In conventional tunnelling, under difficult ground conditions, the fronts are commonly stabilized by employing expensive and time-consuming soil reinforcement techniques. The design of these soil reinforcement techniques is commonly based on the analysis of unreinforced front mechanical response.

In a previous paper, the authors numerically analysed the mechanical response of deep tunnel fronts under undrained conditions and suggested the employment of a non-dimensional front characteristic curve. In this paper, the same approach is employed to study the influence of the excavation rate on the front response. The soil behaviour is modelled by employing the Modified Cam Clay model.

Macroelement plasticity models are being increasingly applied to study non-linear soil–foundation interaction (SFI) problems. Macroelement models are particularly appealing from a computational standpoint, as they can capture the essence of SFI by means of a few degrees of freedom. However, all the macroelement formulations available in the literature suffer from the same limitation, that is the incapability of accounting for changes in both geometry and loading/boundary conditions. Accordingly, macroelement models are usually calibrated to analyse a given boundary value problem, with no chance of handling situations with significant variations in embedment, lateral surcharge and/or phreatic level. The present work shows how standard soil modelling concepts can be exploited to reproduce relevant ‘configurational features’ of non-linear SFI. A macroelement framework is here proposed to simulate the drained load–settlement response of shallow footings on sand in the presence of varying surface/body forces. As a first step, the ideal case of a weightless soil layer is exclusively considered. The macroelement constitutive equations are conceived/calibrated on a minimal set of finite-element results; the satisfactory predictive capabilities of the macroelement model are finally demonstrated by retrospectively simulating selected finite-element tests.

After the abandonment occurred in the '70s, the mining system (rooms and pillars) located in S. Lazzaro di Savena (BO, Italy), grown on three levels with the method rooms and pillars, has been progressively more and more affected by degradation processes due to water infiltration. The mine is located underneath a residential area causing significant concern to the local municipality. On the basis of in situ surveys, laboratory and in situ geomechanical tests, some critical scenarios were adopted in the analyses to simulate the progressive collapse of pillars and of roofs in the most critical sectors of the mine. A first set of numerical analyses using 3D geotechnical FEM codes were performed to predict the extension of the subsidence area and its interaction with buildings. Secondly 3D CFD analyses were used to evaluated the amount of water that could be eventually ejected outside the mine and eventually flooding the downstream village. The predicted extension of the subsidence area together with the predicted amount of the ejected water have been used to design possible remedial measurements.