Impact of Urban

Trac on Fuel

Consumption

Leveraging IoT Data

Case Study of

Riyadh City

Lama Yaseen, Nourah Al-Hosain,

Ibrahem Shatnawi, and Abdelrahman Muhsen

Discussion Paper

December 2024 I Doi: 10.30573/KS--2024-DP72

About KAPSARC

KAPSARC is an advisory think tank within global energy

economics and sustainability providing advisory services to

entities and authorities in the Saudi energy sector to advance

Saudi Arabia’s energy sector and inform global policies through

evidence-based advice and applied research.

This publication is also available in Arabic.

Legal Notice

Center (“KAPSARC”). This Document (and any information, data

or materials contained therein) (the “Document”) shall not be

used without the proper attribution to KAPSARC. The Document

shall not be reproduced, in whole or in part, without the

written permission of KAPSARC. KAPSARC makes no warranty,

representation or undertaking whether expressed or implied,

nor does it assume any legal liability, whether direct or indirect,

or responsibility for the accuracy, completeness, or usefulness

of any information that is contained in the Document. Nothing in

the Document constitutes or shall be implied to constitute advice,

recommendation or option. The views and opinions expressed in

this publication are those of the authors and do not necessarily

reﬂect the ocial views or position of KAPSARC.

3Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Abstract

This study explores the rising trend of trac congestion in Riyadh and its

impact on fuel consumption for passenger cars amid the challenges of rapid

urbanization and increasing vehicle use. By utilizing real-time ﬂoating car data

(FCD) collected by vehicles equipped with Global Positioning System (GPS)

technology and communication systems, this study illustrates the potential

of the Internet of Things (IoT) and smart city technologies in developing

intelligent transportation systems and improving urban mobility management.

A spatial analysis of the trac ﬂow dynamics in Riyadh, focusing on selected

primary highways, reveals that driving on high-capacity roads tends to

increase fuel consumption. We conducted an analysis at a mesoscopic level,

representing trac congestion in Riyadh on high-capacity roads. It shows that

trac congestion leads to up to a 29% increase in fuel consumption, primarily

due to frequent stop-and-go driving behavior, reducing overall fuel eciency.

This study enhances our understanding of urban trac patterns, providing

policymakers with data-driven insights to help them create more sustainable

road planning strategies to address the speciﬁc needs and challenges of

urban mobility in cities.

Keywords: Fuel Consumption, Congestion, Trac, GIS, IoT, FCD

4Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

1. Introduction

Rapid urbanization, coupled with a signiﬁcant increase in vehicle ownership,

has intensiﬁed trac congestion in cities globally, adversely aecting air

quality and substantially increasing greenhouse gas emissions (Sims et al.

2014). In Saudi Arabia (KSA), the reliance on cars for daily commuting, coupled

with rapid urban development, poses signiﬁcant transportation challenges.

Riyadh, as the nation’s capital and largest city, has experienced a marked

annual increase in travel times, driven by escalating trac congestion

(TomTom 2024a). This highlights the critical need for sustainable urban

planning and eective trac management strategies to meet the increasing

mobility demands of its expanding urban population.

Floating car data (FCD) refers to data collected from

moving vehicles equipped with Global Positioning

System (GPS) devices or other location-tracking

technologies. This data includes information about

the vehicle’s location, speed, and direction, which

can be used to analyze trac patterns, monitor road

conditions, and manage transportation systems

(Liu and Ban 2013).

FCD plays a pivotal role in enhancing the Internet of

Things (IoT), smart cities, smart mobility, and research

by providing real-time insights into trac dynamics and

vehicle movements. In this context, smart mobility refers

to the use of advanced technologies and data-driven

solutions to optimize transportation systems, improving

eciency, accessibility, and sustainability (Faria et al.

2017). Smart cities, in turn, are urban environments that

utilize digital technology, data analytics, and the IoT to

integrate and enhance public services, infrastructure,

and governance, creating more ecient, sustainable,

and livable communities (Silva, Khan, and Han 2018).

