Sådhanå (2024) 49:75  Indian Academy of Sciences

https://doi.org/10.1007/s12046-023-02425-4
Sadhana(0123456789().,-volV)FT3](012345
6789().,-volV)

Greenfield settlements due to tunnelling using tunnel boring machine

(TBM) in layered soils: a parametric study
JYOTI AGARWAL* and RAJU SARKAR

Civil Engineering Department, Delhi Technological University, Shahbad Daulatpur Main Bawana Road,
Delhi 110042, India
e-mail: [email protected]; [email protected]

MS received 27 April 2023; revised 20 October 2023; accepted 22 November 2023

Abstract. As India’s cities continue to increase in population, a large number of new metro lines are being
built to accommodate the mobility of residents. For construction of these metro lines, underground tunnels are
being constructed to alleviate traffic problems in congested areas such as Delhi, Chennai, Mumbai, etc., where
land scarcity is a major issue. Although tunnel boring machines (TBMs) are used in the construction of metro
tunnels, ground settlements on the surface are inevitable. Tunnelling induced ground settlements may have
adverse impact on the structures that are above the ground level as well as adjacent to the excavations. In the
present study, ground settlements caused by TBM tunnelling are determined in greenfield conditions, using a
finite element method (FEM) based software, PLAXIS and parametric study is performed after validation of
numerical model with empirical results. Ground settlements obtained using PLAXIS 2D are compared with
those obtained using empirical methods as available in literature. This study captures the effect of governing
tunnel and soil parameters such as diameter of tunnel (D), depth of the tunnel axis from ground surface (z), grout
pressure (GP), undrained shear strength of soil (su) and soil modulus (E’ref), covering the most practical range,
on ground settlements under greenfield condition. The settlement results of this study are validated and com-
pared closely with earlier experimental and observational findings that have been published in the literature. It is
observed from the various parameters studied that they are having a major impact on the ground settlements
during tunnel construction. Therefore, a tunnel designer should consider the effect of all the governing factors in
design prior to the tunnel construction to keep the maximum surface settlements within the permissible limit set
by the standards.

Keywords. Tunnelling; greenfield settlements; volume loss; parametric study.

1. Introduction caused by tunnelling and compared recorded ground

movements with results from existing empirical and ana-
Research in the transportation services for better manage- lytical approaches [2]. Loganathan et al [2] evaluated the
ment and infrastructure development has been prompted by tunnelling-induced ground deformations and their impact
population growth, urbanisation, and the quick expansion on adjacent pile foundations by conducting three centrifuge
of emerging economies. Delhi and Mumbai are predicted to model tests in clays. Jacobsz et al [3] carried out a study
rank among the world’s two most populous cities by the using centrifuge to examine how tunnelling in dense, dry
year 2035. Rapid transit system advancements have fol- sand will affect surrounding single piles and found that
lowed the rise in land scarcity, urbanisation, and industri- volume loss resulted by tunnelling may cause large pile
alisation caused by the rapid population growth rate [1]. settlements within a zone of influence. Marshall et al [4]
With little room for alteration above ground, tunnels have analyse ground movement by conducting centrifuge tests
become a more popular option, making greater use of the due to tunnelling in sands and assessed the impact of tunnel
underground area is a good alternative. Tunnelling induced dimensions and volume loss on soil deformation under
ground settlement is a complex phenomenon that depends greenfield conditions. Many researchers have given ana-
on a variety of factors including the soil and tunnel lytical methods to anticipate tunnelling-induced ground
parameters. movements in clayey and sandy soils [5, 6]. Verruijt and
Many researchers have performed centrifuge model tests Booker [6] proposed an analytical solution for determining
in clayey and sandy soils to estimate ground deformations the uniform radial displacement due to ground loss and
ovalization of tunnel during tunnelling in an elastic soil.
*For correspondence Loganathan and Poulos [7] proposed an analytical solution
 75 Page 2 of 12 Sådhanå (2024) 49:75

