Vehicle Routing Optimization for Improving Fleet Fuel Efficiency: A Case Study in Sydney, Australia

Full text

 Abstract—Vehicle Routing Problem (VRP) is a classic
combinatorial optimization problem involved in many
applications. VRP is even a big concern when the vehicle is a
garbage truck, which travels approximately 100 km/day with
the average consumption of 1 litre/km. For this reason, a small
improvement in collection activities may result in significant
savings in overall cost, fuel and therefore greenhouse-gas
emissions. The primary goal of this research is to find ways to
reduce overall travel distance for collection and transport of
municipal solid waste from residential homes within the
Blacktown City Area in order to reduce the fuel consumption
and therefore greenhouse-gas emissions. Esri’s ArcGIS 10.3
Network Analyst extension has been used in this study. To
calculate optimal routes for solid waste collection, several inputs
to the ArcGIS Network Analyst solver have been used including:
collection points represented the depot start point, the home
rooftops where the garbage is collected and the unload point.
The results of this study show that: Using an optimized route
instead of a regular route can reduce the total travelled distance
by 8 km/day on the pilot site. The optimized route will reduce
that individual truck’s emissions by 5.5 kg CO2 per day for that
collection area. This represents a reduction of about 8% for
that particular collection.
Index Terms—Fleet fuel efficiency, vehicle routing problem,
greenhouse gases, transportation.
I. INTRODUCTION
It has been estimated that of the total amount of money
spent for the collection, transportation, and the disposal of
solid waste around 60-80% is spent on the collection phase
[1]. This provides great opportunity for research to find better
cost saving measures for municipalities. In addition to the
high cost in operation and maintaining municipal vehicles,
there is concern that municipal solid waste (MSW) trucks
have a negative impact on the environment due to the
distances driven, fuel type, engine inefficiency, and exhaust
emissions. Solid waste management comprises the
generation, collection, transport, treatment, and disposal of
solid waste from homes [2]. Routing optimization problems
in waste management have typically been addressed with
different types of mathematical algorithms. Routing
Manuscript received August 21, 2017; revised November 15, 2017. This
work was supported in part by the Australian Research Council (ARC),
Australian Government (No: DP150101015) and data provided by
Blacktown City Council.
Hoda Karimipour and Khoa N Le are with Western Sydney University,
School of Computing, Engineering and Mathematics, Locked Bag 1797,
Penrith, NSW 2751, Australia (e-mail:
h.karimipour@westernsydney.edu.au).
Vivian W. Y. Tam is with Western Sydney University, School of
Computing, Engineering and Mathematics, Locked Bag 1797, Penrith, NSW
2751, Australia. She is also with College of Civil Engineering, Shenzhen
University, China.
Helen Burnie is with Blacktown City Council, Australia.
algorithms use a measuring system called a path length for
determining the ideal route to a defined destination. The
optimal routes are then determined by comparing different
paths. These paths can be calculated by different types of
algorithms. Some of the routing algorithms used include
Simulated Annealing, Tabu Search, Genetic Algorithm, Ant
Colony Optimization, and Dijkstra’s algorithm [1], [3], [4].
Vehicle Routing Problem (VRP) is a classic combinatorial
optimization problem involved in many applications. Since
its introduction by Dantzig and Ramser [5], VRP has been
extensively studied. By considering additional requirements
and various constraints on route construction, different VRPs
have been formulated [6], [7]. VRP is a considerable concern
when the vehicle is a garbage truck, which travels
approximately 100 km/day with the average consumption of
1 litre/km. For this reason, a small improvement in collection
activities may result in significant savings in overall cost, fuel
and therefore greenhouse-gas emissions. Therefore, the
optimization of routes for collection and transportation of
municipal waste can be a crucial issue. This paper proposes
the use of geographic information systems (GIS) to minimize
fuel consumption, while taking into account the truck load
and geographic features (e.g. slopes, relief) of the area where
the waste is being collected.
The primary goal of this research is to find ways to reduce
overall travel distance for collection and transport of
municipal solid waste from residential homes within the
Blacktown City Area in order to reduce the fuel consumption
and therefore greenhouse-gas emissions.
The research for this paper aims to:
 Reduce the overall distance driven to collect and transport
residential waste;
 Reduce fuel consumption during the waste collection
process; and
 Decrease the greenhouse gases emitted by garbage trucks.
