International Journal of Mining Science and Technology xxx (2018) xxx–xxx

Contents lists available at ScienceDirect

International Journal of Mining Science and Technology

journal homepage: www.elsevier.com/locate/ijmst

Prediction of overbreak depth in Ghalaje road tunnel using strength

factor
Ako Daraei ⇑, Shokrollah Zare
Civil Engineering Department, Soran University, Erbil 44008, Iraq
School of Mining, Petroleum & Geophysics Engineering, Shahrood University of Technology, Shahrood 3619995161, Iran

a r t i c l e i n f o a b s t r a c t

Article history: It is well known that the overbreak caused by the blasting damage during tunnel excavation increases
Received 26 January 2018 costs associated with filling the collapsed area with shotcrete and results in filing of a claim by the con-
Received in revised form 28 March 2018 tractor. This paper outlines a new approach for prediction the overbreak depth during tunnel construc-
Accepted 11 April 2018
tion. Hence, firstly excavation damage zone (EDZ) are determined by average specific charge in each
Available online xxxx
zone. Numerical modelling is used to simulate the EDZ around tunnel boundary and the overbreak depth
are calculated by the rock strength factor. The predicted overbreak depth compared with observed field
Keywords:
data from a case study. The results show that there exists an approximately up to 40% difference between
Overbreak
Blasting
the prediction and the observed volume of overbreak depth. Therefore, the method can be well used to
Excavation damaged zone predict the overbreak depth to estimate more precision of shotcrete and concrete volumes in tunnelling
Strength factor cost during design phase.
Tunneling Ó 2018 Published by Elsevier B.V. on behalf of China University of Mining & Technology. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction
The presence of blast induced damaged zone around excava-
tions has been an important concern during tunneling. The term
excavation damaged zone (EDZ) is taken to mean the disturbed
zone that includes the failed and damaged zones closest to the wall
that are caused by the excavation method. The damaged zone
around excavation is dependent upon the strength and deforma-
tion modulus of the rock mass, in situ rock stress, nature and prop-
erties of the discontinuity network, geometry of the excavation and
excavation method [1]. According to Martino, the EDZ are divided
into two main components: disturbed and damaged. In the dis-
turbed zone only the stresses are altered while in the damaged
zone the mechanical, hydraulic and physical properties of the rock
mass are irreversibly changed or altered [2]. Overbreak in tunnel-
ing is defined as a part of damaged zone beyond the profile calcu-
lated for the support elements that falling during rock excavation.
It increases total cost of tunneling to a level of 15% or even more of
the scheduled cost of construction [3]. Many studies have been
performed in order to identify the causes of overbreak during tun-
neling, but, approximately all approaches based on peak particle
⇑ Corresponding author at: Civil Engineering Department, Soran University, Erbil
44008, Iraq.
E-mail addresses: rahman.daraei@soran.edu.iq (A. Daraei), zare@shahroodut.ac.
ir (S. Zare).
velocity (PPV) method that this method is needed to identify wave
and other aspect of seismology. Mohammadi et al. studied mini-
mizing overbreak in underground blasting operations by using
fuzzy set theory [4]. The effects of rock mass quality on overbreak
were investigated by Innaurato et al. [5]. Ibarra et al. studied the
influence of geological conditions and blasting factors on overbreak
and underbreak in tunnel [6]. Mandal et al. studied evaluating
extent and causes of overbreak in tunnels [3]. Using two-
dimensional analytical approach, Mahtab et al. assessed the
overbreak during tunnel design [7]. Miao et al. studied geological
environment and mining conditions of an excavation disturbed
zone in coal mining [8]. Perras and Diederichs studied EDZ depths
in brittle rocks and their results showed numerical limits can be
used for preliminary depth prediction of the EDZs for circular exca-
vations [9]. Using regression analysis, Dey and Murthy established
a relationship between the overbreak percentage and rock mass as
well as charge and blast design parameters [10]. Yang et al. studied
the failure behavior in both inner and outer zones around a circular
opening in a non-persistently jointed rock mass under biaxial com-
pression [11]. The overbreak depth quantity deserves more studies
due to its crucial role in project planning of tunnelling, particularly,
exact estimation of filling volume of damaged zone. In this paper,
firstly EDZ is determined by average specific charge in each zone.
Numerical modelling is used to simulate the EDZ around tunnel
boundary. Rock mass properties in these zone have been assigned
based on Kwon et al. theory, and blast damage factor and the