by redefining the term ground loss as an equivalent ground and building interaction along with the effects of mecha-
loss parameter with regard to gap parameter in order to nized tunnelling on the ground surface during the con-
predict ground deformations due to tunnelling in undrained struction of Rome’s metro line C. Shiau and Sams [18]
condition, and results of this study are found to be satis- developed a numerical method for assessing the settle-
factory especially for the tunnels in stiff clays. Vorster et al ments associated with circular tunnels in soft soils and
[5] presented an analytical method for calculating the presented design charts using dimensionless ratios with
maximum bending moment for pipelines that are rigidly respect to geometry, strength and stiffness parameters.
joined or continuous and affected by ground movement Settlement results of this study compare closely with
resulting from tunnelling. The proposed approach was earlier experimental and observational findings. Kana-
found conservative as compared with centrifuge tests. Fu garaju and Krishnamurthy [19] presented the profiling of
et al [8] proposed an analytical solution to predict ground influence zone in cohesionless deposits using a finite
movement caused by a circular tunnel that is nonuniformly element programme PLAXIS 3D by analysing the tun-
deforming in a heavy elastic half-plane. The proposed nelling-induced settlements under free field or Greenfield
solution was found to be very useful in the preliminary conditions. Rezaei and Adli [20] performed a 3D finite
design of tunnels. Lai et al [9] proposed a new approach to element analysis using ABAQUS for simulation of Tabriz
determine the tunnelling-induced ground settlements by Metro Tunnel mechanized tunnelling and longitudinal
utilizing the centrifuge model test results for loess soils. troughs are obtained. Athar et al [1] performed physical
Hong et al [10] proposed an analytical model based on and numerical modelling to examine the effect of lining
complex variable theory to investigate ground responses shape on ground settlement. It was found that square
due to shallow tunnelling in multi-layered ground with an linings are the least stable, whereas elliptical linings are
arbitrary ground surface load. Based on the principle of best for minimising ground settlement. Aswathy et al [21]
linear-elastic superposition, the multi-boundary problem is performed a numerical study using PLAXIS 3D and cap-
converted into a superposition of multiple single-boundary tured the impact of vital parameters such as grout pres-
problems. The accuracy of the proposed method was vali- sure, depth and diameter of tunnel, etc., on surface
dated using the field data and numerical analysis results. settlement caused by tunnelling in young alluvium deposit.
Sun et al [11] predicted the tunnel settlement in structural Ahmed et al [22] performed finite element analysis to
soft soils during construction by numerically simulated simulate TBM construction processes using PLAXIS 3D
disturbance degree calculated based on effective stress and and estimated the tunnelling-induced surface settlement
developed a two-fold line settlement calculation model by for Metro Rail Line 1 of Dhaka.
combining it with one-dimensional consolidation test
results for disturbed soil.
As computer technology has advanced, numerical mod-
1.1 Ground settlements due to volume loss
elling utilising finite element analysis has become the
dominant method for geotechnical design and analysis. Ground settlements associated with tunnelling using mod-
Results of numerical models can be compared to empirical ern tunnel boring machines (TBM’s) are mainly due to the
and semi-empirical methods that are nonetheless useful and volume loss (VL). When a tunnel is constructed, the soil
capable. Numerous academics have already concentrated surrounding the tunnel releases stress because of the
on surface settlement prediction under greenfield conditions delayed tunnel peripheral lining installation. As a result,
due to tunnelling by employing numerical modelling radial, and longitudinal deformations occur in the soil body
techniques [1, 12–22]. Möller and Vermeer [12] performed that tends the ground surrounding the tunnel moves into the
a 2D finite element analysis to assess lining forces and tunnel cavity [8, 23, 24]. Volume loss (VL) is represented as
deformations caused by closed shield and open face tun- volume of ground loss around a tunnel cavity (VT) with
nelling. Mathew and Lehane [13, 14] used PLAXIS 2D and respect to the tunnel volume per unit length of tunnel
PLAXIS 3D to do a numerical back analysis of greenfield [2, 7, 18, 25–27]. Total volume loss includes ground loss at
deformations caused by twin bored tunnels. Three soil the face of the tunnel (caused by movement of soil towards
constitutive models were used, and parametric study was the excavation chamber), ground loss along the shield (due
performed. Zheng et al [15] proposed a fuzzy statistics- to progressive decrement of shield diameter from the cutter
based method for the calculation of the standard values and head to the TBM tail), ground loss at the tail (Due to the
deviations of mechanical driving parameters. In order to cavity between the tunnel lining and surrounding soil and
study the impact of the mechanical driving parameters on movement of soil into the gap caused by reduction in grout
ground movement, a finite element method was used. pressure), ground loss behind the shield tail (caused by
Golpasand et al [16] calculated the real volume loss using consolidation of grout) and long term volume loss caused
semiempirical, numerical methods via performing back by grout shrinkage and lining deformations [16, 23, 27–29].
analysis on real settlements. Numerical modelling was Volume loss results in ground settlements. In case of clay
performed using FLAC 3D code. Miliziano and Lillis [17] considering non-dilation and undrained condition volume
developed a PLAXIS 3D model to examine the soil, tunnel of surface settlement trough per unit length (Vs) should be
Sådhanå (2024) 49:75 Page 3 of 12 75

equal to the volume of soil moving into the tunnel cavity

(VT) [4, 18, 27]. Cohesionless soils like fills, silts, sands,
and unsaturated clays provide some drainage which causes
actual surface settlement trough to have a different volume
than the volume of ground that moves into the tunnel cavity
as a result of volume change within the soil [24]. Consol-
idation of clays may also cause sublayer subsidence
resulting in additional ground movements along with those
obtained due to volume loss during tunnelling activity [30].
The volume of the surface settlement trough may be less
than the amount of ground that moves into the tunnel due to
dilatation and swelling brought on by unloading that may
cause soil expansion [12].
The survey of the literature reveals that several aca-
demics have worked extensively to estimate the ground
deformations brought on by TBM tunnelling. However, the
majority of the aforementioned research has been con- Figure 1. Tunnel alignment and the study area [31].
ducted outside of India. Large-scale TBM tunnel con-
struction is currently underway in Delhi, India, in order to
establish metro lines. Therefore, a thorough investigation is
present work has been obtained using open-sourced tender
needed to evaluate the impact of TBM tunnelling on the
documents [32]. Tunnel Alignment considered in this study
ground of Indian region. In this paper, ground settlements
is represented in figure 1.
due to TBM tunnelling are obtained using Finite Element
Program PLAXIS 2D for Delhi, capital of India. To do this,
input parameters for soil properties and tunnelling dimen-
sions are used from one of the tunnelling projects ongoing 3. Soil and tunnel parameters for numerical
in Delhi national capital region. Results of finite element modelling
analysis are validated using empirical methods. Finally, a
parametric study has been conducted to evaluate the effects Using field and laboratory test results given in Geotechnical
of tunnel and soil parameters on ground settlements under investigation report, available in tender documents [32]
greenfield condition. interpretation and derivation of geological sub-strata pro-
files with suggested design parameters have been done for
the stretch from Anand Vihar launch shaft to Khichripur
2. Problem statement retrieval shaft. The geology of the subject alignment mainly
consists of two types of strata, fine silty sand (SM) and low
The Delhi–Meerut Regional Rapid Transit System (Delhi– plastic sandy silt (CL). Maximum water table was observed
Meerut RRTS) is an 82.15 km long, semi-high speed rail to be at 15 m below the ground level, as per the geotech-
corridor currently under construction which will connect nical investigation report. The ground profile is made up of
Delhi, Ghaziabad, and Meerut. This corridor has two eight layers that are distinguished by their thickness, soil
underground stretch, one 5.6 km New Ashok Nagar— type, physical and geotechnical properties. Physical prop-
Sahibabad underground package 4 (1 underground station erties and shear strength parameters of soils are taken from
at Anand Vihar) and other 5.68 km Brahampuri DN Ramp - the drained and undrained laboratory test results given in
Begumpul UP Ramp underground package 8 (3 under- Geotechnical investigation report, available in tender doc-
gound stations at Meerut Central, Bhaisali and Begumpul). uments [32]. Soil modulus is obtained using Standard
RRTS underground package 4 stretch will be done by 4 penetration test resistance, N and CIRIA report 143 by
TBMs named as Sudarshan. TBM Sudarshan 4.1 and 4.2 Clayton [33]. Table 1 summarizes the layer-wise physical
will build 2.8 km tunnel from Anand Vihar launch shaft to and geotechnical properties of soil used for numerical
Khichripur retrieval shaft located at the northern end of analysis.
Ashok Nagar DN ramp. TBM Sudarshan 4.3 and 4.4 will Tunnel is having an internal diameter of 6.5 m which is
build 2.09 km tunnels between Anand Vihar launch shaft bigger than the other metro tunnels due to the bigger rolling
and Sahibabad UP ramp [31]. For the present study, tunnel stock and a high design speed of 180 kmph. Depth of
alignment for the stretch from Anand Vihar launch shaft to Tunnel axis is 20 m below the ground surface. Tunnel
Khichripur retrieval shaft is considered for evaluation of lining is assumed to be made of 300 mm thick M50 grade
tunnelling-induced ground settlements under greenfield concrete segments. Table 2 shows material properties of
conditions. Most of the relevant data required for the TBM and lining.
 75 Page 4 of 12 Sådhanå (2024) 49:75