II. MATERIALS AND METHODS
A. Study Area
Blacktown City is the pilot area for this paper. Blacktown
is a modern bustling city of 48 residential suburbs, home to
350,000 people. This is the second largest local government
area by population in New South Wales, Australia. Covering
an area earmarked for considerable residential development,
with more than 4,000 developments approved in 2015-2016
and a population increase of nearly 2.5% in 2016 [8],
Blacktown City Council will experience an increasing
domestic waste stream over the next few years. Australia’s
Nationally Determined Contribution to the 2015 Paris
Agreement is to reduce its total greenhouse-gas emissions by
26-28% below its 2005 level. In line with this, Blacktown
Vehicle Routing Optimization for Improving Fleet Fuel
Efficiency: A Case Study in Sydney, Australia
Hoda Karimipour, Vivian W. Y. Tam, Helen Burnie, and Khoa N. Le
International Journal of Environmental Science and Development, Vol. 8, No. 11, November 2017
776
doi: 10.18178/ijesd.2017.8.11.1056

City Council is working to decrease its emissions. The
biggest opportunities in reductions are from electricity, street
lighting, gas, and transportation through improved fleet fuel
efficiency. This study will investigate the opportunities for
fleet fuel efficiency through route optimization for
Blacktown City Council.
For this study, Council’s waste collection Truck nNo. 634
was selected to investigate the applicability of a route
optimization model for improving fleet fuel efficiency. The
regular route of this truck was used to obtain the current
travel time and travel distance. Figure 1 shows the map of
Blacktown City and the collection area for Truck No. 634.
B. Route Optimization Using ArcGIS Network Analyst
Software, Esri’s ArcGIS Network Analyst extension
allows users to perform complex calculations to solve vehicle
routing problems. The program performs analysis over a
network of connected edges and decides fleet routing, travel
directions, closest facility, service area, and location
allocation. In the application for route optimization, network
dataset edges represent the road network being traversed.
Network Analyst allows the user to dynamically model
genuine network situations. These conditions can include
speed limit, traffic volume at different times of the day,
one-way streets, turn restrictions, obstacles, road conditions,
and limitations.
Network Analyst mainly uses Dijkstra’s algorithm, which
is a simpler algorithm that finds the shortest or lowest cost
path between two points. This algorithm preserves balance
between evaluating a near optimal path to travel with one that
is computationally practical. Dijkstra’s algorithm divides the
network dataset into lines or edges, with each edge
representing a traversable or non-traversable piece of the
network. In addition, each network edge also has an
associated cost, which represents the effort to travel that
specific segment of road. These costs are calculated using
one of two different criteria. The distance criterion is based
on total edge length, and the time criterion measures edge
length and time to traverse a segment [1]. The algorithm
creates nodes or junctions at the start, end, and intersection of
all edges; this defines the network by confirming there is
transitivity between edges and junctions through the entire
connected road network. The software calculates the cost to
reach a node then determines the least cost path to travel to
the next node. This continues until a final destination point is
reached. These steps create only a temporary and partial
solution. Once the initial cost is calculated between all the
stops, the software applies a Tabu-Search heuristic process.
This re-evaluates and confirms, then re-establishes a more
optimal path. This procedure continually runs to optimize the
current path until no further optimization can be performed.
The result is the least cost route to travel, along a path from a
start to end point. Depending on number of points (stops) to
make and complexity of the network, the analysis can take
seconds or hours to complete.
Fig. 1. The Blacktown City location in New South Wales, Australia.
C. Application of GIS in Waste Management
Esri’s ArcGIS 10.3 Network Analyst extension has been
used in this study. In the application for route optimization,
network dataset edges represent the road network being
traversed. Network Analyst allows the user to dynamically
model genuine network situations. These conditions can
include speed limits, one-way streets, turn restrictions,
obstacles, road conditions, and the other limitations.
The most critical functional requirement of the system was
to calculate optimal routes for solid waste collection. The
outputs should show travel distance, drive time and
greenhouse-gas emissions then comparisons among different
resulting routes could be conducted for the different
scenarios. To be able to calculate optimal routes, the system
was also required to store the service order and depot
locations. The service order area is in the suburb of Seven
Hills, located in the South East of Blacktown City. The area
includes about 30 streets/alleys from which household
garbage is collected. A road network dataset representing all
the streets within Blacktown City was also needed to be
compiled in order to properly perform route optimization
calculations through the ArcGIS Network Analyst extension.
