Presented at 1st International Conference on Applied Empirical Design Methods in Mining Lima, Peru June/2014

Empirical Design Methods Update (2014)

Rimas Pakalnis
Pakalnis & Associates, Professor Emeritus/Department of Mining Engineering, University of British
Columbia, Vancouver, BC, Canada
ABSTRACT: Empirical Methods are the most widely used design techniques employed within the mining
industry largely due to their success in the design of mine structures. One of the objectives of this conference
which will ultimately be part of a larger commission on Empirical Design Methods in Mining is to develop
guidelines for the application and development of applied empirical design methods for mining. This paper
forms the basis for the keynote for the Underground Excavations Design session of the conference. It
must be noted that this is the first conference on empirical design in mining. The paper summarizes the applications/implementation of empirical design methods that the author has compiled/coordinated over the past
thirty (30) years with over 150 underground mining operations contributing either through consulting, research, assessment, database, verification and/or implementation. These design guidelines have been implemented throughout the world in association with researchers, mine engineers, operators and legislators to arrive at design methodologies based upon past practice and future implementation and assessment in order to
ensure a safe and cost effective mining operation. This paper summarizes design curves that have been developed at the University of British Columbia over the past twenty (20) years with the methodology for design being a common thread for all the tools presented which requires the operator to address stress, structure
and the rock mass.
1 INTRODUCTION
Empirical derivations have gained acceptance over
1.1 Background
the last thirty years largely due to their predictive
Mining is a dynamic process, which requires in excapability since conventional methods of assessment
cess of thousands of cubic metres of openings to be
have the difficulty of identifying the jointed nature
developed daily over the life of a mining operation.
of the rock material, assigning properties thereto and
Those openings may be for development and/or proestablishing input parameters for subsequent numerduction purposes, however all cases must be deical evaluation. The process that the author has
signed so as to ensure the required behaviour. The
found to be of greatest value is to employ numerical
site engineer therefore, must design openings or pilcodes, analytical tools and observational approaches
lars after addressing all issues relevant to the design
as tools to the overall design process which will inand assessment of the overall behaviour.
corporate an empirical component towards the deA mine structure, whether it is a mine pillar or an
sign. Individually each is only a tool that requires
opening, is generally influenced by numerous blocks
the designer to address the factors most critical to
of intact rock. Individually their properties and bethe stability of the overall underground structure
haviour can be readily assessed on a laboratory
which includes stress, structure and the rock mass as
scale; however, when analyzed on a mine wide
shown in Figure 1.
scale, the interaction of the rock block and the rock
mass behaviour is difficult if not impossible to preThe field of empirical mine design can be broken indict employing solely a deterministic approach.
to the broad categories of pillar design and opening
design. Opening design has evolved from the tunnel
Design can be categorized into three domains:
span design methods of the 1960s and 1970s
(Bieniawski, 1976). Underground openings have
a) Analytical methods in the form of numerical
more complex geometries than found in tunnels,
codes, closed form solutions, classical physiwhich has required different design approaches.
cal and strength models.
Some of the first applied empirical mine design
methods for openings originated with the work of
b) Observational methods which rely on moniMathews (1980) and Laubscher (1977). Applied piltoring of the rock mass and verification of belar design work has developed from early coal minhaviour.
ing with applied design curves developed by Lunder
(1994) and used internationally. Extensive work in
c) Empirical methods which assess the stability
areas of drift and stope openings and pillar design
based upon past practice coupled with existhave been conducted in Australia and Canada with
ing mine behaviour for future prediction.

operate as economically as possible. This paper reviews updates to the design curves presented in
Pakalnis (2002) with subsequent implementation and
observation with respect to their use as design tools
for the mining industry.

