10 Rock Mass Classification Systems 1

Journal of Rock Mechanics and Geotechnical Engineering xxx (2017) 1e17
Contents lists available at ScienceDirect
Journal of Rock Mechanics and

Geotechnical Engineering
journal homepage: www.rockgeotech.org
Full Length Article
Application of rock mass classification systems to rock slope stability

assessment: A case study
Hassan Basahel*, Hani Mitri
Department of Mining and Materials Engineering, McGill University, Montreal, H3A 0E8, Canada
a r t i c l e i n f o a b s t r a c t
Article history: The stability of rock slopes is considered crucial to public safety in highways passing through rock cuts, as
Received 9 June 2017 well as to personnel and equipment safety in open pit mines. Slope instability and failures occur due to
Received in revised form many factors such as adverse slope geometries, geological discontinuities, weak or weathered slope
24 July 2017
materials as well as severe weather conditions. External loads like heavy precipitation and seismicity
Accepted 25 July 2017
Available online xxx
could play a significant role in slope failure. In this paper, several rock mass classification systems
developed for rock slope stability assessment are evaluated against known rock slope conditions in a
region of Saudi Arabia, where slopes located in rugged terrains with complex geometry serve as highway
Keywords:
Rock mass classification
road cuts. Selected empirical methods have been applied to 22 rock cuts that are selected based on their
Graphical slope mass rating failure mechanisms and slope materials. The stability conditions are identified, and the results of each
Continuous slope mass rating rock slope classification system are compared. The paper also highlights the limitations of the empirical
Rock slope stability classification methods used in the study and proposes future research directions.
Ó 2017 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting by
Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).
1. Introduction projection technique. This technique is used to project the orien-

tation of discontinuities by pole, containing information about the
Rock slopes in most road cuts, especially in mountainous areas, dip and dip direction of a joint on a two-dimensional (2D) stereonet
are prone to instability problems due to variation in the rock mass (Price and Cosgrove, 1990). Limit equilibrium method compares the
conditions and external factors induced by the environments such magnitudes of the driving and resisting forces that act along the
as seismic activities and water in the slope (Pantelidis, 2009). The sliding planes to estimate the factor of safety (Coggan et al., 1998),
material characteristics of a rock slope, the height, the face angle, and it is also widely used to examine slope structural stability.
and the rock joint orientations play a significant role in the insta- Numerical modelling is used in more complex slope geometries
bility problem of road cuts and slopes. In addition, the stability of and failure mechanisms. It is particularly useful when the above-
rock slopes may also be influenced by the road curvature, partic- mentioned methods cannot represent the behaviours of the
ularly in rugged terrains (Hoek and Bray, 1981). slope. Numerical modelling provides insight into the effect of stress
Slope stability problems attract major concerns from re- distribution in the slope and displacements on its behaviour
searchers, and consequently, several techniques and methods for (Wyllie and Mah, 2004). Rock mass classification systems or
slope stability evaluation have been proposed. These methods can empirical methods represent an important tool that is often used
be basically grouped into four categories, i.e. kinematic analysis, for preliminary assessment of the engineering behaviours of the
limit equilibrium, numerical modelling, and empirical methods. rock mass (Duran and Douglas, 2000). In this paper, the focus is put
Kinematic analysis is commonly used to predict potential structural on empirical methods, and the goal is to assess their efficiency for
failure mechanisms (planar, wedge, and toppling) using stereonet determination of the rock slope stability.
Rock mass classification systems have been commonly utilised
in the field of geotechnical engineering, especially for design pur-
* Corresponding author. pose. They are widely used due to their simplicity and the limited
E-mail address: Hassan.basahel@mail.mcgill.ca (H. Basahel). need for detailed information (Duran and Douglas, 2000). However,
Peer review under responsibility of Institute of Rock and Soil Mechanics, these classification systems are initially established for
Chinese Academy of Sciences.
https://doi.org/10.1016/j.jrmge.2017.07.007
1674-7755 Ó 2017 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting by Elsevier B.V. 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: Basahel H, Mitri H, Application of rock mass classification systems to rock slope stability assessment: A case
study, Journal of Rock Mechanics and Geotechnical Engineering (2017), https://doi.org/10.1016/j.jrmge.2017.07.007
2 H. Basahel, H. Mitri / Journal of Rock Mechanics and Geotechnical Engineering xxx (2017) 1e17
underground excavations (Hoek, 2007). They are developed based 2. Location of the study area
on the importance of parameters; each is assigned with a numerical
value and a weighting factor. By substituting these weighting The study area is located in the far south-western part of Saudi
values into an empirical formula, one can obtain the final rating for Arabia known as Jazan region, between latitudes 16 and 18 north,
a rock mass (Hack et al., 2003). Such rating is of value in the sense and longitudes 42 and 44 east as shown in Fig. 2a. 22 road cuts are
that it can help the engineering judgment of underground projects selected as case studies along five main roads, which are distrib-
for design purpose (Bieniawski, 1993). uted as follows: five sites along road No. 12, seven sites along road
Even though empirical rock mass systems are established for No. 8, four sites along Al-Hasher road, five sites along Ar-Raith road,
underground stability assessment, some of classification systems and one site along Al-Aydabi road (see Fig. 2a). The selected case
have been applied to the evaluation of rock slopes by calibrating studies are in an area with rugged terrains (Fig. 2b) and are selected
some parameters to make it applicable to surface excavations in based on the type of failure mechanism.
rock (Pantelidis, 2009). Rock mass characterisation and disconti-
nuity conditions represent the backbone of most empirical (1) Structurally controlled failures: different discontinuity ori-
methods. These conditions can be summarised into five categories, entations play a significant role in the instability condition of
i.e. unconfined compressive strength (UCS) of intact rock, rock a structurally controlled slope (Fig. 3). The selected sites are:
quality designation (RQD), spacing between discontinuities, 1, 2, 3, 4, 5, 6, 7, 10, 11, 12, 13, 14, 16, 7, 18, 19 and 22.
discontinuity condition, and groundwater condition. These five (2) Non-structurally controlled failure (stress-controlled): the
factors also refer to as the well-known Bieniawski’s rock mass stability condition of the rock slope is influenced mainly by
rating (RMR) (Bieniawski, 1973, 1976, 1979, 1989). external forces, e.g. groundwater. Also, the degree of
In this paper, five rock mass classification methods are chosen. weathering is high. In non-structurally controlled slopes, the
The selected methods are original slope mass rating (SMR) traces of rock joints are not obvious and have no or little
(Romana, 1985), Chinese SMR (Chen, 1995), continuous SMR effect on the stability behaviour of the slope (Fig. 4). The
(Tomás et al., 2007), graphical SMR (Tomás et al., 2012) and an selected sites are: 8, 9, 15, 20 and 21.
alternative for rock slopes (Pantelidis, 2010). The stability assess-
ment for rock cuts using the selected five empirical methods has
been conducted and their validity was examined. All results for rock 3. Classification systems for rock slopes
slope description and stability categories have also been compared.
The rock slopes of the selected case study are in rugged As previously mentioned, five methods are used and discussed
mountainous terrain with complex geological features. They have in this paper. Since all SMR methods are based on the scores of the
steep slope face angles and sharp curvatures along the road. They basic RMR system, all the case studies are first analysed with the
are subjected to rain storms in different periods of the year, mostly basic RMR value.
in summer and winter seasons. The average annual precipitation of
rainfalls is 500 mm (see Fig. 1). Due to heavy rainfall, most rock cuts, 3.1. Rock mass rating (RMR)
even slightly and moderately weathered, suffer from the stability
problems, thus different failure mechanisms may be encountered. RMR was established by Bieniawski (1973e1989) to evaluate the
In addition, the excavation method used to create the rock cuts quality of rock masses for underground projects. The RMR system
could play a role in the deterioration of the rock slopes. Poor consists of five basic parameters that represent different conditions
blasting methods were used prevalently in most road cuts, without of the rock and the discontinuities. These parameters are: (1) UCS of
indispensable rock support. intact rock, (2) RQD, (3) spacing between discontinuities, (4)
Fig. 1. Annual precipitation map of Saudi Arabian from 1960 to 2014, showing an average of 500 mm.
H. Basahel, H. Mitri / Journal of Rock Mechanics and Geotechnical Engineering xxx (2017) 1e17 3
Fig. 2. (a) Locations of road cuts in the study area, and (b) three-dimensional (3D) topographic map showing the roadway through rugged terrains.
introduced for tunnel and dam foundations but not for slopes
(Aksoy, 2008). Therefore, Bieniawski (1989) implemented more
descriptive details in the fourth parameter of the basic RMR (the
condition of discontinuities). Tables 1e3 show the classification
criteria of RMR and their different rock mass classes (Bieniawski,
1989). Also, in the case of considering the effect of discontinuity
orientation on the slope stability of a rock slope, he recommended
the use of the SMR system proposed by Romana (1985).
Bieniawski (1989) applied a set of discrete functions to the RMR
system, leading to the so-called discrete RMR (DRMR). This could
result in different users deriving different RMR scores, depending on
their own experience and selection of the discrete values. To address
this problem, Sen and Sadagah (2003) modified the RMR system by
converting the classical discrete functions into continuous rating
functions, leading to the continuous RMR (CRMR). In this modifi-
Fig. 3. Example of structurally controlled failure (site 22 in Fig. 1).
cation, the difference between RMR values estimated by different
users can be controlled within 10% (Sen and Sadagah, 2003).
In this paper, both the DRMR and the CRMR were applied to
structurally controlled road cuts, and the resulting scores were
compared. It should be noted that, in both DRMR and CRMR, only
the five parameters (UCS, RQD, spacing of discontinuities, condition
of discontinuities, and groundwater), which are presented by the
basic RMR, are used for rock mass classification of road cuts in the
study area.
3.2. Slope mass rating (SMR)
SMR is a classical lump-rating classification system for rock