These smart cities rely on smart mobility solutions to

reduce congestion, improve transportation options, and

support a greener urban ecosystem. For IoT applications,

FCD supports the ecient monitoring and management

of connected vehicle systems, enabling predictive

maintenance and ﬂeet optimization (Ayfantopoulou

et al. 2022).

In the realm of smart mobility, FCD is crucial for

developing intelligent transportation systems. It provides

granular data on trac patterns, vehicle speeds, and

travel times, which aids in optimizing public transit and

developing more ecient trac management strategies

(Ismaeel et al. 2023). Additionally, FCD is valuable for

research purposes, oering a rich dataset for analyzing

trac behaviors, testing urban planning theories, and

developing new transportation models. By continuously

collecting and analyzing this data, cities and researchers

can create more responsive, ecient, and sustainable

urban environments (Liu and Ban 2013).

The integration of advanced data sources like FCD has

become invaluable for transport demand modeling

by providing real-time, high-resolution data on trac

ﬂow, vehicle speeds, and travel patterns. This detailed

information allows for more accurate predictions of

travel demand, helps identify congestion patterns,

and supports the optimization of trac management

strategies (Croce et al. 2021; Kan et al. 2018; Jiang et al.

2021; Xu, Yue, and Li 2013). Furthermore, the utilization of

spatial analysis techniques will enable the identiﬁcation

of localized increases in congestion and emissions on

urban roads, providing a comprehensive understanding

of trac behaviors and supporting the development of

targeted interventions to improve trac ﬂow and reduce

environmental impacts.

5Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

This paper examines the challenges of trac congestion

and mobility in Riyadh, beginning with a review of

relevant literature. It then outlines the spatial modeling

techniques and FCD data used to analyze urban driving

patterns and how various conditions impact vehicle fuel

eciency, with a particular focus on understanding how

speed variations aect fuel consumption. The study

focuses on selected high-capacity roads (highways) in

Riyadh, examining trac behavior on these key roadways

on a mesoscopic level. Additionally, a spatiotemporal

analysis is conducted to provide various comparable

driving patterns across the city. The discussion

elaborates on the insights acquired from the earlier

analysis to understand the impact of trac congestion

and urban conditions on driving patterns and fuel

consumption. The paper concludes with a summary

of key ﬁndings from the spatial analysis, highlighting

factors inﬂuencing trac congestion and fuel eciency

in Riyadh. Additionally, it emphasizes the signiﬁcance

of the IoT and smart technologies in enhancing urban

mobility management, and suggests potential policy and

infrastructural interventions to address these challenges.

6Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

2. Literature Review

Cities play a pivotal role in the dialogue on climate change, acting both as

major contributors to and frontline responders against the phenomenon.

The concentration of people, industries, and services in urban areas

leads to signiﬁcant energy consumption and, consequently, substantial

carbon dioxide (CO2) emissions. However, urbanization presents a unique

opportunity to mitigate climate change through sustainable city planning,

innovations in public transportation, energy eciency, and the inclusion

of cleaner, more ecient cars. Focusing on the role of vehicles and trac

congestion, it becomes evident that these factors are not just contributors

to urban CO2 emissions but also key areas where interventions can yield

signiﬁcant environmental beneﬁts (Kamal-Chaoui and Robert 2009). By

addressing the challenges posed by cars and trac congestion, cities can

make strides toward reducing their carbon footprint and enhancing their

overall sustainability.

2.1 Trac

Congestion in

Car-Dependent

Cities

Riyadh city has been described as heavily reliant on

private cars for transportation, with studies highlighting

the city’s universal dependence on automobile travel

for all journeys. This reliance on cars has led to issues

such as unrestrained urban sprawl and a focus on

catering to cars more than to any other mode of transport

(Al-Mosaind 1998; Al Zohbi 2021).

