SME Annual Meeting

March 1-3, 1999, Denver, Colorado

Preprint 99-90
DESIGNING AND MANAGING UNPAVED OPENCAST MINE HAUL
ROADS FOR OPTIMUM PERFORMANCE

R. J. Thompson
A. T. Visser
Univ. of Pretoria
Pretoria, South Africa

ABSTRACT inadequate design increase, vehicle operating and maintenance

costs also increase prohibitively.
The expansion of surface mining has led to the development of
very large off-highway trucks currently capable of hauling The operating performance of a pavement can be subdivided into
payloads in excess of 290t. Mine haul roads have historically three distinct design categories as defined in Figure 1. In addition
been designed empirically, relying heavily on local experience; to the mechanical engineering challenges involved in the design
the trend in increasing truck size will thus render the current and construction of large dump trucks, specific mining
pavement systems inadequate. Since truck haulage costs can engineering problems arise, primarily as a result of the gross
account for 10%-50% of the total costs incurred by a surface vehicle mass (GVM) and the resultant loading applied to the
mine, not only would the maintenance costs of existing roads of pavement. Taking a typical example of a 180t capacity rear dump
inadequate design increase, vehicle operating and maintenance truck, the GVM of 317t results in a maximum dual wheel axle
costs would also increase prohibitively. There is thus a need for load of 2 086kN. In comparison, public road authorities in South
improved design technologies encompassing the construction and Africa currently permit a legal maximum dual wheel axle loading
management techniques of mine haul roads, appropriate for the of 90kN.
wheel loads of vehicles now in use.

The aim of the paper is to present the different components of a

mine haul road design and management system and to
demonstrate the value of its application through case studies. The
system has been applied at several South-African surface mines
and distinct benefits have been derived. The improved structural
design of a new road resulted in a 29% saving in construction
costs and also provided better service, whilst the optimal selection
and management of wearing-course materials provided better
functionality at lower total transportation operating costs.

INTRODUCTION

Surface mining in South Africa accounts for over 66% of the

total run-of-mine tonnage mined. Over 70% of copper, 80% of
ferrous metals and 95% or 65 million tons of the land's industrial
minerals are mined by surface mines and quarries. In the coal
mining industry, over of 40% or 106 million tons run-of-mine coal
was produced by opencast methods which require, inter alia, the Figure 1 Three Components of a Total Haul Road Design Strategy
transport of raw coal from the pit to the loading or transfer point.
In any surface mining operation, the transport of ore, and to a
lesser extent waste, is accomplished by large haul trucks running Equally important as the structural strength of the design, is the
on haul roads that have, at best, been empirically designed with functional trafficability of the pavement. This is dictated to a
little or no recognition of the consequences of inadequate design large degree through the choice, application and maintenance of
on cost per ton hauled, operational efficiency or safety. wearing course materials. The current functional performance
Considering that truck haulage costs can account for up to 50% of analysis methods are subjective and localised in nature and any
the total costs incurred by a surface mine, it is of paramount deterioration in pavement condition consequently hard to assess.
importance that these costs are minimised. This becomes all the Poor functional performance is manifest as poor ride quality,
more critical as tonnage increases and larger haul trucks are excessive dust, increased tyre wear and damage and an
deployed. Not only do the maintenance costs of existing roads of accompanying loss of productivity. The result of these effects is

1 Copyright © 1999 by SME

seen as an increase in overall vehicle operating and maintenance the sub-grade to an acceptable level and the upper layers must in
costs. a similar manner protect the layers below. Using this premise,
the road structure should theoretically provide adequate service
The maintenance aspect of haul road design cannot be over its design life.
considered separate from the structural and functional design
aspects since the two are mutually inclusive. Design and In an attempt to obtain satisfactory service over a road's design
construction costs for the majority of haul roads represent only a life, pavement design models can be used to predict performance
small proportion of the total operating and maintenance costs. over a wide range of traffic loads and road structural designs.
Whilst it is possible to construct a mine haul road that requires no Pavement structural design is the process of developing the most
maintenance over its service life, this would be prohibitively economical combination of pavement layers (in relation to both
expensive, as would the converse but rather in terms of operating thickness and type of materials available) that is commensurate
and maintenance costs. An optimal functional design will include with the in-situ material and traffic to be carried over the design
a certain amount and frequency of maintenance (watering, grading life. Typical pavement design terminology is shown in Figure 2.
etc.) and thus maintenance can be planned, scheduled and
optimised within the limits of required road performance and
minimum vehicle operating and road maintenance costs. The
major problem encountered when analysing maintenance
requirements for haul roads is the subjective and localised nature
of the problem; levels of functionality or serviceability being user-
and site-specific. No guidelines exist concerning maintenance
management and scheduling for specific levels of functionality,
nor the cost implications thereof, both in terms of vehicle
operating and road maintenance.