slopes developed by Romana (1985). The SMR system is originally
derived from RMR system, where adjustment parameters that
represent the discontinuity orientations in relation to the slope
attitude are added to the basic RMR, as well as the effect of the
excavation method. The SMR score is obtained by subtracting a
Fig. 4. Example of non-structurally controlled failure due to highly weathering (site 21
in Fig. 1). factor from RMR depending on the joint-slope relationship, and
adding a factor depending on the method of excavation, as
expressed in the following equation:
condition of discontinuities, and (5) groundwater. This RMR system
is known as “the basic RMR” and it gives a value that ranges be- SMR ¼ RMR þ F1 F2 F3 þ F4 (1)
tween 0 and 100 (Bieniawski, 1973).
An additional parameter was proposed by Bieniawski (1976) to where F1 is an adjustment factor, which depends on the parallelism
account for the influence of the discontinuity orientation on the between the joint strike (aj ) (or the plunge direction of the inter-
stability condition (correction factor). However, this parameter is section line of two planes (ai Þ) and the slope face strike (as Þ. It
Table 1
Rock rating system (After Bieniawski, 1989).
Parameter Range of values
1 Strength of intact Point-load strength >10 4e10 2e4 1e2 For the low range, uniaxial
rock mineral index (MPa) compression test is preferred
UCS (MPa) >250 100e250 50e100 25e50 5e25 1e5 <1
Rating 15 12 7 4 2 1 0
2 Drill core RQD (%) 90e100 75e90 50e75 25e50 <25
Rating 20 17 13 8 3
3 Spacing of discontinuities >2 m 0.6e2 m 200-600 mm 60-200 mm <60 mm
Rating 20 15 10 8 5
4 Condition of discontinuities Very rough Slightly rough Slightly rough Slickensided Soft gouge >5 mm
(see Table 2) surfaces surfaces surfaces surfaces, or thick, or
Not continuous Separation Separation Gouge < 5 mm Separation > 5 mm
No separation <1 mm <1 mm thick, or (Continuous)
Unweathered Slightly weathered Highly weathered Separation 1e5 mm
wall rock walls walls (Continuous)
Rating 30 25 20 10 0
5 Groundwater Inflow per 10 m None <10 10e25 25e125 >125
tunnel length (L/min)
Ratio of joint water pressure 0 <0.1 0.1e0.2 0.2e0.5 >0.5
to major principal stress
General condition Completely dry Damp Wet Dripping Flowing
Rating 15 10 7 4 0
Table 2
Guidelines for classification of discontinuity condition in RMR.
Discontinuity length (persistence) Separation (aperture) Roughness Infilling (gouge) Weathering
Value (m) Rating Value (mm) Rating Description Rating Description Rating Description Rating
<1 6 None 6 Very rough 6 None 6 Unweathered 6

