Generally, neglecting Soil Structure Interaction (SSI) in the analysis of bridges is considered to provide a conservative design, owing to increases in the natural period and the damping of the soil-foundation-structure system. But, during several past earthquakes, effect of soil flexibility is found to have played a significant role in the performance of bridges; this is because of amplification of ground motions in the long period range owing to flexibility of soils and proximity of bridge natural period(s) to the site natural period. But, most design codes are yet to incorporate this effect in the design acceleration response spectrum.

This study examines the impact of soil flexibility on design response spectrum. The acceleration response spectrum are derived by one-dimensional site response analysis considering: (a) soil flexibility for different thickness of soil deposits, (b) shear wave velocity of soil deposits in the range 300m/s–1200m/s (resulting different site periods depending on the thickness of soil strata over bedrock), and (c) intensity of shaking with PGA in the range 0.2g to 1.0g. Statistical analysis is conducted with 60 recorded ground motions, and idealised response spectrum are derived, which are compared with design response spectrum given in codes. A proposal is made for revising the existing design response spectrum towards increasing safety of bridges.

Further, in this study, it is proposed to investigate the earthquake behaviour of a bridge using improved response spectrum. The bridge will be designed using results from two separate linear modal structural analyses, namely:
(a) Soil-Foundation-Substructure-Superstructure System subjected to code-based design response spectrum at the base of the soils system, and
(b) Foundation-Substructure-Superstructure System fixed at the foundation using improved design response spectrum at the bottom of the foundation.
Nonlinear dynamic analyses will be performed of these two designed bridges. In both cases, (a) soil will be modelled as a continuum with nonlinear properties, and (b) the structure (including foundation, substructure, bearings and superstructure) by nonlinear finite elements of the structural components that capture the nonlinearities in them. It is proposed to make recommendations on the analysis and design of bridges to enhance their structural safety with relatively reduced computational effort.