Effect of earth reinforcement, soil properties and wall properties on bridge MSE walls

Abstract

Mechanically stabilized earth (MSE) retaining walls are popular for highway bridge structures. They have precast concrete panels attached to earth reinforcement. The panels are designed to have some lateral movement. However, in some cases, excessive movement and even complete dislocation of the panels have been observed. In this study, 3-D numerical modeling involving an existing MSE wall was undertaken to investigate various wall parameters. The effects of pore pressure, soil cohesion, earth reinforcement type and length, breakage/slippage of reinforcement and concrete strength, were examined. Results showed that the wall movement is affected by soil pore pressure and reinforcement integrity and length, and unaffected by concrete strength. Soil cohesion has a minor effect, while the movement increased by 13–20 mm for flexible geogrid reinforced walls compared with the steel grid walls. The steel grid stresses were below yielding, while the geogrid experienced significant stresses without rupture. Geogrid reinforcement may be used taking account of slippage resistance and wall movement. If steel grid is used, non-cohesive soil is recommended to minimize corrosion. Proper soil drainage is important for control of pore pressure.

Graphical abstract

Keywords

mechanically stabilized earth walls / precast concrete panels / backfill soil / finite element modeling / earth reinforcement

1 Introduction

More than 850 m² of mechanically stabilized earth (MSE) retaining wall structures for highway bridges are constructed each year in the US [1], making it an important component of the transportation infrastructure system and therefore the behavior and failures of such walls are important safety considerations. The MSE wall components and the wall configuration are presented in Fig. 1 [2]; Elias et al. [3] found that MSE wall structures perform as a rigid body system, with predictable external behavior similar to that of traditional retaining walls. Hence the primary purpose of this study was to evaluate the internal performance of these structures and verify the integrity of the most recent valid approach to designing them. A case study of a failed MSE wall in Soda Springs, ID, was investigated by Armour et al. [4], and the cause of failure was found to be the sheared panel reinforcement connections that led to some precast concrete panels (PCPs) falling out of the wall. Kibria et al. [5] conducted a comprehensive study of a double-faced MSE wall located on State Highway 342 (SH 342) in Lancaster, TX. The five-year-old wall exhibited lateral movement of 300 to 450 mm. Installed inclinometers recorded additional movement of 38 mm in eight months duration. Numerical modeling results showed that the lateral movement could be significantly decreased by increasing the soil reinforcement lengths by 40%. Another case study by Yoo and Kim [6] investigated varying surcharge load application on an MSE wall with geosynthetic reinforcement. Resulting deformations were within the serviceability limits and led to the conclusion that the Federal Highway Administration (FHWA) [7] design surcharge load is conservative. Abdelouhab et al. [8] performed a numeric parametric study of MSE walls using Fast Lagrangian Analysis of Continuum (FLAC) 2D software. It was found that geosynthetic-reinforced MSE walls had higher lateral deformations than metal-reinforced walls, and the most influential parameters were the backfill soil friction angle and cohesion, interface shear friction and strip stiffness. Fishman et al. [9] performed structural reliability analysis on the recommended resistance factors for MSE wall steel strip corrosion and compared them with observed corrosion rates. The derived resistance factors were lower than that from American Association of State Highway and Transportation Officials (AASHTO) [10], which led to a conclusion that the AASHTO [10] design guide for the corrosion rates is conservative. However, as noted by Hossain et al. [11], existence of fine content in backfill soil may lead to water absorption and corrosion of steel earth reinforcement in MSE wall types. Furthermore, the water retention caused by the fine content could increase pore pressure on the PCPs. Several other monitoring investigations on MSE walls were also undertaken [12,13].

The above discussion indicates that there is a lack of understanding of the internal performance, which is primarily governed by the earth reinforcement design. The objective of this study was to determine the effect of wall dimensions, material properties, support conditions and load conditions on the performance of the soil reinforcement and associated wall safety. A calibrated numerical approach was carried out through involving initial inputs from a real-life MSE wall and a subsequent parametric analysis.

2 Numerical modeling of MSE walls

Hatami and Bathurst [14,15] and Ambauen et al. [16] proved that developing stress or strain distributions through finite element meshing plays a crucial role in predicting performance of earth retaining structures. A 9 m high MSE wall (Fig. 2) located at the intersection of Chisholm Trail Pkwy. and SW Blvd. in Fort Worth, TX was used as the basis for this study.