Geometric design refers to the layout and alignment of the road,

in both the horizontal (curve radius, etc.) and vertical (incline,
decline, ramp gradients, cross-fall, super-elevation etc.) plane,
stopping distances, sight distances, junction layout, berm walls,
provision of shoulders and road width variation, within the limits
imposed by structural, functional and maintenance design
parameters. The ultimate aim is to produce an optimally efficient
and safe geometric design and considerable data already exists
pertaining good engineering practice in geometric design, suffice
to say that an optimally safe and efficient design can only be
achieved when sound geometric design principles are applied in Figure 2 Haul Road Structure Terminology
conjunction with the optimal structural, functional and
maintenance designs (Thompson et al, 1997). Therefore by
addressing the structural, functional and maintenance design In general terms, applied load, subgrade strength and the
problems, an integrated haul road design strategy combining mine pavement structural thickness and layer strength factors
layout, construction techniques, available material and road predominantly control the structural performance of a haul road.
maintenance equipment with hauler choice was developed to The approach adopted in determining a suitable mechanistic
optimise a particular mining situation. structural design involved the quantification of the factors given
above for existing haul roads to determine the efficacy of the
The aim of this paper is to summarise the improved haul road various design options. To fully characterise the structural
design and management techniques which have been developed performance of existing or future designs of haul roads, each
specifically for surface mines. A new mechanistic structural factor was analysed at various levels using a designed factorial
design methodology is presented which can reduce construction experiment. Whilst only the principal developments resulting
cost and improve the structural strength of the pavement. The from this work are presented, further details are given by
improved functionality of a pavement is addressed in terms of the Thompson and Visser (1996).
optimum selection of wearing course materials, based on road-
user acceptability criteria. Finally, the concept of maintenance
management, in which an optimal maintenance strategy is sought Empirical Structural Design Techniques
which minimises both road-user and road maintenance cost
elements is addressed in terms of the various cost models required The findings of a United States Bureau of Mines (USBM) study
to evaluate such a system and the typical cost-benefits associated (Kaufman and Ault, 1977) were that haul road structural design
with its adoption. techniques were entirely empirical, mine operators often electing
to forego the use of sub-base materials and accept infringements
on mobility in the interest of economics. The USBM work was
HAUL ROAD STRUCTURAL DESIGN one of the first annotated applications of the California Bearing
Ratio (CBR) cover thickness design technique for mine haul
Mine haul roads deteriorate with time due to the interactive roads.
effort of traffic load and specific subgrade material strengths and
structural thicknesses. Vertical compressive strains induced in a The CBR method (Boyd and Foster, 1950) has been widely
pavement by heavy wheel loads decrease with increasing depth applied to the design of mine haul roads. The CBR method of
which permits the use of a gradation of materials and preparation haul road design was based on empirical results relating to the
techniques; stronger materials being used in the upper regions of design of asphalt-surfaced airfield pavements for wheel-gear loads
the pavement. The pavement as a whole must limit the strains in up to 4 400kN. When aggregate-surfaced mine haul roads are

