Investigation of the bearing capacity of strip foundations resting on slope with statistical approach

This thesis aims to statistically investigate the bearing capacity of strip foundations resting on slopes. The bearing capacity of foundations on slopes is more complex compared to foundations on flat surfaces and is influenced by numerous factors. Therefore, statistical modeling techniques are employed to overcome the limitations of existing analytical approaches and make more reliable predictions. Within the scope of the study, internal friction angle, cohesion, foundation width, depth of embedment, distance of the foundation to the slope, slope height, and slope inclination angle were determined as independent variables. The effects of the independent variables on the bearing capacity coefficient determined as a response were analyzed statistically. In this context, the Response Surface Method (RSM) was used. With the central composite design approach, 142 cases with different values of the independent variables were created. For each case, basic monotonic loading was applied, and the bearing capacity was determined by the finite element method. The load obtained was converted into a dimensionless bearing capacity coefficient and defined in the test matrix as a response. As a result of the statistical analysis, the appropriate model was determined and validated. Using the obtained statistical model, the individual effects of the independent variables on the bearing capacity coefficient were analyzed. In addition, considering that there may be an interaction between the independent variables, the parameter effect was analyzed with Pareto diagrams. Pareto diagrams were prepared by considering the F-values determined for each independent variable as a result of ANOVA. The results obtained show that the bearing capacity of foundations resting on slope is directly related to the soil properties and slope geometry. Additionally, it was demonstrated that the Response Surface Method is an effective technique for modeling and optimizing the behavior of complex geotechnical systems.