STRESS ANALYSIS

FABRIC ANALYSIS
ROCK MASS CLASSIFICATION

INDUCED STRESS > ROCK MASS STRENGTH

YES

YES

BURST

MODIFY GEOMETRY
MODIFY MINING METHOD
SUPPORT
DESTRESS
SEISMIC
MONITORING
OTHER

ANALYTICAL DESIGN

2. DESIGN RELATIONSHIPS

NO

YIELD

IS STRUCTURE CONTROLLING STABILITY

YES

NO

NUMERICAL MODELLING
STRESS EFFECT

The reader is referred to an earlier paper (Pakalnis,

2002) on the details of the individual design curves
presented in this paper as the focus here is on the
modifications and subsequent use of the proposed
relationships over the past thirty (30) years

EMPIRICAL DESIGN SOLUTION

2.1 Stability Graph Method ELOS

LIMIT SPAN
SUPPORT WEDGE
SEQUENCE
OTHER

EXCAVATION AND MONITORING

RE-EVALUATE MINE PLAN

STRESS
STRUCTURE

ROCK MASS

Figure 1. Design methodology incorporating stress, structure

and the rock mass.

some of the most widely applied methods originating at the University of British Columbia, Canada
(Mathews, 1980, Lang, 1994, Potvin, 1988, Nickson,
1992). The more widespread use of empirical methods inherently made these systems more reliable,
since they are developed and refined only through
use and experience with verification through measurement with inputs from analytical sources coupled
with stress, structure and the rock mass as inputs.
Empirical design methods are evolving and the application of some of these methods is becoming confusing. The methods are being modified and used in
new ground conditions and mining environments,
sometimes without regard to the original conditions
the design methods were developed for. Know Thy
Database. Researchers potentially adding levels of
complexity to the empirical design methods that are
sometimes unwarranted and unnecessarily confusing. The purpose of this commission is to develop
guidelines for the application and development of
applied empirical design methods and to arrive at a
consensus for application.
The design methods summarized in this paper are
based upon a strong analytical foundation coupled
with extensive field observation to arrive at a calibrated empirical approach towards the solution to a
given problem. The methods presented in this paper
are compiled from extensive mine visits, literature
reviews, discussions among researchers and practitioners coupled with analytical and numerical assessments with the successful implementation at
mine operations. Two prime objectives inherent with
the design process are that the mine must be safe and

The Stability Graph Method for open stope design

was initially proposed by Mathews et al. (1980) and
subsequently modified by Potvin (1988) and Nickson (1992) to arrive at the Modified Stability Graph.
Stability was qualitatively assessed as either being
stable, potentially unstable or caved. Research at the
University of British Columbia quantified the degree
of slough by the introduction of the term Equivalent Linear Overbreak/Slough (ELOS) by Clark
(1998). This has resulted in the empirical estimation
of wall slough employing the Stability Number N
as shown in Figure 2 with respect to the Hydraulic
Radius.

Figure 2. Empirical estimation of wall slough (ELOS) after

Clark (1988).

2.2

Span Design Entry Methods

The Critical Span Curve (Lang, 1994) has been

augmented to include 292 case histories of stable
and unstable entry-type stope backs from six (6) different mines by Wang (2000). It expands on the
original database to incorporate lower rock mass ratings. This is shown in Figure 3. The critical span is
defined as the diameter of the largest circle that can

be drawn within the boundaries of the exposed back

as viewed in plan and shown in Figure 3. This exposed span is then related to the prevailing rock
mass of the immediate back to arrive at a stability
condition. The design span refers to spans which
have used no support and/or spans which include
pattern bolting (1.8m long bolts on 1.2m x 1.2m) for
local support. The rock mass rating as proposed by
Bieniawski (1976) is employed with corrections for
the presence of shallow joints (under 30o) by reducing the RMR by 10. The stability of the excavation is classified into three categories:

generally identified in terms of short term stability

(excess of three months) as the database is largely
based upon stoping methods that are of short duration.
Where back movement has been identified critical
magnitudes in excess of 1mm within a twenty-four
hour period dictate that the stope will not be accessed for a further twenty-four hour period. This is
based upon observations from the six (6) mines
comprising the database.
2.3 Pillar Design

a) Stable Excavation
no uncontrolled falls of ground.
no movement of back observed
no extraordinary support measures
have been implemented.
b) Potentially Unstable Excavation
extra ground support may have been installed to prevent potential falls of
ground
movement within back
increased frequency of ground working
c) Unstable Excavation
the area has collapsed
failure above the back is approximately
0.5 x span in the absence of major structure
support was not effective to maintain
stability.