The MSE wall was built in 2013 and experienced some distress, such as differential movement between panels (Fig. 3(a)), opening of panel joints (Fig. 3(b)), and leakage of backfill material (Fig. 3(c)). The model geometry, shown in Fig. 4, was based on the available shop drawings of the PCPs and reinforcement layouts. A 3D finite element model (FEM) of the wall was created in ABAQUS CAE [17] software to simulate the geometric and material properties. As seen in Fig. 4, a typical MSE wall consists of precast panels (gray), backfill soil (green), approach slab (yellow), traffic barrier (white), copings (brown), and embedded earth reinforcement. Visual inspections or field testing for earth reinforcement were not possible due to the inaccessibility of the deeply embedded reinforcement in the backfill soil. Therefore, numerical modeling was used to investigate the effects of internal failures of earth reinforcement, such as pullout or rupture.

2.1 Material properties

Several constitutive models have been developed in the last decade to simulate concrete behavior, including cracking and crushing. Two models are available in ABAQUS to predict concrete behavior: the brittle cracking model and the concrete damage plasticity model [18]. The first model is linear elastic and usually used when tensile cracking is expected to dominate behavior [19]. The damage plasticity model assumes that the two main failure mechanisms of concrete are cracking and crushing, and the crack propagation is modeled by continuum damage mechanics and stiffness degradation [20]. Further details about predicting concrete performance were investigated by Shishegaran et al. [21]. In this study, the concrete material was modeled in ABAQUS using the damage plasticity approach proposed by Lubliner et al. [22].

The soil retained by an MSE wall structure commonly behaves as elastic-plastic. A large number of soil constitutive models are available to predict soil behavior, including the Winkler model, Mohr−Coulomb model, modified Cam−Clay model, Duncan−Chang model, and Elastic Continuum model [23]. In this study, the soil material was modeled using the Mohr−Coulomb method, and the failure criteria was defined by the friction angles and soil cohesion. This method assumes that the failure is controlled by the maximum shear stress, which also depends on the normal stress, as per Eq. (1):

(1)\tau =c-\sigma \cdot \tan \varphi ,

where τ = shear stress, σ = normal stress, c = cohesion of the material, and ϕ = material angle of friction.

The constitutive model used in the FEM to simulate the steel reinforcement was the classic metal linear elastic-plastic model [20]. Welded wire grid (W11 × W11) with a yield and ultimate strengths of 450 and 520 MPa, respectively, was used as earth metal reinforcement based on the reinforcement details and layouts provided by Texas Department of Transportation (TxDOT) for the wall.

High-density-polyethylene (HDPE) reinforcement is a polymeric material that is used extensively as earth reinforcement because the stress-strain response follows a linear elastic-plastic relationship, and the material continues stretching to accommodate the applied tensile load. This study investigated the effect of reinforcement type on MSE wall performance. The geosynthetic reinforcement used in the sample wall was evaluated by performing a tensile strength test according to ASTM D 6637 [24]. The material type used has a yield and tensile strengths of 343 and 848 kPa, respectively.

2.2 Initial assessment

A preliminary 2D MSE wall FEM, shown in Fig. 5, was developed to determine the contact properties that should be used between the different MSE wall components. The preliminary 2D model shown in Fig. 5 consists of facing panel, earth reinforcement, reinforced (backfill) soil, and a loading plate. The lateral movements in two small-scale (1:5.5) experimental MSE walls, tested by Reddy and Navarrete [25], were used for the assessment. The two walls were reinforced with HDPE reinforcement. Instrumentation of the walls included the installation of dial gauges to monitor the lateral movements. A surcharge load of 3.6 kPa was then applied on top of the small-scale walls.

This applied load is equivalent to a surcharge load of 20 kPa applied on a full-scale MSE wall. In the FEM, the shell element type was chosen for the PCPs and backfill soil, while wire elements were chosen for the HDPE reinforcement. The load protocol used for the actual walls was simulated in the model, and the lateral wall movements from the model and the two small-scale walls were compared (Fig. 6).

