Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 105 (2017) 2739 2745

The 8th International Conference on Applied Energy ICAE2016

A MILP model for truck-shovel scheduling to minimize fuel

consumption
D.M. Bajany*, X. Xia, L. Zhang
Department of Electrical, Electronic and Computer Engineering, University of Pretoria, Pretoria 0002, South Africa

Abstract

This paper presents an optimization model, in which fuel consumption of dump trucks and shovels is minimised while
the handling demand of dump sites are met in the case of an under-trucked open-pit mine. The model is build using an
m-trucks-for-n-shovels dispatching strategy so that a truck could be allocated to different shovels during a shift. The
performance of the optimal energy dispatching model was compared to the fixed dispatch method that is still being
used in certain mines. The results show that the average litre of fuel consumption per tonne of mineral transported by
trucks and that for the truck-shovel system are reduced by 8.82% and 4.49%, respectively. This further leads to a 4.64%
fuel savings.
2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
2016 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Selection and/or peer-review under responsibility of ICAE
Peer-review under responsibility of the scientific committee of the 8th International Conference on Applied Energy.

Keywords: Truck-shovel dispatch system, fuel minimization,

1. Introduction

It has been estimated that transportation cost of materials in open-pit mines represents up to 60% of the
total mining costs and 32% of the total energy used in mines is consumed by the haulage operations[1-2].
The capital and the annual operating cost of a dump truck has been estimated to be 1.6 million dollars for a
typical oil sand mine [3]. For this reason, a marginal improvement in the efficiency of the truck-shovel
dispatching system in open-pit mines will result in a noticeably cost savings.

Nomenclature

Truck index
Unloading point index
Shovel index
 Fleet size of trucks
 Number of unloading points
 Number of shovels


* Corresponding author. Tel.: +27786299581; fax: +27123625000
E-mail address:dannybajany@gmail.com

1876-6102 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the scientific committee of the 8th International Conference on Applied Energy.
doi:10.1016/j.egypro.2017.03.925
2740 D.M. Bajany et al. / Energy Procedia 105 (2017) 2739 2745

 Capacity of truck
 The hourly loading capacity of the shovel j [t/h]
 The hourly handling demand of the unloading point  [t/h]
 Shift duration in hour

The loading time of truck y when loaded by the shovel j [h]

Travel time of an empty truck from an unloading point i to a shovel j [h]

Travel time of a loaded truck from a shovel j to an unloading point I [h]

The unloading time of truck [h]

Cycle time of dump truck [h]
 Idle time of shovel [h]
Distance from an unloading point i to a shovel j [km]

The driving speed of an empty truck [km/h]

The driving speed of a loaded truck [km/h]
Fuel consumption of shovel during idle time [l/h]

Fuel consumption of truck y during engine idling [l/h]

Fuel consumption of an empty truck moving from an unloading point to a shovel [l/h]

Fuel consumption of a loaded truck moving from a shovel to an unloading point [l/h]

Fuel consumption of truck y per single cycle. [l]
The total fuel consumed by trucks during a shift [l]

Number of journeys that the truck has travelled from an unloading point to a shovel
during a shift

Number of journeys that the truck has travelled from a shovel to an unloading point
during a shift.
 Liters per tonne moved for trucks and shovels [l/t]
 The engine power of truck y [kW]
LF Load factor

Improving the energy efficiency of the truck-shovel dispatching system in an open-pit mine is an old
hot topic which is still relevant. To this end, several heuristic methods have been presented in the literature.
Three strategies were used for assigning a truck to the right shovel, namely the 1-truck-for-n-shovels, the
m-trucks-for-1-shovel and the m-trucks-for-n-shovels strategies [4]. However, the work done so far does
not minimise simultaneously the fuel consumption of both trucks and shovels.
In this study the m-trucks-for-n-shovels dispatching strategy is formulated as a mixed integer linear
programing (MILP) model which optimizes route choice and minimizes the fuel consumption of both trucks
and shovels with respect to the production goal in the case of an under-trucked mine. The advantage of the
proposed MILP model is that technical specifications of equipment are considered so that this model could
be used in homogeneous and heterogeneous fleets. Indeed, any difference in transportation capacities of
trucks and in shovel capacities involve different power and traction of trucks, as well as the inter-trucks
time variation during a fixed period of time. In the case of an under-truck mine, the inter-truck time variation
influences the utilisation time of shovels therefore their fuel consumptions. Beside this, the loading time of
a truck is directly depending on his capacity and the shovel capacity, thereupon the fuel consumption of a
truck when loaded is function of the loader capacity. For this raison, in the case of a heterogeneous fleet,
technical specifications of equipment must be taken into account for any optimisation problem of haulage
operations in open-pit mines. Optimizing route choice, reduce vehicle emissions and contributes to air
quality improvement [5].

