PERFORMANCE PREDICTION ASSESSMENTS OF A HARD ROCK
TBM USED IN MINING DEVELOPMENT
F. J. Macias1*, L.N.R. Eide2, P.D. Jakobsen1,3, C. Jacobs4 and A. Bruland1
1 Department of Civil and Transport Engineering, NTNU, Trondheim, Norway.
2 Norwegian National Rail Administration, Norway
3 Norwegian Public Roads Administration, Norway
4 Stillwater Mining CO., Montana, USA
*Corresponding author: F.J. Macias
Cell phone: +47 404 91 650
Tunnel boring has shown to be a suitable excavation method on underground expansion. Mining
development requires often long tunnels with limited access points which have to be constructed on a
limited amount of time. In many cases, Tunnel Boring Machines (TBMs) can be a good choice of
tunnelling method. Planning and risk management is of greater importance for technical
successfulness and economic benefit on tunnel boring compared to drill and blast tunnelling.
Therefore, good predictions of performances and excavations costs take great importance for this
issue. A complete analysis of geological mapping from probe drilling cores, rock drillability testing
and TBM log data has been performed in order to analyze performances on a hard rock TBM used in
mining development. Assessments and comparison between actual and performance predictions by
using the NTNU prediction model for hard rock TBMs has been carried out.
Keywords: Hard rock TBMs, Mining development, Net penetration rate, Performance predictions, NTNU
In underground mining there are several methods for development. In hard rock conditions drill and
blast excavation (D&B) and tunnel boring machine (TBM) excavation are the main methods.
Roadheader excavation method might be a possible solution for grounds with suitable rock strength.
The choice of the more suitable excavation method for every project case is more complex than a
simple economic issue. Every project is unique and many parameters with different role are involved
in the decision being necessary to carry out a comprehensive and detailed study (Macias et al., 2014a).
Mining development often requires long tunnels to be performed on a limited amount of time. A
priori, the TBM method might be indicated as the most suitable for mining development tunnels.
Other parameters such as higher geological risk or limited flexibility compared to D&B method,
among others, might also be relevant on the excavation method selection. As mentioned in Macias et
al, (2014a), all the parameters involved in the choice of the excavation method should be considered
and carefully analyzed.
Performance predictions and costs are often decisive on the selection of the excavation method;
furthermore on planning and risk management of TBM projects. Good prediction facilitates the control
of risk and enables delays and budget overruns to be avoided.
For predictions of advance rate, cutter consumption and excavation costs on hard rock TBMs, the
main factor used is the net penetration rate (m/h). Penetration rate on hard rock TBMs is an interaction
of rock mass properties and machine parameters (Bruland, 1998).
Several prediction models for hard rock tunnel boring have been developed in the last few decades.
Some of them are Graham (1976), Farmer and Glossop (1980), Büchi (1984), CSM model (Rostami
and Ozdemir, 1993, Rostami, 1997), Luleå University of Technology (Nelson et al., 1994), NTNU
model (Bruland, 1998), QTBM (Barton, 2000), RME (Bieniawski, 2006), Gong and Zhao (2009) or
Hassanpour et al. (2011). Farrokh et al., (2012) examines the predictive capability of several
At the present, the Colorado School of Mines (CSM) model (1993, 1997) and the Norwegian
University of Science and Technology (NTNU) model (1998) are the prediction models most widely
used for performance and cost estimation on hard rock TBM tunnelling application.
A combination of the decisive rock mass properties and the relevant TBM parameters in order to
determine the main factors influencing the penetration rate predictions are considered in the NTNU
prediction model for hard rock TBMs. Predictions of advance rate, cutter wear and excavation costs
are also achieved using the NTNU prediction model.
The existing edition of the NTNU model (Bruland, 1998) is currently under review and revision. The
main focuses are updating the possible lack of TBM development, extend the range of application for
geology and consider influence of new parameters. A series of studies, included the present research,
is being carried out in order to achieve the goal described previously.
The present paper analyzes the capability of the NTNU prediction model to forecast performance from
continuous probe core drilling of an ongoing 5.5 m diameter open hard rock TBM used in mining
development in Stillwater Mine, Montana, USA. A complete analysis of geological mapping from
probe drilling cores, rock drillability testing and TBM log data has been performed. Assessments and
comparison between actual and predictions by using the NTNU prediction model for hard rock TBMs
has been carried out. The capability and accuracy on performance predictions using continuous probe
core drilling is also analyzed.