The road edges and junctions had to have coincident
International Journal of Environmental Science and Development, Vol. 8, No. 11, November 2017
777

geometry for the route optimization calculations to perform
properly [4]. Finally, the existing collection route needed to be compiled for comparing current and new routes.
Fig. 2. Project’s major components and system design.
D. System Analysis and Design
Displays the project’s major components and how they
integrated into the workflow. To calculate optimal routes for
solid waste collection, there were several inputs to the
ArcGIS Network Analyst solver. Collection points
represented the depot start point, the rooftops of homes where
the garbage is collected and the unload point [4]. The unload
point is the UR-3R, (Urban Resources-Reduction, Recovery
and Recycling), which is an alternative waste treatment
facility located in the southern part of the city. In the UR-3R
process, municipal waste is sorted using the inherent
properties of material such as size, shape and density. There
are three main categories: recyclable, organics and
non-recyclable inorganic material [9]. A waste collection
vehicle in New South Wales is generally known as ‘a garbage
truck’. These side-loading trucks, have automatic ‘arms’ that
pick up a bin at the kerbside and empty it into the truck. The
driver then moves along the street to the next bin. There is no
need for people to run along near the truck to empty the bins
into it [10].
A network dataset of Blacktown City roads was
constructed to ensure all street segments are connected at
junctions. Additional GIS layers were incorporated into the
VRP solver to enable accurate representation of potential
routes travelled. For example, speed limitation and travel
time were included in the model. A raster DTM file was
imported into ArcGIS and included the elevation and
topographic relief information for the study area. These data
were used to perform test calculations of slope for each road
segment within the city. Also, different layers for calculating
greenhouse-gas emissions have been added to the model
including CO2, N2O and CH4 emissions for travelling each
section of the street.
With all the inputs and parameters set up, the VRP solver
can construct a new optimized route with optimal sequencing.
The route was evaluated for accuracy and route parameters
were then adjusted accordingly. Multiple route solver
iterations were run until an optimal collection route was
identified.
E. Research Scope
The scope of the project is to develop a detailed road
network dataset for the study area, which includes total
meters, total minutes, and GHG emission between each two
nodes. This involved exploring how different settings for the
network analyst parameters affect the optimization results.
The results from the new routes were examined and
compared to existing routes. While Blacktown City is divided
into 5 garbage collection areas, this analysis focused on one
collection area.
F. Data Sources
All project data were collected from one of three sources.
The client for this project was the Blacktown City. The city
has its own functioning GIS department which supplied the
collection areas, parcel data, route data, junction’s data, in a
GIS compatible format. Relevant sections of the Council,
which oversees city sanitation collection, supplied data
specific to the garbage collection process. The data included
operational hours, vehicle capacity, fuel type, driver
information, existing collection routes, The location of the
UR-3R waste treatment facility, and current collection
patterns. The QOL supplied the residential collection
locations in table format. The project required road data
features for the entire City of Blacktown. This feature class
provided detailed analysis of each road segment with
matching attribute information such as: road width, lanes,
speed, classification, name, and date of modification
G. GHG Emission Calculation
The Australian National Greenhouse Accounts Factors
were used to calculate the GHG emissions from the truck
transportation along the identified paths. According to this
standard, fuels used for transport purposes produce slightly
different methane and nitrous oxide emissions than if the
same fuels were used for stationary energy purposes. This
Factors list also includes a range of optional emission factors
for post-2004 vehicles and heavy vehicles conforming to
ArcGIS 10.3 Network
Analyst Extension
The street layer of the city
Collection point data
Existing collection route
Speed limitation
Travel time in each street0
GHG emission of travelling
Road network dataset
Optimized
Collection
Route
Vehicle
Routing Problem
Solver
International Journal of Environmental Science and Development, Vol. 8, No. 11, November 2017
778

Euro design standards.