Table 1. Soil properties for numerical analysis [32–34].

Soil layer and parameter Layer-1 Layer-2 Layer-3 Layer-4 Layer-5 Layer-6 Later-7 Layer-8
Layer type SM SM SM CL CL SM SM SM
Drainage type Drained Drained Drained Undrained B Undrained B Drained Drained Drained
Depth (m) 0–5 5–10 10–15 15–20 20–25 25–30 30–40 Below 40
Unsaturated unit weight, !unsat (kN/m3) 17 17 18 19 19 20 20 20
Saturated unit weight, !sat (kN/m3) 18 18 19 20 20 21 21 21
Young’s modulus, E’ref (kPa) 8000 12000 20000 20000 29000 45000 50000 60000
Poisson’s ratio, m (nu) 0.38 0.38 0.38 0.35 0.35 0.38 0.38 0.38
Friction angle, u’ (phi)() 30 30 32 – – 34 35 35
Cohesion, c’ref (kPa) 0 0 0 – – 0 0 0
Undrained shear strength, su (kPa) – – – 90 130 – – –

Table 2. Material properties of TBM and lining [35, 36].

Material parameters Lining TBM

Material type Elastic Elastic
Unit weight, w (kN/m2) 7.5 19.5
Axial stiffness, EA1 (kN/m) 10.6*106 50*106
Bending stiffness, EI (kN-m2/m) 79.5*103 260*103
Poisson’s ratio, m (nu) 0.15 0.3

4. Finite element analysis

Figure 2. PLAXIS 2D model of existing ground conditions with
tunnel.
Even though tunnelling is a three-dimensional (3D) oper-
ation, transverse settlement under greenfield conditions can
be estimated rather adequately in two-dimensional (2D) used for numerical analysis. Table 2 summarizes material
simulation. Numerous parameters are required for 3D properties for TBM, and lining used for numerical anal-
numerical modelling, which makes it complicated, time ysis. Plate elements are used to model TBM and tunnel
consuming and computationally demanding. Moreover, lining. For simulating the interaction between ground and
parameters required for 3D modelling are sometimes TBM as well as ground and linings interface elements
challenging to determine in practice. For the sake of sim- with strength reduction factor (Rinter = 0.67) for all the soil
plicity, it is reasonable to assume that the tunnel is very layers are provided. The water table is assigned to produce
long and at a consistent depth [18]. the hydrostatic initial pore water pressure and phreatic
analysis has been done with the assumption of no
groundwater flow during excavation and lining is imper-
meable. In the proposed study volume loss at tunnel (V L )
4.1 Numerical modelling
is taken as 1.5%.
In this study, numerical simulations have been performed The process of tunnel construction has been modelled
using the finite element program PLAXIS 2D, 2022 to through step-by-step method in 5 phases consists of: (1)
estimate the tunnelling-induced transverse settlements. Initial phase includes setting up of initial conditions (ef-
A plain strain 15-noded element model is used for fective stresses and pore pressures) for the model before
analysis. In order to minimise the boundary effects the excavation with K0 procedure calculation type, (2) TBM
dimensions of the model have been considered 10 times the phase includes the excavation of soil and activation of
diameter of Tunnel in all directions. The model has been TBM, (3) TBM Conicity phase includes contraction of
provided with standard fixities [35, 36]. Model consists of tunnel lining simulation due to the volume loss resulted
650 elements with 5401 nodes after meshing. Final Model from the cone shape of TBM, (4) Tail void grouting phase
geometry and meshing used for FE analysis is shown in includes the simulation of application of grout pressure at
figures 2 and 3, respectively. Elastoplastic soil constitutive the tail of the TBM. The grouting process is simulated by
Mohr–Coulomb model, is used for modelling of soil layers applying a pressure on the surrounding soil, (5) Lining
in the current work. Table 1 summarizes soil properties installation phase includes the activation of final lining.
Sådhanå (2024) 49:75 Page 5 of 12 75

Figure 3. PLAXIS 2D Model after meshing.

Deformed mesh and Total vertical deformations after cal-

culations are shown in figures 4 and 5, respectively.