https://doi.org/10.1016/j.ijmst.2018.04.013
2095-2686/Ó 2018 Published by Elsevier B.V. on behalf of China University of Mining & Technology.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article in press as: Daraei A, Zare S. Prediction of overbreak depth in Ghalaje road tunnel using strength factor. Int J Min Sci Technol (2018),
https://doi.org/10.1016/j.ijmst.2018.04.013
2 A. Daraei, S. Zare / International Journal of Mining Science and Technology xxx (2018) xxx–xxx

overbreak depth are calculated by the strength factor [12,13]. The
predicted overbreak depth compared with observed field data from
a case study. The results show that there exists an approximately
up to 40% difference between the prediction and the observed vol-
ume of overbreak depth. Therefore, the method can be well used to
predict the overbreak depth. To do so, the result of 538 blasting
rounds in Ghalaje road tunnel in west of Iran are studied.
2. Project description
The Ghalaje road tunnel is located in west of Iran on the Eslam
Abad–Eyvan (between Kermanshah and Ilam provinces) main road
with a length of 2500 m, width of 12.2 and height of 8.3 m, with a
cross section of 98 m2. Construction of the tunnel will reduce the
length of this main road more than 12 km, which indirectly facili-
tates all travel to Iraq and will increase the road safety. Fig. 1 rep-
resents the Ghalaje road tunnel location. Due to high strength
properties of surrounding rocks in the tunnel path, the blasting
method has applied to excavate the tunnel.
2.1. Geological and geotechnical investigations
The region is located in the folded sedimentary formations of
Zagros belt, consisting of numerous anticlines and formed syn-
clines. The tunnel path is located in Asmari formation belonging
to the Cretaceous-Eosin period, comprised of massive, thick bed-
ded and jointed varieties of limestone. According to the engineer-
ing geological data and rating scores inferred from surface and
underground mapping, the host rocks were divided into three
zones: zones I, II and III at the construction stage as shown in Fig. 2.
Zone I consists of limestone with some clay interbedding with a
thickness of about 50 cm in some locations. Zone II, assigned to the
most competent part of the host rock, and consists of limestone
and dolomitic limestone. Lastly, limestone with clay and marl
interbedding constitutes the host rocks of Zone III. Main lithology
in the tunnel face is shown in Fig. 3.
The rock mass quality has been scored using the RMR classifica-
tion system. Detailed information of rock mass rating and proper-
ties of rock mass are demonstrated in Table 1. The cohesive,

Fig. 1. A close up view of Ghalaje road tunnel.

Fig. 2. Longitudinal section of Ghalaje tunnel.

Fig. 3. Main lithology in the tunnel face.

Please cite this article in press as: Daraei A, Zare S. Prediction of overbreak depth in Ghalaje road tunnel using strength factor. Int J Min Sci Technol (2018),
https://doi.org/10.1016/j.ijmst.2018.04.013
 A. Daraei, S. Zare / International Journal of Mining Science and Technology xxx (2018) xxx–xxx 3

Table 1
Geological zones properties.

Zone scm (MPa) st (MPa) Em (GPa) Er (GPa) Friction angle (°) Cohesion (MPa) RMR GSI
I 6.8 0.14 5.7 24 35 1.25 52–55 47–50
II 14 0.32 9.0 33 40 1.9 55–60 50–55
III 10.8 0.23 7.0 30 39 1.5 43–52 38–47

Note: Er and Em are the deformation modulus of intact rock and rock mass, respectively; and scm is the uniaxial compression strength of rock mass; and st is uniaxial tensile
strength of rock mass.

Table 2
Joint sets properties in the tunnel path.

