Prediction and optimization of the bearing capacity of strip footing resting on soft soil improved with stone columns using RSM, ANN, and multi-objective GA

This paper studies the multi-objective optimization of the parameters influencing the bearing capacity of a strip footing resting on different soft soils improved with stone columns. Bearing capacity and cost construction of columns are considered objectives. According to an appropriate full factorial design of experiments, the finite difference software FLAC2D can be used for this problem to simulate a two-dimensional numerical model in cross section. However, the effects of column parameters (diameter and depth) and the geometric one of the model are studied using ANOVA analysis. The response surface methodology (RSM) and the artificial neural network (ANN) are considered for developing mathematical models that can be used later for the optimization process in combination with the genetic algorithm (GA). After comparing the two predictive methods, it is important to highlight the fact that the ANN tool provides more accurate models compared to the RSM method. Finally, ANN predicted models coupled with GA are used to determine the optimal values of the design variables of the proposed problem.