2 Copyright © 1999 by SME

considered in conjunction with stabilised bases, albeit at similar Authorities TRH14, 1985) would appear most appropriate, based
load levels, the same approach is of questionable validity. on those mine sites where wearing course layers exhibited
Recourse to a mechanistic analysis was found necessary to adequate structural performance.
quantify the extent of any over- or under-design inherent in the
CBR approach since the latter technique cannot assess each layer The strains induced in a pavement are a function of the effective
independently in terms of its load bearing capacity. If a single elastic modulus values ascribed to each layer in the structure. In
pavement layer is assumed, the technique can be used to generate order to facilitate mechanistic design of mine haul roads, some
reliable estimates of total cover over in-situ material, but where indication of applicable modulus values was required. This was
multi-layer structures are involved and the optimal design is achieved by considering the individual layer modulus values
sought, a mechanistic approach is more appropriate (Thompson generated by the mechanistic analysis of existing pavements and
and Visser, 1998). comparing these values to established modulus values and the
associated material classification. Pavement layers comprised
generally a G4-G6 gravel or low quality gravel where local mine
South African Mechanistic Design Method ferricrete is used and a modulus range of 150-200MPa was
The South African Mechanistic Design Method (SA-MDM) is proposed for these gravels when used as a wearing course and 75-
based on a theoretical model of pavement layer response 100MPa for the same material when used as a base or sub-base
parameters (stresses, strains and deflections in each layer) by layer. Values for the modulus of the in-situ sub-grade material
assuming a linear-elastic multi-layer system. Empirically derived are very much site and material specific and the use of Dynamic
limiting design criteria are then used with which to assess the Cone Penetrometer (DCP) (Kleyn et al, 1982) derived CBR
pavement under the specific loading conditions, thereby values in conjunction with published data provided the most
determining the level of service and in turn, the time at which tractable approach to ascertaining suitable modulus values for this
some maintenance or rehabilitation would be required. The material. Table I relates the recommended elastic modulus values
method has been applied mostly to the design of asphalt for granular and in-situ materials comprising the pavement layers.
pavements on public roads (South African Roads Board, 1994). To facilitate the choice of suitable modulus values for in-situ
materials, the associated range of CBR values (derived from DCP
probing) and abbreviated specifications are given.
Recommended Mechanistic Structural Design Technique
The assessment of existing empirical structural design Recommendations regarding the structural design of surface
guidelines for mine haul roads indicate that a mechanistic mine haul roads are centred on the inclusion of a dumprock layer
approach to structural design would be more appropriate. A within the structure. This design is based upon the findings of the
number of mine roads previously assessed using the empirical mechanistic analysis of roads which incorporate a stabilised layer.
approaches were comparatively assessed using a mechanistic This effect can be adequately reproduced by using mine dumprock
approach. Pavement deflection profiles generated from Multi- or parting material in place of the stabilised layer. The optimal
depth Deflectometer installations in these pavements were location of this layer is immediately below the wearing course
analysed with the aid of the ELSYM5A computer program, the layer, thereby reducing deflections (and consequent deformation)
results of the multi-layer linear elastic analyses were then used to in the lower layers to a minimum. Using this approach, a reduced
deduce acceptable design criteria for haul road structural design. structural thickness was realised without the attendant
A structural performance categorisation was used as a guide to the deformation and reduction in structural performance level that
efficacy of the various existing haul road designs. Strains would otherwise be evident without a rock layer. Figure 3 shows
generated from the multi-layer elastic solution for the particular a comparison between the recommended mechanistically-derived
road test section were then compared with the structural optimal design and that derived from the empirical CBR-based
performance categorisation to established suitable design criteria. cover-curve design approach, based on an in-situ material
modulus value of 85MPa, a 429kN dual rear wheel load and a
630kPa contact stress. It is clear that the optimal mechanistic
Two design criteria were proposed with which to assess the design reduces damaging vertical compressive strains to below
structural performance of mine haul roads, namely factor of safety the critical 2000 microstrain limit and is also associated with
(FOS) for the two uppermost layers and vertical elastic reduced material volumetric and compaction requirements.
compressive strain for each layer below the top layer. The
vertical strain criterion correlated well with structural A cost comparison compiled from contractor tender unit costs for
performance; those mine sites exhibiting poor performance and an the construction of a road designed according to the empirical
associated excessive deformation/maximum deflection were seen CBR cover curve technique and one constructed according to the
to be associated with large vertical compressive strain values in mechanistically derived optimal design was undertaken to
one or more layers. From an analysis of the data it was evident ascertain if any cost advantage was associated with the
that an upper limit of 2000 microstrain should be placed on layer mechanistic design. The variable costs taken into account were
strain values. Strain values exceeding this value were shown to those of the volume and area of materials required and the
be associated with unacceptable structural performance. The associated costs of placing and compaction. An equivalent cost of
depth of influence at which load induced stresses are no longer the optimal mechanistic design was calculated from the actual
felt was identified at approximately 3000mm pavement depth. construction costs of the CBR-based design. It was found that a
The absence of deflections below this depth may be ascribed to 29% variable cost saving was realised, by virtue of the reduced
the low stresses induced in the in-situ material. material volumetric and compaction requirements associated with
the mechanistic optimal design. In terms of total construction cost
The FOS did not appear to correlate with the structural (including preliminary and general costs) a 17% total cost saving
performance classification and it was concluded that the FOS was realised. Subsequent to this analysis several roads were
design criteria in the upper layers is not applicable to haul road constructed according to the mechanistic design and during an
structural design. In the absence of any definitive criteria, a extremely wet 1996 summer, superior performance was reported,
200mm layer of compacted (95-98% Mod. AASHTO) good compared with the existing roads.
quality gravel (G4-G5 following Committee of State Road