THE STILLWATER MINE
The Stillwater Mine is one out of two mines that the Stillwater Mining Company (SMC) operates
along the eastern portion of the J-M Reef in south-central Montana, USA. The reef is the world’s
richest known deposit of platinum group metals. The platinum group metals consist primarily of
palladium, platinum and a small amount of rhodium. The Stillwater Mining Company is the largest
primary producer of platinum group metals in North America (Luxner, 2014). Stillwater Mining
Company (SMC) has an extensive history with tunnel boring machines at the Stillwater and East
Boulder underground PGM mines in Montana.
The Blitz Project
The purpose of the Blitz Project is to develop a new mining area on the J-M Reef for additional PGM
production providing additional ore reserves, and possible increment of the total production of the
current Stillwater Mine. An important part of the Blitz project is to perform mining tunnels for
ventilation and haulage level (Luxner, 2014). The project started in 2010 and the company estimates
that to complete the full Blitz development will take about six years. An overview of the Stillwater
Mine including the Blitz Area is shown in Figure 1
The mining development level has a total length of 6, 975 meters and it is being performed by using a
TBM. In addition, ore body investigation is completed from the bored drive.
The Stillwater complex contains some of the oldest rocks occurring on the North American continent
and is an igneous saucer-shaped layered intrusion, which contains the PGM rich J-M Reef which is
The TBM part of the Blitz Project will be driven within layers of norite and gabbro-norite. The
targeted rock types for the drive are plagioclase rich rocks (more than 60 % plagioclase). Layers of
melano units can also be encountered at times. These melano units consist of more than 50 % dark
minerals such as olivine, bronzite and augite.
The compressive strength of the rocks in the area range from the weakest olivine rocks at
approximately 7 MPa to the hardest plagioclase rocks at about 140 MPa (Stillwater Mining Company,
2011). The norites encountered are typically between 70 and 85 MPa.
Several large transverse cross faults are expected to occur during the project (Luxner, 2014). These are
often highly faulted with small angular block size. The faults will usually contain a core of clay rich
fault gouge, at times as wide as 15 meters. The ground around the faults is also anticipated to be
highly fractured. It is expected that gripping to push the machine through these clay rich zones will be
a challenge, as the strength of the clay is no more than approximately 3.5 MPa. The sections evaluated
on the paper are not influenced by major faults.
The rock of the Stillwater Mine is an igneous rock with low porosity (Stillwater Mining Company,
2011). It is expected to find water in open fractures associated with the faults, and which will need to
be grouted off before allowing the TBM to continue.
In summary the Stillwater Mining Company has a good understanding of the geology encountered
during the Blitz project. However the probe drilling program is vital in order to add more detailed
information about the conditions ahead of the TBM in the areas beyond exiting mine infrastructure.
The machine being utilized in the Blitz Project is a main beam (open gripper) Robbins TBM with a
cutter head diameter of 5.5 m. A summary of some of the most important machine specifications is
given in Table 1.
Cutterhead diameter 5.5 meters
Number of disc cutters 33
Cutter diameter 483 mm (19 inches)
Recommended load per cutter 311 kN
Recommended operating cutterhead thrust 10 885 kN
Cutterhead power 6 x 330 kW = 1 980 kW
Cutterhead speed 0 - 10 RPM
Cutterhead torque 1 900 kNm at 10 RPM
Thrust cylinder stroke length 1.83 m
TBM weight Approx. 360 tons
Back-up weight Approx. 180 tons
For ground support purposes, the machine has been designed to enable installation of roof bolts, wire
mesh slats and possible use of ring beams. Shotcrete or cable bolts can also be applied when
TBM PERFORMANCE PREDICTION FOR HARD ROCK TBMs: THE
The philosophy of the NTNU prediction model is to combine the decisive rock properties and the
relevant machine parameters (Bruland, 1998; Macias et al. 2014c). Net penetration rate, cutter life,
advance rate and excavation costs are estimated on several steps by the NTNU prediction model for
hard rock TBMs. The model has had a successive development since the first version in 1976 by the
NTNU (former NTH). Table 2 shows the successive editions of the NTNU prediction model to date.
