Severe soil movements can be detrimental to the integrity of adjacent piles. In this paper, the simpler two-stage analysis is evaluated to examine its validity to predict the pile behavior due to soil movements. The lateral soil movement profiles obtained from centrifuge model tests are used as the input parameters for the second stage analysis using a beam on elastic foundation program. It has been found that the correct inputs of soil movement magnitude and curvature are crucial in arriving at a reliable prediction of induced pile moment and deflection.
With rapid development of infrastructure works in onshore, marine and offshore environments, there have been reported incidences of foundation damage due to soil movements arising from construction or installation activities in the vicinity, see for example Finno et al. (1991). Such soil movements could induce severe stresses on adjacent piles and may cause excessive pile movements or even pile structural failures. Examples in onshore environment include basement and slope excavation as well as tunneling induced soil movements on piles. For offshore projects, the soil movements due to installation of mobile jackup rig spudcan foundation may induce severe stresses on the piles supporting the adjacent permanent jacket platform. For the past two decades, the National University of Singapore (NUS) has carried out extensive centrifuge model studies to investigate the effects of soil movements on adjacent piles. The results of centrifuge model and numerical studies as well as field study on the effects of excavation induced soil movements on adjacent piles in sand and in clay were reported by Leung et al. (2000, 2003 and 2006) and by Ong et al. (2006, 2009 and 2015). Centrifuge model and numerical studies on the interaction between spudcan and adjacent piles have also been studied at NUS and the results were reported in detail by Xie et al. (2012 and 2017) and Tho et al. (2013).
For the numerical study on spudcan-pile interaction, Tho et al. (2013) established that it is necessary to carry out large deformation finite element analysis to achieve an accurate evaluation of the pile performance due to the very large magnitude of soil movements during spudcan installation. However, the large deformation finite element analysis is very costly and time consuming. Instead a two-stage analysis is often adopted in practice with the first stage being finite element analysis conducted to predict the free-field soil movement due to spudcan installation. The soil movement profile at the pile fictitious location is then used as input parameters on subsequent analysis to evaluate the pile behavior. This paper aims to evaluate the pros and cons of such two-stage analysis and recommendations are put forward on the dos and don'ts when adopting the two-stage analysis.