4.2 Validation
Prior to assessing the influence of governing parameters on
the ground settlements induced by tunnelling, the green-
field surface settlement trough obtained by numerical
modelling has been compared with the surface settlement
trough from empirical methods proposed in literature.
Figure 6 shows the PLAXIS output for the vertical settle-
ments at the ground surface. Here, validation of finite ele-
ment model is done with four empirical relations given in
literature. Figure 5. PLAXIS output—total vertical deformations.
Empirical method proposed by Peck et al [26] and
Attewell and Farmer [27] is most widely used for calcu-
lating tunnelling induced ground settlements because of its
suitability and ease of use. This method typically follows a
Gaussian distribution curve when predicting ground set-
tlements. The primary reasons for its adaptability are its

Figure 6. PLAXIS Output- Surface settlement trough.

shape which shows a close fit with the commonly observed

experimental and field results, and its requirement of few
easy to determine parameters [7]. This method calculates
Figure 4. PLAXIS output—deformed mesh after calculations. ground surface settlement (Sx ) at a horizontal distance (x)
 75 Page 6 of 12 Sådhanå (2024) 49:75

Celestino et al [38] presented settlement trough as a

three-parameter curve equation (8) with Sx , a (length
dimension) and b (dimensionless parameter) as three
parameters.
Smax
Sx ¼  b ð8Þ
1 þ jaxj

i ¼ aB ð9Þ
 1
b1 b
Figure 7. Transverse Surface settlement trough, Gaussian B¼ ð10Þ
bþ1
distribution.
b ¼ 1to3
from the center line of the tunnel (figure 7) using equation a
(1) as follows: ¼ 0:39ðz=DÞ þ 0:31 ðFor Stiff ClayÞ ð11Þ
D
 2
x a
Sx ¼ Smax exp ð1Þ ¼ 0:46ðz=DÞ þ 0:37ðFor Porous ClayÞ ð12Þ
2i2 D
Vorster et al [5] suggested a modified Gaussian curve
D2
Smax ¼ 0:313V L ð2Þ equation (13) to obtain a better fit to the observed soil
i settlement, of the following form with a shape function
VT parameter (n) controlling the width of the profile and a
VL ¼ ð3Þ parameter a to ensure that i remains the distance to the
pD2 =4
inflection point.
i ¼ kz ð4Þ
nSmax
Sx ¼ h  i ð13Þ
Where, k is the trough width parameter which varies 2
ðn  1Þ þ exp a xi
from 0.4 to 0.7 for cohesive soils and 0.2–0.3 for cohe-
sionless soils [37]. z is the depth of tunnel axis. Smax is the
2a  1
maximum surface settlement under greenfield condition, n ¼ ea þ1 ð14Þ
V T is the volume of ground loss around a tunnel cavity per 2a þ 1
unit length, V L is volume of ground loss around a tunnel n ¼ 0:5; 1; 1:5
cavity (VT) with respect to the tunnel volume per unit
length and i is the distance between center point of the Figure 8 compares the surface settlement profiles after
tunnel and the inflection points. Gaussian curve does not excavation of the proposed tunnel from numerical and
always accurately describe the ground surface settlement in empirical methods. By idealizing the settlement trough
several cases and provide inadequate fit to settlement data profile obtained by PLAXIS as a Gaussian curve, the set-
above tunnels in sands [4]. Therefore, many researchers tlement trough parameter (i) is found to be 11 m. The
modified the Gaussian equation of settlement trough to corresponding k-value is 0.55. By taking i equals to 11, the
obtain true surface settlement trough. Gaussian distribution and modified curves have been
Jacobsz et al [3] proposed a modified settlement trough
equation (5) to accommodate the narrower Gaussian shape
x (m)
settlement trough with k0 is the modified trough width
-50 -40 -30 -20 -10 0 10 20 30 40 50
parameter and i0 is the distance between centre point of the
tunnel and the inflection points. 0

  !
Sx (mm)

1 j xj 1:5 -10
Sx ¼ Smax exp  ð5Þ
3 i0 -20 PLAXIS 2D, 2022
Peck, 1969
0 0 Celestino et al., 2000
i ¼kz ð6Þ -30
Jacobsz et al., 2004
Vorster et al., 2005
k0 ¼ 0:85K  0:12 ð7Þ -40

Figure 8. Comparison of surface settlement troughs from

PLAXIS 2D and empirical methods.
Sådhanå (2024) 49:75 Page 7 of 12 75

superimposed on settlement trough obtained by PLAXIS. x (m)

There is an agreement between results of finite element -50 -40 -30 -20 -10 0 10 20 30 40 50

modelling and empirical methods in terms of shape of 0

profile, but finite element modelling shows a narrow set-
-10

Sx (mm)
tlement trough, this is due to the layered soil profile under
water table [39]. Volume of settlement through is not -20
D=5.5m (GP=178.4 kPa)
exactly same as volume loss at tunnel. In this case, volume -30
D=6.5m (GP=175.6 kPa)
of settlement through is 1% corresponding to the 1.5%
-40 D=7.5m (GP=172.5 kPa)
volume loss at tunnel. Mathew and Lehane [13] also
D=8.5m (GP=170.0 kPa)
observed the same phenomenon and stated that this hap- -50
pened due to effects of the rigidity of the tunnel lining as
well as the initial stress state and stiffness of the soil. From Figure 9. Variation of surface settlement trough with tunnel
diameter (D).
figure 8 a minor degree of soil heave is also observed due to
dissipation of small negative excess pore pressure in clay
D (m)
layers. Occurrence of heave may also result due to the equal
4 5 6 7 8 9 10
loading and unloading stiffness moduli in the Mohr-Cou-
lomb model [40].
-10

