Re-Analysis of a Levee on Soft Bay Mud using elasto-viscoplastic constitutive model MIT-SR

Abstract: One of the main problems that still could not be rationally solved in geotechnical engineering is creeping. There is a huge dispute on how rate effects are extrapolated from the lab to the field. Some support the idea that the field soil and the laboratory specimen undergo the same compression strain during consolidation, and others say that the field strains are much larger due to the longer duration of the consolidation process. The recently developed constitutive model MIT-SR (Whittle-Yuan) proves that none of them are wrong. In fact both behavior can be captured and the result will depend on the material properties and the strain rate at which is being deformed. Consequently, my research is based in the incorporation of MIT-SR to a Finite Element software in order to solve a well-known and documented engineering case. It consist on re-analyzing the performance of the New Hamilton Partnership Levee on San Francisco Bay Mud. This structure settled more than 2.5 foot after its construction and unfortunately still any elastoplastic model could accurately predict it. Then, one possible explanation is the huge influence of creep on it.