⚡ electrical engineering

Optimization of Anchor Location and Inclination in Multi-Layered Cohesive–Frictional Soils Using Parametric FEM Analysis

This study utilizes a comprehensive parametric finite element analysis of 216 simulations to optimize anchor location and inclination for sheet pile walls in layered cohesive–frictional soils, demonstrating that strategic placement in the transition zone and inclined configurations significantly reduce wall displacement and bending moments while providing validated empirical design tools for improved geotechnical stability.

Original authors: Mahendra Acharya, Diwash Bhattarai

Published 2026-07-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mahendra Acharya, Diwash Bhattarai

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are building a giant, vertical sandcastle wall to hold back a mountain of dirt. To keep it from collapsing, you stick a long, strong rope (an anchor) into the ground behind the wall and pull it tight. For decades, engineers have been guessing where to tie that rope and at what angle, often assuming the ground behind the wall is one big, uniform block of dirt—like a giant, flavorless block of Jell-O.

But real ground is rarely a single block of Jell-O. It's more like a layer cake: a fluffy top layer of sand, a sticky middle layer of clay, and a crunchy, dense bottom layer of gravel. This paper, run by Mahendra Acharya and Diwash Bhattarai, asks a simple question: If the ground is a layer cake, does it matter exactly where we stick our rope and how we angle it?

The answer is a resounding yes. In fact, sticking the rope in the wrong spot or at the wrong angle is like trying to hold a wobbly tower together with a rubber band tied to the wrong rung of a ladder.

The "Layer Cake" Problem

The researchers used a powerful computer program (a Finite Element Method simulation) to build 216 different digital versions of these walls. They didn't just guess; they ran the numbers on a "sandwich" of soil: a top layer of fill, a middle layer of clayey sand, and a bottom layer of dense sand.

They found that the old way of thinking—assuming the ground is all the same—was misleading. If you treat a layered cake like a single block of Jell-O, your computer model might tell you the wall will move 20% to 35% more than it actually would in real life. That's a lot of unnecessary worry! It also messes up the bending forces inside the wall, making the wall look like it's about to snap when it might actually be fine, or vice versa.

The Sweet Spot: Where to Tie the Knot

The study discovered a "Goldilocks zone" for where to place the anchor.

Too high (near the top): The wall acts like a diving board, bending wildly at the top.
Too low: You aren't getting enough help from the sticky middle layer.
Just right: The anchor needs to be placed in the transition zone—right where the soft, squishy middle layer meets the hard, dense bottom layer.

In the simulations, placing the anchor at a depth ratio of 0.20 to 0.30 (meaning 20% to 30% of the way down the wall's height) was the magic spot. When they moved the anchor there, the wall's sideways wobble dropped by 18% to 24%. It's like finding the perfect spot on a seesaw where you need the least amount of effort to keep it balanced.

The Angle of Attack: Don't Go Flat!

Here is the most playful part: Tilt the rope!
For a long time, engineers mostly used horizontal ropes (parallel to the ground). But this study suggests that angling the rope slightly downward, between 10° and 20°, is a game-changer.

Think of it like this: A horizontal rope just pulls the wall back. But a tilted rope pulls back and pushes down. That downward push acts like a heavy hand pressing on the soil, making it tighter and stronger.

The Result: By tilting the anchor 10° to 20°, the wall's sideways movement dropped by up to 28% compared to a flat rope.
The Catch: If you tilt it too much (like 30°), you lose the benefit. The rope starts pulling up too hard, and the horizontal pull needed to stop the wall from falling isn't strong enough anymore. It's like trying to push a car forward while also trying to lift it into the air; you waste energy.

The Magic Numbers

The researchers didn't just say "it's better." They crunched the numbers to give engineers a cheat sheet. They found that the best angle and depth depend on how "grippy" (friction angle) and "sticky" (cohesion) the soil is.

They created two simple formulas (equations) that act like a GPS for anchor placement:

Depth: The deeper the friction angle (the grippier the soil), the slightly deeper you should put the anchor.
Angle: The grippier the soil, the more you should tilt the anchor downward.

For the specific soil layers they tested (friction angles between 25° and 38°, and cohesion between 0 and 25 kPa), the optimal tilt was found to be 10° to 20°.

What This Means for the Future

The paper is careful to say these results are based on computer simulations of a specific three-layer soil profile. They haven't built a physical wall in the real world to prove it yet (though their computer models matched real-world data from other projects very closely, within 5% to 10%).

However, the implications are huge. If engineers use these "tilted, transition-zone" anchors:

They can reduce the bending stress on the wall by up to 22%. This means they might be able to use thinner, cheaper steel.
The safety factor (how much extra strength the wall has) goes up by 12% to 18%.
They stop wasting money on anchors that are too deep or too shallow.

The Bottom Line

This study suggests that treating the ground like a uniform block is a mistake. Real ground is a layered cake, and your anchor should be placed exactly where the layers change texture, and it should be tilted just enough to press the soil together.

While these findings are currently simulated and specific to the soil types tested, they offer a new, smarter way to design walls that are safer, cheaper, and less likely to wobble. It's a reminder that in engineering, sometimes the best solution isn't to pull harder, but to pull at the right angle, in the right spot.

