Analysis of tunnel considering Modified Mohr-
Coulomb criterions
S Rukhaiyar
Department of Civil Engineering, IIT Roorkee
Roorkee, India
saurav.rukhaiyar@gmail.com
Abstract Analysis of tunnels and underground
excavations requires a realistic prediction of strength and
deformation characteristics of rockmass. The strength
characteristics of rockmass must be determined by laboratory
testing such as triaxial test. The triaxial tests of rockmass are
very cumbersome, difficult and time consuming. Generally,
classification approach in association with the strength
properties of intact rock is used to determine the triaxial
strength of rockmass. A number of empirical criterion had
been developed in past to simulate the triaxial behavior of
rockmass. The Generalized Hoek-Brown (GHB) criterion
(2002) is most widely and generally used. A new triaxial
strength criterion was presented recently known as Modified
Mohr-Coulomb (MMC) criterion. Probabilistic approach
combined with various classification approaches is used to
predict the triaxial strength of rockmass giving much higher
sense of confidence.
These GHB and MMC criterion are nonlinear in nature.
However, the continuum numerical codes and analysis
methods generally do not accept such nonlinear criterion and
may allow only Mohr-Coulomb linear criteria. The researcher
generally fit a linear equation on nonlinear data point with in
certain range of minor principal stress to calculate the
equivalent Mohr-Coulomb criterion parameter i.e. cohesion
and angle of internal friction. This approach retains the
nonlinear behavior of the rockmass.
The present article discusses in detail the calculation of
triaxial data point for three different rockmass data from
tunnel cases published in literature. The triaxial data points
are calculated based on GHB and MMC criterions. Equivalent
Mohr-Coulomb parameters are calculated by fitting a linear
equation on the non-linear data point. Analysis of unsupported
tunnel case is done with the help of convergence confinement
method. A quantitative comparison between MMC and GHB
criterion is done in light of the analysis results. MMC criterion
predict a lower value of equivalent cohesion while a higher
value of angle of internal friction. The response of tunnel
predicted by MMC criterion are on lower side but comparable.
KeywordsModified Mohr Coulomb criterion; Generalized
Hoek Brown criterion; triaxial strength; Convergence
confinement method
I. INTRODUCTION
Elasto-plastic continuum analysis are generally
carried out to predict the response of underground
structures. The analysis require constitutive models for the
N K Samadhiya
Department of Civil Engineering, IIT Roorkee
Roorkee, India
nksamfce@iitr.ac.in
material nearby. Mohr Coulomb linear model are generally
used as contitutive model. The Mohr-Coulomb model
parameters i.e. cohesion (c) and angle of internal friction ()
are calculated based on triaxial test conducted under various
confining pressure. However, the typical triaxial test on
rockmass is very cumbersome, time consuming and
expensive. Generally field observation in association with
classification approach is used to assess the behavior of
rockmass.
However, some researchers are of the opinion that the
behavior of rock mass is non-linear under triaxial state of
stress [1-6]. The non-linearity in strength must be
incorporated while analyzing the rock structure. Hoek -
Brown criterion [4] was the first nonlinear triaxial strength
criterion presented for rockmass. The strength criterion
undergone various modifications and a Generalized Hoek
and Brown (GHB) criterion was presented by Hoek et al.
[7]. The GHB criterion has a history of development which
incorporated the experience of various researchers in real
field condition. The GHB consider only one classification
system which is also based on visual inspection of
rockmass.
Recently a new criterion was presented by Singh and Singh
[6] as Modified Mohr-Coulomb criterion (MMC) which also
considers the nonlinearity in strength of rockmass under
triaxial state of stress. The criterion is based on the critical
state concept [3]. The criteria can consider any classification
system for evaluating the triaxial strength of rockmass.
To incorporate the effect of non-linearity while analyzing
any rock structure, Linearization of triaxial datapoint
obtained from the strength criterion in a given range of
minor principal stress is done. The equivalent Mohr-
coulomb (EMC) parameters are obtained from linearized
data is used for the analysis. Present papers aims to predict
the triaxial strength of the twenty rockmass data collected
from the literature using Modified Mohr-Coulomb (MMC)
criterion. Equivalent Mohr-Coulomb parameters (c and )
were estimated by linearization of triaxial datapoint for a
given range of minor principal stress. An elasto-plastic
analysis of a hypothetical case of tunnel was carried out
using the analytical solution given by Carranza-Torres [8]
also known as convergence confinement method. A
quantitative comparison with GHB criterion was also
carried out. The response of tunnel considering Hoek-Brown
parameters directly using analytical solution given by
 Table 1 Representative values adopted for the analysis
Sl Rock UCS Ei Unit wt RQD RMR GSI Q Name of site Reference
No. Type (MPa) (GPa) (gm/cc) (%)
1 Andesite 93 41.9 2.37 41 40 46 0.56 Urus Dam, Turkey []
2 Basalt 142 40 2.40 15 44 46 0.63 Urus Dam, Turkey []
3 Limestone 62.3 31.42 2.66 69 38 51 1.88 Guledar Dam, Turkey []
4 Sandstone 64.7 27.20 2.70 34 38 41 0.45 Guledar Dam, Turkey []
5 Limestone 43.85 - 2.30 20 42 43 0.88 Raghadan Tunnel, []
Jordan