Figure 3. Critical span curve for mine entry methods employing local support only.

The above approach is based upon rock mass instability and where discrete wedges have been identified they must be supported prior to assessment employing the Critical Span Curve. Stability is

A comprehensive pillar database that relates geometry, loading conditions, in situ rock strength and stability condition has been developed. Analysis of this
database has led to the development of the Pillar
Stability Graph as shown in Figure 4. A total of 178
stability cases have been included where each case
example represents a failed pillar, an unstable pillar
or a stable pillar as compiled by Lunder (1994). The
pillar data used to develop this methodology is a
compilation of seven (7) individual pillar stability
databases that have been published worldwide.
The information available is the pillar stability assessment, the predicted pillar load and the pillar geometry. Five of the seven databases originate from
massive sulphide deposits and all of the databases
have reported rock mass ratings in excess of 65%
representing good to very good quality rock mass
conditions. Figure 4 shows the observed conditions
that would be encountered for each stability
level. This is based upon detailed research work
showing that pillar stability levels could be assessed on a five level scale. This was simplified
in order to assimilate the data with the additional databases that were collected from the literature.
Two primary factors are used in this design
methodology, a geometric term that represents
pillar shape and a strength term that includes the
intact rock strength and the predicted pillar
load. Rock mass strength is dependent upon the
amount of confining stress applied to a sample.
In the case of mine pillars, the more slender a
pillar, the less confining stress will be available
resulting in a lower strength for a given rock
type. The Pillar Strength Graph was developed by
plotting the ratio of pillar load/UCS (unconfined
compressive strength) as shown in Figure 4 to the
Wp/Hp. The pillar width (Wp) is defined as the dimension normal to the direction of the induced stress
whereas

the pillar height (Hp) is measured parallel to the induced stress. The pillar load is measured at the core
of the pillar whereas the unconfined compressive
strength (UCS) is that recorded for the intact rock
comprising the pillar. The Pillar Stability Graph presented here is a collection of a wide range and type
of mine pillar.

3.1 Dead Weight

Dead Weight conditions if identified structurally
by the above test, Figure 5a, would conservatively
estimate that the maximum depth of failure would be
0.5 times the span. Typically a depth of failure
equivalent to either 0.5 to 0.3 times span is employed in practise (Beauchamp, 2006) however,
based upon past observed back failures and modelling (MacLaughlin et al., 2005) one largely observed
an arched back as shown in Figure 5b. It must be
noted that the depth of failure is assumed to be 0.5x
span, however, with adverse structure present one
must assess analytically (ie. Unwedge) the volume
of the wedge to arrive at the true volume which will
generally be less than 0.5 x span. The potential failure zone is then approximated as shown in Figure 5b
with the bolt capacity as defined by the breaking
strength of the bolt as well as the bond strength of
the bolt passing though the failure surface as outlined in Table 1.

Figure 4. Pillar stability graph.

3. SUPPORT DESIGN
The reader is referred to an earlier paper by Pakalnis
(2008) on the details of the methodology towards
support that has been derived from over twenty (20)
mining operations which incorporates analytical and
empirical approaches towards assessing the Factor
of Safety (FS) for a mine opening. The approach is
to assess the potential for adverse structure employing an analytical approach such as provided by the
software program Unwedge (Rocscience, 2008).
The initial assessment as practised at the face is to
define if a dead-weight wedge is possible and this
employs a simple stereoanlysis pole plot as shown in
Figure 5a. This figure shows conditions for gravity
falls and how it is presented onto a stereographic
projection. The plot of great circles representing the
individual planes must encompass the centre in order for a gravity fall to occur whereas for a sliding wedge this does not occur. Similarly the pole
plot shows that a triangle drawn between the three
poles must encompass the centre in order to have a
gravity fall.

Figure 5a. Conditions for a) gravity fall and b) sliding instability for wedge within back of tunnel.