Technical Summary: Optimization of Anchor Location and Inclination in Multi-Layered Cohesive–Frictional Soils

Problem Statement
Conventional design approaches for anchored sheet pile walls often rely on simplified analytical methods or empirical correlations calibrated for homogeneous soil profiles. These methods frequently neglect the complex interplay between anchor geometry (specifically depth and inclination) and soil stratification. In multi-layered cohesive–frictional deposits, where strength and stiffness vary between strata, these simplified assumptions can lead to inaccurate predictions of wall displacement, bending moments, and global stability. The literature lacks a systematic framework for jointly optimizing anchor location and inclination in realistic, heterogeneous soil profiles, often resulting in inefficient load transfer, excessive deformations, or conservative over-design.

Methodology
This study employs a comprehensive parametric Finite Element Method (FEM) analysis to investigate and optimize anchor configurations in layered soils.

Numerical Framework: A two-dimensional plane-strain model was developed using RS2 software, utilizing the Mohr–Coulomb elastoplastic constitutive model for soil and linear elastic beam elements for the sheet pile wall. Zero-thickness interface elements modeled the soil–structure interaction.
Soil Profile: The study focused on a three-layer cohesive–frictional profile (fill, intermediate clayey sand, and dense sand) typical of alluvial and urban reclamation sites.
Parametric Study: A total of 216 simulations were conducted using a full-factorial experimental design. Key variables included:
- Soil friction angle ( $\phi$ ): 25°–38°
- Soil cohesion ( $c$ ): 0–25 kPa
- Anchor depth ratio ( $z_a/H$ ): 0.10–0.30
- Anchor inclination ( $\theta$ ): 0°–20°
Validation: The numerical model was rigorously validated against classical analytical solutions, published numerical benchmarks, and field monitoring data from an instrumented excavation. The model demonstrated good agreement, with errors generally remaining within 5–10% for displacement and bending moment.
Analysis: The results were analyzed using correlation heatmaps, Response Surface Methodology (RSM), and Analysis of Variance (ANOVA) to identify dominant parameters. Dimensionless empirical equations and design charts were developed based on the simulation database.

Key Contributions

Joint Optimization Framework: The study provides a systematic methodology for simultaneously optimizing anchor vertical location and inclination in multi-layered cohesive–frictional soils, addressing a gap where previous studies often varied only one parameter while fixing the other.
Quantification of Stratification Effects: The research explicitly quantifies the error introduced by homogeneous soil assumptions, demonstrating that such models overestimate lateral wall displacement by 20–35% and fail to capture stress redistribution mechanisms at layer interfaces.
Empirical Design Tools: The study derives dimensionless empirical equations and design charts for estimating optimal anchor depth and inclination, as well as normalized responses (displacement, bending moment, anchor force), enabling rapid preliminary design without repeated numerical simulations.
Statistical Sensitivity Analysis: Through ANOVA and RSM, the study identifies anchor depth ratio and soil friction angle as the most influential parameters governing wall performance, followed by anchor inclination.

Key Results

Optimal Anchor Depth: The optimal anchor depth ratio ( $z_a/H$ ) was found to be between 0.20 and 0.30. Placing anchors within the transition zone between soft and stiff soil layers significantly improves performance by enhancing passive resistance mobilization and reducing stress concentrations. Compared to shallow placements ( $z_a/H < 0.15$ ), this optimal depth reduces maximum lateral displacement by 18–24%.
Optimal Anchor Inclination: Moderate downward inclinations of 10°–20° yielded the best structural performance. This configuration reduced maximum lateral wall deflection by up to 28% and peak bending moments by 22% compared to horizontal anchors. The vertical component of the inclined anchor force enhances soil confinement and shifts the wall's rotation center upward. Inclinations exceeding 20° resulted in diminishing returns due to reduced horizontal restraint efficiency.
Impact of Heterogeneity: Stratified soil models revealed that layer interfaces act as preferential zones for shear band initiation but also promote stress arching. Even modest cohesion ( $c > 5$ kPa) in intermediate layers provided a disproportionately stabilizing effect, delaying plastic strain accumulation.
Safety and Efficiency: Optimized configurations (anchors in transition zones with 10°–20° inclination) increased the Factor of Safety (FoS) by 12–18% compared to conventional non-optimized placements.
Predictive Accuracy: The proposed dimensionless equations for optimal depth and inclination achieved high coefficients of determination ( $R^2 = 0.94$ for depth; $R^2 = 0.92$ for inclination) with low prediction errors when validated against the FEM database.

Significance and Claims
The authors claim that this study offers a practical, performance-based methodology for the preliminary design of anchored sheet pile walls in heterogeneous soil conditions. By explicitly accounting for soil stratification and anchor inclination, the proposed framework allows engineers to move beyond the conservative and often inaccurate assumptions of homogeneous limit equilibrium methods. The developed design charts and empirical equations provide a computationally efficient tool for optimizing anchor placement, potentially reducing material usage and construction costs while improving serviceability and global stability.

The study emphasizes that its findings and design relationships are specifically derived for the three-layer cohesive–frictional soil profile investigated under drained conditions. The authors modestly note that these relationships should not be extrapolated to substantially different stratigraphic conditions, undrained scenarios, or multi-anchor systems without further validation. The work serves as a bridge between advanced numerical optimization and practical geotechnical design, highlighting the necessity of stratification-aware methodologies in modern excavation engineering.