Another comparison was made on the strain gauge readings, as shown in Fig. 7. The term “H-5-3 (1)” means strain value at reinforcement layer 5, position number 3 (the third strain gauge away from the panel) and (1) means the result obtained from MSE wall (1). AB stands for ABAQUS output values.

Results showed that the preliminary model with the assumed contact properties was in good agreement with the recorded wall movement (Fig. 6). Moreover, from the strain reading comparison (Fig. 7), both the experimental walls and ABAQUS modeling show similar strain patterns with increase of surcharge live load. Also, the largest strains appeared at layer 5 and the lowest appeared at layer 1, the same was observed from the FE modeling results; thus it was inferred that the model was capable of predicting actual wall response.

2.3 Material interactions

After the initial assessment, the interfaces between the different materials in the original 3D model were developed. Properties at the concrete to concrete interface were defined for the interactions between the PCPs, between PCPs and copings, copings and approach slab, and copings and traffic barriers. Properties at the concrete to soil interface were defined for the interactions between the PCPs and soil, and approach slab and soil. The interaction properties contained both normal and tangential definitions. For mechanical behavior in the normal direction, “hard contact” was set for the pressure overclosure, and the default constraint enforcement method was used [16,26]. For the tangential mechanical behavior, a penalty friction formulation with isotropic directionality was used. The interface shear resistance was modeled as a friction coefficient, using the basic Coulomb friction model (Eq. (2)):

(2)\tau ={\mu }_{\text{S}}\cdot \sigma ,

where µ_S = coefficient of static friction. A 0.7 coefficient of static friction was used for concrete-concrete interface based on the study of Mohamad et al. [27], with a shear stress, due to the interlocking of the shear keys between the concrete layers, of 46 kPa based on the findings of Hatami and Bathurst [14,15]. For concrete-soil interface, the coefficient of static friction can be calculated as tan (δ), where δ is the friction angle between soil and concrete, estimated as 67% of soil friction angle, according to the AASHTO Load and Resistance Factor Design (LRFD) [28]. Based on the initial assessment, this value was set as 0.42 rad in the FEM, which was close to the recommended AASHTO value, as the friction angle was taken as 34°. For both interaction properties, the specified maximum elastic slip, as a fraction of the characteristic surface dimension, was assumed to be 0.005. Separation of the included surfaces was allowed after contact so that the soil elements would not experience tension from the outward movement of the PCPs.

2.4 Reinforcement constraints

Previous work involving physical modeling of the interface between earth reinforcement and soil showed that the interaction is dependent on the soil particle size relative to the reinforcement apertures [29]. It was found that pullout resistance approaches the full shearing soil resistance if the soil particles are small enough to be compacted into the reinforcement apertures. This interlocking behavior was simulated in the FEM, using “embedded region” constraint, and it was assumed that there would be no relative slipping between the reinforcement and soil. This is acceptable in modeling because of the large aperture of the reinforcement relative to the soil particle size. Hence, pullout would manifest as shear in the soil zone that is adjacent to the reinforcement [6].

The steel grid and HDPE reinforcement were modeled as beam elements that were embedded in the backfill soil. The steel reinforcement in the PCPs, copings, and traffic barriers were also modeled as beam elements and embedded into their respective concrete sections. The earth reinforcement was inserted 50 mm into the PCPs, and the reinforcement tips were constrained to the PCPs. This technique simulated a tie connection between the earth reinforcement and the PCPs.

2.5 Boundary conditions

The boundary conditions of the model shown in Fig. 4 were achieved by supporting the back face of the wall with roller connections allowing it to move vertically, while also supporting the sides with roller connections that restrained movement perpendicular to the wall sides. The bottom was assumed to be a pinned connection with no movements or allowed rotations.

2.6 Load applications

The surcharge load “q” in the modeling was assumed to be a uniform pressure that was applied to the entire top surface of the bridge approach slab. The surcharge was applied in a separate step, where the load was increased incrementally up to a maximum pressure of 12 kPa; this value was calculated based on Eq. (3).

(3)q=\gamma \cdot h_{\text{eq}},

where γ = the backfill unit weight and h_eq = an equivalent surcharge soil layer (equal to 0.61 m), as recommended by AASHTO LRFD [28]. On the other hand, a previous study [30] investigated an uncommon case in which PCPs were subjected to blast loadings. It was found that PCPs may exhibit several signs, such as cracking or crushing of PCPs depending on their thickness and compressive strength.