Smax (mm)
5. Parametric study: results and discussion -20

Tunnelling-induced ground settlements are influenced by -30

various parameters. Tunnel parameters for which effect on
ground settlement analysed are tunnel diameter (D), tunnel
-40
axis depth (z) and grout pressure (GP). Soil parameters for
which effect on ground settlement analysed are undrained Figure 10. Variation of maximum surface settlement (Smax) with
shear strength of soil (su) and modulus (E’ref) of soil tunnel diameter (D)
through which tunnel excavation is done. Influences of each
of the above parameters on the tunnelling-induced settle-
ments are analysed. Table 3 highlights the range of 4 cases have been analysed maintaining the z at 20 m below
parameters that are investigated in the present work. The the ground level. It can be observed from figures 9 and 10,
primary goal of selecting a specific range was to compre- that the increase in D significantly increases the Sx and Smax
hend how these parameters affected the settlement. respectively, as reported by many researchers [19, 21].
Maximum surface settlements along with their increment
rates for different tunnel diameters are summarized in
5.1 Effect of tunnel parameters Table 4. An increase of 112.4% in Smax has been found
when D increased from 5.5 m to 8.5 m. GP required for
Effects of tunnel parameters such as D and z are investi- maintaining tunnel VL as 1.5% is also shown in figure 9.
gated by maintaining the VL as 1.5%. Effect of GP has been This is clear from the GP values that with increase in D less
analysed by maintaining the D as 6.5 m and z as 20 m GP is required to maintain the equal VL. It can also be
below the ground surface. Detailed discussion of effects of observed from figure 9, that surface settlement trough peak
tunnel parameters is given in the following paragraphs. becomes sharper with increase in the D.
5.1a Tunnel diameter (D) effect One of the most 5.1b Tunnel axis depth (z) effect The effect of z on Sx has
important factors influencing the ground settlements is the been examined by varying the z from 10 m to 30 m with an
tunnel’s size [21]. To study the effect of D on Sx, D is increment of 5 m. Total 5 cases have been analysed
varied from 5.5 m to 8.5 m with an increment of 1 m. Total maintaining the D as 6.5 m. It can be observed from

Table 3. Range of parameters variation for parametric study.

Parameters Range Increment Base Value

Tunnel diameter, D (m) 5.5-8.5 1 6.5
Tunnel depth, z (m) 10-30 5 20
Grout pressure, GP (kPa) 154.2, 175.6, 200 and 237 – 175.6
Undrained shear strength, su (kPa) 0.75su–1.75su 0.25su su (90 & 130 for layer 4 & 5 respectively)
Young’s modulus, E’ref (kPa) 0.75E’ref–1.75E’ref 0.25E’ref E’ref (20*103 & 29*103 for layer 4 & 5 respectively)
 75 Page 8 of 12 Sådhanå (2024) 49:75

Table 4. Smax versus D. x (m)

-50 -40 -30 -20 -10 0 10 20 30 40 50

Tunnel diameter Maximum surface Rate of increment 0

(D, m) settlement (Smax) (mm/m)
-10

Sx (mm)
5.5 13.62 –
-20
6.5 18.08 4.46 GP=154.2 kPa (VL=2%)
7.5 23.23 5.15 -30 GP=175.6 kPa (VL=1.5%)
8.5 28.93 5.7
-40 GP=200 kPa (VL=1%)

GP=237 kPa (VL=0.5%)

-50
x (m)
-50 -40 -30 -20 -10 0 10 20 30 40 50
Figure 13. Variation of surface settlement trough with Grout
0 Pressure (GP).

-10
GP (kPa)
Sx (mm)

-20
z=10m (GP=57.2 kPa)
100 120 140 160 180 200 220 240 260
0
-30 z=15m (GP=110 kPa)
z=20m (GP=175.6 kPa)
-40 z=25m (GP=221 kPa) -10

Smax (mm)
z=30m (GP=259 kPa)
-50 -20

Figure 11. Variation of surface settlement trough with tunnel

-30
depth (z).
-40
z (m)
0 5 10 15 20 25 30 35 Figure 14. Variation of maximum surface settlement (Smax) with
0
Grout Pressure (GP).

-10
Smax (mm)

Table 6. Smax versus GP.

-20
Grout pressure Maximum surface Rate of decrement
-30
(GP, kPa) settlement (Smax) (mm/kPa)
154.2 25.6 –
-40 175.6 18.08 0.3514
200 11.77 0.25861
Figure 12. Variation of maximum surface settlement (Smax) with 237 6.08 0.15378
tunnel depth (z).

figure 11 that rise in z decreases the Sx near the centreline and their rate of decrement for each tunnel depth. A
of the tunnel however away from the centreline opposite decrease of 62.55% in Smax has been observed when z
phenomenon occurs. Figure 12 shows that increase in z increases from 10 m to 30 m. GP required for maintaining
results in significant decrease in Smax. Previous researchers VL 1.5% is also shown in figure 11. This is clear from the
also saw a decrease in maximum settlements by increasing GP values that with increase in z more GP is required to
the depth of tunnel axis [16, 20, 21, 27]. maintain the equal VL. It can also be observed from fig-
It can also be observed from figure 11, that width of ure 11, that surface settlement trough peak becomes flatter
surface settlement trough increases with increasing z [4]. with increase in the z.
Table 5 shows the values of maximum surface settlements 5.1c Grout pressure (GP) effect To study the effect of GP
on Sx its values are varied as 154.2 kPa, 175.6 kPa, 200 kPa
Table 5. Smax versus z. and 237 kPa corresponding to the VL of 2%, 1.5%, 1% and
Tunnel depth Maximum surface Rate of decrement 0.5% respectively. From the results presented in the fig-
(z, m) settlement (Smax) (mm/m) ures 13 and 14, it can be observed that the increase in GP
results in decrease in VL which reduces the Sx and Smax
10 34.18 – respectively as foreseen by many researchers
15 25.24 1.8
[16–18, 21, 41, 42]. Table 6 summarizes the maximum
20 18.08 1.4
surface settlement values and their rate of decrements for
25 14.87 0.6
30 12.8 0.4 each value of grout pressure. A decrease of 76.25% in Smax
Sådhanå (2024) 49:75 Page 9 of 12 75

has been observed when volume decreases from 2% to x (m)

