Failure Mechanism of a Rainfall-Induced MSE Wall Embankment Revealed by Two-Stage Numerical Back-Analysis

Abstract

Rainfall infiltration is one of the main causes of embankment failure because it changes the engineering properties of soil and reduces slope stability. Conventional stability analyses commonly assume uniform soil conditions, which may not represent the actual field condition after prolonged rainfall. This study investigates the failure mechanism of an MSE wall embankment at Kedungmlati 2 Overpass, Mojokerto–Kertosono Toll Road, using a two-stage numerical back-analysis. Laboratory tests were performed on embankment soils under different degrees of saturation (Sr = 50%, 62.5%, 75%, 87.5%, and 100%) to determine the corresponding engineering properties. The numerical analyses were conducted using the finite element method implemented in PLAXIS 2D. Stage 1 evaluated six numerical models with uniform saturation conditions throughout the embankment. None of these models reproduced the observed failure mechanism. Stage 2 introduced a non-uniform saturation distribution to represent localized rainfall infiltration along the embankment slope The modified model successfully reproduced the observed failure surface. The results demonstrate that the non-uniform saturation scenario adopted in Stage 2 provided a closer agreement with the observed failure mechanism than the uniform saturation scenarios evaluated in Stage 1. In this scenario, a saturated near-surface zone and a relatively unsaturated embankment core were assigned as part of the back-analysis procedure to reproduce the observed failure pattern. The proposed rehabilitation measures further increased the safety factor from 1.085 to 1.765, effectively restoring embankment stability. These findings suggest that considering spatial variations in the degree of saturation can provide a more representative prediction of rainfall-induced failure mechanisms than assuming a uniform saturation distribution. The novelty of this study lies in the proposed practical back-analysis approach, which combines a non-uniform saturation scenario with experimentally determined shear strength parameters at different degrees of saturation to reproduce the observed failure mechanism and evaluate rehabilitation strategies for MSE wall embankments without explicitly modeling transient seepage.