Geostatic pressure is caused by the soil self-weight and tends to increase the out-of-plane PCP movement and the tensile stress in earth reinforcement. This pressure was applied to the backfill soil as both vertical and horizontal pressures at any depth and assigned to the model backfill soil using a predefined “geostatic stress” field.

2.7 Mesh sensitivity

FEM mesh size is important in balancing convergence, model accuracy and runtime. Several basic models were run to determine the optimum mesh size and different mesh sizes were tried for each component. An enhanced hour glass control was selected for the components in ABAQUS to overcome the hour glass concern.

The results of analysis are shown in Fig. 8. For the backfill soil, mesh sizes of 2000, 1500, 1000, 750, 600, 500, and 450 mm were used to determine the optimum mesh size. The out-of-plane and downward movements of PCPs and approach slab was chosen for the investigation, respectively. Optimum mesh sizes shown by red circles in Fig. 8 were selected for each component. Smaller mesh sizes would significantly increase computer run times without much increase in model accuracy.

3 Parametric study

After finalizing the FEM, a parametric study involving the 3D MSE wall was carried out, to investigate the effects of the parameters that may lead to failure. The Design of Experiments (DOE) approach using the full factorial method in Minitab software was used (Table 1) in creating the different models of the parametric study. Six main parameters were chosen that consisted of two cases each, resulting in 64 different models, as follows.

3.1 Pore pressure

It was previously shown that soil pore pressure has an impact on MSE wall performance. Hossain et al. [11] investigated a case study of an MSE wall located at State Highway 342 in Lancaster, TX. The study suggests that pore pressure may be present as a result of perched water zones existing behind the wall that may be present at only certain locations (Fig. 9). Areas A and B in Fig. 9 also indicate the moisture content of soil increased over the time due to excessive fine clay content affecting the pore pressure. In the current study, the pore pressure was initially investigated by applying it to the top-third, mid-third and lower-third heights of the model wall. It was found that the wall performance was mostly affected when the pore pressure was applied at the mid-third height. This is because the maximum recorded out-of-plane PCP movement in a typical MSE wall with no pore pressure applied occurs at mid-height, as will be revealed in Section 3.4. Therefore, the pressure was modeled as a triangle that increased from zero at the top of the 4th panel to 30 kPa at the bottom of the 3rd panel (Fig. 10). The variable pressure was calculated as the unit weight of water multiplied by the water height.

3.2 Earth reinforcement type

The MSE wall performance also depends on the reinforcement extensibility. Two most commonly used soil reinforcement for MSE walls, metal (inextensible) or geosynthetic (extensible) were investigated. A typical steel grid grade 65 and a typical HDPE were modeled, as shown in Fig. 11. Figures 11(a) and 11(c) show the overall reinforcement used for modeling the MSE wall with steel grid and HDPE geogrid reinforcements, respectively, while Figs. 11(b) and 11(d) show the layout of a single reinforcement of both steel grid and HDPE geogrid reinforcements.

3.3 Earth reinforcement length

FHWA [31] specifies 0.7H as the minimum recommended reinforcement length, where H is the wall height. Therefore, the effects of 0.7H and 1.0H earth reinforcement lengths were investigated.

3.4 Breakage/slippage of earth reinforcement

Because the reinforcements are inaccessible and difficult to locate after installation, it is challenging to ascertain their integrity in service. Corroded (broken) or slipped reinforcement may affect the wall stability. Therefore, it was considered as a parameter in this study by eliminating in the model 1/18 of the overall earth reinforcement (Fig. 12). This value was selected because, in reality, when some reinforcement may break or slip, the surrounding reinforcement would carry the extra developed load through redistribution [32]. Furthermore, as previously mentioned, the maximum movement in an MSE wall with no issues occurred at about the mid-height (Fig. 13); hence, the middle reinforcements that were attached to the 4th row panels (Fig. 12) were eliminated, as the wall showed maximum movement when these reinforcements were eliminated.

3.5 PCP concrete compressive strength

The 28-d compressive strength of the PCP concrete was considered as a parameter. Two values of 28 and 35 MPa were investigated. The wall as-built plans mentioned a concrete strength of 28 MPa for the panels “Class H”.

