Foundations behavior is affected by soil behavior which can vary from dilative to contractive depending on the stress level,

particularly in dense frictional soils. The Zero Extension Lines (ZEL) method has been generally developed to predict the

foundations behavior. Knowledge of soil behavior enables the ZEL method to predict the general and local shear failure modes.

In this paper, a relatively simple work hardening/softening soil constitutive model is developed to represent dense frictional soils

behavior under different stress levels. This model is based on the accumulation of the plastic work during a simple direct shear

test and its relationship to stress ratio to establish the hardening law. Verifications have been made for the developed soil model.

The model is then implemented into the ZEL method to theoretically investigate the bearing capacity and load-displacement

behavior of foundations over dense frictional soils. Utilization of this model enables the ZEL method to capture different modes

of failure depending on the foundation size. A numerical study on foundations behavior was performed showing the ability of the

presented approach in capturing both failure modes.

It has been realized that the raft (mat) foundations are capable of bearing very large loads when they are assisted with a pile
group. The contribution of both raft and piles to carry the surcharge loads is taken into account, considering the stiffness and
strength of involved elements in the system, i.e. piles, raft and surrounding soil. The piles are usually required not to ensure the
overall stability of the foundation but to act as settlement reducers. There is an alternative design in which, the piles are nonconnected
from the raft to reduce the settlement, which are then known to be "settlement reducer non-connected piles" to increase
the system stiffness. In this paper, two and three dimensional finite element analysis of connected and non-connected pile-raft
systems are performed on three case studies including a 12-storey residential building in Iran, a 39-storey twin towers in
Indonesia, and the Messeturm tower, 256m high, in Frankfurt, Germany. The analyses include the investigation of the effect of
different parameters, e.g. piles spacing, embedment length, piling configuration and raft thickness to optimize the design. The role
of each parameter is also investigated. The parametric study results and comparison to a few field measurements indicate that
by concentrating the piles in the central area of the raft foundation the optimum design with the minimum total length of piles is
achieved, which is considered as control parameter for optimum design. This can be considered as a criterion for project cost
efficiency. On the other hand, non-connected piled-raft systems can significantly reduce the settlements and raft internal bending
moments by increasing the subsoil stratum stiffness. Finally, the comparison indicates that simple and faster 2D analysis has
almost similar results to the time consuming and complicated 3D analysis.