Carranza-Torres and Fairhurst [9] and by using equivalent GSI 100 (6)
S exp
Mohr-Coulomb criterion by linearization of the generalized 9 3D
Hoek-Brown criterion was carried out using the analytical 1 1 GSI 20
(7)
a e 15 e 3
solution given by Carranza-Torres [10]. 2 6
Where 1 and 3 are the major and minor principal stresses,
I. STRENGTH CRITERIA c is the uniaxial compressive strength of the intact rock
obtained from statistical analysis of triaxial test data, mi is
A. Modified Mohr Coulomb (MMC) Criterion an intact rock constant and D is a rockmass disturbance
Singh and Singh [] proposed a new non-linear criterion for factor. GSI is the Geological strength index obtained by the
rockmass known as Modified Mohr-Coulomb criterion. The visual inspection of rockmass. Some other methods were
strength criterion incorporates the critical state concept also presented by some researchers to calculate the GSI
presented by Barton (1976). The criterion for jointed rock value like rock volume method etc [].
mass is presented as: For the present analysis it is considered that the blasting is
2sin j 0 1 sin j 0 (1) very good with minimal disturbance to the confined
2
1 sin j 0 ci 1 sin j 0 rockmass surrounding the rock structure. Hence, D = 0 has
1 3 cmass 3 3

(for 0 3 ci ) been considered throughout this paper.