3.6 Cohesion of backfill soil

FHWA [31] recommends that the backfill soil for an MSE wall shall be cohesionless. However, the selected backfills may have fine contents that exceed the specified limits [11]. Thus, as parametric variables, backfill soil with cohesion of 20 kPa and without any cohesion were studied.

4 Results and discussion

To investigate the importance of the studied parameters on MSE walls, the maximum out-of-plane (lateral) movement of PCPs (Fig. 13) and maximum stresses developed in earth reinforcement and their respective locations (Fig. 14) were selected for the comparison between the selected parameters of the parametric study in order to assess their effect of wall performance.

Figures 15 and 16 show the out-of-plane movements in walls reinforced with steel grid and HDPE geogrid earth reinforcement, respectively, under the effects of the other variables. The parameter values in the figures were designated based on the acronyms shown in Table 1. The pore pressure is an important factor affecting the wall performance and is applied at the mid-third height of the panels, as explained previously. The figures indicate that the lateral movement of modeled walls for both reinforcement types increases in the following cases: 1) when the pore pressure is applied; 2) when reinforcement failure is considered; and 3) if the reinforcement length is taken as 0.7H. As expected, soil reinforcement failure results in increased wall movement. Comparison of Figs. 15 and 16 shows that the PCP compressive strength has no bearing on the wall movement regardless of earth reinforcement type. It is observed from Figs. 15(a) and 15(c) and Figs. 15(b) and 15(d) that the walls exhibit less lateral movement when the soil cohesion is increased to 20 kPa. Similar inferences may be made for the HDPE geogrid walls in Fig. 16. However, these models showed overall wall movements that were 13–20 mm greater than the steel grid reinforced walls. Clearly, the HDPE geogrid is more flexible in nature.

Figures 17 and 18 list the maximum stresses developed in earth reinforcement for the models with 28 MPa concrete strength. These figures state that the reinforcement length does not affect the ultimate stresses developed in the earth reinforcement (both types). When pore pressure is present, the breakage/slippage of reinforcement significantly increases the tensile stresses in the steel grids, as expected. Figures 18(a) and 18(b) suggest that the tensile stresses in HDPE geogrids increase significantly when pore pressure is applied, while it increases slightly when the breakage/slippage of reinforcement is accounted for.

The ultimate stresses developed in each reinforcement types were then compared with their respective yield, F_y, and ultimate tensile strength, f_u. This comparison was conducted to examine any rupture of the earth reinforcement which would occur when the actual tensile stresses surpass their ultimate tensile strength. Figure 19(a) shows that the tensile stresses in steel grids were about 50% less than their yield strength, while Fig. 19(b) indicates that the tensile stresses in HDPE geogrids surpassed their yielding strength. However, the stress did not exceed the ultimate strength, meaning that the HDPE geogrid did not fail due to rupture.

After that, the FHWA [31], the most recent code used for designing MSE walls, was compared with the findings of the study; the internal stability serviceability limits (pullout or rupture of earth reinforcement) were checked with the reinforcement ultimate stresses obtained from the FEM. Based on FHWA [31], the resistance of rupture failure of reinforcement requires that (Eq. (4)):

(4)T_{\text{MAX}}⩽T_{\text{r}},

where T_MAX = maximum tensile stress (force per unit length, Eq. (5)), and T_r = factored tensile resistance in reinforcement (Eq. (6)).

(5)T_{\text{MAX}}={\sigma }_{\text{H}}{S}_{\text{V}},

(6)T_{\text{r}}=\varphi T_{\text{al}},

where σ_H = horizontal stress (Eq. (7)), S_V = reinforcement vertical spacing, ϕ = rupture resistance factor, and T_al = nominal long-term strength of reinforcement (Eqs. (8a) and (8b)).

(7){\sigma }_{\text{H}}=K_{\text{r}}\left[{\gamma }_{\text{x}}\left(z+h_{\text{eq}}\right){\gamma }_{\text{(EV - MAX)}}\right].