1e3 4 <0.1 5 Rough 5 Hard filling < 5 mm 4 Slightly weathered 5
3e10 2 0.1e1.0 4 Slightly rough 3 Hard filling > 5 mm 2 Moderately weathered 3
10e20 1 1e5 1 Smooth 1 Soft filling < 5 mm 2 Highly weathered 1
>20 0 >5 0 Slickensided 0 Soft filling > 5 mm 0 Decomposed 0
Table 3
Guidelines for classification of discontinuity condition in RMR. Table 5
Adjustment factor F4 for the method of excavation (Romana, 1985).
Rating Class Description
Method of excavation F4 value
100e81 I Very good rock
80e61 II Good rock Natural slope 15
60e41 III Fair rock Pre-splitting 10
40e21 IV Poor rock Smooth blasting 8
<20 V Very poor rock Normal blasting or mechanical excavation 0
Deficient blasting -8
ranges from 1, when the joint and the slope face strikes are near
slope height factor (z), and the discontinuity factor (l), as shown in
parallel, to 0.15, when the angle between strikes is 30 (see Table 4).
the following equation (Chen, 1995):
F2 refers to joint dip angle (bj Þ in the planar failure or the plunge
of the line of intersection of two planes (bi Þ in the wedge-type
CSMR ¼ zRMR þ lF1 F2 F3 þ F4 (2)
failure mode. It varies from 1 for joints dipping more than 45 to
0.15 for joints dipping less than 20 (see Table 4). where z is a non-dimensional parameter accounting for the effect
F3 reflects the effect of the angle between the slope face dip (bs ) of slope height, and is given by
and the joint dip (bj Þ or the plunge of the intersection line of two
planes (bi Þ. It ranges from 0 (very favourable) when bj bs or bi z ¼ 0:57 þ 0:43 80=H (3)
bs is more than 10 , to 60 (very unfavourable) when bj bs or bi
bs is less than 10 . where H is the slope height in meter. l is a parameter accounting for
F4 is an adjustment factor that depends on the excavation the effect of discontinuity type, and is defined as follows:
method. The values are selected empirically as shown in Table 5.
Planar-type and toppling-type failure modes are considered in (1) l ¼ 1 for faults of long weak seams filled with clay;
the SMR system. Anbalagan et al. (1992) added a wedge failure to (2) l ¼ 0.8e0.9 for bedding planes of large scale joints with
Romana’s classification. Wedge-type and planar-type failure modes gouge; and
will be addressed separately, where the inclination and the direc- (3) l ¼ 0.7 for joints of tightly interlocked bedding planes.
tion of the line of intersection of two planes are considered for
wedge failure analysis. This modified SMR (Anbalagan et al., 1992) Regarding the slope height, Eq. (3) is applicable to slope height
is used in this paper, and three types of the structurally controlled (H) > 80 m. For a slope height (H) 80 m, the parameter is equal to 1.
failures, i.e. planar, wedge, and toppling, are used. These are iden-
tified in Table 4 as ‘P’ for planar, ‘W’ for wedge and ‘T’ for toppling.
Different classes of SMR scores that describe the slope and its sta- 3.4. Continuous slope mass rating
bility condition are listed in Table 6.
The continuous SMR was proposed by Tomás et al. (2007). In
3.3. Chinese slope mass rating (CSMR) this system, continuous functions were proposed to replace the
lump-rating system adopted in Romana (1985)’s system. The
CSMR rating was proposed by Chen (1995), where two co- proposed F1, F2, and F3 continuous functions are expressed as
efficients were added to Romana’s system (SMR). These are the follows.
Table 4
Adjustment ratings for F1, F2, and F3 (Romana, 1985, modified by Anbalagan et al., 1992).
Case of slope failure Very favourable Favourable Fair Unfavourable Very unfavourable

P aj as >30 30 e20 20 e10 10 e5 <5

T aj as 180
W jai as j
P/T/W F 1 0.15 0.4 0.7 0.85 1
P bj <20 20 e30 30 e35 35 e45 >45
W jbi j
P/W F2 0.15 0.4 0.7 0.85 1
T F2 1 1 1 1 1
P bj bs >10 10 e0 0 0 to 10 <10
W bi bs
T bj þ bs <110 110 e120 >120 e e
P/T/W F3 0 6 25 50 60
Table 6 3.6. Hazard index (HI)

SMR classes according to Romana (1985).
Class SMR Description Stability Pantelidis (2010) proposed a quantitative system, in which the
I 81e100 Very Good Completely stable
hazard associated with rock slope failure is assessed by considering
II 61e80 Good Stable two separate functionsenormal condition (fNC ) and triggering
III 41e60 Normal Partially stable mechanism (fTM ) for surface water and groundwater conditions.
IV 21e40 Bad Unstable These two functions contain sub-factors that describe the level of
V 0e20 Very bad Completely unstable
hazard by assigning a score to each level. By selecting the appro-
priate score for each category of normal and triggering factors, a
final score is obtained to represent the level of hazard of a rock
slope failure.

16 3 1 The quantitative attribution of the normal condition (fNC ) is
F1 ¼ arctan ðjAj 17Þ (4)
25 500 10 classified into four sub-factors based on the failure mechanism. A
structurally controlled rock slope can be assessed by the factor of

where jAj ¼ aj as for planar failure; jAj ¼ aj as safety (Goodman and Bray, 1976; Hoek and Bray, 1981), and by the
180 for toppling failure, jAj ¼ jai as j for wedge failure: The relationship between slope and joint orientations. Non-structurally
parameters aj , as and ai are the joint dip direction, slope dip di- controlled slope is examined by using the geological strength index
rection, and the trend of the line of intersection of two planes, (GSI) (Hoek and Marinos, 2000), and by calculating the volume of
respectively. suspended rock (hanging rock mass due to undercutting) if the rock
slope is affected by weathering. The rating scores for each sub-

9 1 17 factor are 1, 3, 6 and 10, where “1” represents favourable condi-
F2 ¼ þ arctan B5 (5)
16 195 100 tion of the rock slope and “10” is unfavourable.
The quantitative attribution of the triggering mechanism (fTM ) is
where B is (in degree) the dip angle of the joint for planar-type and assessed by the influence of surface and groundwater (drainage
toppling-type failure modes ðbj Þ, and the plunge of the line of factor), and the ratio of the mean annual precipitation to the critical
intersection of two planes for wedge-type failure modeðbi Þ. annual precipitation against the stability behaviour of a rock slope.
The drainage factor fD can be examined by field inspection of the
1 slope materials and structures of the rock cut or slope (see Table 7);
F3 ¼ 30 þ arctan C (6)
3 it is then assigned by a score (1, 3, 6 or 10). The score of drainage
factor is multiplied by the ratio of the mean annual precipitation to
1 the critical annual precipitation to obtain the final score of the
F3 ¼ 13 arctan ðC 120Þ (7)
7 triggering mechanism fTM of a rock cut as follows:
where C is defined in degree as follows: Im