Where 1 and 3 are the major and minor principal stresses. II. ROCKMASS DATA
Parameter ci and cmass are uniaxial compressive strength of For the present study, The database of rockmass had been
intact and jointed rockmass. j 0 is angle of internal friction selected for which the rockmass classification data is
of jointed rockmass corresponding to very small confining available in term of RQD [], RMR [], Q system [] and GSI [].
Table 1 shows the database of the rockmass used in the
pressure ( 3 0 ) and is estimated from the relationship:
present study.
s in i 0
(1 SRF ) (2)
1 s in i 0 The classification of rockmass was presented based on the
sin j 0
s in i 0 RQD, RMR and Q system. Table 2 show the classification of
(2 SRF )
1 s in i 0 the rockmass of the database considered for the present
i 0 is angle of internal friction of intact rock corresponding study. An overall classification is also presented based on
other classification system provided and judgement.
to very small confining pressure i.e. 3 0 []
Table 2 Classification of the Rockmass considered for the
SRF is strength reduction factor and is calculated as:
present study
(3)
SRF cmass Sl Based on Based on Based on Q Overall
ci
No. RQD RMR value
B. Generalised Hoek and Brown (GHB) Criterion 1 Poor Poor Very Poor Poor
The generalised Hoek and Brown criterion [] for the 2 Very Poor Fair Very Poor Poor
rockmass is presented as: 4 Fair Poor Poor Poor
a 5 Poor Poor Very Poor Poor
3
1 3 c mb S (4) 11 Very Poor Fair Very poor Poor
c
GSI 100 (5)
mb mi exp
28 14 D
 principal stress at failure have been estimated corresponding
A. Triaxial Simulation of intact rock to minor principal stress of 0, 2, 4, 6 and 10 MPa for each
Both the rockmass strength criteria under consideration need rock. The average value of major principal stress from the
parameters which require triaxial testing of the intact rock two strength criterion [] is used to calculate the value of mi,
specimen. The parameters are io in the Modified Mohr c and Sinio. Roclab software has been used to calculate mi
Coulomb criterion while parameters c and mi in the and c values whereas Sinio is calculated using Microsoft
generalized Hoek-Brown criterion are calculated based on Excel spreadsheet. The value of Sinio, mi and c is
triaxial test results of intact rock. presented in Table 5. This approach is used so as to have
The c and mi value can be calculated using the roclab uniformity between GHB and MMC criterion in terms of
software. Singh et al. [] reported that io may be obtained intact rock strength parameters.
from triaxial test data as:
Sin
Bi (8) III. CALCULATION OF ELASTIC AND PLASTIC PARAMETERS
2 Bi
i0
OF THE ROCKMASS

Where, A. Calculation of uniaxial strength of rockmass (cmass) and

Elastic Modulus of rockmass (Emass)
Bi 2 Ai ci (9)
The modified Mohr Coulomb (MMC) criterion requires SRF
and value as explained in previous section (Equation 3). The
SRF calculation requires an estimate of cmass i.e. uniaxial
Ai
(
1 3 ci ) for 0 3 ci (10)
compressive strength of rockmass. A number of criteria
( 2
2
3 ci 3)
were presented by previous researchers for the calculation of
A number of triaxial strength criteria have been presented uniaxial strength of jointed rockmass, which are based on
by various researchers to predict the value of major principal various classification systems. Table 3 present twelve such
stress at failure for intact rock corresponding to a minor strength criteria considered in present study. The strength of
principal stress [] (Hoek and Brown, 1980; Ramamurthy, rockmass is calculated from all the twelve criteria which
1993; Singh and Singh, 2005; Singh et al., 2011; Rafiai, consider different classification system. The value of
2011). Most recently Shen et al. [] presented a simplified rockmass strength obtained from the twelve equations is not
strength criterion for intact rock based on rock type and a unique value and found to be very scatter. Probabilistic
uniaxial compressive strength. analysis is in vogue to have a better representation of
The Shen strength criterion is presented as: variability []. To get a unique representative value of
rockmass strength, Weibulls probability distribution was
0.5
3 (11) fitted into the all these twelve value. This approach of using
1 3 ci B ci
ci probability distribution gives a confidence level in the
representative value of parameter.
where,
B a cib1 (12) The Weibulls probability distribution function is defined
ci is the uniaxial compressive strength of intact rock in as:
m
u
MPa. a and b are constants which depend upon rock type.
a
Shen et al. [] presented a table for a and b for eight specific P 1 e (16)
rock type. However the value of a and b may be taken as 2.0
and -0.97 respectively in general, if no data regarding rock Where, P = Probability of occurrence, = variable
type is available or for rock type other than the eight specific parameter (stress), u = minimum value of the variable
ones. below which the probability of occurrence is zero. a is the
Singh and Singh (2005) also presented a parabolic scale parameter and m is the shape parameter representing
strength criterion for intact rock. The strength criterion is the steepness of the plot.
also based on critical state concept (Barton 1976) as: The Weibulls probability distribution is converted into
linear function form as:
1 A 32 2 A 3 ci ci 3 for 0 3 ci (13) 1 (17)
ln ln m ln u m ln a
for 3 ci
1 A ci 3
2 (14) 1 p
The value of u was assumed from dataset and parameter m
ci