For metal reinforcement:

\begin{array}{}\text{(8a)}& T_{\text{al}}=\frac{F_{\text{y}}{A}_{\text{c}}}{b}.\end{array}

For geosynthetic reinforcement:

\begin{array}{}\text{(8b)}& T_{\text{al}}=\frac{T_{\text{ult}}}{R{F}_{\text{ID}}R{F}_{\text{CR}}R{}_{\text{FD}}},\end{array}

where K_r = lateral earth pressure coefficient, γ_r = backfill soil unit weight, and γ_(EV-MAX) = maximum load factor (1.35). F_y = yield tensile strength of steel, A_c = design cross-sectional area considering the corrosion loss, and b = width of reinforcement. T_ult = the geosynthetic ultimate tensile strength, and RF_ID, RF_CR, R_FD = reduction factors for installation damage, creep and durability, respectively (1.60, 2.60, and 1.00, as provided by the HDPE manufacturer, representing reduction factors for actual long-term strength losses for the FHWA pullout serviceability checks). The pullout serviceability limits must be checked so that the pullout resistance, P_r, on the effective pullout length, L_e, should be at least equal to the tensile stress, T_MAX, developed in the reinforcement. The pullout resistance of each reinforcement layer must be calculated using Eq. (9) and checked with the respective T_MAX value.

(9)P_{\text{r}}=F^{\ast }\alpha {\sigma }_{\text{v}}^{\mathrm{\prime }}{L}_{\text{e}}C,

where F* = pullout resistance factor, α = scale correction factor, σ'_v = vertical stress at the reinforcement level and C = reinforcement effective unit perimeter (= 2 for strip, grid and sheet reinforcements).

The comparison between the developed stresses from FEM and the calculated FHWA [31] serviceability limits is presented in Fig. 20. The calculated maximum tensile forces in reinforcement layers from FHWA [31] were slightly lower than the maximum tensile forces from the FEM for both reinforcement types. For the steel grid reinforcement, both the FHWA pullout and rupture limits are mostly much larger than the actual stresses, while the rupture limits for the steel grids are conservative (Fig. 20(a)). The variation in the pullout limits is caused by the varying wall parameters for the respective models. The code pullout limits become critical for steel grid reinforced models when pore pressure is applied, reinforcement length is taken as 0.7H, and breakage/slippage of reinforcement is considered. For the HDPE geogrid models, the pullout limits are more conservative than the rupture limits (Fig. 20(b)), the latter being close to the actual stresses.

The corrosion potential in steel grid reinforcement is an important consideration that is outside the scope of this study. Such soil conditions may lead to premature steel corrosion, reduction of strength, increased elongations and long-term durability issues. Geogrid reinforcement is not affected by corrosion. However, the latter type is more expensive than the steel grid type. The cost/benefit aspect is important, but outside the scope of this study.

5 Conclusions and recommendations

A 3D FE model was developed in this study to reasonably predict the MSE wall performance. The model accounts for the material properties and interactions, modeling calibration, reinforcement constraints, boundary conditions, loading application and mesh sensitivity. A parametric study was carried out to investigate the performance of the MSE wall as affected by pore pressure, reinforcement type and length, reinforcement breakage/slippage, PCP compressive strength and soil cohesion. The following conclusions and recommendations may be made based on the findings from the numerical model.

1) Overall lateral wall movements with geogrid reinforcement were 13–20 mm greater than that in the steel grid reinforced walls. It can be noted HDPE geogrids are more flexible in nature than steel grids.

2) The out-of-plane movement of walls for both steel grid and HDPE geogrid earth reinforced walls increased in the following cases: when the soil pore pressure is applied; when reinforcement breakage/slippage is considered; and if the reinforcement length is taken as 0.7 times the wall height.

3) Variations in the concrete compressive strength for the MSE wall facia PCPdo not affect the wall movement, regardless of the earth reinforcement type. Hence, it is apparent that the wall panels move as rigid bodies, only affected by the soil properties and the earth reinforcement behavior.

4) Soil cohesion has a minor effect on the wall movements. The movements decrease slightly when the soil cohesion is increased from zero to 20 kPa for both reinforcement types. However, cohesive backfill soils may lead to drainage issues that may add extra lateral earth pressure acting on the wall.

5) In steel grid reinforced wall, pore pressure and reinforcement length have a crucial effect on wall movement as the lateral movement is recorded between 5 and 10 mm and 6–12 mm, respectively, with these ranges being dependent on the other variables.