Estimates of emissions from the combustion of individual
fuel types are made by multiplying a (physical) quantity of
fuel combusted by a fuel-specific energy content factor and a
fuel specific emission factor. This is performed for each
relevant greenhouse gas (in this case, carbon dioxide,
methane and nitrous oxide). Total greenhouse emissions are
calculated by summing the emissions of each fuel type and
each greenhouse gas. The following formula has been used to
estimate greenhouse gas emissions from the combustion of
each type of fuel listed used for transport energy purposes
[11].
𝐸𝑖𝑗 = 𝑄𝑖 × 𝐸𝐶𝑖 × 𝐸𝐹𝑖𝑗𝑜𝑥𝑒𝑐
1000
where:
Eij is the emissions of gas type (j), carbon dioxide,
methane or nitrous oxide, from fuel type (i) (CO2-e tonnes).
Qi is the quantity of fuel type (i) (kilolitres or gigajoules)
combusted for transport energy purposes
ECi is the energy content factor of fuel type (i) (gigajoules
per kilolitre or per cubic metre) used for transport energy
purposes If Qi is measured in gigajoules, then ECi is 1.
EFijoxec is the emission factor for each gas type (j)
III. RESULTS AND DISCUSSIONS
The result of the route optimization for the study truck has
been shown in Fig. 3
This figure shows the location of the bins that need
emptying and the path that should be travelled from
Council’s depot to the collections area and then to the UR-3R
facility at Eastern Creek, where the truck unloads. Compared
to the regular route this optimized one will reduce the travel
distance by 8 km/day for the pilot truck
In addition, the CO2, N2O and CH4 emissions from this
optimized route have been calculated. To calculate the
emissions of the total route, the emissions for each section of
each street were calculated and then the optimized route
pieced together. This enabled the total emissions of the route
to be extracted. As shown in the Fig. 3, the total emissions of
the route are 69.78 Kg. This optimized route, will reduce the
daily CO2 emissions by 5.5 kg for this collection area. That
means if this path is travelled by the pilot truck one day per
week for 52 weeks per year, it annually reduces emissions by
1386 Kg of CO2 compared with the regular route. If the
routes of all 400 trucks owned by Blacktown City Council
can be optimized between 4 to 8 Km per day like the pilot
truck, the average greenhouse-gas emissions reduction would
be between 300 to 600 tonnes per year.
IV. CONCLUSION
This paper examined the applicability of vehicle route
optimization on improving fleet fuel efficiency and reducing
greenhouse-gas emissions. The pilot area was one garbage
collection area of Blacktown City, Sydney, New South Wales,
Australia. For this purpose, Esri’s ArcGIS 10.3 Network
Analyst extension was used. The most critical functional
requirement of the system was to calculate optimal routes for
solid waste collection. The outputs showed travel distance,
drive time and greenhouse emissions before and after route
optimization. The results of this study show that:
 Using the optimized route instead of the regular route can
reduce the total travelled distance by 8 km/day on the
pilot site.
 The optimized route will reduce that individual truck’s
emissions by 5.5 kg CO2 per day for that collection area.
This represents a reduction of about 8% for that particular
collection.
 The total reduction in fuel consumption for a similar
garbage truck for approximately the same distance will be
193 litres per four-weeks, per truck.
Fig. 3. The result of route optimization for waste collection.
International Journal of Environmental Science and Development, Vol. 8, No. 11, November 2017
779

Future studies in this field can consider the fuel efficiency
effects of other factors, such as load and type of trucks.
Future work could focus on the effects of elevation, traffic
load and the weight of the truck on the optimized routes. The
impacts of driving behaviour on the fuel consumption is an
additional area for future research.
REFERENCES
[1] N. V. Karadimas et al., "Routing optimization heuristics algorithms for
urban solid waste transportation management," WSEAS Transactions
on Computers, vol. 7, no. 12, pp. 2022-2031.
[2] Z. Zsigraiova, V. Semiao, and F. Beijoco, "Operation costs and
pollutant emissions reduction by definition of new collection
scheduling and optimization of MSW collection routes using GIS. The
case study of Barreiro, Portugal," Waste Manag., 2013, vol. 33, no. 4,
pp. 793-806.
[3] G. Ristić et al., "Methodology for rout optimization for solid waste
collection and transportation in urban areas," FACTA UNIVERSITATIS,
2015, vol. 12, no. 2, pp. 187-197.
[4] D. L. O'Connor, Solid Waste Collection Vehicle Route Optimization for
the City of Redlands, California, 2013, University of Redlands.
