---

Pushover analysis is a simplified nonlinear analysis technique that can be used to estimate the dynamic demands imposed on a structure under earthquake excitations. One of the first steps taken in this approximate solution is to assess the maximum roof displacement, known as target displacement, using the base shear versus roof displacement diagram. That could be done by the so-called dynamic pushover analysis, i.e. a dynamic time history analysis of an equivalent single degree of freedom model of the original system, as well as other available approximate static methods. In this paper, a number of load patterns, including a new approach, are considered to construct the related pushover curves. In a so-called dynamic pushover analysis, the bi-linear and tri-linear approximations of these pushover curves were used to assess the target displacements by performing dynamic nonlinear time history analyses. The results obtained for five different special moment resisting steel frames, using five earthquake records were compared with those resulted from the time history analysis of the original system. It is shown that the dynamic pushover analysis approach, specially, with the tri-linear approximation of the pushover curves, proves to have a better accuracy in assessing the target displacements. On the other hand, when nonlinear static procedure seems adequate, no specific preference is observed in using more complicated static procedures (proposed by codes) compared to the simple first mode target displacement assessment.

---

The nonlinear static procedures (NSPs) proposed by design codes do not lead to reliable results especially for tall buildings.

They generally provide inconsistent estimates of inelastic seismic demands, especially for the top floors due to their inabilities in

considering the higher modes effects. In this paper, a new enhanced pushover procedure is proposed which is based on the

envelope of the structural responses resulting from two separate pushover analyses as a combination rule. Also, the suggested

pushover analyses are performed using a newly proposed modal load pattern, i.e., the Modal Spectra Combination (MSC), and

the ASCE41-06 required first mode load pattern. The MSC load pattern is consisted of a number of mode shapes combined with

appropriate weighting factors that depend on their modal participation factors, modal frequencies and design spectral values. A

number of 2-D steel moment resisting frame models with different number of stories are used to investigate the efficiency of the

proposed method. The inter-story drifts and the maximum plastic beam moment and curvature responses are used as a measure

to compare the results obtained from the nonlinear time-history analyses (NL-THA) and some other NSPs. The results obtained

through rigorous nonlinear dynamic analyses show that the application of the proposed method leads to acceptable results for

steel MRF systems in comparison to other available enhanced NSPs. The OpenSees program is used for numerical analysis.