6) In HDPE reinforced walls, pore pressure increases wall movement significantly, by 14−25 mm, while reinforcement length causes an increase in movement by 6–13 mm, depending on the other variables.

7) When 1/18 of overall earth reinforcement is eliminated due to breakage/slippage, the lateral movement increases but it is slightly higher in HDPE geogrid walls than that of steel grid walls.

8) The maximum stresses in steel grids were about 50% less than the yield strength. The final stresses in HDPE geogrids surpassed the yield strength, but did not exceed the ultimate strength, meaning that the HDPE geogrid may experience large extensions without rupture. However, possible corrosion issues may affect the steel grid strength and long-term durability because of pore pressure concerns.

9) The reinforcement length also does not affect the stresses developed in the earth reinforcement, for both types. The reinforcement length seems to affect only the pullout resistance, as shown by the comparison with the serviceability limits from FHWA. It was determined that the pullout limits are more conservative than the rupture limits. The pullout limits were found to be critical for the steel grid reinforcement type when pore pressure, breakage/slippage and short reinforcement length parameters are considered.

Base on this study, the following recommendations are made for future work.

1) Geogrid earth reinforcement should be used for MSE walls with proper considerations for breakage resistance and excessive wall movement. For steel grid reinforced MSE walls, adequate measures should be put in place to minimize any pullout or corrosion failure. This could be achieved through the use of cohesionless soil and reinforcement length at least equal to the wall height.

2) MSE walls with different reinforcement types and wall heights can be evaluated through the installation of strain gauges and inclinometers during the construction phase to check for the validity of the wall design.

3) Additional wall parameters can be considered in the parametric study to evaluate their effect on the wall movement and the developed stresses in the reinforcement.

4) The FHWA approach for MSE walls can be investigated using the structural reliability analysis method.

References

[1] Berg R R, Christopher B R, Samtani N C, Berg R R. Design of Mechanically Stabilized Earth Walls and Reinforced Soil Slopes—Volume I. Report No. FHWA-NHI-10–024. Federal Highway Administration, 2009

[2] Tarawneh B, Al Bodour W, Masada T. Inspection and risk assessment of mechanically stabilized earth walls supporting bridge abutments. Journal of Performance of Constructed Facilities, 2018, 32( 1): 04017131–

[3] Elias V, Christopher R, Barry P E. Mechanically Stabilized Earth Walls and Reinforced Soil Slopes Design and Construction Guidelines: FHWA Demonstration Project 82. Report No, FHWA-SA-96-071. Federal Highway Administration, 1997

[4] Armour T A, Bickford J, Pfister T. Repair of failing MSE railroad bridge abutment. In: GeoSupport 2004: Drilled Shafts, Micropiling, Deep Mixing, Remedial Methods, and Specialty Foundation Systems. Florida: ASCE, 2004, 380–394

[5] Kibria G, Hossain M S, Khan M S. Influence of soil reinforcement on horizontal displacement of MSE wall. International Journal of Geomechanics, 2014, 14( 1): 130– 141

[6] Yoo C, Kim S B. Performance of a two-tier geosynthetic reinforced segmental retaining wall under a surcharge load: Full-scale load test and 3D finite element analysis. Geotextiles and Geomembranes, 2008, 26( 6): 460– 472

[7] FHWA. Mechanically Stabilized Earth Walls and Reinforced Soil Slopes Design and Construction Guidelines. Report No. FHWA-NHI-00-043. Federal Highway Administration, 2001

[8] Abdelouhab A, Dias D, Freitag N. Numerical analysis of the behavior of mechanically stabilized earth walls reinforced with different types of strips. Geotextiles and Geomembranes, 2011, 29( 2): 116– 129

[9] Fishman K L, Withiam J L, Gladstone R A. Metal loss for metallic reinforcements and implications for LRFD design of MSE walls. Earth Retention Conference, 2010, 3 : 844– 853

[10] AASHTO. AASHTO Load and Resistance Factor Design Movable Highway Bridge Design Specifications. Washington, D. C.: AASHTO, 2007

[11] Hossain M S, Kibria G, Khan M S, Hossain J, Taufiq T. Effects of backfill soil on excessive movement of MSE wall. Journal of Performance of Constructed Facilities, 2012, 26( 6): 793– 802