[5] G. B. Dantzig and J. H. Ramser, "The truck dispatching problem,"
Management Science, 1959, vol. 6, no. 1, pp. 80-91.
[6] Y. Suzuki, "A new truck-routing approach for reducing fuel
consumption and pollutants emission," Transportation Research Part
D: Transport and Environment, 2011, vol. 16, no. 1, pp. 73-77.
[7] Y. Xiao et al., "Development of a fuel consumption optimization
model for the capacitated vehicle routing problem," Computers &
Operations Research, 2012, vol. 39, no. 7, pp. 1419-1431.
[8] Population and Household Forecasts, 2011 to 2036, 2016.
[9] C. Georgiadis, "Costa visits Australia's biggest compost heap, in
gardening Australia," 2013.
[10] SUEZ Environnement. Putting Your Waste to Good Use, 2015,
Australia.
[11] Department of the Environemnt, National Greenhouse Accounts
Factor, 2015, p. 78.
Hoda Karimipour is a PhD research scholar in
environmental engineering at Western Sydney
University, Sydney, Australia. She received her master
of science in environmental management from
University of Tehran in 2009. From June 2009 to
February 2016, she had been senior technical expert at
United Nations Development Programme. From
February 2016 to now, she has been lecturer/tutor at
Western Sydney University in different aspects of
engineering. Her current research project is “The Carbon Footprint
Reduction Toolkit for Blacktown City Council”. Her research interests are
dynamic modeling for environmental and ecological changes, designing
dynamic monitoring systems for climate change, GIS and remote sensing
application in climate change studies, decision support systems for natural
resource management, bayesian network models for natural resource
problems. She has been editorial board member for 3 international journals.
She has published over 4 books, 4 book chapters, 8 referred journal articles
and 10 referred conference articles.
Vivian W. Y. Tam is the director of research quality
and innovation and director of higher degree research
student excellence at School of Computing,
Engineering and Mathematics, Western Sydney
University, Australia and honorary professor at College
of Civil Engineering, Shenzhen University, China. She
received her Ph.D. in sustainable construction from the
Department of Building and Construction at City
University of Hong Kong in 2005. Her research
interests are in the areas of environmental management in construction and
sustainable development. She is currently the editor of International Journal
of Construction Management and the Research Group Leader for
Sustainable Construction Management and Education Research Group under
the School. She has published over 3 books, 19 book chapters, 196 referred
journal articles and 91 referred conference articles. She has been awarded
thirty-one research grants (totalled AU$2.1 million), including the first of
two Australian Research Council (ARC) Discovery Projects, awarded under
FoR 1202 (Building).
Helen Burnie is a senior environment officer with
Blacktown City Council. Her focus is sustainability,
especially climate change policy and its
implementation. She gained her PhD in environmental
studies from University of Western Sydney in 2014.
Her research interests are in social science and policy
that support individuals and organizations in
transitioning to more sustainable practice. In
2008-2011 she worked on the Regenesis carbon
reforestation project for Blacktown City Council. In 2013-15 she worked
with NSW Office of Environment and Heritage to engage regional
communities on improving energy efficiency and accessing renewable
energy. She has tutored, lectured and worked on research projects for
Western Sydney University’s School of Science and Health, and School of
Education.
Khoa N. Le received his Ph.D. in October 2002 from
Monash University, Melbourne, Australia. From April
2003 to June 2009, He was a lecturer at Griffith
University, Gold Coast Campus, Griffith School of
Engineering. From January to July 2008, he was a
visiting professor at Intelligence Signal Processing
Laboratory, Korea University, Seoul, Korea. From
January 2009 to February 2009, he was a visiting
professor at the Wireless Communication Centre,
University Technology Malaysia, Johor Bahru, Malaysia. He is currently a
senior lecturer at School of Computing, Engineering and Mathematics,
University of Western Sydney. His research interests are in wireless
communications with applications to structural problems, image processing
and wavelet theory. Dr. Le is the editor in Chief of International Journal of
Ad Hoc, Sensor & Ubiquitous Computing (IJASUC). He has also been on
the Editorial Board of International Research Magazine of Computer
Science and other wireless communications journals.
International Journal of Environmental Science and Development, Vol. 8, No. 11, November 2017
780