Predicted and observed behaviour of pre-installed barriers for the mitigation of tunnelling effects

Tunnelling activities in densely urbanised areas are likely to interfere with existing structures, often requiring the adoption of techniques to mitigate the effects of the induced ground movements. If the tunnel passes to the side of an existing structure, embedded barriers pre-installed between the tunnel and the buildings can act as a passive intervention reducing the displacements that would occur in greenfield conditions. A reliable prediction of the displacement field induced in the presence and the absence of the barrier is, therefore, of primary importance for the design of such a mitigation measure. In this respect, this paper discusses the efficiency of a barrier made of a line of bored piles through the comparison of field monitoring data from a test site in Rome with the results of soil-structure interaction (SSI) FE analyses and simple empirical relationships. Three numerical procedures to simulate plane-strain tunnel excavation were adopted to back-analyse the settlement trough induced at the ground surface by the first passing tunnel. The comparison between the observed and computed settlement profiles permitted to identify the most reliable procedure for 2D simulations of tunnelling excavation which are usually adopted in routine analyses, thus providing some useful guidance for a rational design of the intervention.