0.5% for the grout pressure values of 154.2 kPa and 237 -50 -40 -30 -20 -10 0 10 20 30 40 50

kPa, respectively. 0

-10

Sx (mm)
5.2 Effect of soil parameters -20
0.75su (VL=2.34%)
-30 su (VL=1.5%)
Effects of soil parameters were investigated by maintaining
1.25su (VL=1.22%)
the D as 6.5 m and z as 20 m below the ground surface to -40
1.5su (VL=1.13%)
make it pass through layers 4 and 5. For parametric study 1.75su (VL=1.09%)
-50
parameters of soil from which tunnel pass through have
been varied within a practical range. Soil parameters effects Figure 17. Variation of surface settlement trough with
are studied in two cases, first by maintaining the VL as Undrained shear strength (su) under constant GP.
1.5% and second by maintaining the GP as 175.6 kPa.
Detailed discussion of both the cases with respect to soil su Times (kPa)
parameters is given in following paragraphs. 0 0.5 1 1.5 2 2.5
5.2a Undrained shear strength (su) effect To investigate 0

the effect of su its values for layers 4 and 5 have been

modified from 0.75 times to 1.75 times with an increment -10

Smax (mm)
of 0.25 times, with respect to the actual value considered in
-20
Table 1. Impact of su is illustrated in the figures 15, 16, 17
and 18. It can be observed that increase in su results in
-30
decrease in Sx and Smax in both the cases under constant VL
(1.5%) as well as constant GP (175.6 kPa). Same scenario
-40
of decreasing surface settlement with increase in su was
observed by Kanagaraju and Krishnamurthy [19]. Table 7 Figure 18. Variation of maximum surface settlement (Smax) with
summarizes maximum surface settlements along with their Undrained shear strength (su) under constant GP.

Table 7. Smax versus su under constant VL.

x (m)
-50 -40 -30 -20 -10 0 10 20 30 40 50
Undrained shear Maximum surface Rate of decrement
0 strength (su, kPa) settlement (Smax) (mm/0.25 times su)
0.75 su 21.53 –
Sx (mm)

-10
1 su 18.08 3.45
-20
0.75su (GP=202 kPa) 1.25 su 16.03 2.05
su (GP=175.6 kPa) 1.5 su 15.16 0.87
1.25su (GP=158.5 kPa) 1.75 su 14.55 0.61
-30
1.5su (GP=149.4 kPa)
1.75su (GP=144 kPa)
-40
rate of decrements for each value of su under constant VL of
Figure 15. Variation of surface settlement trough with 1.5%. GP required for maintaining VL as 1.5% is also
Undrained shear strength (su) under constant VL. shown in figure 15. This is clear from the GP values that
with increase in su value less GP is required to maintain the
equal VL. Table 8 summarizes maximum surface settle-
su Times (kPa) ments along with their rate of decrements for each value of
0 0.5 1 1.5 2 2.5
0

-10
Table 8. Smax versus su under constant GP.
Smax (mm)

Undrained shear Maximum surface Rate of decrement

-20
strength (su, kPa) settlement (Smax) (mm/0.25 times su)
-30 0.75 su 34.64 –
1 su 18.08 16.56
-40 1.25 su 12.78 5.3
1.5 su 11.22 1.56
Figure 16. Variation of maximum surface settlement (Smax) with 1.75 su 10.49 0.73
Undrained shear strength (su) under constant VL
 75 Page 10 of 12 Sådhanå (2024) 49:75

su under constant GP of 175.6 kPa. Volume loss caused by E'ref Times (kPa)
0 0.5 1 1.5 2
constant grout pressure is also shown in figure 17. This is 0
clear from the VL values that with increase in su value VL
decreases with constant GP. When su increases upto 1.75su

Smax (mm)
from 0.75su, a 32.42% and 69.71% decrease in Smax have -10

been found under constant VL and constant GP

respectively.
-20
5.2b Soil modulus (E’ref) effect To investigate the effect
of E’ref its values for layers 4 and 5 have been modified
from 0.75 times to 1.75 times with an increment of 0.25 -30
times with respect to the actual value considered in the
Table 1. Impact of E’ref is illustrated in figures 19, 20 and Figure 21. Variation of maximum surface settlement (Smax) with
21. It was observed that under constant VL of 1.5% Sx Soil modulus (E’ref) under constant GP.
remains nearly equal for all values of E’ref.. GP required for
maintaining tunnel VL 1.5% is also shown in figure 19. This Table 9. Smax versus E’ref under constant GP.
is clear from the GP values that with increase in E’ref value
less GP is required to maintain the equal VL. Young’s
modulus (E’ref, Maximum surface Rate of decrement
Under constant GP of 175.6 kPa settlement decreases
kPa) settlement (Smax) (mm/0.25 times E’ref)
with increasing E’ref. Table 9 summarizes the maximum
surface settlements and their rate of decrements for each 0.75 E’ref 21.12 –
value of E’ref under constant GP. VL caused by constant GP 1 E’ref 18.08 3.04
is also shown in figure 20. This is clear from the VL that 1.25 E’ref 16.03 2.05
with increase in E’ref VL decreases under constant GP. 1.5 E’ref 15.08 0.95
1.75 E’ref 14.23 0.85
Similar trend of decrease in surface settlements and volume
loss with increase in modulus was seen by many
researchers [19, 21]. A decrease of 32.6% in Smax has been 6. Conclusion
observed when E’ref increases from 0.75 times to 1.75 times
of its base value. In the current study TBM tunnelling induced ground set-
tlements under greenfield condition are estimated by sim-
x (m) ulating step-by-step tunnel construction process in finite
-50 -40 -30 -20 -10 0 10 20 30 40 50
element software PLAXIS 2D. A parametric study has been
0 conducted to analyse the effect of different vital tunnel and
soil parameters on ground settlements. Prior to conducting
-10 the parametric study, the results of numerical model were
Sx (mm)