Parameter Zone I Zone II Zone III

Dip/dip direction 72/302 67/224 70/178 30/040 60/223 76/ 306 72/182 27/041 78/215 80/115 82/032 17/180
Type Js Js Js B Js Js Js B Js Js Js B
Spacing (cm) 40–80 50–100 50–90 80–120 50–100 40–100 50–90 80–120 50–100 40–80 80–120 80–120
Persistence (m) 3–10 3–8 1–6 >10 3–10 3–8 1–6 >10 >8 3–8 1–6 >10

35 mm in diameter as primer explosives and using the electric ini-

tiation system, with 1.5–3 m blast hole length, while blast holes
are 45 mm in diameter. Depth of round and holes per round are
two important factors in blasting which directly affect the excava-
tion unit price. Description relevant to this issue in the tunnel will
be given in the following subtopics.

3.1. Depth of round

At first sight, increasing the tunnel advance per round will be

Fig. 4. Stereographic projection of discontinuties.
friction angle and deformation modulus of the intact rock have
been determined by laboratory tests and rock mass properties
instead of overpriced and time consuming in situ tests, estimated
by RocData Software which have been developed based on the
Hoek–Brown criterion.
During surface and underground geological mapping, four joint
sets were mapped in the tunnel path, in each zone. Joint set prop-
erties in each zones are presented in Table 2.
Based on surveying along the tunnel, joints have different per-
sistence of almost 3–10 m and their spacing is between 40 and
100 cm. In order to determine the most important discontinuities
for joint network modeling, pole concentration of joint sets were
drawn in dips software. Plot of pole concentration frequency
shown two main joint sets and a random joint set in tunnel district
such 040/27 and 201/75, as seen in Fig. 4.
achieved by applying a greater number of holes with longer length
and wider diameter. Consequently, this will increase the specific
charge, but on the other hand such an approach will cause unfavor-
able effects such as fly rocks, ground and air vibration [15]. The
depth of a round is an important parameter in tunnel blasting, as
most of the excavation engineers desire a higher tunneling rate.
A rough guideline on the length of the blast holes in the cut and
easer holes of an angle cut was provided by Pokrovsky [16]. Con-
sidering the tunnel area in the top heading equal to 59 m2, the
cut hole would be 5.7 m.
For cut holes, l = 0.75(A)0.5 = 0.75(59)0.5 = 5.7 m
For easer holes, le = 0.5(A)0.5 = 0.5(59)0.5 = 3.8 m
where A is the tunnel area, m2; lc the length of cut hole, m; and
le the length of easer hole, m.
Considering the drilling machine specifications and crushed
zone, the blasting pattern in our case was executed with cut holes
and easer holes with a length between 1.5 and 3 m, in order to
minimize unfavorable effects of blasting waves. Table 3 represents
the number of blasting round in each zone.
3.2. Holes per round

3. Blasting design in different zones of the tunnel path The number of holes per round is decided mainly by the tunnel
size and hole diameter. Ziegler reports that the number of holes per
One of the most important factors influencing on a tunnel blast round in a drift is reduced by 3% with every 0.001 m increase in
efficiency is the proper design of blasting pattern. Among blasting
parameters, blast hole diameter and tunnel face area are more sig- Table 3
nificant so that any change in these parameters could finally effect Number of blasting round in the top heading at different zones.
on specific charge and specific drilling [14]. In blast design, increas- Zone Number of blasting Chainage Length (m)
ing the face advance in order to minimize the cost is considered as round
the main target. In other words, other consequences of the blasting From To
procedure such as drilling cost and time usually depend on the face I 127 2 + 100 1 + 420 680
advance. The drill pattern selected is based on the size of the face, II 252 0 + 650 1 + 420 770
the type of material being drilled and the blasting geometry. The 2 + 100 2 + 620 520
III 159 0 + 520 0 + 650 130
basic characteristics of drilling and blasting in this tunnel are as
2 + 620 3 + 003 383
follows: blasting is done using ANFO as main charge and Emulite

Please cite this article in press as: Daraei A, Zare S. Prediction of overbreak depth in Ghalaje road tunnel using strength factor. Int J Min Sci Technol (2018),
https://doi.org/10.1016/j.ijmst.2018.04.013
4 A. Daraei, S. Zare / International Journal of Mining Science and Technology xxx (2018) xxx–xxx

Fig. 5. Typical blasting pattern.

