The optimal design of supported excavations

In this thesis, the design variables of braced supported bored pile retaining structures have been investigated using the JAYA metaheuristic optimization algorithm with continuous data for the purpose of cost optimization. Plaxis 2D software has been selected due to its ability to model realistic soil/structure interaction conditions such as soil heterogeneities, anisotropy, and nonlinear behavior. The integration of Plaxis software with the Python programming language enabled the automatic execution of optimization processes and a more effective analysis of the results, owing to its ability to work directly with the JAYA metaheuristic algorithm. In the first part of the study, a parametric optimization of the bored pile shoring system supported by steel pipes in the head beam was conducted. Design parameters such as pile diameter, steel pipe support diameter/wall thickness, and pile socket length were optimized for cost minimization across different soil types and excavation depths. In the second part of the study, the temporary excavation support structure of a metro line in Istanbul was evaluated in terms of cost optimization through a real-world application. The results demonstrate that effective and reliable optimization solutions can be achieved for internally supported bored pile retaining structures by using the JAYA algorithm in conjunction with Plaxis software. This method has shown a high potential for cost savings, particularly in complex ground conditions.