger catenary action.

Numerical analysis on collapse performance of post-tensioned concrete frames with post-tensioning BFRP bars under an interior-column-removal scenario

Basalt fiber reinforced polymer (BFRP) bars are ideal substitutes for steel reinforcement due to their excellent mechanical performance, which could avoid energy consumption and carbon dioxide emission during the production of construction steel. Concrete structures reinforced with BFRP bars encounter the collapse risk under an occasional accident, such as losing one or several load-bearing members. To date, little research has been performed on the collapse behavior of post-tensioned concrete frames with bonded post-tensioning BFRP bars. This paper developed finite element (FE) models of frame substructures with post-tensioning BFRP bars to investigate their collapse behavior under an interior-column-removal scenario. Parametric analyses show that arranging post-tensioning BFRP bars at the top and bottom of the frame beams or increasing the post-tensioning BFRP bars at the beam bottom could effectively improve the collapse resistance and deformability of the substructures, with the 96.6% and 51.9% improvement at the most, respectively. Reducing initial effective prestress is helpful to the deformability of the frame beams, resulting in a 56.9% increase in ultimate vertical deformation. A mixed reinforcement of BFRP bars and steel reinforcing rebars in the beams is beneficial in preventing collapse. In most cases, the beams’ flexural mechanism and catenary action are essential in resisting the collapse load. The ratio of the maximum vertical displacement at the failed interior column to the net span of the frame beam varies between 0.051 and 0.1. The calculation model for equivalent dynamic increase factors (DIFs) of collapse load was proposed, which could provide a reference for determining DIFs when static collapse analysis is conducted. The DIFs of collapse load for the substructure models were discussed in the proposed method based on the analytical results, indicating that DIFs approached 2.0 at the initial collapse stage and varied between 1.26 and 1.70 at the final collapse failure.

Effects of Floor System on Progressive Collapse Behavior of RC Frame Sub-Assemblages

The ability to predict the resistance of reinforced concrete (RC) structures to progressive collapse as a result of an interior column removal has become a need in structural design. In general, three resistance mechanisms characterize the structure resistance to progressive collapse, flexural action, compressive arch action, and tension catenary action. The objective of this study is to investigate the effects of floor system configurations on the progressive collapse-resistance of RC frame sub-assemblages and the amount of energy dissipated in each resistance mechanism. This investigation employs a fiber element-based modeling technique to present findings into the effects of beam size and reinforcement details on the progressive collapse-resistance and energy dissipation of RC beam-column sub-assemblages with four equal spans. Three different span lengths of 5, 6, and 7 m were considered. A total of 38 floor system designs for gravity loads were performed in accordance with the ACI 318-14 design code. The modeling technique employed in this study was validated and utilized by the authors in previously published works. The study shows that beam size and the presence of slab are critical as they significantly affect the energy dissipation and progressive collapse-resistance and failure pattern of the sub-assemblage frames. Moreover, the presence of a slab was found to increase the energy dissipation by around 28%.

Analytical model on progressive collapse resistance of prestressed precast concrete frame under middle column removal scenario

This paper investigates an analytical model on progressive collapse resistance of prestressed precast concrete (PC) frame structures. The analytical model is established based on a sub-frame with two stories and two spans under middle column removal scenario, in which the vertical resistance of two-span beam in each story is obtained according to force equilibrium and deformation compatibility of the beam. The tension-stiffening effect and bond slip of top bars near side joint interface are considered in the model to predict the collapse resistance of the beam at compressive arch action (CAA) stage and tensile catenary action (TCA) stage accurately. Besides, to consider the distinct mechanical behavior of beams in failed span in different stories in the process of progressive collapse, the model is related to the restrained stiffness of beam ends, which is determined according to deformation compatibility and force equilibrium of side column in frame. The accuracy of the analytical model is verified according to the results of tests and numerical simulation. Parametric studies are conducted based on the proposed analytical model to investigate the effects of structural parameters and boundary conditions on vertical resistance of the two-span beam in the failed span. The results show that increasing the length of dissipated segment of top bars can improve the structural ductility in progressive collapse, and increase the TCA capacity. Increasing the initial tension of steel strands can enhance the CAA capacity, but the effect of enhancement is alleviated with the increase of initial tension. The larger restrained stiffness can increase the CAA capacity of beam, and the translation stiffness has a more prompt effect on the vertical resistance of beam than the rotation stiffness.