Numerous studies in academic literature have examined

the growing problem of trac congestion in Riyadh,

with many proposing potential solutions to address this

challenge. Al-Mosaind (1998) recognized the growing

issue of trac congestion over two decades ago,

emphasizing the need for integrated land-use policies to

address and alleviate the problem. Youssef, Alshuwaikhat,

and Reza (2021) explored the potential for alleviating

trac congestion by promoting mode shifts, such as

the introduction of a public transport system to reduce

the reliance on private vehicles and improve the trac

ﬂow in urban areas. Other studies have suggested more

technical solutions, utilizing emerging IoT technologies.

For instance, Al-Majhad et al. (2018) proposed a new

trac congestion framework based on IoT services to

provide residents with real-time trac information and

improve trac management in the city. Collectively, the

body of research highlights the importance of adopting a

comprehensive approach to address trac congestion,

which is crucial for mitigating the adverse environmental

and economic impacts that urban centers face (Al Zohbi

2021).

The relationship between trac congestion and CO2

emissions has been extensively examined in the

literature. Barth and Boriboonsomsin (2008) examine

the relationship between congestion levels, average

trac speeds, and CO2 emissions, and show that

deteriorating trac conditions directly contribute to

rising CO2 emissions. They also detail in the article

how, as congestion increases and average speeds

decrease, vehicles exhibit more ‘stop-and-go’ driving

7Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

patterns, leading to higher CO2 emissions. Zhang and

Batterman (2013) further highlight the link between

prolonged vehicular congestion and increased emissions,

which pose signiﬁcant air quality and public health

challenges. Moreover, Sitati et al. (2022) arm that

congestion signiﬁcantly inﬂuences CO2 emissions at

the street level, which has detrimental environmental

impacts. Such studies emphasize the critical need for

eective congestion mitigation strategies to reduce the

environmental burden of urban trac.

2.2 The Role of

Smart Mobility

Smart cities and smart mobility are interconnected

because both use technology and data-driven solutions

to optimize urban infrastructure and transportation

systems. These solutions promote sustainability by

reducing congestion, lowering emissions, and improving

resource management and quality of life. A report

addressing the United Nations Sustainable Development

Goals (SDGs) for transport and development (United

Nations 2021) highlights the critical role of smart cities

and intelligent transport systems (ITS) in advancing

sustainable urban development. As cities house over half

of the world’s population – a number expected to rise –

they are key players in promoting sustainable transport

initiatives. The implementation of smart city technologies,

including ITS, is essential for improving the quality of

life in cities, enhancing their operational eciency, and

maintaining their competitive edges.

These technologies streamline urban transport

management through sophisticated applications like

e-hailing, real-time trac navigation, and smart parking

systems. Moreover, the integration of IoT technologies

and big data, including FCD, is essential for enhancing

services, particularly in sustainable trac planning,

which plays a key role in advancing the environmental

sustainability of urban transport.

It is essential to integrate diverse sources of mobility

data, including public transportation, road sensors,

surveys, and social media, into integrated databases to

develop an urban mobility atlas for smart cities (Faria et al.

2017). Smart mobility has the potential to reduce trac

congestion, commuting times, and trac collisions, by

oering passengers the ability to customize their journeys

(Bıyık et al. 2021).

In 2024, The Saudi Data and Artiﬁcial Intelligence

Authority (SDAIA) established the Center of Excellence

in Congestion Solutions Using Data and Artiﬁcial

Intelligence (AI) as part of its eorts to leverage advanced

technologies to tackle urban challenges. The center

focuses on crafting intelligent solutions that utilize

data and AI to improve trac management and reduce

congestion, aiming to signiﬁcantly enhance the quality

of life for urban dwellers (SPA 2024). By fostering smart

solutions across various government sectors, the center

supports data-driven policymaking, which is crucial

for enhancing urban sustainability and eciency. This

development serves as a prominent example of Saudi

Arabia’s commitment to integrating smart AI solutions in

addressing urban issues.

8Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

3. Methodology

In this methodology, FCD is employed to estimate fuel consumption and

vehicle eciency under speciﬁc driving conditions. The spatial distribution

of the collected data is leveraged to analyze various road segments, with a

focus on selected highways throughout the city of Riyadh. Additionally, the

selection of these roads was based on high-capacity routes with a maximum

speed of 120 kilometers per hour (kph) to ensure comparability. King Fahad

Road was also included, despite its lower speed limit of 100 kph, due to its

recognition in the literature as one of the roads in Riyadh most aected by

trac congestion issues (Al-Mosaind 1998).

3.1 Floating Car

Data (FCD)

FCD is collected from GPS-equipped vehicles and

provides detailed information on vehicle movement. This

data includes latitude, longitude, speed, and time stamps,

which can be used to analyze trac patterns and estimate

fuel consumption and emissions.

3.1.1 Challenges with Commercial FCD

One of the major challenges associated with FCD is

the variability in dataset penetration, which signiﬁcantly

inﬂuences the accuracy and reliability of the data. The

penetration rate refers to the proportion of vehicles

equipped with data transmission capabilities within the

overall trac ﬂow. Low penetration rates result in sparse

data that may not accurately represent trac conditions,

leading to potential biases and inaccuracies in the

analysis (Fourati, Dabbas, and Friedrich 2021). Altintasi,

Tuydes-Yaman, and Tuncay (2022) emphasized the

importance of combining FCD data with ‘ground truth’

data. They noted that datasets obtained from commercial

entities often lack standardization due to varying

technologies and methodologies, which complicates the

integration and comparison of data from multiple sources.

Moreover, commercial datasets may have proprietary

constraints, which limit the transparency and veriﬁability

of the data (Leduc 2008; Gitahi et al. 2020). However,

the concern over penetration rates does not render the

dataset unreliable. In fact, some studies have found that

penetration rates as low as 10%-14% are sucient for

accurately measuring parameters like travel time, even

though measuring density or volume typically depends

on the total number of cars captured on the roads

(Altintasi et al. 2022; Wang et al. 2015). In this paper, we

concentrate on the increase in fuel consumption due to

increased congestion and reduced speeds under the

assumption of a typical passenger vehicle navigating the

selected road segments.

3.1.2 Collected Data Characteristics

and Geographical Scope

The trac data utilized in this study is sourced from

TomTom (2024a). The analysis in this paper speciﬁcally

focuses on passenger car trac data collected on

Saudi weekdays (Sunday to Thursday). For this case

study, we limited the geographical scope to Riyadh city

only, following the ocial boundaries shown in Figure 1.

Additionally, we processed the trac data to obtain hourly

average metrics for each road segment, providing a

detailed hourly proﬁle of urban trac ﬂows in Riyadh.

TomTom’s Functional Road Class (FRC) categorizes roads

based on the service they provide, facilitating vehicle

movement across various locations such as cities, towns,

and recreational areas. The FRCs are deﬁned as follows:

FRC 0 includes motorways and major roads; FRC 1 covers

major roads slightly less important than those in FRC

0; FRC 2 includes other major roads; FRC 3 represents

secondary roads; FRC 4 consists of local connecting

9Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

roads; FRC 5 refers to local roads of high importance;

FRC 6 and FRC 7 cover local roads of varying importance;

and FRC 8 includes other roads (typically pedestrian

and cycling paths) (TomTom 2024c). For our analysis,

we excluded FRC 8 due to its focus on non-vehicular

pathways.

The analysis is conﬁned to a subset of FCD data focused

on passenger vehicles, operating under the following

deﬁned assumptions and constraints:

• The data is spatially organized as a network and is

recorded at the segment level. Each record represents

a speciﬁc road segment, capturing details such

as average speed variations, road classiﬁcations,

average travel time, road names, and segment lengths

(Figure 1).

• The data collected only includes passenger vehicles

for 20 working days in October 2022 (Sunday, Monday,

Tuesday, Wednesday, and Thursday).

• We assume that the subset of the private cars

extracted from the FCD data only represents gasoline

consumption in that window.

• The functional road classes in the subsets range from

FRC 0 to FRC 7.