fTM ¼ f (9)
Icr D
(1) The difference in angle between the joint dip and slope dip,
i.e. bj bs , for planar-type failure mode; where Im is the mean annual precipitation, Icr is the critical annual
(2) The difference in angle between the plunge of the line of precipitation, and fD is the drainage factor.
intersection of two planes and the dip of the slope, i.e. bi bs , The hazard index, HI, for a rock cut is calculated as
for wedge-type failure mode;
(3) The sum of the two dip angles of the joint and slope, i.e. HI ¼ ðfNC fTM Þ1=2 (10)
bj þ bs , for toppling-type failure mode.
The HI is given on a scale of 1e10 defining 4 intervals, i.e. 1e4
(good), 4e6 (fair), 6e8 (poor) and 8e10 (very poor).
Eq. (6) is used for slopes with planar- or wedge-type failure
mode, whilst Eq. (7) is used for slopes with toppling-type failure
mode. 4. Case study
In this study, 22 road cuts have been chosen in the study area to
3.5. Graphical slope mass rating examine their stability conditions using the five selected empirical
methods. Table 8 summarises the basic features of the selected sites
A graphical method was proposed by Tomás et al. (2012) for and field observations.
determination of the correction factors of SMR. The correction The methodology used in this study is demonstrated in Fig. 5. As
factors F1, F2, and F3 for the joint sets are determined graphically can be seen, the methodology begins with the selection of road cuts
by using stereo plots, which are designed for the failure modes (22 sites) from the region based on anticipated failure mechanism.
(planar, wedge, and toppling). The correction factors F1 and F2 are The second step is field investigation and data collection. In the
grouped into one term j. The stereo plots determine the position current study, all data and measurements related to rock mass and
of the discontinuity pole (for planar-type and toppling-type discontinuities characteristics were gathered. This includes the
failure modes) or the intersection of planes (for wedge-type state of weathering, RQD, joint spacing, joint surface conditions,
failure mode). The proposed stereo plots for F3 depend on the effect of water, and attitudes of different joint sets, as well as the
failure mode. Therefore, the modified SMR equation is expressed slope geometrical properties such as slope height, face dip and dip
as follows: direction. Also, a visual assessment of the rock mass was carried out
using GSI to characterise the rock mass condition. The GSI is an
SMR ¼ RMR þ jF3 þ F4 (8)
important factor especially in the assessment of highly weathered
rock slopes (non-structurally controlled).
Table 7
Rating criteria and factors for the drainage sub-factors fD;n (Pantelidis, 2010).
Drainage sub-factors Rating criteria and score
1 3 6 10
Groundwater
Case 1
Potential for water Favourable conditions Two favourable and one One favourable and two Unfavourable conditions
infiltration from the Mantle: Impermeable unfavourable conditions are unfavourable conditions are Mantle: Permeable
upslope area, f D; 1a Upslope gradient: Abrupt fulfilled fulfilled Upslope gradient: Gentle
Coverage: The upslope area is or or Coverage: The upslope area is
covered by dense evergreen One favourable and two One unfavourable and two bare or almost bare (e.g. sparse
forest or urban development intermediate conditions are intermediate conditions are vegetation, deciduous
fulfilled. fulfilled. development)
Potential for build-up of Water circulation seems Free drained cutting (wide (I) Fair drainage of groundwater (II) Poor drainage of
hydrostatic pressure, impossible (very tight or hard- unfilled discontinuities) through unfilled or soft-filled groundwater through narrow
f D; 1b filled discontinuities or intact discontinuities unfilled or soft-filled
rock) and discontinuities
(III) No or short freezing and
periods. (IV) Long freezing periods
Surface water
Case 2
Potential for water Surface water flow is unlikely to Minor inflow which can be (I) Fairly drained rock mass (II) Poorly drained rock mass
inflow through trigger the type of failure drained through narrow with regards to the expected with regard to the expected
exposed studied (water inflow seems discontinuities inflow quantity inflow quantity
discontinuities and impossible: very tight or hard- or and and
build-up of filled discontinuities or intact Free drained cutting (wide (III) No or short freezing (IV) Long freezing periods.
hydrostatic pressure, rock) unfilled discontinuities) periods.
f D; 2
Case 3
Instabilities due to Surface water flow is unlikely to Minor instabilities are likely Instabilities are likely due to Major instabilities are likely due
surface water flow, trigger the type of failure due to wash-out of infilling wash-out of infilling material to wash-out of infilling material
f D; 3 studied (no loose stones or material or or
blocks, insignificant surface or Water flow from upslope may A large amount of water
water flow, etc.) Water flow from upslope may cause the transportation of a flowing from upslope (e.g. from
cause the instability of only few loose blocks or stones. ravine) may cause the
some small loose stones. transportation of several loose
blocks or stones.
Table 8
Summary of the basic features of selected case studies.
Site Slope Slope face (dip ( )/ Joint sets (dip ( )/dip direction ( )) Potential of failure mechanism Field observations
No. height dip direction ( ))
Set 1 Set 2 Set 3 Set 4 Set 5
(m)
1 18 60/40 52/43 75/357 63/238 39/63 e Structurally controlled Partially stable (disjointed rock mass
in the upper portion)
2 19 69/27 73/242 83/134 45/342 e e Structurally controlled Partially stable (dispersed blocks
in the upper portion)
3 15 77/55 17/16 54/220 80/210 41/95 83/95 Structurally controlled Stable
4 8 60/269 55/283 65/229 64/116 84/137 e Structurally controlled Unstable (failed)
5 19 78/20 78/252 25/78 84/163 e e Structurally controlled Stable
6 11 80/26 39/260 20/16 79/229 e e Structurally controlled Partially stable
7 18 80/10 57/244 62/195 70/50 85/343 e Structurally controlled Stable
8 13 60/10 e e e e e Non-structurally controlled Unstable in rainy season
(highly weathered slope)
70/285
67/30
17 23 76/154 43/292 49/72 45/102 64/186 78/130 Structurally controlled Partially stable
18 15 74/8 11/95 83/30 73/38 e e Structurally controlled Partially stable (small blocks)
19 36 74/190 53/219 63/133 53/113 e e Structurally controlled Unstable (failed)
20 11 52/45 e e e e e Non-structurally controlled Stable
The third step in the methodology (see Fig. 5) is to perform However, the results of CRMR show observable changes on RMR
laboratory testing. Rock samples were collected from each site for ratings with increased RMR values. The RMR classifications for sites
rock testing to determine the physico-mechanical properties. The 5, 10 and 14 have changed from “fair” to “good”. The scores for sites
bulk density was identified for all rock specimens. The point-load 11 and 12 have also increased to 91 and 84, respectively, and were
test was carried out for each rock sample to determine the UCS of classified as “very good”.
the rock.
The fourth step of the methodology shown in Fig. 5 is for 4.2. Kinematic analysis
structurally controlled slopes. Two types of analyses are employed
in this stage: the kinematic analysis and the RMR. The kinematic The kinematic analysis of rock cuts was conducted for the
analysis is performed with the DIPS (stereographic projection structurally controlled failure using DIPS 6.0 software (Rocscience
software) software (Rocscience Inc., 2014). The likelihood of failure Inc., 2014). The analysis results indicate that there is potential for
and its types (planar, wedge and toppling) are determined. In planar failure in 5 locations, one of which is site 4 that already failed
addition, the most critical failure surface in the rock slope is iden- as shown in Fig. 7a. Nine locations show potential for wedge failure
tified. The RMR scores were calculated for all sites. As mentioned mechanism, where site 19 already failed (see Fig. 8a). For toppling
previously, all rock mass classification systems adopted in this failure, five sites show potential for flexural toppling (Fig. 9), with
study basically use the RMR system, except for the HI method high probability recorded for site 3, and five locations indicate
(Pantelidis, 2010). Two versions of RMR systems were used: the potential for block toppling (Fig. 10) with the maximum probability
DRMR (Bieniawski, 1989) that is used in all SMR methods, and the of 25% at site 2. All results of the kinematic analysis are listed in
CRMR (Sen and Sadagah, 2003) that is used only for continuous Table 9.
SMR classification system.
The fifth step is applying the empirical methods to the 22 sites 4.3. Rock slope stability assessment
based on the results of the kinematic analysis represented by the
type of failure and the joint-slope relationship and RMR score. All The stability of the rock cuts in the case study area was assessed
SMR systems (original, Chinese, continuous, and graphical SMR) with five rock mass classification systems developed for rock
have been applied to structurally controlled rock slopes. The HI slopes, i.e. original SMR (Romana, 1985), Chinese SMR (Chen, 1995),
method (Pantelidis, 2010) has been applied to both structurally and continuous SMR (Tomás et al., 2007), graphical SMR (Tomás et al.,
non-structurally controlled failures of rock cuts. All the results of 2012) and the HI method (Pantelidis, 2010). Three main failure
empirical methods have been compared and their pros and cons in mechanisms were defined using kinematic analysis for the struc-
terms of application to stability assessment of road cuts are turally controlled rock cuts (planar, wedge and toppling), and their
highlighted. results have been used in the selected rock classification systems.
The stability of rock cuts for the selected rock slopes has been The rock cuts with non-structurally controlled failure mechanisms
assessed through application of the rock mass classification sys- were assessed only with the HI method, because this system was
tems, especially the ones developed for rock slopes. To do this, RMR designed to address both structurally and non-structurally
should be determined, and the type of failure mechanism for jointed controlled failures, unlike the SMR method that is only applicable
rock mass (structurally controlled failure) needs to be identified. to structurally controlled failure.
4.1. Rock mass rating system 4.3.1. Structurally controlled failure
The results of the DRMR system (Fig. 6) indicate that the rock (1) Planar-type failure mode
cuts sites (1, 2, 3, 4, 5, 6, 10, 14, 16, 17 and 18) have RMR values of
41e57 and are classified as “fair”. The four locations (7, 11, 12 and SMR (Romana, 1985) results for rock cuts indicate that all scores
13) have RMR values of 63e79 and are classified as “good”. Site 19 are below 40 and are categorised from unstable to completely
gives poor rock quality with RMR value of 37. unstable. When the strike angle between the slope face (as Þ and
Site selection
Field investigation
Laboratory testing
Kinematic analysis (Structurally RMR calculations

