Grant: Strategic Projects - Group

Bridge behaviour during an earthquake depends on the design of the superstructure and the effects of foundations and abutment embankments on the bridge. These effects are called soil-structure interaction effects. The best approach to predicting how the bridge will behave involves a very complicated analysis that includes bridge, foundation soils, foundations and abutments. As well as being complicated the analysis is very time-consuming and does not encourage studying the behavior of the bridge under different conditions. Therefore designers have adopted simple approximate ways of including the effects of foundations and abutments on the response of the bridge to earthquake shaking. The problem is that while these methods have some rational basis they have never been validated by comparing them with results from full scale analyses. One of the aims of this study is to remedy this serious omission by testing how these simpler models compare with the more exact analyses. Then, if they prove deficient, we propose to either improve existing models or develop new ones. This work should lead to better and more cost effective designs. Effective simpler modes are imperative because fullscale analyses are not a viable design tool. Their primary functions should be restricted to evaluating a design reached by simpler methods. During a major study of this kind, we learn a lot about how soil-structure interaction affects the response of different kinds of bridges. Currently this information is largely lacking to the practising engineer and we expect this data will lead to recommendations for seismic provisions in the bridge code that will lead to improved design generally.