Figure 5b. Factor of safety analysis Dead Weight

The above approach assumes all bolts are plated

with the support capacity defined by the value of the
sum of the individual bolts which are defined by either the breaking strength or the bond strength of the
bolts passing through the failure surface whichever
is the lower of the two. The above is a conservative
approach as the analysis assumes a centre bolt at
mid span. The potential for structural instability was
identified as dead weight, however, a detailed analytical assessment such as provided by Unwedge
will result in a more realistic and less conservative
assessment as the depth of failure is most likely
much less than 0.5 x span but has to be identified in
the field as having this potential. A further constraint
is that the corner bolt should be in excess of 0.5 x
bolt spacing from the adjacent wall in order to ensure effectiveness unless strapped/shotcreted.
3.2 Rock Mass Assessment
The potential for rock mass instability is defined by
the Span Design Curve which employs local/no
support as shown in Figure 3 and detailed by Pakalnis, 2007. The depth of failure has been observed
in the absence of structural instability to be equal to

0.5 times span for the Unstable zone and approaching from 0-0.5 times the span within
the potentially unstable zone.

Figure 5c. Shotcrete as confining the rock mass into a

single unit.

Table 1. Support properties.

3.3 Surface Support

This is detailed in Pakalnis (2010) and summarized in this paper where the use of shotcrete as
confining the rock mass fabric into a single
support unit is assessed as shown in Figure 5c.

It must be recognized that a weak rock mass

will likely result in the individual rock
blocks falling between the bolts and therefore surface support is required to confine
the rock mass as a single support unit. It is
also critical to bolt through the shotcrete
membrane within a weak rock mass (RMR76
< 45%) to ensure that the bolts and surface
support are confined and do not act independently of each other. Surface support requirements for temporary and permanent
(storage rooms, crusher chambers and
maintenance shops) support correspond to
an Excavation Support Ratio (ESR) of 3
and 1.3, respectively (Grimstad and Barton,
1993). This is shown in Figure 5d with the
range of weak rock masses observed within
Nevada operations (20%-45%) and associated spans of 6m (20ft) and summarized in
Table 2. It must be recognized that the shotcrete is reinforced with bolts which confine
the potential block.
Table 2. Fabric support requirements (after Grimstad and
Barton, 1993) for 6m span.

Figure 5e. Intersection support Dead Weight.

4. WEAK ROCK MASS DESIGN

Figure 5d. Surface support (after Grimstad and Barton, 1993)

3.4 Intersection Support

Intersection support as shown in Figure 5e employs
the methodology outlined in this section in terms of
dead weight support and surface support as shown
in Figure 5d. The volume of failure is approximated
by a right angle cone having a height equal to 0.5 x
diameter with support capacity defined by the value
of the sum of the individual bolts which are defined
by either the breaking strength or the bond strength
of the bolts passing through the failure surface
whichever is the lower of the two. The above is a
conservative approach as the analysis assumes a
centre bolt at mid span. The potential for structural
instability was identified as dead weight, however,
a detailed analytical assessment such as provided by
Unwedge will result in a more realistic and less
conservative assessment as the depth of failure is
most likely much less than 0.5 x span but has to be
identified in the field as to this potential. A further
constraint is that the bolt past the cone boundary
must be greater than 0.5 x bolt spacing since one
would largely negate the support capability with
having a corner bolt unless confined by
straps/shotcrete.

A major study was undertaken by the US National

Institute for Occupational Safety and Health
(NIOSH) and detailed in Brady (2005) to address the
high frequency of roof-fall injuries being recorded in
Nevada mines. The Spokane Research Laboratory
and the University of British Columbia Geomechanics group focused on the development of safe and
cost-effective underground design guidelines for
weak rock masses having an RMR in the range of
15% to 45%. Figure 6a shows the existing empirical
database prior to the study where the rock masses
were largely classified in excess of 65% (RMR76).

Figure 6a. Schematic showing transition of weak rock mass to

stronger and existing database.