2. Truck-Shovel dispatching problem

Fig.1 displays a generic open pit mine which has  unloading points,  shovels and   transport
routes. During each shift, empty trucks located at unloading points are assigned to shovels and those who
 D.M. Bajany et al. / Energy Procedia 105 (2017) 2739 2745 2741

were loaded by shovels return back to unloading points to complete their cycles. Shovels consume fuel
during their working and idling periods, and dump trucks consume fuel during their waiting, loading,
unloading and traveling periods. Fuel consumption of each equipment (truck and shovel) depends on its
size, type, operating time, operating conditions and technical specifications. In this work, the following
assumptions are made: Firstly, in haulage operations, depending on the mine topography, truck engine
specifications and payload capacity, each truck is driven at its optimum speed that leads to the most efficient
fuel consumption. Secondly, all unloading points are wide enough, so that loaded trucks could dump their
loads at the same time. Therefore, no queuing of trucks can happen at a dumping point. Lastly, no waiting
time of truck at a shovel can happen because the case considered in this work concerns an under-trucked
mine.

Fig.1. Transport routes of an open-pit mine

3. Mathematical formulation

The travel time of truck from the unloading point to the shovel and that for the return

trip ( ) are calculated by equations (1) and (2). The loading time of truck with the shovel and the
duration of a single cycle of truck (y) are determined by equations (3) and (4). The idle time of the shovel
for a complete shift is given by equation (5). The hourly fuel consumption by a truck (y) is estimated by
equation (6) [6] and its fuel consumption per single cycle is calculated by equation (7). For the entire shift
the total fuel consumed by trucks is calculated by equation (8). For a complete shift the total fuel consumed
by shovels is calculated by equation (9).

y
d ij (1)
t ij y
v ij

d ji
t yji (2)
v yji
Cy (3)
t l,y j
Cj
y
tcycle tijy  t yji  tl,yj  t uy (4)
Ni Ny
(5)
Ij sh  Z
i 1 y 1
y
ji t l,y j [h]

f y 0.3u P y u LF [l/h] (6)

F y
ij
y y
f t f t f
ij ij
y y
ji ji
y
idle t y
l, j t y
u  (7)
Nj Ni Ny
Ft >Z
j 1 i 1 y 1
y y y
f t  Z yji f jiy t yji  Z yji f idle
ij ij ij
y
t l,y j  t uy  @ (8)

N N
j
Ni y
(9)
Fs f j, u
Z yji t l,y j  I j f j, idle
j 1 i 1 y 1
2742 D.M. Bajany et al. / Energy Procedia 105 (2017) 2739 2745

Where:  , 0.3 and LF represent the engine power (kW) of truck y, the unit conversion factor (L/kW/h)
and the engine load factor respectively. The load factor is the portion of full power required by the truck.
In order to determine the fuel consumption of the dump truck, the following load factor values have been
considered: 35% for a loaded truck (normal load), 20% for an empty truck and 10% for a truck while idling
[7]. The hourly fuel consumption of shovel (j) during its working time is estimated by assuming that
the shovel works at high load factor (90%). The hourly fuel consumption of shovel (j) during its
idle time is assumed to be 10% of the hourly fuel consumption of that shovel when it works with a low load
factor (less than 50%).
In this study we use the litres of fuel per tonne moved as a performance indicator for the truck-shovel
dispatching system. The litres per tonne moved at the end of a shift for trucks and shovels is evaluated by
equation (10).
Ft  Fs
LTts N j Ni N y (10)
Z yji C y
j 1 i 1 y 1

4. Optimization problem

As we previously mentioned, the objective of this study is to minimize the fuel consumption of both
trucks and shovels. The model is built in such a way that all demands of dump sites are met, the optimal
number of trips that each truck realizes on each route of the pit is determined and the optimum idle time of
each shovel is estimated. The technical specifications such as: payload of trucks, loaded capacity of shovels,
and fuel consumption in function of the operating conditions of equipment are directly considered in the
mathematical model. The formulation of the optimisation problem is given in the following.