3 Copyright © 1999 by SME

 Table I Suggested Modulus Ranges for Granular Materials

Recommended Effective Elastic Modulus (E) MPa (1)

Material Abbreviated Specification(5)
Code(2) Over Wet state Wet state
granular (good (poor support)(4)
layer support)(3)

G1 Dense-graded unweathered crushed stone. Max size 150 50-250 40-200

37,5mm. 86-88% of apparent density. Fines PI<4

G2 Dense-graded unweathered crushed stone. Max size 100-400 50-200 40-200

37,5mm. 100-102% Mod AASHTO. Fines PI<6

G3 Dense-graded stone and soil binder. Max size 100-350 50-150 40-200
37,5mm. Min 98% Mod AASHTO. Fines PI<6

G4 Natural gravel base, CBR$80, PI#6 75-350 50-150 30-200

G5 Gravel CBR$45, PI#10, max size 63mm 40-300 30-200 20-150

G6 Gravel, low quality sub-base. CBR$25. 30-200 20-150 20-150

Max size < b layer thickness

In situ(6) CBR$25 135 105

24$CBR$15 125 85
14$CBR$10 120 65
9$CBR$7 95 55
6$CBR$3 65 45

Notes
1. Using a Poisson's ratio of 0,35 2. Following CSRA TRH14 (1985) classification
3. In-situ material (dry state) 4. In-situ material (wet state)
5. Using soaked CBR values 6. Using CBR values at in-situ density

HAUL ROAD FUNCTIONAL DESIGN

The functional trafficability of a pavement is dictated to a large

degree through the choice, application and maintenance of
wearing course materials. Poor functional performance is
manifest as poor ride quality, excessive dust, increased tyre wear
and damage and an accompanying loss of productivity. The
corollary of these effects is seen as an increase in overall vehicle
operating and maintenance costs.

Previous work on functionality was found to be highly subjective

and localised in nature (Thompson, 1996) and it was thus
necessary to develop a qualitative functional performance
assessment methodology in which typical functionality defects
could be classified and ascribed a defect score in terms of extent
and degree (severity) of a defect. The defects of potholing,
corrugating, rutting, loose material, dustiness, fixed and loose
stoniness, longitudinal, slip and crocodile cracking, erosion and
skid resistance were evaluated monthly in terms of degree of
severity and extent. In addition, the maintenance interval was
also recorded to enable predictions to be made concerning the rate
Figure 3 Results in Terms of Vertical Strain Design Criteria for of overall and individual defect functional performance
CBR and Mechanistic Optimal Designs deterioration with time and traffic. This approach reduced
subjectivity and, by virtue of the degree and extent classifications,
provided a universal basis for the assessment and reporting of
functionality.

The experimental approach adopted in determining suitable

material selection guidelines involved the analysis and
quantification of a range of factors which determine the efficacy
of the various wearing course material types, in terms of
functionality. In general, these factors are the material type itself,