A weak rock mass upon review of site conditions

observed was presented as that having an RMR76 of
less than 45% and/or a rock mass Quality (Q) of under 1.0. This is based upon a database in excess of
twenty mining operations as detailed in Pakalnis
(2007). Weak ground conditions, ground support,
and mining methods used in several Nevada underground mines were observed. The RMR values were
calculated to update both span design calculations
and stability graphs, and a data-base on underhand
mining methods was developed to reflect existing
Nevada mining conditions. The immediate rock
mass was also characterized and analyzed in terms
of prevailing type of ground support, potential failure mechanisms, and rock behavior which resulted

in modifications/augmentation to the design curves

as shown in Figure 6b for the wall stability graph for
weak rock masses. An initial observation from Figure 6b is that the classical design curves (ELOS) as
shown in Figure 2 are inaccurate at low N' and hydraulic radius values. If hydraulic radius is kept below 3.5 m in a weak rock mass, the ELOS value was
recorded to be under 1 m. It appears a hydraulic radius less than 3 m would not result in ELOS values
much greater than 1 m.

Figure 6d. Loading of 5m x 5m face at Barrick Goldstrike.

The effect of blasting on weak rock masses was further assessed through a NIOSH sponsored initiative
and detailed by Caceres (2011) in his PhD dissertation Quantifying the Effect of Rock Mass Quality
on Peak Particle Velocity (PPV) for Underground
Drift Development. Figure 6e shows a relationship
derived from the PPV and Scaled Distance (SD) as a
function of rock mass quality obtained from multiple
records and operations to arrive at critical thresholds
for blast design.

Figure 6b. Wall stability graph as developed for weak rock

masses (Pakalnis, 2007)

4.1 Operations Guidelines

With weak rock masses, blast control is critical to
ensure that the rock mass is not further disturbed
from over blasting. Both Queenstakes SSX Mine
and Barricks Goldstrike operation guidelines for
blasting based on RMR values have been summarized in Figure 6c. In addition, the length of round
pulled is related to the rock mass rating for a 5m
supported back span, Figure 6c. An example is
where the RMR is 15-20% this would dictate that
only a 1.2m advance is possible otherwise failure of
the unsupported back will result. Spiling is recommended at these RMR thresholds. Figure 6d shows
the degree of loading of a development round at Barricks Goldstrike Mine for a 5m x 5m heading with
respect to the RMR.

Figure 6c. RMR versus round advance at Queenstake

(Pakalnis, 2007).

Figure 6e. PPV versus scaled distance for varying rock qualities.

A further observation amongst weak rock mass

mines was the success achieved through arching of
the back. It was found that for weak rock masses that
arching the back dramatically increased the overall
stability for a given span. This is schematically
shown in Figure 6f whereby the potential wedge
volume is significantly decreased employing an
arched back and the effectiveness of the applied
support increased as a greater length of bolt passes
beyond the failure plane. This has been shown to be
a major contributing factor to the overall stability for
mines operating within weak rock masses.

Figure 6f. Effect of arching on back of tunnel.

5. OTHER OBSERVATIONS
This section refers to empirical approaches developed with UBC researchers and industry sponsors to
address problems that assisted in understanding the
overall concerns and subsequent modification thereof in order to arrive at a safe and cost effective operation.
5.1 Rock Mass Assessment
A concern is the evaluation of RQD for weak rock
masses which requires a definition and interpretation
for sound core. The ISRM(1978) definition:
pieces of sound core over 10cm long that are expressed as a percentage of the length drilled. This is
further defined by Deere(1988) to only incorporate
good rock recovered from an interval of a borehole and not to include problematic rock that is
highly weathered, soft, fractured, sheared and jointed and counted against the rock mass. The ISRM
further identifies material that is obviously weaker
than the surrounding rock such as over consolidated
gouge is discounted as it is only able to be recovered
by advanced drilling techniques. The author incorporates the above augmented by weak rock mass operations within this database to arrive at the following as detailed in Milne et al, 1998. In practice, a
high RQD value does not always translate to high
quality rock. It is possible to log 1.5 metres of intact
clay gouge and describe it as having 100% RQD.
This may be true based on the original definition of
RQD, but is very misleading and gives the impression of competent rock. A parameter called
'Handled' RQD (HRQD) was introduced, Robertson
(1988). The HRQD is measured in the same way as
the RQD, after the core has been firmly handled in
an attempt to break the core into smaller fragments.
During handling, the core is firmly twisted and bent,
but without substantial force or the use of any tools.
This attempts to quantify sound core.