Fig. 6. Relationship between specific charge and advance per round at zones I, II and III.

diameter of the explosive cartridge, and the hole concentration is Table 4

greater in small openings compared to large ones [17]. In this case, Excavation damaged zone depth.

wedge cut is used in all rock categories and drilling patterns for the Zone Average specific charge (kg/m3) Q (kg/m) EDZ (m)
tunnel containing 102 blast holes on average, as shown in Fig. 5. I 2.92 1.32 2.79
II 2.53 1.40 2.99
III 2.53 1.39 2.96
3.3. Specific charge

One of the main challenges involved in blasting engineering is

the effect of specific charge on the amounts of EDZ. In order to
determine EDZ, it is necessary to obtain the specific charge in each
zone. In all zones which has been excavated with 127,252 and 159
different blast rounds, respectively, specific charge and advance
were recorded. According to Fig. 6, results show that in order to
reach 3 m advance per round, specific charge was between 1 and
4 kg/m3. Also, it can be understood from the graph that the amount
of specific charge and increasing advance per round, reduced expo-
nentially. In zone I the difference in clay and bed rocks strength
causes heterogeneity in media and consequently reducing the
advance per round because of damping gas pressure. Regarding
the Ghalaje tunnel area in the top heading cross section (59 m2),
specific charge was obtained as 2.92, 2.53 and 2.53 kg/m3 on aver-
age in zones I, II and III respectively.
4. Determination of EDZ
The EDZ will be defined as the zone beyond the final excavation
boundary, where the rock has been damaged by blasting. The
dimensions of EDZ depend on the size of explosive charge deto-
nated, rock’s dynamic strength and density, wave velocity propa-
gation, and vibration velocities transmitted to the rock mass [18].
In general, the factors influencing the magnitude of that can conve-
niently be grouped into two categories, which are geological fac-
tors and blasting factors. One of the most common for
calculating EDZ depth is empirical Eq. (1) proposed by Pusch and
Stanfors based on charge concentration per meter in blast hole
[19]. Therefore, based on specific charge, the charge concentration
was calculated, then the EDZ depth that are illustrated in Table 4
was determined.
EDZ ¼ 2  Q 1:2 ð1Þ
where Q is charge concentration, kg/m.
5. Numerical modeling
The two dimensional software program Phase2 was used to pre-
dict overbreak depth at the tunnel. The model dimensions was
considered as 110 m  106 m. In order to state strength parame-
ters, generalized Hoek-Brown failure criterion was applied [20].
2D numerical models that were developed for the purposes of this
analysis can be seen within Fig. 7. The stresses were applied on
model in a gravity form, and vertical to horizontal ratio coefficient
K was determined based on Rummel Equations [21,22]. The
amount of stress relaxation in models for considering the three-
dimensional effect, and also expressing stress relief due to excava-
tion step in the model, was considered using the Panet Equation
and the ground reaction curve [23,24]. According to study by Kwon
et al., in relation to the blast damage effect on rock mass
properties, deformation modulus up to 56% was reduced during

Please cite this article in press as: Daraei A, Zare S. Prediction of overbreak depth in Ghalaje road tunnel using strength factor. Int J Min Sci Technol (2018),
https://doi.org/10.1016/j.ijmst.2018.04.013
 A. Daraei, S. Zare / International Journal of Mining Science and Technology xxx (2018) xxx–xxx 5

where Jkn is the joint normal stiffness; Jks the joint shear stiffness;
and Gr and Gm the shear modulus of intact rock and rock mass,
respectively; and S the average joint spacing.
After modeling, the values of strength factor at tunnel crown
and walls in each zones were obtained according to Fig. 8. The
strength factor represents the ratio of material strength to induced
stress, at a given point. If the strength factor is less than 1, this indi-
cates that the stress in the material exceeds the material strength.
Hence, where the depth with strength factor lower than 1, were
considered as predicted overbreak depth. Given Eq. (1) value and
identifying it on the strength factor graphs, the distributed area
safety factor were determined as 1.2, (1.9,1.7 and 1.55) and 1.2,
Fig. 7. Tunnel model in Phase 2.
at tunnel crown and walls in zone I, II and III, respectively.