4 Copyright © 1999 by SME

together with road geometrics, climate and traffic volumes. To Recommended Wearing Course Material Selection Guidelines
fully characterise the functional performance of existing or future
selections, each factor was analysed at various levels by means of The derivation of wearing course material selection guidelines
a designed factorial experiment. was based on the identification, characterisation and ranking of
haul road functional defects, together with an assessment of the
From the statistical analysis of deterioration and maintenance suitability of existing wearing course material selection guidelines
effects, a defect progression model was derived from and classifications as a basis for mine haul road material
consideration of the rate of change in overall defect score over a specification. The TRH14 material classification system was
maintenance cycle. The model incorporated wearing course found to be inadequate as a base for haul road wearing course
material properties, especially grading and plasticity parameters, material selection due to its inability to adequately differentiate
together with traffic volume. By analysing the rate of change of between critical defects over the parameter ranges represented by
individual defect score beyond a minimum value encountered at typical haul road construction materials.
each site, models were developed for the rate of change in defect
score with time and the propensity of specific material properties The TRH20 (CSRA TRH20, 1990) wearing course material
to affect the progression. In determining suitable wearing course selection guidelines were found to be a more suitable source for
material selection guidelines this work confirmed qualitative the specification of mine haul road wearing course material
observations that grading and plasticity parameters will parameter requirements. A revised range of parameters was
adequately anticipate the functional performance of a wearing derived based on the road-user preference for much reduced wet
course material. slipperiness, dustiness and dry skid resistance defects, as shown
in Figure 5. The specification included the parameters of
In parallel with a functional performance assessment exercise shrinkage product (Sp) and grading coefficient (Gc) and limits of
carried out at a number of mine sites, a functionality acceptability 85-200 and 20-35 respectively were proposed. In addition, from
questionnaire was completed by mine and equipment suppliers to analysis of the range of material property parameters assessed and
indicate the performance levels required by operators, again, their association with the functional defects analysed, additional
according to the qualitative functional performance assessment parameter ranges were specified as given in Table II. By
methodology. Each functional defect was ascribed a range of analysing the trends evident in the individual defect rankings, the
scores in terms of degree and extent covering desirable, predictive capability of the specification was enhanced by
undesirable and unacceptable performance, as illustrated in depicting the variation in functional defects which would arise
Figure 4. This enabled a comparison to be made between the when departures are made from recommended parameter limits.
functionality of the various types of wearing course material In addition, by incorporating material property values into the
surveyed. In addition to assigning acceptability ranges to each defect score progression models, operators can determine the
type of defect, the impact and accident potential of each defect practical implications of using a sub-standard wearing course
was categorised and ranked according to the total impact and material or blend of materials, in terms of their impact on overall
accident potential on the components of hauling, namely haul road functionality, individual defect scores and acceptability,
operation, truck and tyre. It was clear from the ranking exercise together with the required maintenance interval for user-defined
that wet skid resistance, dustiness, erodibility and ravelling and functionality defect score upper limits.
corrugating are critical defects which control the functionality of
mine haul roads. An understanding of the functional
consequences, in terms of the possible generation of these defects, Wearing Course Performance Improvement Through Dust
was therefore incorporated into the selection criteria established Palliation
for mine haul road wearing course materials.
Since the wearing course material selection guidelines include
a predictive component, the practical implications of using sub-
standard wearing course materials can be assessed and the
anticipated functional defects pro-actively managed. Most mines
source wearing course materials from local borrow-pits and very
little control can be exercised over the quality of these materials.
Dustiness is a common and critical defect on most mine haul
roads, effecting safety, productivity and total operating costs
(Thompson, 1996, Amponsah Dacosta, 1998). Using the limits
of dust defect desirable performance (as shown in Figure 4),
should a particular wearing course material exhibits an
undesirable or unacceptable dust defect, a direct quantitative

Figure 4 Limits of Defect Functional Performance Acceptability

5 Copyright © 1999 by SME

 Table II Recommended Wearing Course Material Parameter Specifications for Optimal Functional Performance

WEARING COURSE MATERIAL SELECTION PARAMETERS

Material Parameter (following Range Impact on Functionality

CSRA TRH20, 1990) Min Max

Shrinkage Product 85 200 Reduce slipperiness but prone to ravelling and corrugation

Grading Coefficient 20 35 Reduce erodibility of fine materials, but induces tendency to ravel

Dust Ratio 0,4 0,6 Reduce dust generation but induces ravelling

Liquid Limit (%) 16 26 Reduce slipperiness but prone to dustiness

Plastic Limit (%) 12 17 Reduce slipperiness but prone to dustiness

Plasticity Index 4 9 Reduce slipperiness but prone to dustiness and ravelling

CBR at 98% Mod AASHTO 80 Resistance to erosion, rutting and improved trafficability

Maximum Particle Size (mm) 20 Ease of maintenance and vehicle friendly ride

Figure 6 Increase in Road Dustiness (x100 mg/m3 minus 10µm

Figure 5 Optimum Material Selection Ranges and General
dust) Following a Single Application of 0,5l/m2 Water
Trends of Increasing Defect Scores