A is the criterion parameter, which can be estimated from ci

and a were obtained by fitting a straight line to Equation
as:
17. Several trials were made to fix the value of u such that
A 3.97 ci1.10 (15) best coefficient of correlation is achieved.
Where ci is uniaxial compressive strength of intact rock in The value of variable parameter corresponding to 50% of
MPa within a range of 7-500MPa. occurrence is considered as representative value of
The above mentioned two strength criteria for intact rock parameter. The rockmass strength corresponding to 50% or
have been used to simulate five triaxial data point. The major 0.5 probability of occurrence is designated as representative
value rockmass strength (cmass). Figure 1 shows the typical 2 [] RMR 10
Weibulls distribution fitted for the calculation of uniaxial Emass 10 (GPa)
40

strength of rockmass. Table 5 present the representative For RMR < 50

value of cmass for all the rockmass. 3 [] Emass = Ei * RF (%)
Where,
Table 4 Various uniaxial rockmass strength criterion RMR

RF (%) = 0.0028( RMR) 0.9e

2 22.82
considered for present study
Sl. Reference Criteria 4 [] RMR 25

No. Emass 10 40
(GPa)
Based on RMR classification 5 [] RMR
7.65 RMR 100 Ei 0.5 1 cos
100
Emass
1 [] cm
e 100
GPa
ci 6 [] 3
RMR
2 [] cm RMR 100 Emass 0.1
e 18.75 10 GPa
ci RMR 100
7 []
3 [] cm 0.5 e0.06 RMR MPa Emass Ei e 17.40

Based on RQD Classification

RMR 100
4 [] cm 8 [] Emass Ei 100.0186 RQD 1.91
e 24
ci Based on Q value classification
RMR 100
5 [] cm 9 [] 25 log Q
e 20 For Q > 1
ci 1
10 [] Q ci 3
6 [] cm RMR Emass 10
100
ci RMR 6(100 RMR )
11 [] Emass Ei e0.8625Q2.875

Based on RQD Classification Based on GSI

cm 12 [] ci GSI 10
7 [] Emass 10 40
100.013RQD1.34 100
ci
For ci < 100MPa
Based on Q value classification
GSI 10
8 [] cm = 7fcQ1/3 (MPa) Emass 10 40

9 [] cm = 7Q1/3 (MPa) For ci > 100MPa

13 [] D
1
10
2 2
10 [] 1 Emass 75 25 D GSI
cm = 5 Qc (MPa) 1 e 11
3
GPa
100
14 [] D
Based on GSI 1
Emass Ei 0.02 2
60 15 D GSI
1 e 11
11 [] GPa

12 []

Table 5 Various Rock Mass Modulus criterion considered

for present study
Sl. Reference Criteria
No.
Based on RMR classification
1 [] Emass 2RMR 100
(GPa)
For RMR > 50
Similar to the uniaxial compressive strength of rockmass,
the elastic modulus (Emass) of the rock mass is also
calculated using fourteen equations presented in Table 4.
The values of elastic modulus obtained from various
equations using different classification system is very
scatter. To obtain a single representative value of Emass,
Weibulls probability distribution as explained earlier is
used. The value of Emass corresponding to 50% of
occurrence is considered as representative value of the
parameter. For some rockmass data elastic modulus of intact
rock (Ei) is not provided. So only those equations which can
predict the elastic modulus of rockmass without considering
Ei have been used. Figure 2 shows the typical Weibulls
distributions fitted for calculation of elastic modulus of
rockmass. The value of Emass for the rockmass is presented
in Table 5.
For the GHB criterion only GSI classification system is
used. The rockmass strength (cmass) and elastic modulus of
rockmass (Emass) were calculated using equation which
consider the GSI classification only. Table 5 present the
value of both cmass_HB and Emass_HB calculated considering
only GSI classification system.
B. Calculation of Modified Mohr-Coulomb Parameters
The Sinio value as presented in Table 5 is used to calculate
the value of Sinjo using equation 2. The SRF value is
obtained using equation 3. Table 5 presents the parameters
obtained for MMC criterion for all the rockmass.
C. Using Generalised Hoek and Brown criterion
The Generalised Hoek and Brown (GHB) criterion
considers only GSI classification of the rockmass. The input
parameters are GSI value for rockmass, mi and c value for
the intact rock. The Roclab software is used to calculate the
parameters of GHB criterion i.e. mb, S and a value of the
rockmass. The mi and c value as presented in Appendix-II
and GSI value as presented in Table 1 are supplied to the
software for calculating the parameters of GHB criterion for
all the rockmass. Table 7 present the parameters obtained
for GHB criterion for all the rockmass.