Robustness of Post-Tensioned Concrete Beam-Column Subassemblies under Various Column Removal Scenarios

To fully recognize the load-resisting mechanisms of post-tensioned concrete (PC) structures with realistic boundary conditions against disproportionate collapse, four beam-column subassemblies were extracted from a prototype building and the side columns and joints were reproduced to reflect the actual boundary condition. The parametric analysis was conducted, including location of the removed column (middle or penultimate) and strand profile (straight or parabolic). In addition, two RC counterparts were tested as a control group. Test results indicate that the unbonded post-tensioning strand (UPS) was able to enhance the structural robustness by increasing compressive arch action capacity of RC beams and developing catenary action. Compared with RC specimens, both PC specimens achieved much higher load resistance; herein, the PC specimen with straight strand profile obtained the highest load resistance due to two strands used, while the PC specimen with parabolic profile had higher deformation capacity. However, the existence of UPS increased the tensile force demand to the side column, leading to the flexural tension failure of the side column when the loss of a penultimate column was considered. Finally, an analytical study was carried out to quantify the load resistance from each dominant load-resisting mechanism.

Damage assessment of perforated steel beams subjected to blast loading

Despite the many studies that discussed the structural performance of steel beams with solid web subjected to blasts, the dynamic responses of steel beam with perforated web (PSBs) are not studied up under blast loads till now. In this research, the damage evaluation was numerically investigated under the combination of gravity and blast loading to simulate the realistic case of the external blast near PSBs. The damage indices that will be generated from the residual flexural and shear capacities could give an accurately express about the damage indices than the immediate deformation response after the direct blast shock. Extensive explicit FE analyses using LS-DYNA software are accomplished in order to examine the effects of several parameters including boundary conditions, web opening shapes, and strengthening techniques on toughness and the proposed damage index of the PSBs compared to the parent steel sections. From the comparative results, it was observed that PSBs provide a higher toughness in comparison to their parents through the overall stages of loadings. By changing the boundary condition from pinned to fixed ends, the deformations of PSBs decrease during blast shock directly, despite the toughness increases post the blast effects due to the catenary action that generated by the beam flanges. The beams with sinusoidal web openings and those with elongated circular shape get more toughness than other shapes in the case of fixed and pinned ends respectively. Furthermore, the strengthening strategy by continuous vertical stiffeners has a reliable improvement to get higher toughness in several levels of blast.

Simulation and analysis on improving progressive collapse resistance of RC assemblies by embedding locally debonded rebars near beam-ends

A novel method has been experimentally validated using locally debonded short rebars embedded at the beams' half-height near beam-ends to improve the progressive collapse resistance of reinforced concrete (RC) beam-column assemblies. As an extended study, this paper numerically investigated the load transfer mechanism of RC beam-column assemblies, the principle of resistance improvement by the novel method, and the influence of critical design parameters on the effectiveness of the novel method. First, the finite element models were created, with emphasis on the modeling of the bond-slip behavior of reinforcing steel bars in pre- and post-yield ranges. This behavior was found to have significant impacts on the deformation and load-carrying capacities of the RC beam-column assemblies. Second, the finite element models were validated against test data and then employed to understand the load transfer mechanisms. The resistance of the RC beam-column assemblies at the stage of small deformation was found to primarily provided by flexural/arch mechanism, reaching a maximum at middle column displacement (MCD) ≈ 0.5 h (i.e., 150 mm). Once the value of MCD exceeded h (i.e., 300 mm), the catenary mechanism was mobilized and approximately 90% of the resistance at this stage was contributed by the axially tensile force of the beams. On this basis, the principle of resistance improvement using the novel method was discussed, i.e., the progressive collapse resistance of the RC beam-column assemblies was improved through fully mobilizing catenary action and maintaining the flexural mechanism in the beams. Finally, parametric studies were conducted with consideration of the debonded length (lpd) and the reinforcement ratio (ρs,mid) of locally debonded short rebars. Results indicated that the appropriate lpd ranged from 600 to 1000 mm, and the proper ρs, mid varied from 1.5% to 2.2% when used in engineering practice.