By monitoring the performance of a palliated test section,

cost-based comparison can be made between the various dust similar quantitative data was generated and the degree of dust
management solutions; water-based dust suppression, palliative- palliation and its variation with time and traffic volume was
based dust suppression or that acheived through the use of a more found. A suppression efficiency was determined by comparing
appropriate wearing course material or mix of materials. the palliated test section suppression performance to the dust
suppression acheived by water spraying alone. For the particular
Typical results are presented based on a test program embarked combination of palliative treatment, traffic volume and type,
upon by Pennzoil Products (SA) (Pty) Ltd and a local mining climatic conditions and wearing course material type, suppression
company (Thompson and Visser, 1998(a)). In order to provide efficiencies of over 200% were maintained for up to 15 days
the mine with quantitative data regarding the efficacy of the dust following application as shown in Figure 7. This performance
palliative tested (PennzSuppress® D spray-on mixtures), the mine was correlated with the maximum (allowable mine) dust defect
assessed its own dust defect intervention level score (maximum score of 15 and a palliative performance model derived. This
desirable dust defect score as shown in Figure 4). Figure 6 indicated that re-treatment of the road was required every five
illustrates the typical increase in dustiness (x100 mg/m3 minus days for a palliative and water mix, compared to every 70 minutes
10µm dust) as measured with a Hund Tyndalometer on the mine with water alone. Further cost-benefits were hypothised based on
haul road, following a single application of 0,5l/m2 water, and its the reduced loss of fine (binding) materials, reducing the need for
correlation to dust defect score. Given the mine's preference for periodic re-gravelling, reduced road maintenance intervals,
a maximum dust defect score of 15, water spraying of the haul reduced vehicle operating costs and a reduction in visibility and
road was thus required every 70 minutes. health hazards on the road. Testwork is currently in progress to
develop models of dust suppression efficiencies for a range of
treatments, wearing course material types and traffic volumes, by
means of which operators can determine the cost-effectiveness of

6 Copyright © 1999 by SME

 operating costs and maintenance requirements, as illustrated in
Figure 9 in which the reduced defect scores and rate of
deterioration in functionality between the original and optimum
wearing course material selection are seen. In order to quantify
this cost benefit, the road maintenance and vehicle operating costs
(VOC) associated with existing wearing courses were assessed
and compared to those estimated from models.

Vehicle Operating and Road Maintenance Cost Models

Two elements form the basis of the economic evaluation, namely

pavement roughness (rolling resistance) progression and vehicle
operating and road maintenance costs. The proposed mine haul
road MMS was developed from established MMS applied on
public roads, together with specific modifications which reflect
the requirements of mine haul road-users. The road roughness
progression model forms the basis of the MMS since roughness
(or rolling resistance) is the principal measure of pavement
condition that can be directly related to both VOC and the
frequency of maintenance activities. A qualitative road roughness
Figure 7 Typical Suppression Efficiency Curve and Comparison
evaluation technique was developed, based on the functional
with Mine Maximum Allowable Dust Defect Score of 15
assessment methodology of defect degree and extent, as a
precursor to the development of a model for roughness
such treatments for their own particular combination of progression. The progression model was found to be a function
circumstances. of wearing course material parameters, traffic volumes and
maintenance interval. To facilitate portability and comparison of
HAUL ROAD MAINTENANCE SCHEDULING AND the qualitative assessment technique, expressions were developed
MANAGEMENT to enable direct comparison to be made between roughness defect
score, the International Roughness Index (m/km IRI) and rolling
Maintenance design concerns the optimal frequency of wearing resistance (N/kg vehicle mass).
course maintenance commensurate with minimum vehicle
operating and road maintenance costs. Through the development The second element of a MMS for mine haul roads was based on
of a maintenance management system (MMS), the optimum models of the variation of vehicle operating and road maintenance
maintenance frequency for a mine haul road network can be costs with road roughness. VOC models included fuel, tyre and
determined, commensurate with the lowest total vehicle operating maintenance cost and labour. For the fuel consumption model,
and road maintenance costs. Existing MMS's are based on the simulations were conducted from which four universal equations
optimisation of these elements, as shown schematically in Figure were derived by means of which vehicle speed could be predicted
8. for any combination of total road rolling resistance and truck
loading.