Table 5 Parameters for the Modified Mohr-Coulomb Criterion, Generalised Hoek-Brown Criterion and equivalent Mohr
Coulomb parameters (for 0 < 3 < 10MPa)
Modified Mohr Coulomb criterion
DB ci cmass Sinio SRF Sinjo Emass cMMC MMC c_1
No. (MPa) (P50) (P50) (Mass) (MPa) () (MPa)
(MPa) (GPa)
1 93 6.825 0.7834 0.0734 0.8196 5.375 0.264 1.21 54.22 7.48
2 142 10.216 0.8046 0.0719 0.8346 5.556 0.36 1.63 56.05 10.69
4 62.3 6.163 0.7614 0.0989 0.8036 6.128 0.264 1.21 52.19 7.04
5 64.7 4.792 0.7801 0.0741 0.8173 3.645 0.336 0.94 53.61 5.71
11 43.85 3.935 0.7557 0.0897 0.8001 5.328 0.25 0.91 51.26 5.15
Generalised Hoek Brown Criterion
DB cmass_HB Emass_HB c mb s a cHB HB c_1
No. (MPa) (GPa) (Mass) (MPa) (MPa) () (MPa)
1 4.425 7.799 0.264 96.417 2.183 0.0025 0.508 2.23 41.31 9.85
2 6.757 7.943 0.36 146.94 2.327 0.0025 0.508 2.74 45.10 13.28
4 3.977 8.361 0.264 64.489 2.463 0.0043 0.505 2.08 38.97 8.72
5 2.276 4.791 0.336 67.314 1.932 0.0014 0.511 1.67 37.65 6.90
11 1.743 4.502 0.25 45.368 1.933 0.0018 0.509 1.49 34.42 5.66