1st edition 1976
2nd edition 1979
(published in 1981)
3rd edition 1983
4th edition 1988
5th edition 1994
6th edition 1998
The last prediction model edition (Bruland, 1998) is based on data from almost 250 km. of tunnels
Rock boreability can be defined as the resistance (in terms of ease or difficulty) encountered by a
TBM when it penetrates in a rock mass composed of intact rock properties and rock mass parameters
(Macias et al. 2014a).
The rock parameters of most of the prediction models for hard rock TBMs can be divided into intact
rock and rock mass; NTNU model (Bruland, 1998), QTBM (Barton, 2000), RME (Bieniawski, 2006),
Gong and Zhao (2009) or Hassanpour et al. (2011).
The NTNU prognosis model for tunnel boring relies on the drillability of the rock through the Drilling
Rate IndexTM (DRI) as the main intact rock descriptor. Drillability influences penetration rate, cutter
life and thereby excavation costs.
DRITM is evaluated on the basis of two laboratory tests, the Brittleness Value (S20) test and the
Sievers’ J-Value (SJ) miniature drill test. S20 indicates the amount of energy required to initiate cracks
for crushing the rock. Regarding surface hardness, the Sievers’ J-Value (SJ) constitutes a measure of
the rock surface hardness or resistance to indentation. DRITM is an indirect measure of the required
breaking work and a good representation of the rock-breaking process under a cutter.
Related to the rock mass influence on performance prediction, it has been found that the main rock
mass boreability parameters are type, degree of fracturing, and orientation of the fracture system(s)
(Bruland, 1998; Macias et al. 2014a).
The rock mass fracturing factor (ks), characterized by the degree of fracturing and the angle between
the tunnel axis and the plane of weakness, is found to be the geological factor applying the greatest
influence on net penetration rate, and thus it has a major influence on tunnelling costs in TBM hard
rock (Bruland, 1998; Macias et al.,2014a). High rock mass fracturing means greater rock mass
boreability during hard rock TBM excavation.
Parameters for performance prediction
Rock parameters consist of intact rock and rock mass parameters. The rock parameters are combined
in a rock mass boreability parameter, the equivalent fracturing factor (kekv) while the machine
parameters are combined into one machine parameter, the equivalent thrust (Mekv).
Table 3 shows the rock properties and machine parameters influencing the net penetration rate.
ROCK PARAMETERS MACHINE
Index, DRITM Rock Mass
Number of cutters
The main machine parameter boring in hard rock is the average cutter thrust. When the thrust is
increased, the cutter will indent deeper into the rock surface and therefore transmit the energy from the
cutterhead to the rock more efficiently. Gross average cutter thrust is used in the NTNU prediction
model. This means that the total cutterhead thrust is divided by the number of cutters on the cutterhead
and also averaged over the time.
Other input of the model is the cutter diameter. Increasing the cutter diameter enables an increase of
the applicable cutter thrust. A correction factor for the cutter diameter is included in the model. It is
related to cutter ring edge width since variations in the cutter diameter means variations of the edge
width of the standard rings. The NTNU model does not consider the cutter edge width as an
Other parameters included in the model are the average spacing of the cutters on the cutterhead. The
average cutter spacing means the cutterhead radius divided by the number of cutters on the cutterhead.
The model does not consider the possible influence of the TBM cutterhead shape, e.g. flat and domed.
Most of the model database is associated with TBMs which have a cutterhead RPM according to the
maximum rolling velocity of the outer gauge cutter. Therefore the model does not consider the
influence of the cutterhead RPM on the penetration, despite that the cutterhead RPM has an influence
on the penetration rate per revolution (Bruland, 1998; Macias et al., 2014c).
As previously indicated, a research in order to include the RPM influence on performance predictions
and cutter life is in progress.
The NTNU model estimates net penetration rate (m/h), cutter life (h/cutter, sm3/cutter), machine
utilization (%), weekly advance rate (m/week) and finally excavation Costs (NOK/m).
The model allows developing sensitivity analysis of influence of one or more factors on the output
parameters. For example the geological risk can be analyzed keeping constant the TBM parameters
and varying the rock parameters.