Figure 8 Minimisation of Road Maintenance and Vehicle

Operating Costs
Figure 9 Predicted Reduction in Haul Road Functional Defects
for Existing (old) and Optimised (new) Wearing Course Material
Additionally, by combining the functional design
recommendations within a MMS, new specifications for wearing
course materials or mixes of materials may reduce vehicle

7 Copyright © 1999 by SME

 With regard to the tyre, vehicle maintenance parts and labour
models, existing models developed for public commercial trucks
were used as a basis for the development of mine haul truck
models (Chesher and Harrison, 1987). Although the parameter
ranges bore little resemblance to those of mine haul trucks, when
coupled with a hypothesis of the influence road roughness and
geometry on these cost components, a basic model was developed
in each case. These models were then compared with the limited
mine data available to verify the order of magnitude of the costs
modelled and, more critically, to indicate the likely rate of change
of these costs with road roughness or rolling resistance. These
models are discussed in more detail by Thompson and Visser
(1996(a), 1997).

MMS Computer Model

The interaction and influences of the various models proposed

to represent VOC, road maintenance costs and the progression of
road roughness can only be effectively analysed using a systems
analysis approach as shown in Figure 10. Figure 11 Segment and Total Road-user Cost Variation with
Maintenance Interval

Cost savings associated with the adoption of a maintenance

management approach are dependant on the particular hauling
operation, vehicle types, road geometry and tonnages hauled, etc.
From an analysis of the rate of change in vehicle operating and
road maintenance costs for individual segments of one particular
mine road with reductions in the frequency of maintenance
beyond the optimal maintenance interval, these sub-optimal
maintenance strategies were seen to be associated with in excess
of R310 000 annual over-expenditure on total road-user costs.
The adoption of the MMS model program for mine haul roads has
the potential to generate significant cost benefits when used
dynamically, in conjunction with production planning, to optimise
mine haul road maintenance activities for particular combinations
of wearing course material, traffic volumes and vehicle types.

CONCLUSIONS

Figure 10 Flow Chart of Proposed MMS for Mine Haul Roads The world-wide expansion of surface mining has led to the
(for a single maintenance strategy iteration) development of very large off-highway trucks currently capable of
hauling payloads in excess of 290t. Typical axle loads ranging
from 110t to 200t are applied to haul roads that have been, at
The MMS model program for mine haul roads was developed to best, empirically designed on the premise of "satisfactory" or
evaluate alternative maintenance intervals and the associated "failed", both in terms of structure and function of the road. The
effect on total road-user costs, comprising vehicle operating and use such ultra-heavy haul trucks on surface mines, in conjunction
road maintenance cost elements. Road maintenance costs and with the current empirical mine haul road design techniques have
fleet productivity was assessed by means of user specified data in been shown to be inadequate; not only do the maintenance costs
conjunction with a basic grader productivity model. An of existing haul roads of inadequate structural or functional design
evaluation of the total cost variation with maintenance interval increase, vehicle operating and maintenance costs also increase
enabled the optimum maintenance interval to be determined, both prohibitively. Under these conditions the need for improved
on a minimum total road-user cost basis and in terms of technologies encompassing the construction and management
maintenance equipment available operating hours. When techniques of flexible mine haul roads, appropriate for the wheel
analysing the results of individual mine simulations, the actual loads of vehicles now in use was recognised. Whilst the research
mine operating practice was seen to closely resemble that was based on South African road construction materials and
predicted by the model, especially with regard to increased climatic conditions, the principal findings are nevertheless
maintenance interval on lightly trafficked roads. A typical result internationally applicable.
is illustrated in Figure 11 from which it is seen how total vehicle
operating costs for two sections of a mine haul road network are A new mechanistic structural design methodology was presented
minimised at the optimal road maintenance frequency interval. which has the potential to reduce road construction cost and
improve the structural strength of the pavement, primarily by
virtue of reduced pavement layer thickness and compaction
requirements. The improved functionality of a pavement was
addressed by defining the optimum wearing course material