confinement method. Three different analyses had been

D. Poissons ratio of Rockmass
The Poissons ratio for rockmass is considered as 1.2 times
of poisons ratio of intact rock. The rockmass for which
poisons ratio is not provided, a value of 0.25 is considered.
Table 6 and 7 also present the value of poisons ratio.
IV. ANALYSIS OF TUNNEL
For the present paper, stability analysis of an unsupported
circular tunnel has been done with the help of convergence
done in the present study. Firstly the GHB criterion had
been applied directly using the analytical solution given
Carranza-Torres and Fairhurst (1999). Secondly, the
analysis is done considering the equivalent Mohr-Coulomb
parameters obtained by linearization of GHB criterion
(EMC_GHB). Thirdly the analysis is done considering the
equivalent Mohr-Coulomb parameters obtained by
linearization of MMC criterion (EMC_MMC). For second
and Third case, the analytical solution given by Carranza-
Torres (2003) has been used which consider Mohr-Coulomb
Material.
pictorial representation of problem considered.
B. Analysis using GHB criterion directly
Carranza-Torres and Fairhurst (1999)[] presented an
analytical solution for elasto-plastic analysis of circular
tunnel subjected to uniform far-field stress. The strength
criterion considered is Hoek-Brown criterion. The details of
the method can be studied in detail in Carranza-Torres and
Fairhurst (1999). The analysis of tunnel is done for non-
associated case i.e. ( = 0) as well as for associated case (
= ) where is the dilatation angle. For the associated case
the value is considered as angle of internal friction
obtained from linearization of GHB criterion. The response
of tunnel in terms of radius of plastic zone (Rpl), tunnel
convergence (Ur) and tunnel convergence strain (Ur/a) for
both non-associated and associated cases is presented in
Table 9 (a) and (b).
using linearization in existing range, artificial range of
stresses and equating model response as stated by earlier
researchers. Jimenez et al. (2008) presented a technique for
linearization of Hoek and Brown Criteria to calculate the
equivalent cohesion and angle of internal friction by
applying a best fit in lambes p-q field.
The linearization of triaxial data of rockmass is done for the
minor principal stress in range of Support pressure applied
i.e. zero to Far field stress i.e. 10MPa. A best fit line
A. Tunnel dimension and far field stress
The tunnel considered for the present study is of radius (a)
5m and subjected to far field stress (0) of 10MPa. Figure 3
shows the
C. Analysis using Linearization of GHB criterion
Hoek et al. (2002), Sofianos and Halakatevakis (2002),
Sofianos and Nomikos (2006) and Jimenez et al. (2008)
worked on the Hoek and Brown criteria for rockmass for
deducing the equivalent cohesion and angle of internal
friction. Earlier Hoek et al. (1997) stated that linearizing the
eight triaxial data for minor principal stress in range of zero
to 25% of UCS of intact rock can serve the purpose.
However the case is very different with tunnel as support
pressure plays an important role in development of plastic
zone around the opening. Artificial range of minor principal
stress was defined by Hoek et al. (2002) over which the
linearization has to be done. Sofianos and Halakatevakis
(2002) stated that in real field condition existing range of
stresses has effect on equivalent Mohr-Coulomb criteria.
Sofianos and Nomikos (2006) compared various values of c
and calculated
equation is obtained using excel spreadsheet. The value of
equivalent Mohr-Coulomb Criterion is calculated as:
c _ l (18)
c
2 K
K 1 (19)
sin 1
K 1
where, equation of best fit line is presented as:
1 K3 c _ l (20)

Table 9 Response of Tunnel considering various conditions of strength criterion

DB Generalized Hoek and Brown Equivalent Mohr-Coulomb Equivalent Mohr-Coulomb
No. (GHB) Corresponding to Generalized Corresponding to Modified Mohr-
Hoek-Brown (GHB) Criterion Coulomb (MMC) Criterion
Non Associated Case
Rpl(m) Ur (mm) Ur /a (%) Rpl(m) Ur (mm) Ur /a (%) Rpl(m) Ur (mm) Ur /a (%)
1 5.63 9.32 0.186 5.71 9.07 0.181 5.52 13.08 0.262
2 5.33 9.04 0.181 5.33 8.82 0.176 5.29 12.70 0.254
4 5.85 9.00 0.180 5.96 8.78 0.176 5.63 11.78 0.235
5 6.16 17.71 0.354 6.39 17.61 0.352 5.71 21.35 0.427
11 6.73 21.34 0.427 7.04 21.45 0.429 5.91 14.66 0.301
Associated Case
1 5.63 12.33 0.25 5.71 11.56 0.23 5.52 21.60 0.43
2 5.33 10.38 0.21 5.33 9.54 0.19 5.28 15.55 0.31
4 5.85 12.60 0.25 5.96 12.01 0.240 5.63 20.32 0.41
5 6.16 27.04 0.54 6.39 27.51 0.55 5.71 42.79 0.85
11 6.73 37.33 0.75 7.04 39.30 0.79 5.90 33.59 0.67

