Reference: A. Palmström, www.rockmass.

net

An introduction to the Rock Mass index (RMi) and its applications

by Arild Palmström, Ph.D.

1 Introduction
Construction materials commonly used in civil engineering are mostly characterized by their strength
properties. In rock engineering, however, no such specific strength characterization of the rock mass is
in common use. Most engineering is carried out using various descriptions, classifications and
unquantified experience. Hoek and Brown (1980), Bieniawski (1984), Nieto (1983) and several other
authors have, therefore, indicated the need for a strength characterization of rock masses.

The Rock Mass index, RMi, system has been developed to meet this need. It was developed between 1986
and 1995. The main development is presented in the Ph.D. thesis of Arild Palmström from 1995. A
reference list of the RMi publications is shown at the end of this article.

2 The rockmass index, RMi

An important issue in the development of RMi system has been to use input parameters which have the
greatest significance in rockmass behaviour. The main principles in the RMi value and the input data
used are shown in Figures 1 and 2.

Figure 1.
The layout of the Rock
Mass index, RMi. Input
for interlocking was
introduced in 2005.

Figure 2
The main parameters in
the rock mass are
included in the
RockMass index, which
approximately
characterizes the
uniaxial compressive
strength of a rock mass.

Dec. 2014
 2

RMi is based on the principle that the joints intersecting a rockmass tend to reduce its strength.
Consequently, it is expressed as: RMi = σc × JP
Here σc = the uniaxial compressive strength of intact rock (in MPa), measured on 50 mm samples. (Often,
UCS is used instead of σc)
JP = the jointing parameter, expressing the reduction in strength of the intact rock caused by the
joints.

The jointing parameter (JP) is composed of the joint condition factor, jC, and the block volume, Vb. The
joint condition, jC = jR × jL/jA,
where
• jR, the joint roughness factor, determined by
- js = smoothness of joint surface factor and
- jw = joint plane planarity or waviness factor,
• jA, the joint condition or alteration factor, and
• jL, the joint size and continuity factor.

As shown in Figure 1, JP incorporates the main joint features in the rock mass. From the test results
presented in Figure 3 the Jointing Parameter was found as
JP = 0.2 jC × VbD where D = 0.37jC -0.2

Figure 3.
Test results from 8 large
scale compressive strength
tests or back calculations
were used to find the
expression for the jointing
parameter, JP. The known
data for the samples were
plotted in the diagram and
the lines for the joint
characteristics, jC, were
draw as shown. These lines
represent the expression
for JP.

Unfortunately, it was not

possible to detect more than
the 8 large scale test results
at the time when RMi
system was developed,
though many organisations
and companies were
contacted.

The input parameters to RMi are shown in Table 1. They can be determined by commonly used measure-
ments and observations, and from empirical relationships. The RMi system has some features similar to
www.RockMass.net
Dec. 2014
 3

those of the Q-system. Thus, jR and jA are almost the same as Jr and Ja in the Q-system. RMi requires more
calculations than the RMR and the Q system, but spreadsheets can preferably be used, from which the RMi
value can be found directly. This is presented in Figure 4.

Table 1. The input parameters to RMi

Uniaxial compressive strength of rock (UCS or σc ) value in MPa (from lab. tests or assumed from handbook tables)

Block volume ( Vb ) value in m³ (from observations at site or on drill cores, etc.)

Joint condition factor ( jC ) jC = jR x jL / jA (ratings of jR, jA and jL from the tables below)

Joint roughness factor ( jR ) Large scale waviness of joint plane

composed of large scale and small scale undulations Slightly Stepped or
(The ratings in bold italic are similar to Jr in Q-system) Planar undulating Undulating Strongly undulating interlocking
m

m
S

s
c
a

o
o

h
n
e

e
s

s
s

c
r
l
l

f
j

Very rough 2 3 4 6 6
Rough 1.5 2 3 4.5 6
Smooth 1 1.5 2 3 4
Polished or slickensided*) 0.5 1 1.5 2 3
For filled joints jR = 1 For irregular joints a rating of jR = 6 is suggested
*) For slickensided surfaces the ratings apply to possible movement along the lineations

Joint alteration factor ( jA ) (the ratings are based on Ja in the Q-system)

bet
we

wa
en

lls

Healed or welded joints filling of quartz, epidote, etc. jA = 0.75

CLEAN Fresh joint walls no coating or filling, except from staining (rust) 1
JOINTS: - one grade higher alteration than the rock 2
Altered joint walls
Conta