the modeling of EDZ [12]. To quantify damage in the excavation

walls, Hoek introduced the blast damage factor D and incorporated
it into Hoek-Brown’s failure criterion [13]. Since rock mass in a
blast damaged zone becomes more jointed, the GSI value for this
part of rock mass should be lower from the adjacent undisturbed
rock mass. Hence, the GSI value of disturbed rock mass is being
decreased for 10 points and distribution factor was used 0.8 in
EDZ. (If blasting is poorly done, it is suggested that D = 0.8.).
Cross-jointed network was used to establish the discontinuities
modelling in Phase 2 software. Also using Eqs. (2) and (3) normal
and shear stiffness of discontinuity in each zone was calculated
that results are illustrated in Table 5.
Er  Em
Jkn ¼
SðEr  Em Þ
Gr  Gm
Jks ¼
SðGr  Gm Þ
6. Discussion
In order to evaluate the strength factor precision relevant to
overbreak depth in tunnel blasting, the results compared with
observed data that have been surveyed during tunnel construction
after each blasting round. Based on survey of profiles after each
blasting round, the observed depth of the overbreak profiles was
obtained according to Fig. 9.
The difference between predicted and observed overbreak
depth are shown in Table 6. The results show that there exists an
approximately up to 40% difference between the prediction and
the observed overbreak depth. Therefore, the method can be well
ð2Þ used to predict the overbreak depth. That is to say, in some parts
of each zone, wedge failure was occurred, but they were local,
hence the area wasn’t considered in the calculation.
ð3Þ

Table 6
Table 5 Predicted and observed overbreak depths.
Normal and shear stiffness of discontinuities.
Zone Aver. predicted overbreak Aver. observed overbreak Difference
Zone Jkn (GPa/m) Jks (GPa/m) Aver. spacing (m) depth (m) depth (m) (%)
I 9.96 3.98 0.75 I 0 0.4 40
II 16.5 6.60 0.75 II 0.60 0.6 0
III 12.1 4.86 0.75 III 0.77 0.7 9

Fig. 8. Strength factor in crown and walls.

Fig. 9. Observed overbreak depths in zone I, II and III.