The machine utilization derives from the estimated boring time divided by the total available time per
shift, day or week. It is based on the estimation of the time consumption in h/km. Machine utilization
must be analyzed from the TBM side and not for the tunnel as a whole.
The advance rate is estimated by the net penetration rate (m/h) and the machine utilization (%). The
model is based on averaged data over the complete tunnel length.
Related with excavation costs, the cost model is directly related to the models for penetration rate,
cutter life and advance rate. Normalized TBM costs, consumables costs and interest rates are
The cost model is based on detailed estimations of all excavation costs. In the 1998 version of the
model, the cost and possible extra time for rock support measures are not considered.
A complete analysis of field data was carried out. The field data comprises a laboratory testing,
geological mapping for performance estimation purposes from core sampling and gathering of TBM
Laboratory testing for drillability, abrasivity and strength has been performed at the NTNU/SINTEF
laboratory. Rock samples from rock chips and core samples were selected at tunnel site. The testing
comprised Drilling Rate Index ™ (DRI), Cutter Life Index™ (CLI), point load strength (Is50), Cerchar
Abrasivity Index (CAI), see Table4.
Laboratory index Value
Quartz content (%) < 1
Normalized Index, Is50 (MPa) 5.0
Indicated Compressive Strength, σc (MPa) 80.8
Since the rock mass is reasonably well known based on previous mining operations, rock samples
were selected from two locations along the tunnel.
According to Bruland (1998), the rock (norite) is classified in terms of drillability as a rock with a
“medium” DRITM that means high boreability and low resistance to boring. The Cutter Life IndexTM is
considered as representative for the abrasion process of the cutter ring. The rock is defined as “high”
CLITM. The Cerchar test defines the rock is defined as “high” abrasivity according to Alber et al., 2014
and “High” strength rock according to the definition given by ISRM (1978).
Correlation of Core Sampling Information to Tunnel Chainage
The TBM was allowed to bore 372 meters before the probe drilling program was started (Luxner,
2014). The presented research is based on a total TBM bore of 1456 meters which has been related to
eight probe drillings.
Geological and geotechnical data from all of these eight probe stops was treated to compile necessary
information to assess the geological conditions in the tunnel. The first point of treatment of the core
data was to correlate the information to a tunnel chainage. The cores are drilled in front of the tunnel at
an angle of approximately 5 degrees relative to the tunnel axis obtaining good assessments of the
geological conditions in the tunnel drive.
Figure 2 shows a simplified version of how the core drilling is situated related to both the tunnel axis
and fracturing structures.
The rock structure is well known due to the mining operation. So, it is possible to assume that the
geological and geotechnical properties of the rock mass will follow the structures that the tunnel drives
through; it is possible to correlate the properties of a certain point in the core sample to a point in the
tunnel. The angle α is stated to be 5 degrees for the straight ahead probes. The angle β between the
core samples and the fracturing structure is given for various intervals in the diamond drill logs from
each probe stop. For the practical use in the model established in this research, the possible error
related to the angles was considered neglected.
An overview of the probe stops with core drill hole numbers as well as approximated tunnel chainage
represented by each probe is shown in Table 5
Probe stop Core drill hole no. Tunnel chainage represented [m]
1 36 409 372 - 594
2 36 894 594 - 818
3 37 013 818 - 1025
4 37 293 1025 - 1198
5 37 412 1198 - 1344
6 37 487 1344 - 1471
7 37 591 1471 - 1695
8 37 682 1695 - 1828
As can be seen in Table 5, the different cores represent various lengths of tunnel, while normal
geological investigation usually divides the tunnel into intervals of equal length. As it is showed in
Figure2, there is an overlap between core samples. For geological mapping purposes, it has therefore
been decided to only include information from a core until the point where the next core started. The
margin of error in the correlation between core sample and tunnel chainage will increase with the
length of the core sample since the distance to the tunnel always increases.
Rock mass fracturing factor (k
) along the tunnel
The fracturing factor was obtained based on the calculated average spacing between fractures for each
core interval and the plane of weakness orientation based on Bruland (1998).
