Behaviour of a piled raft foundation during its construction and first operating conditions The expansion of Customs and Excise Station in Fiumicino (Rome, Italy) involved the construction of some buildings in proximity to Tevere river delta and Leonardo da Vinci international airport. With reference to figure 1, the project included the design and the construction of: two main buildings (A and B, whose dimensions in plan are about 14 m x 58 m and 18.5 m x 48 m, respectively); a technological building (E); two smaller structures (not in the figure). The poor mechanical properties of the subsoil (about 70 m of a thick deposit of normally consolidated alluvial clay resting on a gravel layer; Figs. 2 and 3) suggested for the main buildings the use of piles of displacement type. In particular, jacked steel tubular Soles piles were driven up to the gravel. The choice of a proper pile type was attentively considered in order to fulfil two conflicting requirements: to install very long piles (70 m and more) with a very limited in height equipment (15 m) in order not to interfere with air space of the nearby landing strip n. 3 of the airport. Both the requirements were accommodated by adopted pile, installed by a 3 m height equipment. In the paper, after a short description of subsoil profile and soil properties, some information are given about the project of the piled foundations. Moreover, a description of the monitoring system is also given (Fig. 5); it consists of: high precision topographic survey of 18 benchmarks installed on the rafts (9 for each raft, Figs. 8a and 12b); pore pressure measurements by an electrical piezometer transducer (Figs. 8b and 12d); several vibrating wire transducers placed along 2 pile shafts and at the head of other 5 piles. Due to the specific pile installation procedures, monitoring started during the installation of the piles in order to measure displacements of the raft and pore pressure increase. Details about pile behaviour have been obtained by performing 2 load tests on production piles (1.5 times the expected maximum design axial load); one of them (pile 1A; Figs. 10 and 11) has been instrumented along the depth in order to get information about the load transfer along the shaft and at the base. The whole set of information obtained from the monitoring system provided useful indications about the qualitative behaviour of the foundation system. With particular reference to the complex raft-piles-soil interaction, it has been highlighted the major role played by the specific installation procedure of Soles piles: the excess of pore pressure generated during their installation and the continuously variable constraint conditions due to the progressive structural pile-raft connection have determined very different residual loads for the piles, therefore influencing the final distribution of the load among the piles as a function of their installation time, position and connection. If compared with classical analyses of piled rafts, for which the load sharing between raft and piles, as well as load distribution, is mainly governed by soil, raft and pile properties, it is clear that the case under examination is much more complex because of the influence of the specific installation procedure adopted case by case. Further theoretical and experimental investigations are therefore needed and will be carried out in a next future.