Degree Name

Doctor of Philosophy


School of Civil, Mining and Environmental Engineering


Rapid population growth and the correspondingly increased demand for infrastructure necessitate effective ground improvement to support surface infrastructure. Due to the highly compressible and problematic soft soil that commonly exists in the coastal areas of Australasia, improving the ground before construction by giving it sufficient time to reach post-consolidation settlement is imperative. But these time constraints must be weighed against the urgency of the project and its projected completion date. Certainly there are various types of ground improvement techniques available for soft soil, but prefabricated vertical drains (PVDs) combined with surcharge and vacuum preloading is one of the most popular, effective, and environmentally friendly techniques. Here, the drains are driven into the ground to a considerable depth (normally the thickness of soft clay) to accelerate the dissipation of pore water pressure by promoting radial consolidation with a shortened drainage path length. The vacuum used in this system also accelerates dissipation by creating suction in the drain which helps the soil reach postconstruction settlement much earlier.

Soft soil, possesses time-dependent stress-strain behaviour due to its viscous nature, and this visco-plastic behaviour has an effect on long term settlement and pore water dissipation. A novel mathematical model has been developed to describe the viscoplastic behaviour of soft clay with a non-Darcian flow function; it was developed by coupling the basic radial consolidation equation developed by Barron combined with Bjerrum’s time-equivalent (Bjerrum, 1967) concept that incorporates Yin and Graham (1989b) visco-plastic parameters. The settlement and excess pore water pressure obtained from this model are compared with pre-existing models as a Class C prediction for the Ballina trial embankment. The proposed elastic visco-plastic model gave better results in terms of settlement and pore water pressure with the field data, although the excess pore water pressure that did not dissipate after 1 year or so is mainly due to the biological and chemical clogging of piezometers in acid sulphate soil (ASS) terrain. These predictions were also compared with the results from finite element method using PLAXIS by performing both plane strain and actual 3D models, respectively.

Similarly, to investigate the elastic visco-plastic behaviour of soft soil more clearly, the concept of strain rate dependency of pre-consolidation pressure was used along with the isotache concept. Since pre-consolidation pressure is a function of the strain rate (which was high in the laboratory experiment but very low in the field), a pressure ratio can be obtained from the strain rate and pre-consolidation pressure plot to convert laboratory pre-consolidation pressure into field pressure. Delayed consolidation could then be discovered using the change in effective stress from a pre-consolidation pressure change which actually retarded the dissipation of excess pore water pressure. This model was then used to validate the different case histories (Pacific Highway, Australia, SBIA, Bangkok and Muar, Malaysia); the results obtained using the isotache model are more promising than the other pre-existing radial consolidation model.

Laboratory tests using samples of remoulded and undisturbed Ballina clay were carried out in UOW laboratory and tested using a Rowe cell and large scale consolidometer. This 350mm diameter large scale consolidometer cum corer was used to extract an undisturbed sample from the field, and then transported into the laboratory and tested with appropriate instrumentation to determine the behaviour of this undisturbed marine soft clay containing random seashells and natural partings. Different consolidation and smear parameters were obtained from the large scale consolidometers, which were then used to predict the exact behaviour in the field, and they differed from the parameters obtained using the remoulded sample.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.