controlled failure) (DRMR & CRMR)
1. Original SMR
Empirical methods 2. Chinese SMR Structurally controlled
3. Continuous SMR
4. Graphical SMR
Results Structurally and
5. Hazard Index non-structurally controlled
failure
Conclusion
Fig. 5. Methodology used in the present study.
100
90
80
RMR value % 70
60
50
40
30
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Discrete RMR (1989) 57 48 51 44 55 54 64 55 79 74 63 54 52 52 41 37 58
Connuous RMR (2003) 59 52 54 48 66 52 72 67 91 85 63 62 59 56 45 40 61
Fig. 6. Discrete and CRMR results for jointed rock cuts.
dominant joint (aj ) is 20 and less, SMR will yield significantly All the analysis results for the potential of planar failure are
reduced scores. Thus, the stability condition of the rock cut is documented in Table 10.
decreased as can be seen from the bar chart in Fig. 11. However, the
values of RMR scores are above 60, suggesting “good” quality rock. (2) Wedge-type failure mode
This will improve the SMR values.
The Chinese SMR system is applied to the case studies without The results of original SMR indicate partially stable condition at
the slope height factor because the heights of all sites are less than sites 3, 7, 10-1, 16 and 17. Rock cuts 1, 2, 10-2 and 13-1 are classified
80 m, thus only the discontinuity factor (l) is considered. The re- as unstable, and the results of wedge failure analysis for sites 4, 13-
sults of this method show that the SMR scores have been increased 2, 18 and 19 indicate completely unstable rock cuts. The analysis
compared to Romana’s system. The reason for this increase is the results for sites 18 and 19 have negative SMR values. This is because
low values of the discontinuity factor (l) for most rock cuts with an the RMR values are below 40 indicating poor quality rocks, and the
average value of 0.7. This value implies tight joints with high difference in the angle between the strikes of the slope and the line
cohesion, which leads to increased SMR values and thus an increase of intersection of the two joints that form the wedge failure is less
in the degree of stability condition. While this may not be the case than 10 (see Fig. 12). In addition, by field observation, sites 18 and
in reality, it can be concluded that the Chinese SMR tends to be less 19 show partially stable and unstable (failed) rock cuts, respectively
conservative than Romana’s SMR. (see Table 8). The graphical SMR analysis has given the same results
The continuous SMR (Tomás et al., 2007) results show that there and conclusion as the original SMR.
is a noticeable increase in the SMR scores, where most of the sta- The results of the Chinese SMR show a slight increase in SMR
bility conditions have switched from unstable to partially stable. scores. For example, in site 2, the SMR classification is unstable,
However, when the CRMR results were used with the continuous while the Chinese SMR gives the result of partially stable. Also, at
SMR functions, only a slight change in the SMR scores was sites 10-2 and 22, the SMR scores are classified as completely un-
observed, and the stability condition of only two sites, 7 and 10, stable in the original SMR, and they are unstable in the Chinese
changed from partially stable to stable. Sites 7 and 10 show stable SMR.
and partially stable conditions, respectively (see Table 10). Two kinds of results are calculated for continuous SMR by using
The graphical SMR method (Tomás et al., 2012) results show no DRMR and CRMR. Once again, the results of the continuous SMR
difference with Romana’s SMR results. This is not surprising as the with DRMR indicate an increase in SMR ratings compared to the
underlying concepts used in the graphical SMR method are the original SMR. Sites 10-2, 13, and 22 switch from completely un-
same as the ones in the original discrete SMR (Romana, 1985) for stable and unstable to partially stable. After the continuous func-
the correction factors F1, F2 and F3. The difference is only in the tions of RMR in combination with the continuous SMR are applied,
method of application, which is based on using the stereo plots to the SMR ratings of some locations change, such as site 2 where the
determine the correction factors in the graphical SMR method. stability condition goes from unstable to partially stable. Also at
The results obtained from the HI indicate that sites 1 and 13 are sites 7 and 10, the SMR predictions switch from partially stable to
categorised as “fair”, and sites 4, 7, 10 and 22 are classified as “poor”. stable. Sites 10 and 22 are in partially stable conditions, while sites
As can be seen from Table 8, the field observation data show that 7 and 13 are in stable conditions (see Table 11).
site 1 is partially stable (especially the upper portion), while site 13 Using the HI method, the results for all rock cuts are classified as
is in a stable condition. Also, site 4 has already failed in reality, and “poor”, except for sites 2, 3 and 10 where the results show “fair”
site 7 shows a stable condition; however, sites 10 and 22 are rock slopes. However, by field observation, sites 2 and 10 indicate
partially stable. partially stable and site 3 shows a stable condition.
Fig. 7. (a) Planar failure at site 4, and (b) Stereographic projection showing the potential for planar failure.
All the analysis results of potential of wedge failure for the except for sites 1 and 12 where the stability conditions are partially
selected cuts sites are listed in Table 11. stable. The continuous SMR results show that the stability condi-
tions for the rock cuts of sites 1, 3, 6, 7 and 12 are partially stable; for
(3) Toppling-type failure mode site 11, the stability condition is improved to good or stable, while
site 4 remains unstable. The SMR scores increase when using the
The toppling failure analysis was conducted for two failure CRMR with continuous SMR functions. For example, for sites 4, 7
mechanisms, which are flexural and block toppling. The flexural and 12, the stability conditions change to partially, good and very
toppling can be treated in all rock slope classification systems good rock slopes, respectively.
selected in this study. However, the HI method (Pantelidis, 2010) The results of HI method for flexural toppling show that sites 1, 3, 4,
was basically developed to assess all possibility of rock failures 7-1 and 11 are fair rock cuts, and sites 6 and 12 are good rock cuts,
based on failure mechanism, thus this method can analyse both whilst site 7-2 is a poor rock cut. Regarding the block toppling analysis
kinds of toppling-type failure modes. carried out by the HI method, sites 10-1, 11, 12, 14, 17 and 22 are
The results of SMR’s classification systems indicate that seven classified as good rock cuts, and sites 10-2 and 2 are classified as poor;
rock cut sites are prone to flexural toppling failure. Based on the while in site 5, the rock cut is classified as fair rock cut. In field
original and graphical SMR systems, the rock cuts are classified as observation, as shown in Table 8, sites 1, 2, 3, 4, 7,10 and 11 are partially
follows: sites 1 and 4 show unstable conditions, sites 3, 6 and 7 are stable rock cuts, while site 5 is potentially in a stable condition.
partially stable, and sites 11 and 12 are classified as stable. The All the results for toppling failure analysis are described in
Chinese SMR results show no change in the stability categories, Table 12.
Fig. 8. (a) Wedge failure at site 19, and (b) Stereographic projection showing the potential for wedge failure.
4.3.2. Non-structurally controlled failure The analysis results have revealed a noticeable contrast between
As previously mentioned, this type of failure mechanism can the discrete and continuous functions, whether in the RMR or in the
only be analysed by the HI method. The results indicate that sites 8, SMR’s methods. As noted in Fig. 6, the RMR scores increase when
9, 15, 20 and 21 are classified as poor rock cuts as shown in Table 13. calculated with the continuous functions. This is because in the
DRMR, the evaluation process for basic RMR parameters is based on
5. Discussion step functions, which can increase the level of subjectivity. The
advantage of the CRMR is that it gives a distinctive score for each
The results of the study show that all SMR methods can be used RMR parameter, thus it can reduce subjectivity.
for structurally controlled failure of rock slopes (jointed rock mass), The same can be said for the continuous SMR, where the scores
while the HI method proposed by Pantelidis (2010) is applicable to increase compared to those obtained from the discrete functions
both structurally and non-structurally controlled failure mecha- such as original SMR, Chinese SMR and graphical SMR. The original
nisms. In addition, the HI method is the only method that considers and graphical SMR methods show similar results, and the classes of
the effect of external factors, such as water, that can trigger failure. rock slope assessment are almost the same. This similarity is not
Fig. 9. (a) Potential for flexural toppling failure at site 5, and (b) Stereographic projection showing for flexural toppling failure.
surprising as the underlying concepts used in the two methods are conditions. It can be concluded that the Chinese SMR tends to be
the same for the correction factors F1, F2 and F3. less conservative than Romana’s SMR.
The Chinese SMR, a modified version of the original SMR, adds The effect of the angle between the slope face and the
two factors, i.e. the height factor (z) for slopes higher than 80 m and critical failure surface in case of plane failure (or the intersec-
the discontinuity condition factor (l). When applying in this study, tion line in wedge failure) is obvious in the values of discrete
the height factor is eliminated because all road cuts in this study are SMR methods. When the angle is less than 10 , the SMR scores
less than 80 m high. The Chinese RMR ratings are found to be become very low, suggesting an underestimation of the stability
higher than the original SMR. This is due to the discontinuity condition. This scenario can be seen in sites 1, 4, 13-2 and 22
condition factor (l), especially for discontinuities with favourable for planar failure (see Fig. 11) and in sites 4 and 13-2 (see
Fig. 10. (a) Potential for block toppling failure at site 5, and (b) Stereographic projection showing for block toppling failure.
Fig. 12) for wedge failure. Furthermore, if the quality of the rock positive quantity regardless of the value of the said angle. This
mass in DRMR is classified as “poor” or DRMR 40, the discrete may suggest that the discrete SMR is not appropriate for slopes
SMR score may be negative (DSMR < 0), as the cases in sites 18 with weak rock mass, especially when the angle between the
and 19 (see Fig. 12). However, the continuous functions of SMR slope face and the critical joint surface is less than 10 . It also
do not have this peculiarity and the total score is always a suggests that the continuous SMR might be considered as a
Table 9 better method for stability assessment of rock slopes amongst

