Reliability assessment and forecasting of moment ratio/factor of safety for sheet pile walls utilizing hybrid ANFIS enhanced by optimization techniques

Abstract

Addressing the inherent uncertainties in geotechnical engineering, particularly concerning natural materials, this study focuses on the crucial aspect of reliability analysis in geotechnical structures. The research delves into the Factor of Safety, specifically examining the Moment Ratio, for cantilever sheet pile walls in cohesionless soil. The study employs both the First-Order Reliability Method (FORM) and Second-Order Reliability Method (SORM), integrating various optimization techniques such as Genetic Algorithm, Particle Swarm Optimization, Firefly Algorithm, and Biogeography-Based Optimization. The research systematically evaluates the performance of a model, utilizing the widely-adopted Adaptive Neuro-Fuzzy Inference System (ANFIS) to predict pile moment ratios based on soil properties such as the angle of shearing resistance and unit weight. Rigorous criteria are applied to assess the model's efficacy, revealing the superior predictive capabilities of the hybrid ANFIS and Particle Swarm Optimization (ANFIS-PSO) model. This abstract encapsulates a comprehensive methodology for evaluating and quantifying risk in civil engineering projects related to cantilever sheet pile walls, providing valuable insights for the development and implementation of robust structures in the field.