Performance-Based Seismic Design (PBSD) approach embraces explicit assessment of the response of structural components with target building performance objectives. As shallow foundation exhibits inelastic behavior at the interface of the soil-foundation system upon excessive load, realistic assessment of cyclic inelastic rocking response of foundation is recommended [1,2]. Two mechanisms of nonlinearity take place between the soil and foundation; geometric nonlinearity (i.e. rocking response) and material nonlinearity (i.e. yielding of soil). The rocking response of shallow foundations has been one of the key research areas that has gained interest in recent years. Various experimental studies, such as a large-scale shaking table test of a bridge column [3], a small-scale shaking table test of a 3-storey building [4] and a centrifuge modelling of rocking-isolated inelastic RC bridge piers [5], have captured the rocking of the shallow foundation and found that this behavior reduces the residual drift and seismic demand of the structure. However, rocking shallow foundation may also experience large differential settlement during excessive cyclic loads which mainly result from yielding of near-field soil. On this basis, rocking of the foundation and yielding of soil should be carefully analyzed in the PBSD of shallow foundations [4,5]. There exists literature which provides guideline on modelling of shallow foundations. For example, ASCE 41-13 [6] provides a component action table with modelling parameters and acceptance criteria for nonlinear and linear analysis of shallow foundations. The values in the component action tables for nonlinear analysis procedures are based on the analysis of rocking shallow foundation, which was observed from experimental model tests. For the linear analysis procedure, the empirical coefficient, m-factor, is revised to reflect the allowable rotation of the rocking foundation from the nonlinear analysis procedure [7]. Unlike ASCE 41-06 [8] where rocking foundation and yielding at the soil-foundation interface are uncoupled and checked separately, ASCE 41-13 [6] considers the coupled behavior of foundation rocking and yielding of the soil. This approach is more realistic as the failure of the foundation is governed by stiffness degradation and yielding of the soil [9]. Kutter et al. [7] provided a rationale for the revisions made in ASCE 41-13 for rocking shallow foundation and validated these revisions with extensive experimental results [7,10]. The numerical modelling of dynamic soil-structure interaction (SSI) between an inelastic soil domain and a structural model for shallow foundation is a complicated task. The model entails an infinite soil domain, interface property between the structural foundation and soil, and verification and benchmark analysis. The Finite Element (FE) analysis is a rigorous method which is able to model the infinite soil domain with an arbitrary geometry and diverse soil layers. However, special measures should be taken to accurately model the boundaries of the numerical model. The scattered waves from a structure should be dissipated or absorbed at the boundaries in order to avoid the wave reflection. This means that the numerical domain should be large enough to avoid the negative effects of the reflected wave to the structural responses.