Document Type : Research Paper
Authors
1
Geotechnical Department, Faculty of Civil Engineering, Semnan University, Semnan, Iran.
2
Faculty of Civil Engineering, Semnan University, Semnan, IRAN.
3
M.Sc. student of Geotechnics, Faculty of Civil Engineering, Semnan University, Semnan, IRAN.
10.22075/jtie.2025.38752.1734
Abstract
Geosynthetic-reinforced soil walls, due to advantages such as ease of construction, cost-effectiveness, and favorable performance, are widely used in civil engineering projects. However, the presence of uncertainties in the strength parameters of the soil and reinforcing materials poses a significant challenge in the analysis and design of these structures. The present study, aimed at improving the accuracy of safety and performance assessment of reinforced soil walls, employs a probabilistic approach using the Limit Equilibrium Method (LEM) combined with Monte Carlo Simulation (MCS). Input parameters (such as dry unit weight, friction angle, and cohesion) were considered as random variables, and their influence on the factor of safety was evaluated. The results indicate that increasing the soil internal friction angle from 19° to 29° raised the factor of safety from 0.8 to 1.1, while reducing the probability of failure from about 0.05 to 0.0005. A geosynthetic length-to-wall height ratio (L/H) greater than 0.8 and geosynthetic layer spacing of less than 0.6 m also improved stability (probability of failure decreased from 0.0007 to 0.001). In contrast, increasing the soil unit weight of abutment up to 22 kN/m³ resulted in a reduction of the factor of safety. Comparison of the study results with code-based criteria (F.S = 1.5) revealed that, in cases such as an internal friction angle of 29° and L/H = 1, the obtained factor of safety was about 15% higher than the reference code value, whereas in conditions such as increasing the backfill unit weight to 22 kN/m³, it was about 20% lower than the code requirement. Therefore, adopting probabilistic analysis can provide a more realistic and accurate estimation of uncertainty compared to deterministic code-based design.
Keywords
Main Subjects
)