Please cite this article in press as: Daraei A, Zare S. Prediction of overbreak depth in Ghalaje road tunnel using strength factor. Int J Min Sci Technol (2018),
https://doi.org/10.1016/j.ijmst.2018.04.013
6 A. Daraei, S. Zare / International Journal of Mining Science and Technology xxx (2018) xxx–xxx
7. Conclusions
Based on the results and discussion presented in this paper, the
following conclusions can be drawn:
(1) Based on Fig. 6, the findings showed that with an increase in
face advance, the specific charge decreased exponentially
per round to a limited threshold and then tended towards
a constant value.
(2) Increasing the face advance per round will be obtained by a
nominal decrease in specific charge.
(3) In Ghalaje tunnel, the overbreak area was measured before
and after each blasting round in different zones, and can
be compared as III > II > I. It can be inferred that the exis-
tence of the interbedding layer which is located mainly in
the crown of the tunnel causes a lower degree of damage
beyond the designed counter blast holes. Thus, the over-
break area in zone I is lower than the other zones.
(4) According to results, strength factor can be well approached
to predict the overbreak depth to estimate more precision of
shotcrete and concrete volumes in the tunnelling cost during
design phase.
References
[1] Emsley S, Olsson O, Stenberg L, Alheid H, Falls S. ZEDEX-a study of damage and
disturbance from tunnel excavation by blasting and tunnel boring. Svensk
Kärnbränslehantering AB/Swedish Nuclear Fuel and Waste Management
Co: Stockholm; 1997.
[2] Martino JB. The excavation damaged zone and recent studies at the URL. In:
Proceedings of the international EDZ workshop, Toronto, 2003.
[3] Mandal S, Singh M. Evaluating extent and causes of overbreak in tunnels.
Tunnel Undergr Space Technol 2009;24(1):22–36.
[4] Mohammadi M, Hossaini MF, Mirzapour B, Hajiantilaki N. Use of fuzzy set
theory for minimizing overbreak in underground blasting operations–a case
study of Alborz tunnel, Iran. Int J Min Sci Technol 2015;25(3):439–45.
[5] Innaurato N, Mancini R, Cardu M. On the influence of rock mass quality on the
quality of blasting work in tunnel driving. Tunnel Undergr Space Technol
1998;13(1):81–9.
Please cite this article in press as: Daraei A, Zare S. Prediction of overbreak depth in Ghalaje road tunnel using strength factor. Int J Min Sci Technol (2018),
https://doi.org/10.1016/j.ijmst.2018.04.013
[6] Ibarra JA, Maerz NH, Franklin JA. Overbreak and underbreak in underground
openings Part 2: and implications. Geotech Geological Eng 1996;14:325–40.
[7] Mahtab MA, Rossler K, Kalamaras S, Grasso P. Assessment of geological
overbreak for tunnel design and contractual claims. Int J Rock Mech Min Sci
1997;34(3–4):1–13.
[8] Miao SJ, Lai XP, Cui F. Top coal flows in an excavation disturbed zone of high
section top coal caving of an extremely steep and thick seam. Min Sci Technol
(China) 2011;1:99–105.
[9] Perrasa M, Diederichsa M. Predicting excavation damage zone depths in brittle
rocks. J Rock Mech Geotech Eng 2016;8(1):60–74.
[10] Dey K, Murthy V. Prediction of blast-induced overbreak from uncontrolled
burn-cut blasting in tunnels driven through medium rock class. Tunnel
Undergr Space Technol 2012;28:49–56.
[11] Yang XX, Jing HW, Chen KF. Numerical simulations of failure behavior around
a circular opening in a non-persistently jointed rock mass under biaxial
compression. Int J Min Sci Technol 2016;26(4):729–38.
[12] Kwon S, Lee CS, Cho SJ, Jeon SW, Cho WJ. An investigation of the excavation
damaged zone at the KAERI underground research tunnel. Tunnel Undergr
Space Technol 2009;24:1–13.
[13] Hoek E, Blast Damage Factor D. Technical note for Rocknews available in
https://www.rocscience.com, 2012.
[14] Khalili S, Yavary M, Ebrahimabadi A. Trend analysis and comparison of basic
parameters for tunnel blast design models. Int J Min Sci Technol
2015;25:595–9.
[15] Olafsson O. Applied explosive technology for construction and
mining. Sweden: APPLEX; 1990.
[16] Pokrovsky N. Driving horizontal workings and tunnels. Moscow: Mir
Publishers; 1980.
[17] Ziegler K. Blasting techniques in coal mines of the Republic of Germany. Proc.
International Symposium on Mining Tech. & Sci.: China Coal Industry
Publishing House, Trans Tech Publication; 1985.
[18] Da Gama C. Prediction of EDZ from explosive detonation in underground
openings. ISRM International Symposium on Rock Engineering for
Mountainous Regions-EUROCK: Madeira; 2002. p. 25–7.
[19] Pusch R, Stanfors R. The zone of disturbance around blasted tunnels at depth.
Int J Rock Mech Min Sci 1992;29(5):447–56.
[20] Hoek E, Carranza-Torres C, Corkum B. Hoek-Brown failure criterion. NARMS-
TAC Conference, Toronto, 2002.
[21] Rummel F. Stresses and tectonics of the upper continental crust- a
review. Proc. Int. Symp. on Rock Stress and Rock Stress
Measurements: Stockholm, Centek Publ; 1986.
[22] Rummel F. Rock mechanics with emphasis on stress. A.A.BALKEMA Publisher;
2005.
[23] Panet M. Recommendations on the convergence-confinement method. AFTES;
2001.
[24] Carranza-Torres C, Fairhurst C. Application of the convergence-confinement
method of tunnel design to rock masses that satisfy the Hoek-Brown failure
criterion. Tunnel Undergr Space Technol 2000;15(2):187–213.