• The total length of the roads in the datasets amounts to

957,721 kilometers, as deﬁned in Figure 1.

Figure 1. Scope and time of the data for the study of Riyadh city.

Road network

Riyadh boundry

522

6330

740 m

5318

5309

533

535

nillaa

al Park

505

509

10 20 40 Km

Source: Network data from TomTom (2024a) clipped to the deﬁned boundaries.

10Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

3.1.3 Measuring Trac Congestion

Quantifying congestion can be done in many ways,

with numerous measures developed to assess dierent

performance criteria. These include speed, travel

time, delay, level of service, and congestion indices.

While congestion is often measured through various

means such as delay ratios and travel time indices

(TTI), our methodology focuses speciﬁcally on speed-

related indicators (Afrin and Yodo 2020). We will be

using the Speed Reduction Index (SRI) to measure

trac congestion, as it aligns with our objective of

understanding how variations in speed inﬂuence fuel

consumption. This is critical for quantifying the impact

of speed ﬂuctuations on overall fuel eciency.

The SRI we are calculating is based on the dierence

between the average vehicle speed and the speed limit,

expressed as a ratio of the speed limit. The method

allows for a straightforward and intuitive measure of trac

congestion, reﬂecting to what extent trac has slowed

relative to ideal free-ﬂow conditions. The SRI is calculated

as follows:

SRI

max 0, Speed Limit

−

AverageSpeed

Speed Limit

⎛

⎝

⎜⎞

⎠

⎟ (1)

This approach is aligned with the methodology presented

in Afrin and Yodo (2020) and Rao and Rao (2012) as

the formula captures the degree of speed reduction

under congested conditions. However, one adjustment

we are considering is to avoid multiplying the results

by 10, maintaining a scale from 0 to 1 instead of 0 to 10.

Additionally, we want to ensure the measure cannot be

negative, as we aim to exclude instances where vehicles

are traveling faster than the road’s speed limit, focusing

on free-ﬂow trac conditions.

3.2 Estimating Fuel

Consumption

3.2.1 Incorporating Urban Driving

Conditions

Research has demonstrated that driving behaviors such

as accelerating, idling, speeding, and stop-and-go trac

patterns signiﬁcantly inﬂuence a vehicle’s emissions.

Aggressive driving, for instance, typically leads to

greater fuel consumption and increased emissions

compared to steady, moderate driving (Wang et al.

2019). Frequent idling and constant variations in speed,

common in congested urban settings, also contribute to

greater emissions. Thus, understanding and modifying

driving behaviors can be a key strategy in reducing

vehicular carbon emissions (Wang et al. 2019). Similarly,

reducing trac congestion, which leads to stop-and-go

trac patterns, can also reduce vehicle emissions.

In this paper, we adopt a methodology similar to previous

studies that have examined the relationship between

average speed and both fuel consumption and emissions.

We utilize average speeds obtained from FCD data from

TomTom (2024a) to conduct our analysis. Previous studies

have developed speciﬁc parameters, primarily focusing

on vehicle speed and vehicle characteristics.

The relationship between fuel consumption and speed, as

highlighted in studies such as Barth and Boriboonsomsin

(2008), Sobrino, Monzón, and Hernández (2014), Song, Yu,

and Wu (2016), and Ricardo-AEA (2014), generally follows

a U-shaped curve. This means that fuel consumption

is higher at very low speeds, decreases as speed

approaches an optimal range, and then increases again

at higher speeds. This U-shape is inﬂuenced by factors

such as engine eciency, aerodynamic drag, and rolling

resistance, all of which vary based on vehicle type, as

these studies demonstrate. At lower speeds, engines are

less ecient, and more fuel is consumed to overcome

inertia and friction. As speed increases, engines enter

their optimal operating range, leading to reduced fuel

consumption per distance unit. However, at higher speeds,

aerodynamic drag becomes more dominant, demanding

more energy and causing fuel consumption to rise again.

