Inverse Analysis of a Supported Excavation in Chicago

Abstract

Adaptive management is a design approach which allows one to change support and/or construction procedures during excavation based on observed responses to meet specific design objectives. To obtain the most benefit from an adaptive management approach, inverse analysis procedures are used to make parameter updates based on observations collected during the early stages of a project and applied to later stages to predict future performance. This paper describes the inverse analysis of lateral wall deformation data collected from a supported excavation in Chicago. The Hardening Soil (HS) model and the HS model with small-strain stiffness are used to represent stress-strain behavior of medium clay and stiff clay layers of an excavation in Chicago. Initial parameters for the soil models are based on previous inverse analysis studies conducted for excavations in Chicago soils. Deformations computed using the initial parameters are compared with the field measurements to evaluate the capabilities and limitations of each model. Optimization during inverse analysis is based on a gradient method using measured lateral wall deformations. Changes in the optimized parameters during various levels of excavation are assessed. Computations made with the optimized soil parameters are compared and their efficacies for prediction purposes are evaluated.