Summary | The effects of different beam-column plastic-hinge modelling assumptions on the seismic behaviour of steel frames are studied through Incremental Dynamic Analysis (IDA). The wellknown 9-story LA9 2D steel frame is used as a testbed by adopting multiple possible momentrotation relationships for the beam plastic-hinges. These are modelled as full quadrilinear backbones with pinching hysteresis, including an elastic, a hardening, a negative stiffness and a residual plateau branch, terminating with a final drop to zero strength. The properties considered include the post-yield hardening ratio, the end-of-hardening rotation, the residual moment capacity and the ultimate rotation reached. The hinge parameters are varied one at a time, globally throughout the building, generating several plausible structural models which differ only in the adopted connection model, some being more brittle and others more ductile. The seismic performance of each resulting frame is then evaluated using IDA, i.e., by performing multiple nonlinear time history analyses for a suite of ground motion records appropriately scaled to several intensity levels. By appropriately post-processing the results the median IDA curves are estimated, forming a solid basis for comparing the different models. Thus, we are able to evaluate the influence of several plastic-hinge modelling assumptions to the seismic performance of the frame and understand the sensitivity of such results that are usually estimated on the basis of a single structural model.