Robustness analysis for innovative tall composite modular buildings with composite shear walls

This research investigates the robustness of tall composite modular buildings using alternate load path method (ALP) subjected to sudden element removal under the effect of gravity loads. A finite element model of a 50-storey modular building was developed with concrete filled steel tubular (CFST) columns and composite shear walls (CSW) in Abaqus. Core CSW were designed instead of conventional reinforced concrete shear walls to benefit with their modularity, effective time frame and better load sharing. Conventional steel columns were replaced with CFST columns to provide improved buckling and post buckling resistance. Non-linear dynamic and static analyses were conducted to investigate the real-time response, load transfer behavior and the ultimate failure scenarios. The 50-storey composite modular building showed sufficient resistance against progressive collapse under column and module loss scenarios. It was observed that the inter-module connections play a critical role in a module loss scenario, whereas the adjacent beams play a crucial role in the event of column removal. Dynamic amplification factors (DAF) were interpreted using the non-linear static analysis and compared with the available guidelines by general service administration (GSA). DAF of 1.20 was calculated for the 50-storey modular building subjected to critical corner module removal which was significantly lower than the value of 2.0 as suggested by GSA guidelines. A detailed parametric study was conducted to evaluate the DAF for various locations in the floor layout and the elevation of building. DAF of 1.223 and 1.216 were observed for a 30-storey and 40-storey modular building, respectively. • Robustness analysis was conducted for tall composite modular buildings with composite shear walls. • Critical initial damage scenarios were investigated for progressive collapse behavior. • Dynamic amplification factors were suggested from a detailed parametric study. • The new modelling methods were very accurate and computationally efficient.