Three-dimensional finite element analysis and optimization of raft foundation systems supported with rammed aggregate piers

In this thesis, numerous three-dimensional finite element analyzes were conducted to reach the optimum design within the framework of safe design and cost balance in raft foundation systems supported by rammed aggregate piers (RAP). In the first stage, a finite element analysis was performed, and the model was verified by three-dimensional modeling of a single RAP, which was subjected to a loading test in the field. Behind verification, 81 different combinations of raft foundation-RAP group systems were created and analyzed by varying the distance between the RAPs, the raft foundation thickness, and the RAP elasticity modules. Deformations of all RAPs and raft foundations in all systems were examined by creating detailed graphics. An optimization study was conducted utilizing the novel Goal Attainment Method supported by artificial intelligence and the Response Surface Method with the data obtained from the analysis results. In the optimization study, all the inputs were in range, while the dependent variables of cost and deformation were minimized in both methods to reach a balanced design. The effects of varying parameters on settlement and cost were examined. Two different methods specified in the optimizations performed as multi-objectives were compared with each other. The solutions offered by the artificial intelligence-supported novel Goal Attainment Method, which has a higher desirability value, were determined as the optimum design parameters. The optimization analysis was verified by performing a new three-dimensional finite element analysis based on the data obtained from the optimization study. Accordingly, optimum design parameters for raft foundation-RAP systems have been received, with a cost-safety balance that will guide project practitioners and researchers.