- two grades higher alteration than the rock 4

joint
ct

COATING or Frictional materials sand, silt calcite, etc. without content of clay 3
THIN FILLING OF: Cohesive materials clay, chlorite, talc, etc. 4
w
P
a

n
o

o
n

a
y

c
r

r
t

t
l

l
l

Thin filling (< 5mm) Thick filling

Frictional materials sand, silt calcite, etc. (non-softening jA = 4 8
THICK FILLING OF: Hard, cohesive materials clay, chlorite, talc, etc. 6 5 - 10
Soft, cohesive materials clay, chlorite, talc, etc. 8 12
Swelling clay materials material exhibits swelling properties 8 - 12 13 - 20

Joint size factor ( jL ) composed of the length and continuity of the joint Continuous joints Discont. joints*)
Bedding or foliation partings length < 0.5 m jL = 3 jL = 6
with length 0.1 - 1 m 2 4
Joints with length 1 - 10 m 1 2
with length 10 - 30 m 0.75 1.5
(Filled) joint, seam or shear **) length > 30 m 0.5 1
*) Discontinuous joints end in massive rock **) Often a singularity and should in these cases be treated separately

Interlocking of rockmass structure ( IL ) (the ratings are based on the interlocking used in the GSI system)
Very tight structure undisturbed rock mass, well interlocked IL = 1.3
Tight structure undisturbed rock mass with some joint sets 1
Disturbed / open folded / faulted with angular blocks 0.8
Poorly interlocked broken with angular and rounded blocks 0.5

3 RMi in massive rock

In massive rock1 the RMi value is found from
RMi = σc × fσ (where fσ > JP)
The massivity parameter, fσ, represents the scale effect of the uniaxial compressive strength (which for
intact rock samples or massive rock has a value of approximately fσ ≈ 0.5).

1 Massive rock is here defines as rocks with low degree of jointing, i.e. block volumes are larger than a few m 3
www.RockMass.net
Dec. 2014
 4

4 RMi in weakness zones

Weakness zones should in many cases be treated individually without using classification systems for support
estimate. Support assessments for crushed zones may, however, be carried out using the support chart for
blocky ground in Figure 6 and input parameters as for blocky ground. In zones with thickness less than
approximately 20 m), the stability is influenced by the interplay between the zone and the adjacent
rockmasses. Therefore, the stresses in such zones are generally lower than in the adjacent ground, which will
reduce the effect of squeezing.

For crushed weakness zones, some typical RMi values for the most common conditions are given in Table
2. They may be used for estimates at an early stage of a project, or for cases where the composition of the
zone is not known. The approximate RMiz values are based on assumed representative block volumes for
the various types of zones.

Table 2. Typical RMi values for various types of crushed zone (assumed common values)
Average uniaxial Average joint Approx. block Approx. typical Approx. block
Weakness zone compressive condition factor, volume, value, diameter,
strength, σc: MPa jC Vb: m3 RMiz Db : m
Coarse fragmented zones 100 0.5 0.01 2 0.2
Small fragmented zones 100 0.5 0.0001 0.3 0.06
Clay-rich (simple) zones 80 0.1 0.01 0.3 0.2
Clay-rich (complex) zones 40 0.1 0.001 0.03 0.12
Clay zones* 0.1 0.1 (nominal) 1 cm3 (nominal) 0.05 0.01
*For zones with mainly clay, approximate support estimates may be carried out using a nominal block volume of Vb = 1 cm3 = 0.000001 m3

5 Finding RMi graphically

The diagram in Figure 3 can be used to find JP when Vb and jC are known from field observations or
estimated from site descriptions. The values or ratings of the input joint features incorporated in JP
are shown in Table 1.

Figure 4.
Diagram to
find RMi

Example: With block volume Vb = 10dm3 and jC = 0.2 (for clay coated joints), RMic = 1. As the uniaxial
compressive strength of rock in this example is σc = 150MPa and tight interlocking (IL = 1), the RMi = RMic x
150/100 x 1 = 1.5 (if the joints had been undulating, rough (jC = 3), RMic = 8 and the RMi = 12)
www.RockMass.net
Dec. 2014
 5

As RMi is a measure for the strength of a rock mass it can be applied in several applications. The main ones
are shown in Figure 5.