0.75E'ref (GP=184.5 kPa)

compared to the existing empirical approaches for valida-
-20
E'ref (GP=175.6 kPa) tion. Major findings of this study are summarised as
-30
1.25E'ref (GP=169 kPa) follows:
1.5E'ref (GP=164 k)
1.75E'ref (GP=159.5 kPa) • PLAXIS 2D can provide a reasonable prediction of
-40
transverse surface settlements in layered soils of
Figure 19. Variation of surface settlement trough with Soil varying stiffnesses.
modulus (E’ref) under constant VL. • Maximum surface settlements and their rate of incre-
ments increase with increasing the tunnel diameter
under constant volume loss. When the diameter of
x (m)
-50 -40 -30 -20 -10 0 10 20 30 40 50 tunnel was increased from 5.5 m to 8.8 m a significant
112.4% increase in the maximum surface settlement
0 was observed.
• Maximum surface settlements and their rate of decre-
-10
ments decrease with increasing the tunnel depth under
Sx (mm)

0.75E'ref (VL=1.71%)
-20 E'ref (VL=1.5%)
constant volume loss. When the tunnel’s axial depth
1.25E'ref (VL=1.36%) was increased from 10 m to 30 m from the ground
-30 1.5E'ref (VL=1.28%) surface a significant 62.55% decrease in the maximum
1.75E'ref (VL=1.22%) surface settlement was observed.
-40
• Increase in grout pressure results in decrease in volume
Figure 20. Variation of surface settlement trough with Soil loss. Volume loss is an important parameter which
modulus (E’ref) under constant GP. affects the maximum surface settlements. Maximum
Sådhanå (2024) 49:75 Page 11 of 12 75

surface settlements and their rate of decrements [8] Fu J, Yang J, Klapperich H and Wang S 2016 Analytical
decrease with increasing the grout pressure. When prediction of ground movements due to a nonuniform
the grout pressure was increased from 154.2 kPa to 237 deforming tunnel. Int. J. Geomech. 16: 04015089
kPa a significant 76.25% decrease in maximum surface [9] Lai H, Zhang J, Zhang L, Chen R and Yang W 2019 A New
Method based on Centrifuge Model Test for Evaluating
settlement was observed.
Ground Settlement Induced by Tunnelling. KSCE Journal of
• Maximum surface settlements and their rate of decre-
Civil Engineering 23: 2426–2436
ment decrease with increasing the undrained shear [10] Hong X, Zhang D and Sun Z 2023 Mechanical responses of
strength of soil through which tunnel is excavated multi-layered ground due to shallow tunnelling with arbitrary
under constant volume loss as well as constant grout ground surface load. Frontiers of Structural and Civil
pressure. When undrained shear strength of soil Engineering. 17: 745–762
through which tunnel is excavated is increased from [11] Sun F, Jin Z, Wang C, Gou C, Li X, Liu C and Yu Z 2023
0.75 to 1.75 times its base value a significant 32.42% Case Study on Tunnel Settlement Calculations during
and 69.71% decrease in maximum surface settlement Construction Considering Shield Disturbance. KSCE Journal
was observed under constant volume loss and constant of Civil Engineering. 27: 2202–2216
grout pressure, respectively. [12] Möller S C and Vermeer P A 2008 On numerical simulation
of tunnel installation. Tunn. Undergr. Space Technol. 23:
• Maximum surface settlements and their rate of decre-
461–475
ment decrease with increasing the soil modulus
[13] Mathew G V and Lehane B M 2013 Numerical back-
through which tunnel is excavated under constant analyses of greenfield settlement during tunnel boring. Can.
grout pressure. When soil modulus through which Geotech. J. 50: 145–152
tunnel is excavated is increased from 0.75 to 1.75 times [14] Mathew G V and Lehane B M 2014 Measured and Back
its base value a significant 32.6% decrease in maxi- Analysed Soil Structure Interaction Effects in a Layered
mum surface settlement was observed under constant Stratigraphy During Tunnel Boring. Geotech Geol. Eng. 32:
grout pressure. However, under constant volume loss, 873–884
soil modulus does not affect the ground settlements. [15] Zheng G, Lu P and Diao Y 2015 Advance speed-based
• Soil modulus has a little impact on surface settlements parametric study of greenfield deformation induced by
as compared to the other parameters. EPBM tunnelling in soft ground. Computers and Geotech-
nics 65: 220–232
Above conclusions demonstrate a substantial correlation [16] Golpasand M R B, Nikudel M R and Uromeihy A 2016
between surface settlement and tunnel parameters as well Specifying the real value of volume loss (VL) and its effect
as surface settlement and soil parameters, highlighting the on ground settlement due to excavation of Abuzar tunnel,
significant impact of such modifications on ground stability. Tehran. Bull. Eng. Geol. Environ. 75: 485–501
[17] Miliziano S and Lillis A D 2019 Predicted and observed
settlements induced by the mechanized tunnel excavation of
metro line C near S. Giovanni station in Rome. Tunn.
Undergr. Space Technol. 86: 236–246
References [18] Shiau J and Sams M 2019 Relating volume loss and
greenfield settlement. Tunn. Undergr. Space Technol. 83:
[1] Athar M F, Sadique M R, Alsabhan A H and Alam S 2022 145–152
Ground Settlement Due to Tunneling in Cohesionless Soil. [19] Kanagaraju R and Krishnamurthy P 2020 Influence of
Appl. Sci. 12: 3672 Tunneling in Cohesionless Soil for Different Tunnel Geom-
[2] Loganathan N, Poulos H G and Stewart D P 2000 Centrifuge etry and Volume Loss under Greenfield Condition. Advances
model testing of tunnelling-induced ground and pile defor- in Civil Engineering 1–11
mations. Geotechnique 50: 283–294 [20] Rezaei A H and Adli M A 2020 The Volume Loss: Real
[3] Jacobsz S W, Standing J R, Mair R J, Hagiwara T and Estimation and Its Effect on Surface Settlements Due to
Sugiyama T 2004 Centrifuge modelling of tunnelling near Excavation of Tabriz Metro Tunnel. Geotech. Geol. Eng. 38:
driven piles. Soils and Foundations 44: 49–56 2663–2684
[4] Marshall A M, Farrell R, Klar A and Mair R 2012 Tunnels in [21] Aswathy M S, Vinoth M and Mittal A 2022 Impact of
sands: the effect of size, depth and volume loss on greenfield Governing Factors on Prediction of Tunneling Induced
displacements. Ge´otechnique 62: 385–399 Surface Settlement in Young Alluvium Deposit. Indian
[5] Vorster T E B, Klar A, Soga K and Mair R J 2005 Estimating Geotech. J. 52: 13–27
the Effects of Tunneling on Existing Pipelines. J. Geotech. [22] Ahmed K S, Sharmin J and Ansary M A 2023 Numerical
Geoenviron. Eng. 131: 1399–1410 investigation of tunneling induced surface movement: A case
[6] Verruijt A and Booker J R 1996 Surface settlements due to study of MRT Line 1. Dhaka. Underground Space. 12:
deformation of a tunnel in an elastic half Plane. Ge´otech- 116–136
nique 46: 753–756 [23] Al-Omari R R, Al-Soud M S and Al-Zuhairi O I 2019 Effect
[7] Loganathan N and Poulos H G 1998 Analytical prediction of Tunnel Progress on the Settlement of Existing Piled
for tunnelling induced ground movements in clays. J. Foundation. Studia Geotechnica et Mechanica 41: 102–113
Geotech. Geoenviron. Eng. 124: 846–856
 75 Page 12 of 12 Sådhanå (2024) 49:75