Kinematic analysis results for structurally controlled failure mechanism. the SMR methods.
Site No. Planar failure (%) Wedge failure (%) Toppling failure (%) All SMR methods utilised in this study apply only to structurally
Flexural Block
controlled failure mechanisms. The HI method (Pantelidis, 2010) is
a
applicable to both structurally and non-structurally controlled
1 47.5 65.98 15 0.64
slopes, which considers the normal condition and the triggering
2 0 29.1 7.5 25
3 0 7.05 40 1.92 mechanism. Note that the triggering mechanism does not exist in
4 17.5 29.49 5 4.23 all SMR methods. Also, the HI method considers block toppling-
5 0 3.47 0 14.14 type failure mode, which SMR methods do not account for. It is
6 5 8.97 27.5 4.10
for these reasons that the HI method can be considered as the
7 0 17.72 20 5.78
10e1 32.5 58.61 0 0.13 preferred empirical method used in this study.
10e2 32.5 82.39 0 0.13 Although the HI method is superior to other empirical methods,
11 0 2.71 30 17.68 it has some limitations. Firstly, the notion of “hazard” suggests the
12 0 2.57 30 4.24 use of statistical methods to estimate the likelihood of occurrence
13e1 22.5 34.53 0 0.77
based on the probability distribution function of random variables
13e2 30 40.82 0 1.28
14 0 15.64 0 18.08 such as the joint properties. In the HI method, the normal condition
16 0 13.08 0 6.41 function (fNC ) is calculated with a deterministic approach, and
17 0 39.72 0 8.35 hence it does not consider the inherent uncertainty of the basic
18 0 22.21 0 0.51
input parameters like joint dip and dip direction. Furthermore, the
19 5 62.64 0 0.26
22 35 40.38 0 17.69
HI method relies on the interpretation of field observations and
a
linking such observations to the descriptions reported in Table 7 to
Percentage of joints mapped.
determine the triggering mechanism function (fTM ) due to water
and slope drainage condition. The approach is subjective and could
Fig. 11. Results of original SMR, DRMR, and strike difference of slope-joint for planar failure.
Table 10
Stability analysis results for planar failure using rock slope classification methods.
Site No. Original SMR Chinese SMR Continuous SMR using DRMR Continuous SMR using CRMR Graphical SMR Hazard Index
Rating Stability Rating Stability Rating Stability Rating Stability Rating Stability HI Category
1 7 C. unstable 22 Unstable 46 P. stable 48 P. stable 7 C. unstable 6 Fair