5.2 Underhand Cut and Fill

A major focus of study at UBC is the development
of design guidelines for mining under fill. This is detailed by Hughes (2014) in his Phd Dissertation
Underhand Cut and Fill Cemented Paste Backfill
Sill Beams. It incorporates empirical data along
with analytical assessments to arrive at design relationships that enable one to understand the overall
failure mechanism and extend upon the existing state
of knowledge in sill mat design (Pakalnis et al,
2006). Figure 8 relates the database of underhand
mining operations with vertical sidewalls with a FS
of 2.0. The chart is based on flexural instability employing fixed beam analysis with surcharge loading.
The observe is generally lower than strengths dictated by analytical method and this could be due to
QA/QC, confinement/other. The relationship found
between tensile and unconfined compressive
strength (UCS) for consolidated paste was well in
excess of 0.1 UCS which has been employed historically in the literature which indicates that the sill
mat is much stronger than assumed (Hughes, 2014).

Figure 8a. Design of undercut sill spans as function of sill mat

thickness, unconfined compressive strength and stope span.

Figure 7. Handled RQD.

MINE

%CEMENT SPAN
(m)

SILL THICKNESS UCS

(m)
(MPa)

COMMENTS
PASTE
DESIGN STRENGTHS GOVERN TIME
TO MINE UNDER(14D-28D)

REASON
UNDER FILL

RED LAKE MINE

10

6.1

3
(~0.6m gap)

1.5

STRESS ~2000m DEPTH

2a
2b
2c

ANGLOGOLD(1999 VISIT)
MURRAY MINE
(QUEENSTAKE-2004)

6.5
8%
8%

7.6
9.1
21

4.6
4.6
4.6

5.5
6.9
6.9

CRF
CRF DESIGN
MINED REMOTE - NO CAVE
2" MINUS AGG
GO UNDER A MIN OF 14D,
WALL CRF 5-6% BINDER
JAM TIGHT TO BACK/STEEP

WEAK RMR ~25%+

ESKAY

4 - 12

CRF(4MPa Design)
UCS is 11MPa(28Day)

WEAK RMR ~25%+

4a
4b
4c

TURQUOISE RIDGE

9
9
9

13.7
3.7
7.3

4
3
3

8.3
8.3
8.3

CRF TEST PANEL

CRF DRIFT & FILL
CRF PANEL

WEAK RMR ~25%+

MIDAS

2.7

3.4

CRF

WEAK RMR ~25%+

DEEP POST

6.75

4.9

4.3

4.8

WEAK RMR ~25%+

7a
7b
7c
7d
7e
7f
7g
7h

STILLWATER - NYE

10

1.8
2.4
3
3.7
4.3
4.9
5.5
6.1

2.7
2.7
2.7
2.7
2.7
2.7
2.7
2.7

0.7
0.3
0.5
0.7
1
1.4
1.8
2.3
2.9

CRF
GO UNDER IN 28DAYS
PASTE (FS=1.5)
GO UNDER IN 7DAYS-28DAYS)
(5% BINDER-0.5MPa UCS 28D)
(7% BINDER-0.7MPa UCS 28D)
(10% BINDER-1MPa UCS 28D)
(12% BINDER-1.2MPa UCS 28D)

MIEKLE STH
BARRICK

4.6-6.1

4.6

5.5

CRF

WEAK RMR ~25%+

Gold Fields - AU

10

4.45

CRF

WEAK RMR ~25%+

10

Stratoni Mine
TVX

12.8

6-9

WEAK RMR ~25%+

11

Galena - Coeur de Alene

10

3
(includes 0.9m air gap)