Table 7 presents the equivalent c and values obtained the analytical solution given by Carranza-Torres (2003) for
from linearization of triaxial data for the minor principal Mohr-Coulomb material. The detail of the method can be
stress in the range of 0 to 10MPa for GHB criteria. The studied in literature itself. The analysis has been done using
Analysis of the tunnel case considered has been done using Excel spreadsheet. The analysis of tunnel is done for non-
associated case i.e. ( = 0) as well as for associated case (
= ). The response of tunnel in term of radius of plastic
zone (Rpl), tunnel convergence (Ur) and tunnel convergence
strain (Ur/a) for both non-associated and associated cases is
presented in Table 9.
D. Analysis using Linearization of MMC criterion
Table 6 presents the equivalent c and values obtained
from linearization of triaxial data for the minor principal
stress in the range of 0 to 10MPa for MMC criteria. The
Analysis of the tunnel case has been done using the
analytical solution given by Carranza-Torres (2003) for
Mohr-Coulomb material. The analysis is done using Excel
spreadsheet. The analysis of tunnel is done for non-
associated case i.e. ( = 0) as well as for associated case (
= ). The response of tunnel in term of radius of plastic zone
(Rpl), tunnel convergence (Ur) and tunnel convergence strain
(Ur/a) for both non-associated and associated cases is
presented in Table 9.
V. DISCUSSIONS
A. Ratio of equivalent cohesion and angle of internal
friction
Table 8 present the ratio of equivalent cohesion and angle of
internal friction obtain by linearization of GHB and MMC
criterion for all the rockmass considered. The ratio of the
equivalent cohesion of rockmass obtained by considering
MMC criterion to that of cohesion obtained by considering
GHB criterion is 0.0.57. While the ratio of equivalent angle
of internal friction is 1.36.
Table 8 Ratio of equivalent Mohr-Coulomb Parameters
considering MMC and GHB Criterion
DB No. CMMC/CHB MMC/HB
1 0.54 1.31
2 0.59 1.24
4 0.58 1.34
5 0.56 1.42
11 0.61 1.49
Average 0.57 1.36
B. Radius of Plastic zone
Figure 4 shows the radius of plastic zone for various
rockmass cases. It can be clearly interpreted that for the
MMC criterion underestimate the radius of plastic zone as
compared with that of GHB criterion. The underestimation
of radius of plastic zone is very significant or high in the
case of very poor rockmass for all the cases. However for
poor rockmass, the underestimation is very small except for
the two case of DB No. 11 and 16. For the fair rockmass, the
radius of plastic zone predicted by MMC criterion is lower
but close to that predicted by GHB criterion
C. Tunnel convergence Strain
Figure 5 and 6 shows the tunnel convergence strain
calculate for various rockmass cases for associated and non-
associated case. The tunnel convergence strain predicted by
MMC criterion is significantly underestimated as compared
to that predicted by GHB criterion for both non associated
and associated case. For some cases (DB No. 15 and 18), the
underestimation is very large. It may be because the
rockmass for these two cases are very poor as well as the
uniaxial compressive strength of intact rock is very near to
far field stress. For the case of poor and fair rockmass, the
tunnel convergence strain predicted by MMC criterion is
comparable to that predicted by GHB criterion except for
one rockmass case (DB No. 16). It may be because the Q
value for the rockmass is on higher side while the GSI value
is in lower side.
VI. CONCLUDING REMARKS
The present paper attempts to predict the value of triaxial
strength of rockmass as well as an analysis of hypothetical
tunnel case using Modified Mohr-Coulomb criterion. The
rockmass data from twenty different sites were collected
from literature. Weibulls probability distribution is used for
determining a unique representative value of rockmass
uniaxial strength and its elastic modulus. The triaxial