[24] Islam M S and Iskander M 2021 Twin tunnelling induced [34] Bowles J E 1988 Geotechnical and index properties:
ground settlements: A review. Tunn. Undergr. Space Tech- laboratory testing; settlement and strength correlations,
nol. 110: 103614 Foundation analysis and design. 5th edition, New York,
[25] EIMouchi A M, Hassan A M and Amer M I 2018 McGraw-Hill, 2: 123
Performance of an existing raft foundation rested on granular [35] PLAXIS 2D- Tutorial Manual. Edition V22
soils due to TBM tunnelling process. Arabian Journal of [36] PLAXIS 2D- Reference Manual. Edition V22
Geosciences 11: 108 [37] O’Reilly M and New B 1982 Settlements above tunnels in
[26] Peck R B 1969 Deep excavations and tunneling in soft the United Kingdom: their magnitude and prediction. In:
ground. In: Proceedings of 7th International Conference on Proceedings of Tunnelling ’82, Brighton pp. 173–181
Soil Mechanics and Foundation Engineering, Mexico City, [38] Celestino T B, Gomes R A M P and Bortolucci A A 2000
State-of-the-art Volume, pp. 225–290 Errors in Ground Distortions Due to Settlement Trough
[27] Attewell P B and Farmer I W 1974 Ground disturbance Adjustment. Tunn. Undergr. Space Technol. 15: 97–100
caused by shield tunnelling in a stiff, overconsolidated clay. [39] Lee S W 2002 The use of compensation grouting in
Engineering Geology 8: 361–381 tunnelling: a case study. Proceedings of The Institution of
[28] Vu M N, Broere W and Bosch J 2016 Volume loss in Civil Engineers- Geotechnical Engineering 155: 101–109
shallow tunnelling. Tunn. Undergr. Space Technol. 59: [40] Obrzud R F and Truty A 2018 Introduction The hardening
77–90 soil model - a practical guidebook. Z Soil.PC 100701 report.
[29] Dias T G S and Bezuijen A 2015 TBM Pressure Models – Switzerland: Zaca Service Ltd. pp. 2–10
Observations, Theory and Practice. In: Proceedings of 15th [41] Fargnoli V, Boldini D and Amorosi A 2013 TBM tunnelling-
Pan-American Conference on Soil Mechanics and Geotech- induced settlements in coarse-grained soils: The case of the
nical Engineering - Geotechnical Synergy in Buenos Aires, new Milan underground line 5. Tunn. Undergr. Space
pp. 347–374 Technol. 38: 336–347
[30] Wu H N, Shen S L and Yang J 2017 Identification of Tunnel [42] Jenck O and Dias D 2003 Numerical analysis of the volume
Settlement Caused by Land Subsidence in Soft Deposit of loss influence on building during tunnel excavation. In:
Shanghai. J. Perform. Constr. Facil. 31: 04017092 Proceedings of the third international FLAC Symposium-
[31] The Metro Railguy Home page https://themetrorailguy.com/ FLAC and FLAC3D Numerical Modeling in Geomechanics,
delhi-meerut-rrts-information-route-maps-fares-tenders-upda Sudbury, Ontario, Canada pp. 135–144
tes/, Last accessed 03/03/2023
[32] BidAssist Home page https://bidassist.com/delhi-tenders/
national-capital-region-transport-corporation/detail-2a42fa Springer Nature or its licensor (e.g. a society or other partner)
4a-5b37-46b5-98b1-7e3bdeb4585b, Last accessed 05/05/ holds exclusive rights to this article under a publishing agreement
2022 with the author(s) or other rightsholder(s); author self-archiving of
[33] Clayton C R 1995 The standard penetration test (SPT): the accepted manuscript version of this article is solely governed
methods and use. Construction Industry Research and by the terms of such publishing agreement and applicable law.
Information Association. 143