2.5

12

Lucky Friday - Hecla

(Gold Hunter)

2.4-4.6

3
(includes 0.6m air gap)

4.8

13

Newcrest
(Kencana Mine

12-24
vs dry tuff

6-8

1.2-1.5

14

Lanfranchi Nickel Mines

(Helmuth South)

4-8

6-12*
*inters

1.2-2

15

Cortez Hills
(Barrick)

7.8

6-11*
*inters

4.6

16

Andaychagua Mine
(Volcan)

14

5-15

3.5

16+

High Density Slurry

(78% WT SOLIDS)
10% Cemented Hydraulic Fill
(73-75% Wt Solids)
(UCS after 7 days)
GO UNDER IN 3 DAYS(2.4MPa UCS)
8% Paste(COARSE TAILS)
(no free water)
1.2MPa IN BACK AND 0.5MPa IN WALLS
DESIGN STRENGTHS GOVERN TIME
TO GO UNDER PASTE 7D-28D
SPAN 6m UNDER PASTE
SPAN 12m INTERSECTIONS CABLED(6m)
TO GO UNDER PASTE 14D
GO UNDER IN 28 DAYS
SPAN IS 6m WITH 11m AT INTERSECTIONS
MAXIMUM TOP SIZE 5cm(2")
CEMENTED AGGREGATE FILL
SPAN IS 15m
AGGREGATE FILL -3/4"

Figure 9. Effect of freezing on RMR76 (Roworth, 2013).

6. CONCLUSIONS
STRESS~ 800m

STRESS ~1000m DEPTH

STRESS ~2000m DEPTH

WEAK RMR ~25%+

STRESS ~850m DEPTH

WEAK RMR ~15%+

WEAK RMR ~15%+

Figure 8b. Underhand cut and fill database.

5.3 Effect of Freezing on Rock Mass Rating
The Cameco operations among others such as the
Kupol Mine of Kinross rely on frozen ground in part
for stability and/or water control. A Cameco sponsored initiative and detailed by Roworth (2013) in
her MASc dissertation Understanding the Effect of
Freezing on Rock Mass Behaviour as Applied to the
Cigar Lake Mining Method is detailed in this paper. Figure 9 shows a relationship whereby frozen
versus unfrozen ground was recorded in terms of
RMR76 and related to exposed span. The influence
of freezing on the rock mass quality was found to be
significant for very weak rocks and decreases exponentially with increasing rock mass strength. The
unfrozen RMR76 for RMR under 50% upon freezing
had a gain of up to 40 points whereas unfrozen RMR
greater than 50% had a frozen RMR of less than 10
point gain. Conservatively an increase of 20 occurs
ie. 30% unfrozen RMR is 50% frozen.

The approach towards design summarized in this

paper is one whereby existing databases have been
calibrated to analytical and empirical approaches
and modified according to observed mine behaviour. The tools have been used successfully to predict levels of dilution, pillar stability, opening stability, support requirements among factors that
have been trialed in the field. The methodology towards design must identify the potential for stress,
structural and rock mass instability and therefore
the design curves present only one part of the overall design process. It is critical that empirical tools
be employed to predict rock response by interpolating and not extrapolating where minimal data is
available. The approaches presented in this paper
are to be employed as a tool for the practitioner and
to augment the methodology with his/her own database and decision making process in order to arrive at a workable solution. The relationships developed here assist the operator/engineer to identify potential concerns thereby developing a safer work environment. These design guidelines have been
implemented throughout the world in association
with researchers, mine engineers, operators and legislators to arrive at design methodologies based upon past practice and future implementation and assessment in order to ensure a safe and cost effective
mining operation. This conference on empirical design is the first of many to come and allows one to
define concerns, find solutions and discuss similar
problems.
7. ACKNOWLEDGEMENTS
The students, the mines and the colleagues who
have contributed to the knowledge base.
8. REFERENCES