[12] Allen T M, Bathurst R J. Design and performance of 6.3-m-high, block-faced geogrid wall designed using k-stiffness method. Journal of Geotechnical and Geoenvironmental Engineering, 2014, 140( 2): 04013016–

[13] Allen T M, Bathurst R J. Improved simplified method for prediction of loads in reinforced soil walls. Journal of Geotechnical and Geoenvironmental Engineering, 2015, 141( 11): 04015049–

[14] Hatami K, Bathurst R J. Development and verification of a numerical model for the analysis of geosynthetic-reinforced soil segmental walls under working stress conditions. Canadian Geotechnical Journal, 2005, 42( 4): 1066– 1085

[15] Hatami K, Bathurst R J. Numerical model for reinforced soil segmental walls under surcharge loading. Journal of Geotechnical and Geoenvironmental Engineering, 2006, 132( 6): 673– 684

[16] Ambauen S, Leshchinsky B, Xie Y, Rayamajhi D. Service-state behavior of reinforced soil walls supporting spread footings: A parametric study using finite-element analysis. Geosynthetics International, 2016, 23( 3): 156– 170

[17] ABAQUS. Dassault Systems Simulia Corporation. 2014

[18] Simulia D S. Abaqus 6.11 Theory Manual. 2011

[19] Martin O. Comparison of different constitutive models for concrete in ABAQUS/explicit for missile impact analyses. JRC Scientific and Technical Reports, 2010

[20] Obaidat Y. Structural retrofitting of concrete beams using FRP-debonding issues. Dissertation for the Doctoral Degree. Skane: Lund University, 2011

[21] Shishegaran A, Varaee H, Rabczuk T, Shishegaran G. High correlated variables creator machine: Prediction of the compressive strength of concrete. Computers & Structures, 2021, 247 : 106479–

[22] Lubliner J, Oliver J, Oller S, Oñate E. A plastic-damage model for concrete. International Journal of Solids and Structures, 1989, 25( 3): 299– 326

[23] Abdel-Mohti A, Khodair Y. Analytical investigation of pile–soil interaction in sand under axial and lateral loads. International Journal of Advanced Structural Engineering, 2014, 6( 1): 54–

[24] ASTM. Standard Test Method for Determining Tensile Properties of Geogrids by the Single or Multi-Rib Tensile Method, ASTM D6637/D6637M–15. West Conshohocken, PA, 2015

[25] Reddy D V, Navarrete F. Experimental and analytical investigation of geogrid MSE wallsIn: Symposium Honoring Dr. John HSchmertmann for His Contributions to Civil Engineering at Research to Practice in Geotechnical Engineering Congress 2008. Louisiana: ASCE, 2008, 277– 291

[26] Ambauen S J. Numerical simulation of mechanically stabilized earth walls for parametric evaluation of behavior under surcharge loading. Thesis for the Master Degree. Oregon: Oregon State University, 2014

[27] Mohamad M E, Ibrahim I S, Abdullah R, Rahman A A, Kueh A B H, Usman J. Friction and cohesion coefficients of composite concrete-to-concrete bond. Cement and Concrete Composites, 2015, 56 : 1– 14

[28] American Association of State Highway and Transportation Officials. AASHTO LRFD Bridge Design Specifications, Customary U. S. Units, 7th ed, with 2015 and 2016 Interim Revisions. Farmington Hills: AASHTO, 2016

[29] Lee W F. Internal stability analyses of geosynthetic reinforced retaining walls. Dissertation for the Doctoral Degree. Seattle: University of Washington, 2000

[30] Shishegaran A, Khalili M R, Karami B, Rabczuk T, Shishegaran A. Computational predictions for estimating the maximum deflection of reinforced concrete panels subjected to the blast load. International Journal of Impact Engineering, 2020, 139 : 103527–

[31] FHWA. Design and Construction of Mechanically Stabilized Earth Walls and Reinforced Soil Slopes—Volume I. Washington, D. C.: Federal Highway Administration, 2009

[32] Zevgolis I E, Bourdeau P L. Stochastic modeling of redundancy in mechanically stabilized earth (MSE) walls. In: GeoCongress 2008: Geo sustainability and Geohazard Mitigation. New Orleans: ASCE, 2008, 1179–1186