The exact nature of this curve can vary depending on

vehicle type and engine size, as shown in the previous

studies. Barth and Boriboonsomsin (2008) created the

curves by using second-by-second vehicle velocity data,

integrating it with real-time trac data, segmenting it by

consistent levels of service, applying the Comprehensive

Modal Emissions Model to estimate CO2 emissions, and

ﬁtting the results to a fourth-order polynomial.

Sobrino, Monzón, and Hernández (2014) collected

consumption factors from the COPERT IV database,

establishing fuel consumption curves based on speed

for each vehicle type, customizing these curves using

non-linear regression techniques, and adapting them

to speciﬁc case studies by considering the circulating

vehicle ﬂeet composition and annual mileage

correction factors.

Song, Yu, and Wu (2016) collected real-world vehicle

activity data, developing speed-speciﬁc vehicle-speciﬁc

11Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

power distribution models, and calculating speed

correction factors by combining these distributions with

emission rates.

Ricardo-AEA (2014) developed the parameters by ﬁtting

speed-emission factors to vehicle categories using a

We use the parameters from the Ricardo-AEA’s (2014)

report for the United Kingdom (U.K.) National Atmospheric

Emissions Inventory (NAEI), which reﬂects the U.K.’s

country-wide ﬂeet composition, enabling us to develop

a similar consumption curve for Saudi Arabia. The U.K.

study includes parameters for petrol cars running on

gasoline, which aligns with our study, given its focus on

gasoline consumption in passenger vehicles.

polynomial equation. These factors were scaled for fuel

quality and emission degradation, and the coecients

were weighted according to the ﬂeet composition for each

vehicle category. The ﬁnal curves were then normalized

to match real-world data. Table 1, below, shows the curves

developed by the relevant studies.

The parameters developed are formulated

based on a function of speed, embodying the

dynamics of urban driving conditions as outlined

in Table 2. They are typically established by

considering factors such as engine size, vehicle

weight, and other relevant variables that can impact

fuel consumption.

Table 1. Overview of studies developing curves for estimating fuel consumption and emissions.

Curve type Data source Optimal

speed (when

consumption

at lowest)

Low-speed

impact (when

consumption

increases)

Reference

Fuel consumption

(various fuels,

including gasoline

and diesel)

Fleet composition

of the United

Kingdom (U.K.) for

various years

Around 75-85 kph 45 kph and below Ricardo-AEA 2014

Fuel consumption

for gasoline

Collected

passenger vehicles

in Spain with

engines smaller

than 1.4 liters (L)

Around 65-80 kph 40 kph and below Sobrino, Monzón,

and Hernández

2014

CO2 emissions Vehicle activity

data collected from

probe vehicles on

Southern California

freeways,

2005-2007

Around 72-80 kph

(originally reported

as around 45 miles

per hour [mph])

Around 48 kph

(originally reported

as around 30 mph

and below)

Barth and

Boriboonsomsin

2008

Hydrocarbons,

nitrogen oxide

(NOx) and carbon

monoxide (CO)

Vehicle data

collected from taxis

from 2005 to 2012

on speciﬁc roads in

Beijing

For hydrocarbons:

around 75-80 kph

20 kph Song, Yu, and Wu

2016

12Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

The function is represented as:

(2)

where

• FC = fuel consumption (liters per kilometer [L/km])1

• v = average traveling speed (kilometers per hour

[km/h]).

• The parameters a, b, c and d for each vehicle category

incorporate diverse speed scenarios, from slow-paced

urban driving to faster motorway conditions.

These parameters reﬂect the variation in fuel usage

corresponding to dierent vehicular speeds in the U.K.,

based on the 2015 ﬂeet composition data. The speed

spectrum between 10 kph and 130 kph is designed to

mirror the real-world driving scenarios encountered

within city limits, where vehicles navigate a mix of slow-

moving trac, stop-and-go conditions, and occasional

stretches of open road that allow for higher speeds. The

lower end of the spectrum (10 kph) represents congested

urban trac or residential areas, while the upper limit

(130 kph) accounts for urban expressways or major

roads where vehicles can travel at higher speeds. This

range ensures that the emission curves derived from

the test are representative of the vast majority of driving

conditions in urban areas, providing a comprehensive

overview of vehicle emissions across dierent speeds.