Nj

min Ft  I jf j, idle (11)
j 1

Subject to:

N j Ny

Z y
C y t D ish (12)
ji
j 1 y 1
N j Ny
(13)
Z
j 1 y 1
y
ji C y d C jsh
N j Ny
(14)

j 1 y 1
Z yji t l,y j d sh
N j Ny

>Z
j 1 y 1
y y
ij ij @
t  Z yji t yji  Z yji t l,y j  t uy  d sh (15)

Ni Ni
(16)
Z
i 1
y
ij Z
i 1
y
ji

Nj Nj
(17)
Z yji  Zijy d 1
j 1 j 1
Nj Nj

Z  Z y
ij
y
ji d1 (18)
j 1 j 1

Zijy N; Zyji N; i 1,..., Ni ; j 1,..., N j ; y 1,..., Ny (19)

The first constraint (12) ensures that the material transported from all shovels to an unloading point is
greater than or equal to the handling demand of that unloading point. Constraint (13) ensures that the
material transported from each loading point is less than or equal to the capacity of the shovel allocated to
that pit. The utilisation time of shovels is constrained by the shift duration time as specified by constraint
 D.M. Bajany et al. / Energy Procedia 105 (2017) 2739 2745 2743

(14). The sum of cycle time of all cycles that a truck realises during a shift is constrained by the shift
duration time as specified by constraint (15). The continuity of loading and transportation is maintained by
constraints (16), (17) and (18). The last constraint (19) ensures that the trip numbers of trucks are positive
and integer.

5. Case study

A case study of a hypothetical downgrade open pit mine with two unloading points and three shovels is
considered for optimization and simulation of the model. Ten trucks of 56t, 65t and 96t are used in the
haulage operations. The effectiveness of the proposed model is verified by comparing the results obtained
with those obtained with a fixed assignation of trucks. The design specifications and operating parameters
of loading and haulage equipment used as input parameters of the model are shown in Table 1. The MILP
model is solved using Intlinprog algorithm in Matlab.

6. Results and discussion

Consider a shift of 8 hours, solving the MILP model given in Section 3 with the operating data given in
Table 1, the optimum number of trips that each truck has realized when loaded or empty on each route of
the mine is given in Table 2. The results obtained with the MILP model show that: for the entire fleet of
trucks and shovels used in the haulage operations the energy efficiency indicator is 0.1307 liters per tonne.
4915.6 liters of fuel are consumed for the whole haulage operations (loading and transport). To show the
effectiveness of the proposed MILP model, the same dispatching problem has been solved by using the
fixed dispatching policy. This results in 0.1370 liters per tonne and 5154.15 liters of fuel consumption for
the whole haulage operations. Comparing the results obtained with both models, the following observations
were made. The liters per tonne for mixed capacity dump truck (56t, 65t, and 96t) operating in this mine
are reduced by 8.82 %, the liters per tonne of both trucks and shovels are reduced by 4.60 %. 4.64 % saving
of fuel is achieved with the proposed energy optimization model.