strength of rockmass predicted by both GHB and MMC are
compared based on equivalent Mohr-Coulomb (EMC)
parameters. The MMC criterion predicts lower value of
cohesion and higher value of angle of internal friction as
compared to GHB criterion. The analysis of an unsupported
case of tunnel has been carried out using the three different
approaches. The analysis showed that for MMC criterion
predict very less radius of plastic zone compared to GHB
criterion for very poor rockmass. For poor and fair
rockmass, the radius of plastic zone is comparable.
The tunnel convergence strain predicted for both non
associated and associated case, the MMC criterion predict
lower value as compared to GHB for very poor rockmass.
For Poor and fair rockmass, the tunnel convergence strain is
quite close to each other. Squeezing condition in rockmass
as predicted by MMC condition is similar to that predicted
by GHB criterion except for few cases.
The analysis of tunnel presented here above using MMC
criterion consider a number of equation presented for
estimation of strength, elastic modulus. These equations
were presented by a number of authors with their vast
experience in field. The analysis also takes into account of
various rock classification systems. The incorporation of
probabilistic approach gives a much higher sense of
confidence.
Acknowledgement
The authors are grateful to Prof. Mahendra Singh,
Department of Civil Engineering, IIT Roorkee for
explaining Modified Mohr Coulomb Criterion in detail and
suggestions regarding it.
REFERENCES
[1] A. Heim ,Untersuchungen ber den Mechanismus der
Gebirgsbildung. vol. 2, Basel, 1878.
[2] K.Terzaghi and F.E. Richart, Stresses in rock about cavities.
Geotechnique, vol. 3, pp 57-90, 1952.
[3] Wegener Alfred, The Origin of Continents and Oceans, Book
published in 1915.
[4] N. Hast, The measurement of rock pressure in mines, Sver. Geol.
Under. Ser. C. vol. 52, pp 1-152, 1958.
[5] J.F. Dewey, Plate tectonics, Scientific American, vol. 226, pp. 56-
68. 1972.
[6] D.P. Mckenzie and J.C. Sclater, The evolution of the Indian Ocean,
Sci. Amer. Vol. 228, no.5, pp. 63-72, 1973.
[7] V. Courtillot and G.E. Vink, How continents breakup, Sci. Am.,
vol. 249, pp. 42-49, 1983.
[8] E.T. Brown and E. Hoek, Trends in relation-ships between measured
rock in situ stresses and depth, Int. J. Rock Mech. Min. Sci. &
Geomech. Abstr. Vol.15, pp.211-215, 1978.
[9] G. Herget, Stresses in rock, Rotterdam: Balkema, 1988.
[10] M. L. Zoback, Stress field constraints on intraplate seismicity in
eastern North America, J. Geophys. Res.vol. 97, pp. 1176111782,
1992.
[11] E. Hoek, Structurally controlled instability in underground
excavations, Proc. 19th Rock Mechanics Symposium, Keystone,
Colorado, 1977.
[12] P. Croney, T.F. Legge and A. Dhalla, Location of block release
mechanisms in tunnels from geological data and the design of
associated support, Computer Methods in Tunnels Design. The
Institution of Civil Engineers, London, pp. 97-119, 1978.
[13] P.R. Sheorey, A theory for in situ stresses in isotropic and
transversely isotropic rock, Int. J. Rock Mech. Min. Sci. &
Geomech. Abstr. Vol. 31, no.1, pp. 23-34, 1994.
[14] Hari Dev, Rajbal Singh and Santosh K. Sati, Orientation of Large
Underground Caverns based on In-Situ Stresses and Wedge
Analysis, Indorock-2013: Fifth Indian Rock Conference, 29-31 May
2013, Solan, India, 2013.
[15] B.C. Haimson and C. Fairhurst, Initiation and extension of hydraulic
fractures in rocks, Society of Petroleum Engineers Journal, Sept. pp.
310-318, 1967.
[16] ISRM, International Society for Rock Mechanics (ISRM) suggested
methods for rock characterization, testing and monitoring, Editor:
E.T. Brown, Published by Pergamon press, pp. 1-211, 1981.
[17] CSMRS, Report on Hydrofracturing tests in Desilting Chamber of
Punatsangchhu-II H.E. Project, Bhutan, June 2013.
CSMRS, Report on Hydrofracturing tests in Downstream Surge
Chamber of Punatsangchhu-II H.E. Project, Bhutan, March 2013