8 Copyright © 1999 by SME

selection parameters, based on both road-user acceptability roads". Jnl. South African Institute of Civil Engineering, vol 38,
criteria and models of functional defect progression. Where sub- n2, second quarter, Johannesburg, South Africa, pp13-21.
standard wearing course materials are encountered, the functional Thompson, R.J., and Visser, A.T. 1996(a). "Maintenance
performance assessment methodology can also be applied in a scheduling and management of surface mine haul roads". Haul
quantitative assessment framework to assess dust palliative roads research project, Stage III contract report for AMCOAL,
performance in comparison with other dust remediation options. University of Pretoria, South Africa.
Finally, the concept of maintenance management was addressed Thompson, R.J., and Visser, A.T. 1997. "Material selection for
in which a maintenance management system model was optimal haulroad performance". Proceedings of the Institute of
developed as an aid in identifying the most appropriate haul road Quarrying (SA) Annual General Meeting and Conference,
maintenance schedule commensurate with minimum total road- Durban, South Africa, pp 57-69.
user costs. The adoption of the MMS model program for mine Thompson, R.J., and Visser, A.T. 1998. "Pavement design for
haul roads has the potential to generate significant cost benefits ultra-heavy mine haul trucks". Proceedings of the 5th Intl. Conf.
when used dynamically, in conjunction with production planning, on the Bearing Capacity of Roads (BCRA'98), July, Trondheim,
to optimise mine haul road maintenance activities. Norway.
Thompson, R.J., and Visser, A.T. 1998(a). "Evaluation of
A total haul road design strategy combining mine layout, Pennzsuppress® D road dust suppressant at Iscor Quarries
construction techniques, available material and road maintenance Mooiplaas Mine". Investigation report, Pennzoil Products (SA)
equipment with hauler choice has been developed. Application (Pty) Ltd, Johannesburg, South Africa.
of these general design and operation optimisation strategies will Thompson, R.J, Visser, A.T, Smith, R.A.F & Fourie, G.A.F.
enable mine operators to realise a significant reduction in haulage 1997. "An investigation into the influence of haul road design,
and road maintenance costs. construction and maintenance practices on transport accidents for
surface and underground mines, and to determine further research
to improve safe transportation operating conditions". Final report
ACKNOWLEDGEMENTS for project SIMOT 308. Safety in Mines Research Advisory
Committee, Other Mines Sub-committee project report OTH308,
Acknowledgment is given to AMCOAL Colliery and Industrial Department of Minerals and Energy, Pretoria, South Africa.
Operations Ltd., for permission to publish these results, their
financial support and provision of facilities for the research upon
which these developments are founded. The assistance and co-
operation received from mine personnel is also acknowledged.

REFERENCES
Amponsah Da-costa, F. 1998. "Cost effective strategies for dust
control in an opencast coal mine". MSc Project Report,
University of the Witwatersrand, South Africa.
Boyd, W.K., and Foster, C.R. 1950. "Design Curves for very
heavy multiple wheel assemblies, development of the CBR
Flexible pavement design methods for airfields, a symposium".
Transactions, American Society of Civil Engineers, vol 115,
pp534.
Chesher, A., and Harrison, R. 1987. "Vehicle operating costs -
evidence from developing countries". Highway Design and
Maintenance Standards Series, Transportation Department, World
Bank, Washington, DC, USA.
Committee of State Road Authorities. 1985. "Technical
recommendations for highways TRH14: Guidelines for road
construction materials". NITRR TRH14, Pretoria, South Africa.
Committee of State Road Authorities. 1990. "Technical
recommendations for highways TRH20: The structural design,
construction and maintenance of unpaved roads". NITRR TRH20,
Pretoria, South Africa.
Kaufman, W.W., and Ault, J.C. 1977. "The design of surface
mine haul roads manual". USDOI Information circular 8758,
USBM, pp2-7, 19-30.
Kleyn, E.G., Maree, J.H., and Savage, P F. 1982. "Application
of a portable pavement dynamic cone penetrometer to determine
the insitu bearing properties of road pavement layers and
subgrades in South Africa". ESOPT II, Amsterdam, Holland.
South African Roads Board. 1994. "The South African
mechanistic pavement rehabilitation design method". Research
report RR 91/242, Department of Transport, South Africa.
Thompson, R.J. 1996. "The design and maintenance of surface
mine haul roads". PhD Thesis, University of Pretoria, South
Africa.
Thompson, R.J., and Visser, A.T. 1996. "Towards a
mechanistic structural design method for surface mine haul

9 Copyright © 1999 by SME