Probabilistic stability analysis of anchored cantilever sheet pile walls using fuzzy set theory

Abstract

Ensuring the safety and stability of waterfront retaining structures is of utmost importance. However, uncertainties may lead to imprecise analysis and inadequate or uneconomical design. Fuzzy set theory is one of the computationally efficient methods of evaluating the effects of uncertainties on systems and has a great potential of making probability-based stability analysis of sheet piles considerably less cumbersome. This paper analyzes the effects of uncertainty on the stability of anchored cantilever sheet pile walls in the sand using the fuzzy set theory and genetic algorithm. The uncertain variables include soil properties as well as structural and geometrical parameters. Different levels of uncertainty are assigned to the governing parameters using the triangular fuzzy membership functions and the α-cut method. To generate the fuzzy responses for wall stability, a search for extreme (minimum and maximum) safety factors is carried using evolutionary algorithm at various cut levels. The proposed method is used in this study to investigate the effects of uncertainty in soil and wall parameters on overturning, sliding, and flexural stabilities, as well as the overall wall stability. In addition, the effects of changes in water level, anchor tensile force and wall section modulus on the probability of different failure modes were determined. Results confirm the previous findings in the literature on significant impact of uncertainty in these parameters on the wall stability. Conducting a sensitivity analysis, it was found that the uncertainty in water level is an effective factor on all failure modes. Reduction of the water level leads to a decrease in the probability of flexural failure, while it increases the probability of overturning and sliding failures. The provided diagrams are also useful for probability-based design against various modes of failure, in engineering practice. The results emphasize the robustness of the presented fuzzy approach to account for the uncertainties of input variables in the sheet pile wall design procedure.