Table 1. Mine topography and resources

Parameter Specifications
Trucks [8] Model 773D 775D 777D
Gross power 509kW 541 kW 699 kW
Capacity 56 t 65 t 96 t
Empty speed 50 km/h 50 km/h 50 km/h
Speed (loaded) 39 km/h 35 km/h 36 km/h
Number of trucks 3 3 4
Shovels [8] Number of shovels 3
Model 992G 992G 5130B ME
Capacities 1600 t/h 1600 t/h 2000 t/h
Fuel consumption of shovel (idle time) 6.6 l/h 6.6 l/h 9.5 l/h
Fuel consumption of shovel (working time) 117 l/h 117 l/h 130 l/h
Mine topography Downgrade mine Total resistance 4%
(Grade & Rolling)
Distance between dumping points and shovels 1.5 km 2 km 3 km
3.5 km 2 km 1.5 km
Hourly load demand at each unloading points  2325 t/h  2375 t/h
2744 D.M. Bajany et al. / Energy Procedia 105 (2017) 2739 2745

Table 2. Optimum trip number of trucks

Unloading Shovel 1 Shovel 2 Shovel 3

Truck No - Capacity
Points Empty Loaded Empty Loaded Empty Loaded
1 -56t 1 42 42 0 0 0 0
2 0 0 0 0 32 32
2 -56t 1 41 41 0 0 0 0
2 0 0 0 0 33 33
3 -56t 1 42 42 0 0 0 0
2 0 0 0 0 32 32
4 -65t 1 31 31 0 0 0 0
2 0 0 1 1 0 0
5 -65 t 1 53 53 0 0 0 0
2 0 0 0 0 0 0
6 65 t 1 0 0 0 1 0 0
2 0 0 1 0 0 0
7 - 96 t 1 22 22 0 0 0 0
2 0 0 0 0 30 30
8 -96 t 1 0 0 2 3 0 0
2 0 0 18 17 32 32
9 96 t 1 0 0 33 34 0 0
2 0 0 5 4 9 9
10 96 t 1 3 3 2 2 0 0
2 0 0 9 9 39 39

7. Conclusion

A model for fuel consumption minimization of dump trucks and shovels in an open pit mine has been
developed. A case study of an under trucked mine with two unloading points and three shovels has been
considered. The optimal number of trips that each truck should realize on each route of the mine during a
shift is optimally determined. The results when compare to a fixed dispatch show that the developed
optimization model decreased, the liters per tonne of fuel consumption of trucks by 8.82%, and that of both
trucks and shovels by 4.60 %. Overall, 4.64 % saving of fuel is achieved. The fact that fuel saving is
realized, vehicle emission is reduced and air quality is improved.
For an operational open-pit mine with certain topography and resources, the liters per tonne obtained
from the proposed MILP model can be used as a reference for evaluating the efficiency of his truck-shovel
dispatching system. One of the potential direction for the future research is to extend the applicability of
the proposed MILP model to a case of an over trucked-mine.

References

[1] Zhang L, Xia X. An Integer Programming Approach for Truck-Shovel Dispatching Problem in Open-Pit Mines. Energy
Procedia. 2015; 75:1779-84.
[2] L. K. Sahoo, S. Bandyopadhyay, R. Banerjee, Benchmarking energy consumption for dump trucks in mines, Applied Energy
2014; 113:1382-1396.
[3] M. Delasay, B. Kolfal, A. Ingolfsson, Maximizing throughput infinite-source parallel queue systems, European Journal of
Operational Research 2012; 217:554-559.
 D.M. Bajany et al. / Energy Procedia 105 (2017) 2739 2745 2745

[4] S. Alarie, M. Gamache, Overview of solution strategies used in truck dispatching systems for open pit mines, International
Journal of Surface Mining, Reclamation and Environment 2002; 16:59-76.
[5] Zeng, W., Miwa, T., & Morikawa, T. (2016). Prediction of vehicle CO 2 emission and its application to eco-routing navigation,
Transportation Research Part C: Emerging Technologies 2016; 68, 194-214.
[6] I. C. Runge, Mining economics and strategy, SME, 1998.
[7] V. Kecojevic, D. Komljenovic, Haul truck fuel consumption and CO2 emission under various engine load conditions, Mining
Engineering 2010; 62:44-48.
[8] Caterpillar.Inc, Caterpillar Performance Handbook, Peoria, Illinois, USA, 2009.

Biography
Mr Mpyana Bajany has two B.Eng, one in electromechanical from University of
Lubumbashi, D.R. Congo, and one in mechanical from University of Pretoria, South
Africa. He is currently a master student at University of Pretoria. His research is principally
focused on improving the energy efficiency of the dispatch system in open-pit mines.