4 9 C. unstable 13 C. unstable 34 Unstable 38 Unstable 9 C. unstable 6.5 Poor
7 22 Unstable 34 Unstable 53 P. stable 61 stable 22 C. unstable 7.4 Poor
10e1 34 Unstable 34 Unstable 59 P. stable 71 stable 34 Unstable 6.5 Poor
10e2 34 Unstable 34 Unstable 53 P. stable 65 stable 34 Unstable 6.5 Poor
13e1 21 Unstable 33 Unstable 52 P. stable 52 P. stable 20 C. unstable 6 Fair
13e2 3 C. unstable 21 Unstable 52 P. stable 52 P. stable 3 C. unstable 6 Fair
22 7 C. unstable 22 Unstable 42 P. stable 45 P. stable 8 C. unstable 7.4 Poor
*P: Partially; **C: Completely.
Fig. 12. Results of original SMR, DRMR, and the strike difference of slope-joint for wedge failure.
Table 11
Stability analysis results for wedge failure using rock slope classification methods.
Site No. Original SMR Chinese SMR Continuous SMR using DRMR Continuous SMR using CRMR Graphical SMR Hazard Index (HI)
Rating Stability Rating Stability Rating Stability Rating Stability Rating Stability Rating Stability
1 33 (IV) Unstable 40 (IV) Unstable 33 (IV) Unstable 35 (IV) Unstable 33 (IV) Unstable 6 Fair
2 40 (IV) Unstable 43 (III) P. stable 37 (IV) Unstable 41 (III) P. stable 40 (III) Unstable 6 Fair
3 43 (II) P. stable 46 (III) P. stable 39 (IV) Unstable 42 (III) P. stable 43 (III) P. stable 5.1 Fair
4 2 (V) C. unstable 6 (V) C. unstable 33 (IV) Unstable 37 (IV) Unstable 2 (V) C. unstable 6.5 Poor
7 55 (III) P. stable 58 (III) P. stable 53 (III) P. stable 61 (II) stable 55 (III) P. stable 7.6 Poor
10e1 50 (III) P. stable 50 (III) P. stable 60 (III) P. stable 71 (II) stable 50 (III) P. stable 6.5 Poor
10e2 28 (IV) Unstable 28 (IV) Unstable 59 (III) P. stable 71 (II) stable 28 (IV) Unstable 6.5 Poor
13e1 21 (IV) Unstable 33 (IV) Unstable 52 (III) P. stable 52 (III) P. stable 21 (IV) Unstable 6.3 Fair
13e2 3 (V) C. unstable 21 (IV) Unstable 52 (III) P. stable 52 (III) P. stable 3 (V) C. unstable 6.3 Fair
16 44 (III) P. stable 46 (III) P. stable 41 (III) P. stable 48 (III) P. stable 44 (III) P. stable 6.5 Poor
17 44 (III) P. stable 46 (III) P. stable 41 (III) P. stable 45 (III) P. stable 43 (III) P. stable 7.4 Poor
18 9 C. unstable 6 (V) C. unstable 31 (IV) Unstable 35 (IV) Unstable 9 C. unstable 7.4 Poor
19 15 C. unstable 3 (V) C. unstable 29 (IV) Unstable 32 (IV) Unstable 15 C. unstable 8.5 Poor
22 16 (V) C. unstable 29 (IV) Unstable 48 (III) P. stable 51 (III) P. stable 16 (V) C. unstable 7.4 Poor
*P: Partially; **C: Completely.
Table 12
Stability analysis results for toppling failure using rock slope classification methods.
Site No. Original SMR Chinese SMR Continuous SMR using DRMR Continuous SMR using CRMR Graphical SMR Hazard Index (HI)
Flexural Block Flexural Block Flexural Block Flexural Block Flexural Block Flexural Block
1 40 (IV) NA 45 (III) NA 53 (III) NA 55 (III) NA 40 (IV) NA 5 e

2 e NA e NA e NA e NA e NA e 6.5
3 41 (III) NA 44 (III) NA 46 (III) NA 49 (III) NA 41 (III) NA 5.8e4.1 e
4 34 (IV) NA 35 (IV) NA 39 (IV) NA 43 (III) NA 34 (IV) NA 4.1 e
6 44 (III) NA 47 (III) NA 49 (III) NA 47 (III) NA 44 (III) NA 3.9 e
7e1 54 (III) NA 57 (III) NA 58 (III) NA 66 (II) NA 54 (III) NA 4.8 e
7e2 43 (III) NA 49 (III) NA 54 (III) NA 62 (II) NA 43 (III) NA 7.6 e
10e1 e NA e NA e NA e NA e NA e 3.4
10e2 e NA e NA e NA e NA e NA e 6.2
11 66 (II) NA 87 (I) NA 82 (I) NA 94 (I) NA 66 (II) NA 4.6 2.5
12 72 (II) NA 75 (II) NA 77 (II) NA 88 (I) NA 72 (II) NA 2.9 1.5
14 e NA e NA e NA e NA e NA e 1
17 e NA e NA e NA e NA e NA e 1
lead to wide variation of results depending on the observer’s which is more often suited to describe the rock mass quality.
interpretation. A more robust approach is to derive a range of fTM Rather, it would be more representative to adopt descriptions such
values based on characteristic cases with numerical simulation as high, moderate, and low hazard for slope stability as done by
while accounting for the groundwater effect, especially in the case others (Anbalagan, 1992; Pachauri and Pant, 1992).
study area where most failures are known to occur shortly after Finally, it is worth mentioning that all empirical methods used
heavy rainfalls, i.e. under undrained condition. The HI method in this study share another shortcoming, i.e. they are all developed
describes the slope stability as good, fair and poorea classification for linear structures. They are not designed for mountainous roads
Table 13
Stability analysis results for non-structurally controlled failure using Hazard Index.
Site No. Slope face (dip ( )/dip direction ( )) Structure Surface condition GSI Normal condition (fNC) Triggering factor (fTM) Hazard index
HI Category
8 60/10 Disintegrated Very poor 10 10 7.1 8.5 Poor

