Integrity and progressive collapse resistance of RC structures with ordinary and special moment frames

•Discusses difference between response of structures to earthquakes and element loss.•Special frames may not perform better than ordinary frames after element loss.•Longer span special frames perform better than ordinary frames after element loss.•SDOF model closely estimates nonlinear response of frames following element loss.•Element torsional stiffness & concrete tensile strength affect response considerably.

General building codes and standards include structural integrity requirements as an indirect approach to mitigate the likelihood of collapse of structures due to abnormal loading conditions. Structures are designed for load combinations including live, dead, wind, earthquake, and other loads. Particularly for earthquake loads, different design approaches are used with the general understanding that for sites with high seismicity, structures need to be ductile and in turn may be designed for relatively smaller base shear. In this paper the effects of seismic design and structural integrity requirements on progressive collapse resistance of reinforced concrete frame structures is evaluated. The relative importance of ductility (deformation) capacity and strength is discussed for response of structures subjected to severe seismic ground motions and to loss of a column. The effects of span length on a building’s response after column removal are discussed. It is demonstrated that for buildings with shorter spans at sites with low to medium seismic severity, designing for higher seismicity does not necessarily lead to a better performance and smaller vertical displacement following loss of an exterior column. Effects of other parameters, such as the joist torsional stiffness and concrete tensile strength are also discussed. An approximate method using equivalent single degree of freedom systems is presented for evaluating maximum displacement response of structures after element failure, which estimated the building responses studied in this paper with a maximum error of 13%.