Figure 5.
Applications of the
RMi

The RMi value expresses the quality (approximate strength) of the rockmass (rock penetrated by joints) as
a material in dry condition as in principle shown in Figure 2.

The site specific ground condition (similar to the Q-value) is expressed in the Ground condition factor
as Gc = RMi × GW × SL × C
where GW = groundwater conditions given as the water inflow into the underground opening
SL = stress level
C = an adjustment factor for wall or inclined roof (as in a shaft)

Table 3. The adjustment parameters used in the RMi support method. Note that the use of unit values = 1 for normal
or common conditions
roof 45o *) 60o *) wall
Roof and wall ( C )
1 2.2 3 5
K1 Stress level ( SL ) very low
0.1 0.5
low moderate
1
high
1.5
very high
**)

influence on stability → low moderate significant

Groundwater ( GW )
1 2.5 5
very slightly un- un- very un-
favourable
favourable favourable favourable favourable
Orientation of joints and zones ( Co )
0.75 1 1.5 2 3
K2 2+rando
1 set 1+random 2 sets m 3 sets 3+random 4 sets 4+random
Number of joint sets
( Nj ) 3 2 1.5 1.2 1 0.85 0.75 0.5
**) in massive rock, very high stresses may cause rock
K1 = C x SL x GW; K2 = Co / Nj *) roof in inclined shaft burst (covered in the support for continuous ground)

6 Support estimate by the RMi system in discontinuous (blocky) ground

The following two support parameters are used in the support chart in Figure 6:
I. The ground condition factor. The adjustments ( K1 ) to RMi may be found from Table 3: RMi × K1
II. The geometrical factor, expressed as the size ratio between the size of the opening (tunnel etc.) and
the rock blocks with adjustments for orientation and joint pattern (number of joint sets), given as

www.RockMass.net
Dec. 2014
 6

Sr = (Dt/Db) × (Co/Nj) = (Dt/Db) × K2

where Dt = The diameter or span of the tunnel or cavern (m). (For walls, the wall height Wt is used instead of Dt).
Db = The equivalent block diameter Db ≈ 3 Vb (in metre).
Co = An adjustment factor for orientation of the main joint set related to the tunnel or cavern, see Table 3.
Nj = an adjustment factor for the number of joint sets; and hence the freedom for the blocks to fall. Its
ratings in Table 3 can also be found from Nj = 3/nj, where nj is the number of joint sets (nj = 1 for
one set; nj = 1.5 for one set plus random joints; nj = 2 for two sets; nj = 2.5 for two sets plus random
joints; etc.) .

Table 4. Classification of joint orientation.

In one wall In opposite wall In roof
Term
Strike: ° Dip: ° Strike: ° Dip: ° Dip: °
Very favourable ≥ 70 All > 60 All > 60
Favourable < 70 ≤ 20 30-60 All 45-60
50-70 > 20 ≤ 30 ≤ 45 30-45
Fair All strikes
≤ 50 20-45 ≤ 30 ≤ 45
Unfavourable 30—50 ≥ 45 ≤ 30 > 45 15-30
Very unfavourable ≤ 30 ≥ 45 ≤ 30 > 45 ≤ 15

Figure 6. The RMi charts for estimates of rock support in blocky ground and weakness zones and
continuous ground (massive or highly jointed)

The support chart in Figure 6 shows the estimated total amount and types of support. It is based on installed
support in several tunnels in addition to the authors' experience from several tunnels and other underground
drill and blast excavations in Scandinavia.
www.RockMass.net
Dec. 2014
 7

7 Support estimate by the RMi system in continuous ground

Massive ground (i.e. rockmass with few joints) in moderate stress conditions has generally stable conditions
(see Figure 7), and does generally not need any support, except for some scaling work in drill and blast
tunnels. Massive, overstressed ground, however, requires support because the following time-dependent
types of deformation and/or failures may take place:
• squeezing in overstressed ductile rocks (such as schists, clayey rocks) and particulate rockmasses
(broken rocks)
• slabbing (spalling) or rock burst in overstressed brittle, hard rocks (such as granite, quartzite, marble and
gneiss).