9 55/60 Very blocky Very poor 35 10 7.1 8.5 Poor
15 66/95 Disturbed/Seamy Very poor 17 10 7.1 8.5 Poor
20 52/45 Disintegrated Very poor 10 10 7.1 8.5 Poor
21-1 65/10 Disintegrated Very poor 10 10 7.1 8.5 Poor
21-2 60/310 Disintegrated Very poor 10 10 7.1 8.5 Poor
which are often curved and result in rock cuts of concave and as the Jazan region in the sKingdom of Saudi Arabia where the case
convex surface shapes. The introduction of an adjustment factor to study is carried out.
account for the slope face curvature would make a valuable
improvement to the existing empirical methods. The above- Conflict of interest
mentioned shortcomings are the subject of research currently un-
derway by the authors. The authors wish to confirm that there are no known conflicts of
interest associated with this publication and there has been no
6. Conclusions significant financial support for this work that could have influ-
enced its outcome.
This paper examines the rock mass classification systems,
especially those developed for the assessment of rock slope sta- Acknowledgment
bility. Five empirical methods, original SMR, Chinese SMR, contin-
uous SMR, graphical SMR and HI method, are applied to 22 sites of This work is financially supported by the Saudi Geological Sur-
rock cuts located in a rugged area along mountainous roads in the vey through a doctoral fellowship at McGill University. The authors
southwestern part of the Kingdom of Saudi Arabia. The following are grateful for their financial as well as technical support. Special
conclusions can be drawn from this study. thanks are due to the Saudi Arabian Cultural Bureau in Canada.
SMR methods consider the condition of the rock mass presented
by the RMR system, and the relationship between the dip and the
References
dip direction of both the slope face and the joint. These relation-
ships are presented by the correction factors of F1, F2 and F3 as well Aksoy C. Review of rock mass rating classification: historical developments, appli-
as the effect of the excavation method F4. SMR methods can only be cations, and restrictions. Journal of Mining Science 2008;44(1):51e63.
applied to slopes that exhibit structurally controlled failure Anbalagan R. Landslide hazard evaluation and zonation mapping in mountainous
terrain. Engineering Geology 1992;32(4):269e77.
mechanisms. Anbalagan R, Sharma S, Raghuvanshi TK. Rock mass stability evaluation using
On the other hand, the HI method considers the influence of the modified SMR approach. In: proceedings of the 6th National Symposium on
normal rock cut condition and the triggering factors (e.g. the Rock Mechanics; 1992. p. 258e68.
Bieniawski ZT. Engineering classification of jointed rock masses. Civil Engineer in
presence of water). It can be applied to both the structurally and South Africa 1973;15(12):335e43.
non-structurally controlled failure mechanisms. Bieniawski ZT. Exploration for rock engineering: proceedings of the Symposium on
The original and graphical SMR methods show similar results exploration for rock engineering. Cape Town, South Africa: A.A. Balkema; 1976.
Bieniawski ZT. The geomechanics classification in rock engineering applications. In:
and the resultant stability conditions are almost the same because
Proceedings of the 4th ISRM Congress. International Society for Rock Me-
the underlying calculation formulae of F1, F2, F3 and F4 are the same. chanics; 1979.
However, application of the technique is different. Bieniawski ZT. Engineering rock mass classifications: a complete manual for engi-
neers and geologists in mining, civil, and petroleum engineering. John Wiley &
The continuous functions, such as the CRMR and continuous
Sons; 1989.
SMR, are relatively less subjective classification systems because Bieniawski ZT. Classification of rock masses for engineering: the RMR system and
they give specific scores for each input parameter value. However, future trends. Rock Testing and Site Characterization 1993:553e73.
the effects of triggering mechanisms such as water and earthquakes Chen Z. Recent developments in slope stability analysis. In: Fujii T, editor. Pro-
ceedings of the 8th International Congress of Rock Mechanic, vol. 3; 1995.
are absent in the SMR methods. p. 1041e8.
The Chinese SMR method is not suitable for rock cuts below Coggan JS, Stead D, Eyre JM. Evaluation of techniques for quarry slope stability
80 m, where the effect of slope height is ignored. In the case of the assessment. Transactions of the Institutions of Mining and Metallurgy, Section
B: Applied Earth Science 1998:107.
rock slopes located in the study area, all rock cuts are less than 80 m Duran A, Douglas K. Experience with empirical rock slope design. In: proceedings of
high. ISRM International Symposium; 2000.
The HI method is as good as continuous SMR, which does not Goodman RE, Bray JW. Toppling of rock slopes. In: Rock engineering for foundations
& slopes; 1976.
underestimate nor overestimate the stability evaluation. However, Hack R, Price D, Rengers N. A new approach to rock slope stability-a probability
the HI method is superior as it takes into account the effect of water, classification (SSPC). Bulletin of Engineering Geology and the Environment
and this is observed in the comparison of analysis results of HI 2003;62(2):167e84.
Hoek E, Bray J. Rock slope engineering. Revised 3rd ed. London: the Institution of
method with SMR methods. Mining and Metallurgy; 1981.
In conclusion, the continuous SMR and the HI could be the Hoek E, Marinos P. Predicting tunnel squeezing problems in weak heterogeneous
proper methods for the stability assessment of the rock cuts. These rock masses. Tunnels and Tunnelling International 2000;32(11):45e51.
Hoek E. Rock mass properties. Practical rock engineering 2007. www.rocscience.
methods could be further improved by adding the factor of slope
com/learning/hoek-s-corner.
height below 80 m, and by analysing the effect of pore water Pachauri AK, Pant M. Landslide hazard mapping based on geological attributes.
pressure on the stability behaviour of a rock slope through incor- Engineering Geology 1992;32(1e2):81e100.
poration of the numerical modelling technique with the analysis Pantelidis L. Rock slope stability assessment through rock mass classification systems.
International Journal of Rock Mechanics and Mining Sciences 2009;46(2):315e25.
and derived adjustment factors. The influence of slope curvature Pantelidis L. An alternative rock mass classification system for rock slopes. Bulletin
should also be incorporated especially in mountainous areas such of Engineering Geology and the Environment 2010;69(1):29e39.
Price NJ, Cosgrove JW. Analysis of geological structures. Cambridge, England: Hassan M. Basahel is currently a PhD student at the
Cambridge University Press; 1990. Department of Mining and Materials Engineering, McGill
Rocscience Inc. Dip v. 6.0 - graphical and statistical analysis of orientation data. University. He obtained his BSc and MSc degrees in engi-
Toronto, Canada. 2014. neering geology from King Abdulaziz University, Jeddah,
Romana M. New adjustment ratings for application of Bieniawski classification to Saudi Arabia in 1998 and 2008, respectively. Since 2000,
slopes. In: proceedings of international symposium on the role of rock me- he has been working with the Geohazards Department,
chanics. Zacatecas: ISRM; 1985. p. 49e53. Saudi Geological Survey, Jeddah, Saudi Arabia. His research
Sen Z, Sadagah BH. Modified rock mass classification system by continuous rating. interests include rock slope stability assessment using
Engineering Geology 2003;67(3e4):269e80. statistical and numerical methods, as well as geological
Tomás R, Delgado J, Serón JB. Modification of slope mass rating (SMR) by continuous hazards.
functions. International Journal of Rock Mechanics and Mining Sciences
2007;44:1062e9.
Tomás R, Cuenca A, Cano M, Garcßa-Barbra J. A graphical approach for slope mass
rating (SMR). Engineering Geology 2012;124(4):67e76.
Wyllie DC, Mah C. Rock slope engineering. 4th ed. CRC Press; 2004.

10 Rock Mass Classification Systems 1

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

10 Rock Mass Classification Systems 1

Diunggah oleh

Hak Cipta:

Format Tersedia

Journal of Rock Mechanics and Geotechnical Engineering xxx (2017) 1e17

Contents lists available at ScienceDirect

Journal of Rock Mechanics and

Full Length Article

Application of rock mass classiﬁcation systems to rock slope stability

1. Introduction projection technique. This technique is used to project the orien-

3.2. Slope mass rating (SMR)

SMR is a classical lump-rating classiﬁcation system for rock

Parameter Range of values

Discontinuity length (persistence) Separation (aperture) Roughness Inﬁlling (gouge) Weathering

<1 6 None 6 Very rough 6 None 6 Unweathered 6

Table 6 3.6. Hazard index (HI)

where C is deﬁned in degree as follows: Im

Drainage sub-factors Rating criteria and score

4.1. Rock mass rating system 4.3.1. Structurally controlled failure

Kinematic analysis (Structurally RMR calculations

Fig. 5. Methodology used in the present study.

Fig. 6. Discrete and CRMR results for jointed rock cuts.

Table 9 better method for stability assessment of rock slopes amongst

1 7 C. unstable 22 Unstable 46 P. stable 48 P. stable 7 C. unstable 6 Fair

*P: Partially; **C: Completely.

*P: Partially; **C: Completely.

1 40 (IV) NA 45 (III) NA 53 (III) NA 55 (III) NA 40 (IV) NA 5 e

8 60/10 Disintegrated Very poor 10 10 7.1 8.5 Poor

Anda mungkin juga menyukai