Distinguishing the type of weakness, fissure or joint, on core samples may result in errors due to the
difficulty to extrapolate the discontinuities continuity from the cores to the tunnel size. According to
the experience, fissure type is the most common type of plane of weakness. Hence, it was assumed
fissure for the entire section studied according to the observations on the core drills. Fissures are
considered to have less influence on the performances than joints (Bruland, 1998). By choosing
fissures, it will be avoided an overestimations of the influence of natural fractures in the rock mass,
thus it will lead to more conservative performance estimations.
Figure 3 and Figure 4 show the rock fracturing factor (ks) for the total length studied.
ANALYSIS AND DISCUSSION
Rock mass influence on the net penetration rate
Figure5 shows the variation of net penetration rate (m/h) and rock mass fracturing factor (ks) along the
tunnel along 50 m section lengths. High fracture factor (ks) values indicate easier rock mass boreability
giving rise to higher net penetration rate.
Direct comparison between the rock mass fracturing factor without DRI correction (Bruland, 1998) is
shown due to the constant DRI considered for the section studied.
Figure6 shows correlations of the rock mass fracturing factor (ks) with the net penetration rate (m/h)
and cutter thrust (kN/cutter). Good correlations are obtained in both cases which conform the findings
by Bruland (1998) and Macias et al. (2014b).
The significant decreasing of the thrust relation is due to operational reasons to avoid damage and
excessive wear to the cutters as well as due to avoid the collapse of the muck transportation system.
Figure7 (left) shows the correlation between the rock mass fracturing factor and the cutterhead speed.
The figure indicates how the RPM is adjusted according the rock mass degree of fracturing. Figure7
(right) shows the relation between net penetration rate (m/h) and penetration rate (mm/rev) with the
Figure7 shows as higher the ks, lower the rpm. This is mainly due to avoid machine damages and
collapse of the much transportation system.
Performance estimation by using the NTNU model
Estimation of net penetration rate (m/h) by using the NTNU model prediction model for hard rocks
has been achieved. The software Fullprof (NTNU, 2009) was used for this purpose.
The NTNU model philosophy and input parameters have been discussed previously. Several
assumptions are needed for the estimations:
‐ Estimations are performed for the sections defined previously (29 sections of approximately
‐ The same rock type is encountered, so constant drillability values are considered for the
‐ Weighted average of the applicable gross thrust is used on the estimations according to
‐ Weighted gross average cutter thrust is used for the estimations.
In addition, performance estimations with an “adjusted” applied cutter thrust have been carried out.
The thrust reduction is based on a proposed “propel react test”. The procedure of the proposed “propel
react test” is basically to push the TBM while rotating the cutterhead after retraction of the machine.
Frenzel et al, (2012) has proposed a procedure to consider the friction during boring. The necessary
thrust to overcome the weight and friction resistance is estimated. Cutterhead vibrations produced
during excavation will facilitate the machine movement which will lead to a reduction of the react
thrust (Frenzel et al, 2012).
The prediction cases analysed are:
Prediction A case: Using the total cutter thrust applied.
Prediction B case: Adjusting the cutter thrust applied with the “propel react test” value (42 kN/cutter).
Table 6 displays the statistics values of actual performances, prediction performances case A and B
using the SPSS PASW software (IBM, 2009)
Actual data Prediction A Prediction B
Weighted Average 4.79 5.22 4.22
Mean 4.89 5.30 4.31
Prediction deviation - 9 % -12%
Std. Error 0.12 (2.5%) 0.15 (2.9%) 0.13 (3.1%)
Std. Deviation 0.62 (13%) 0.78 (15%) 0.68 (16%)
Minimum 3.66 4.02 3.16
Maximum 6.53 7.04 5.79
Lower 4.66 5.01 4.05
Upper 5.13 5.60 4.57
The prediction deviations are 9% over and 12% underneath the actual values. Even considering the
worst case, 10%, does not mean a great deviation for hard rock TBM performance prediction
considering the uncertainties involved. Standard error and deviation of the performance predictions
presents the same range of values as actual penetration rates.
