A numerical comparison of installation sequences of plain concrete rigid inclusions

Soil displacement induced when installing controlled modulus columns (CMC) as ground reinforcement could affect the columns installed close by. Realising numerical analyses may provide useful insights, this paper describes a numerical approach to investigate how groups of CMC installed in different sequences could affect columns installed previously. Coupled consolidation analyses in large strain mode and incorporating soil-CMC interaction were carried out using the three-dimensional finite difference software package FLAC3D. The CMCs were modelled using advanced non-linear Hoek-Brown material with a tensile yield criterion while soils with a typical profile were characterised using the modified Cam Clay and the elastic-perfectly plastic material with a Mohr-Coulomb yield criterion. Where possible, the predicted responses of ground surrounding the CMCs were compared to a number of existing analytical methods. Predictions revealed that lateral soil movement and soil heave near existing CMCs induced by installing new CMCs towards the existing CMCs were approximately 15% and 25% greater than corresponding predictions when a reverse installation sequence was adopted. The maximum excess pore water pressures, induced near existing columns due to installing new columns towards the existing ones, were almost twice more than those caused by the reverse sequence of installation. Moreover, the predicted bending moments generated in the existing columns induced by installing new columns towards the existing CMCs were almost 22% greater than the corresponding values when the reverse installation sequence was adopted. This shows the importance of selecting an appropriate installation sequence in the CMC construction process as well as considering the initial stress field and bending moments in the surrounding soil and CMCs, respectively when designing embankments on improved soft soils.