Beauchamp, L. 2006. Ground support manual. Mines and

Agregates Safety and Health Association, Ontario.
Brady, T., Martin, L., and Pakalnis, R., 2005. Empirical approaches for opening design in weak rock masses. Mining
Technology (Trans. Inst. Min. Metall. A), March 2005. Vol.
114 A1.
Caceres , C. 2011. Quantifying the effect of rock mass quality on peak particle velocity for underground drift development. PhD Thesis, University of British Columbia, Vancouver, B.C.
Clark, L 1998. Minimizing dilution in open stope mining with
focus on stope design and narrow vein longhole blasting.
MASc Thesis, University of British Columbia, pp. 336.

low 1000 m in hard rock. Canmet DSS serial no: 0SQ8000081., Canada.
Pakalnis, 2002. R. Empirical design methodsUBC Geomechanics update. In NARMS-TAC : Mining and Tunneling
Innovation and Opportunity, R. Hammah W. Bawden, J. Curran, and M. Telesnicki, eds.,Toronto, Canada
Pakalnis, R., Caceres,C.,Clapp,K,Brady, T., Williams,
T.,Blake,W and M. MacLaughlin. 2006. Design spans underhand cut-and-fill mining., CIMM Bulletin, Vol. 99, No
7pp
Pakalnis, R. 2008. Methodology towards ground support.
Strategic versus tactical approaches in mining, Quebec, Canada.

Deere, D.U. & D.W. Deere D.W. 1988. The rock quality index in practice. rock classification systems for engineering
purposes. ASTM STP 984, l. Kirkaldie ed. pp. 91-101.

Pakalnis R., Roworth M., Caceres C. and Lourence P. 2010.

Ground support methodology employing shotcrete for underground mines. 100 years of mining research symposium SME
Phoenix.

Grimstad,E. and Barton, N. 1993. Updating the Q-System for

NMT. Proc. Int. Symp. On Sprayed Concrete, Fagernes, (eds.
Kompen, Opsahl and Berg), Oslo: Norwegian Concrete Association.

Pakalnis, R., Brady, T., Hughes, P. and McLaughlin, M.,

2007. Design guidelines for underground mining operations
in weak rock masses. 1st Canada-US Rock Mechanics Symposium, Vancouver, Canada.

Hughes, P. 2014. Underhand cut and fill cemented paste

backfill sill beams. PhD Thesis, University of British Columbia, Vancouver, Canada.

Potvin, Y. 1988. Empirical open stope design in Canada. PhD

Thesis, University of British Columbia, pp. 276.

ISRM, 1978. Suggested methods for rock mechanics commission on standardization of laboratory and field tests. Int. J.
Rock Mech. Min. Sci. & Geomech. Abstr. Vol. 15. Pp. 319368.

Robertson, A.M., 1988. Estimating weak rock strength,

AIME-SME Annual Meeting, Phoenix, AZ., Preprint #88-145.
Rocscience (2008) UNWEDGE software package. Toronto,
Ontario

Lang, B., 1994. Span design for entry type excavations.

M.Sc. Thesis, University of British Columbia, Vancouver,
Canada.

Roworth, M. 2013.Understanding the effect of freezing on

rock mass behaviour as applied to the cigar lake mining
method. MASc Thesis, University of British Columbia, Vancouver, Canada.

Laubscher, D.H. 1977.Ggeomechanics classification of jointed rock masses - mining applications. Trans. institution of
mining and metallurgy (sect. a: mining industry). 86, p. A1A8.

Wang, J., Pakalnis, R., Milne, D. and B. Lang. 2000. Empirical underground entry-type excavation span design modification. 53rd Annual Conference of the Canadian Geotechnical
Society, pp. 8.

Lunder, P. 1994. Hard rock pillar strength estimation: an applied empirical approach. MASc Thesis, University of British
Columbia, pp.166.
MacLaughlin, M.M., R. Pakalnis, and Brady, T. 2005. A distinct element parametric study of failure modes around an
underground opening in rock masses of varying quality, Proceedings of the 40th Symposium on Engineering Geology &
Geotechnical Engineering, Utah..

Matthews, K.E.., Hoek,E., Wylie,D., and Stewart, 1981.

Prediction of stable excavation spans for mining at depths be-