Figure 8 and Figure 9 show a good fitting of the performance predictions with the actual net
penetration rate. Uncertainties on rock testing and geological mapping from core samples should be
considered as causes of inaccuracy. It is shown slightly over trend performance predictions in
comparison with the actual data. The slightly overestimation of the performances might be explained
due to an overall increase of the rock fracturing on the core samples. The scale effect of the core
diameter on the fissure considerations might lead to a greater number of fissures than in a geological
back-mapping on the tunnel. Figure5 shows how the net penetration rate (m/h) and the rock mass
fracturing factor (ks) have a good correlation and the rock mass fracturing overestimation might be
general and not for specific fracturing values. A further geological back mapping on the tunnel is
planned in order to clarify and asses the possibility of overestimation of the rock mass fracturing factor
due to the core samples.
Other possible explanation might be due to the applied machine thrust used on the predictions.
Comparison of performance prediction using the total cutter thrust applied and using the cutter thrust
adjusted with the propel react thrust has been carried out.
The cutter thrust applied has a great influence on the performance predictions as is shown in Figure9.
For planning purposes, the cutter thrust should be adjusted according to the rock mass fracturing
values which are expected to be encountered. Higher ks gives lower thrust and vice versa. As a part of
the revision of the current edition of the NTNU model (Bruland, 1998), the cutter thrust to be used for
performance predictions is being analyzed.
Box plots showing the median, quartiles, extreme values and outliers. Figure 10 shows the box plots
and the error bar graph of the actual performance and prediction cases data. The calculations have
been carried out by using the SPSS PASW Statistics software (IBM, 2009).
The data analyzed is used for the updating and adjusting process on the NTNU prediction model for
hard rock conditions.
Risk analysis of performance predictions
A risk analysis considering the variability of the rock mass properties, intact rock (DRITM) and rock
mass (ks), and the uncertainty of the machine parameters (cutter thrust) has been completed. The
Fullprof software (NTNU, 2009) has been used to generate the estimations.
Table7 shows the distribution type and main values of the parameters used for the risk analysis as well
as the net penetration rates (m/h) estimated and actual.
deviation Min Max
DRITM Normal 54.5 5.5 49 60
ks Triangular 2.16* - 1,28 3,36
(kN/cutter) Triangular 291* - 195 338
predictions (m/h) Normal 4.94 0,89 2,44 8,16
(m/h) Normal 4.79* 0,62 3,66 6,53
*Weighted average over boring time
According to Macias et al. (2014), a triangular distribution can be assumed for the rock mass
fracturing factor (ks). For the cutter thrust applied, due to the displaced weighted average, a normal
distribution does not fit with the data and a triangular distribution is more adequate.
As it is show on Table7, the risk analysis result in a good prediction of the net penetration rate but
more scattered. This is mainly due to the combination of high rock fracturing and high cutter thrust
values and low rock fracturing and low cutter thrust values which do not match with the reality as
previously discussed, Figure6. Relations between rock mass fracturing factor and cutter thrust applied
should be considered on performance predictions.
The following conclusions are listed following:
‐ The rock mass fracturing factor (ks) has been proven to be a good representative of the
rock mass influence on performance prediction in agreement with Macias et al.,
‐ Good level of predictions has been obtained by using the NTNU model for the section
studied. Performance predictions slightly overestimated (9 %)
‐ The rock mass fracturing factor (ks) has been indicated to be suitable for rock mass
assessments on rock cores. A further work planned consists in a complete geological
back-mapping in terms of performance predictions in order to assess the geological
assessment based on core samples.
‐ The cutter thrust applied is the main machine parameter on performance predictions
by using the NTNU model. A brief discussion is presented and used for further update
and revision of the current edition of the model.
‐ A risk analysis considering rock properties variability and machine uncertainties has
been carried out. The results show a good prediction of the net penetration rate but
more scattered. Combination of high rock fracturing and high cutter thrust values and
low rock fracturing and low cutter thrust values do not usually happen on tunnel
boring and result on a wider range of predictions.
‐ It should been considered that the present research has been completed with a high
level of detail which is not always possible in the early stage of the TBM projects. A
risk analysis including variability and uncertainties should be always carried out.
The authors would like to thank the research project “Future Advanced Steel Technology for
Tunnelling” (FAST-Tunn). This project is managed by SINTEF, and funded by the Research Council
of Norway. The Robbins Company, BASF Construction Chemicals, the Norwegian Railroad
Authorities, Scana Steel Stavanger, BMS steel, the LNS Group and Babendererde Engineers are
industrial partners and co-founders. NTNU is a researcher partner in the project responsible for three
We also extend our thanks to the Stillwater Mining Company staff for their support during the field
work and for sharing data.