Particulate materials (highly jointed rocks) generally require immediate support. Their initial behaviour is often
similar to that of blocky ground, i.e. the support chart in Figure 6 can be used. In overstressed (incompetent)
ground, time-dependent squeezing may, in addition to the initial instability, take place. However, for this type of
ground the support chart in Figure 7 needs updating, when more experience in this type of ground is available; or
separate calculations and convergence measurements should be performed.

Figure 7. RMi support chart for continuous ground.

8 RMi input applied in Hoek-Brown faiure criterion for rockmasses

The criterion applies the two factors m and s. Hoek and Brown (1980) adapted the RMR and/or the Q
classification systems as input of the ground conditions. However, it is easier to use the parameter JP of the
RMi for this, as shown below.

As RMi is similar2 as the unconfined compressive strength expressed in the criterion the factor s can be
expressed by the jointing parameter (JP) as:
s = JP2

In the beginning of the 1990s Hoek et al. introduced the ratio mb / mi , where the constant mb is the same as m
in the original criterion. It varies with the jointing and can be mathematically expressed as:

2 ½
The unconfined compressive strength of a rock mass according to the criterion is: σcm = σc × s and RMi = σc × JP
www.RockMass.net
Dec. 2014
 8

a) for undisturbed rock masses mb = mi × JP 0.64

b) for disturbed rock masses mb = mi × JP 0.857
mi varies with the intact rock and can be found from laboratory tests or from published tables (Hoek et al,
2002,or 2006, see reference list).

It should be born in mind that the Hoek-Brown failure criterion is only valid for continuous rock masses
(Hoek and Brown, 1980), i.e. massive rock or highly jointed rock masses.

8.1 Papers presented on the RMi system

Palmström A.: RMi – a rock mass characterization system for rock engineering purposes. PhD.
thesis, Oslo University, Norway, 1995, 400 p.

Palmström A.: Characterizing the strength of rock masses for use in design of underground structures.
Int. Conf. on Design and Construction of Underground Structures, New Delhi, 1995.

Palmström A.: Characterizing rock burst and squeezing by the rock mass index.
Int. Conf. on Design and Construction of Underground Structures, New Delhi, 1995.

Palmström A.: RMi - a system for characterizing rock mass strength for use in rock engineering.
Journal of Rock Mechanics and Tunnelling Technology, Vol. 1, Number 2, 1995, pp. 69-108.

Palmström A.: The weighted joint density method leads to improved characterization of jointing.
Int. Conf. on Recent Advances in Tunnelling Technology, New Delhi, India, 1996, pp. 9-14.

Palmström A.: Application of seismic refraction survey in assessment of jointing.

Conference on Recent Advances in Tunnelling Technology, New Delhi, 1996.

Palmström A.: RMi - a new practical characterization system for use in rock engineering.
Conf. Svenska Bergmekanikdagen 1996, Stockholm, pp. 39-63.

Palmström A.: The rock mass index (RMi) applied in rock mechanics and rock engineering.
Journal of Rock Mechanics and Tunnelling Technology, Vol. 2, Number 1, 1996

Palmström A.: Characterizing rock masses by the RMi for use in practical rock engineering. Part 1: The
development of the rock mass index (RMi). Tunnelling and Underground Space Technology, Vol. 11,
No. 2, pp. 175-186, 1996

Palmström A.: Characterizing rock masses by the RMi for use in practical rock engineering. Part 2: Some
practical applications of the rock mass index (RMi). Tunnelling and Underground Space Technology,
Vol. 11, No. 3, pp. 287-303, 1996

Palmström A.: A new method to characterize rock masses for applications in rock engineering.
Norwegian conference Bergmekanikkdagen, 1996, Oslo, 27 p.

Palmström, A.: Collection and use of geological data in rock engineering. ISRM News, 1997, pp. 21- 25

Palmström A.: Characterization of rock masses by the RMi for use in practical rock engineering (in Spanish).
Ingeo Tuneles, volume 2, in series Ingenieria de tuneles, Madrid, 1999, pp. 79 – 107.

Palmström A. and Nilsen B.: Engineering Geology and Rock Engineering. Handbook.
Norwegian Tunnelling Society, 2000, 250 p.

Palmström A.: Recent developments in rock support estimates by the RMi.

Journal of Rock Mechanics and Tunnelling Technology, vol. 6, no. 1, May 2000, pp. 1 – 19
www.RockMass.net
Dec. 2014