Abstract:  Determination of the specific area replacement ratio is crucial in construction， considering the uncertainty of mechanical parameters of soft soil itself and the balance between postconstruction settlement satisfaction and cost. Based on the robust geotechnical design principle, the pile and soil constrained modulus is regarded as noise factors with uncertainty, and by using the MonteCarlo simulation method the value of several possible noise factors is determined; Further, the standard deviation of postconstruction settlement is calculated by the oneway layerwise summation method as the measure of robustness; Finally, the area replacement ratio of gravel pile could be optimized with considering safety, cost and robustness simultaneously. In this study, a robust design framework of stone column was proposed and its application was illustrated through an engineering example. Based on this example, the influence rules of settlement limit, coefficient of variation of compression modulus, and parameter distribution mode on the optimization were discussed. The results show that the optimal replacement rate is 29%; The improvement of settlement limit increases the engineering cost; The compressive modulus of soil has a greater influence on the final optimization than that of gravel pile; When using normal instead of lognormal distribution to simulate the noise factor distribution, the results of this calculation are more uncertain.

Key words: constrained modulus, stone column, composite foundation, optimization design, area replacement ratio, robustness