Alber M., Yarali O., Dahl F, Bruland A, Käsling H, Michalakopoulos T N, Cardu M, Hagan P, Aydin H,
Özarslan A (2014). ISRM Suggested Method for Determining the Abrasivity of Rock by the CERCHAR
Abrasivity Test. Rock Mech Rock Eng (2014) 47: 261-266.
Barton N (2000). TBM Tunneling in Jointed and Faulted Rock. Balkema, Rotterdam.
Bieniawski von Preinl ZT, Celada Tamames B, Galera Fernandez JM, Alvarez Hernandez M (2006). Rock mass
excavability indicator: new way to selecting the optimum tunnel construction method. Tunnelling and
Underground Space Technology 21 (3-4), 237
Bruland A (1998). Hard Rock Tunnel Boring: Vol 1 – 10, Ph.D. Thesis. Norwegian University of Science and
Technology (NTNU), Trondheim, Norway.
Dahl F, Bruland A, Jakobsen PD, Nilsen B, Grøv E (2012). Classifications of properties influencing the
drillability of rocks, based on the NTNU/SINTEF test method. Tunnelling and Underground Space Technology
Eide, LNR (2014). TBM Tunnelling at the Stillwater Mine. Master Thesis. . Norwegian University of Science
and Technology (NTNU), Trondheim, Norway.
Frenzel C, Galler R, Käsling H, Villeneuve M (2012). Penetration tests for TBMs and their practical application.
Geomechanics and Tunnelling 5 (2012), No. 5: 557-566.
Farrokh E, Rostami J, Laughton, C (2012). Study of various models for estimation of penetration rate of hard
rock TBMs. Tunnelling and Underground Space Technology 30: 110-123
Gong QM, Zhao J, (2009). Development of a rock mass characteristics model for TBM penetration rate
prediction. International Journal of Rock Mechanics and Mining Science 46 (1), 8–18.
Hassanpour J, Rostami J, Zhao J (2011). A new hard rock TBM performance prediction model for project
planning. Tunnelling and Underground Space Technology 26: 595-603
IBM (2009). SPSS PASW Statistics 18 software. Polar engineering and consulting.
Luxner T (2014). Stillwater's Blitz TBM Project Update. SME Annual Meeting, Feb. 23-26 2014, Salt Lake
Macias FJ, Bruland A (2014a). D&B versus TBM: Review of the parameters for a right choice of the excavation
method. The 2014 ISRM European Rock Mechanics Symposium (Eurock 2014), Vigo, Spain.
Macias FJ, Jakobsen, P.D., Seo, Y, Bruland, A (2014b). Influence of the rock mass fracturing on the net
penetration rates of hard rock TBMs. Tunnelling and Underground Space Technology 44: 108-120.
Macias FJ, Jakobsen PD, Bruland A, Log S, Grøv E (2014c). The NTNU Prediction Model: A Tool for Planning
and Risk Management in Hard Rock TBM Tunnelling. Proceedings of the World Tunnel Congress 2014 -
Tunnels for a better Life, Foz do Iguazu, Brazil.
Macias FJ, Jakobsen PD, Bruland A (2014d). Rock mass variability and TBM prediction. The 2014 ISRM
European Rock Mechanics Symposium (Eurock 2014), Vigo, Spain.
NTNU (2009). Fullprof software 2009. Norwegian University of Science and Technology (NTNU), Trondheim,
The Robbins Company (2011). Technical Proposal #032911-2O-T-REV 3 for the Supply of a Remanufactured
18'-0" (5.5 m) Hard Rock Open Gripper TBM, Back-up System and Auxiliary Equipment for the Footwall Drive
Tunnels, Stillwater Mine Project.
Stillwater Mining Company (2011). Geotechnical Baseline Information Regarding the Footwall Drive Tunnels,
Stillwater Mine Project. Blitz Tunnel Construction, Exhibit C – Scope of Work.
Stillwater Mining Company (n.d.) Available at: http://www.stillwatermining.com (Visited 20.11.14)