However, while the approach of Ricardo-AEA (2014)

provides valuable insights, it is important to note that

mesoscopic modeling, as used in this paper, does

not capture the individual vehicle interactions seen

in microscopic models (Ferrara, Sacone, and Siri

2018). Although we can identify congestion patterns

on high-capacity roadways, The mesoscopic nature

of the methodology adopted in this paper does not

account for the distinctive behaviors of cars in smaller

neighborhoods, where vehicles take turns and stop for

trac lights. However, it is suited for examining city-

wide eects and larger areas. To make the model more

comprehensive, a hybrid mesoscopic-microscopic

trac simulation model could improve fuel consumption

analysis by bridging the gap between large-scale

network behavior and localized driving dynamics. The

hybrid model could have the capability to address these

limitations by combining broader network simulations with

detailed vehicle interactions, providing more accurate

insights into how congestion on major roads and speciﬁc

driving behaviors in smaller areas, such as frequent

stopping or turning, impact fuel consumption (Burghout,

Koutsopoulos, and Andreasson 2005). This integrated

approach oers a more comprehensive understanding

of trac conditions and is a key consideration for future

work within the context of this study.

3.2.2 Estimating Fuel Consumption

for Saudi Arabia’s Vehicle Composition

To be able to utilize the parameters to estimate the fuel

consumption for Riyadh, we assume that the composition of

the nation’s vehicle ﬂeet at the national level is equivalent

to that of Riyadh. The NAEI (2019) report categorizes car

composition for the U.K. by engine size, and we replicated

this using Saudi Arabia’s 2023 car type distribution from the

Industrial Cluster (IC) automotive market report (IC 2024).

We then mapped the relevant car types to engine sizes

based on commercial car market data from Motory (2022).

Table 2. Fuel consumption parameters.

Parameters

Vehicle category a b c d Min speed

(kph)

Max speed

(kph)

Petrol car 0.45195 0.09605 −0.00109 0.000007 10 130

Sources: NAEI (2019); UK Department of Transport (2018).

13Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

The car market data included over 2,000 cars (with car

speciﬁcations) sold in Saudi Arabia between 2020 and

2023. Data from the industrial cluster, which includes

the percentage distribution of purchased car brands,

allowed us to reﬁne our calculations. We then weighted

the distribution of car brands w_brandi and body types

w_bodyTypei according to data from IC (2024) and

correlated these weights with engine sizes based on the

most popular brands and vehicle types in the market. This

approach allowed for a more accurate representation of

the local vehicle ﬂeet in our calculations.

KSA

percentage

for

each

engine

category j

∈

∑

brandi

bodyTypei

( )

∑

∈

∑

brandi

bodyTypei

( )

⎛

⎝

⎜

⎞

⎠

⎟

KSA

percentage

for

each

engine

category j

∈

∑

brandi

bodyTypei

( )

∑

∈

∑

brandi

bodyTypei

( )

⎛

⎝

⎜

⎞

⎠

⎟

(3)

Where:

• j is the engine category.

• i ∈ j represents the cars in engine category j

• k represents all engine categories.

Table 3 shows the fuel consumption in gallons per

kilometer (g/km) for three engine size categories,

alongside the ﬂeet composition percentages for KSA and

the U.K. The weighted average fuel consumption and

the ratio of the KSA’s to the U.K.’s weighted average fuel

consumption are also shown.

In the table, the fuel consumption values, expressed in

grams per kilometer (g/km), represent the average amount

of fuel consumed for every kilometer driven by vehicles of

dierent engine sizes. These numbers were adopted from

the 2019 European Monitoring and Evaluation Programme

(EMEP)/European Environment Agency (EEA) air pollutant

emission inventory guidebook published by Ntziachristos

and Samaras (2019).

The U.K. ﬂeet composition was retrieved for the

year 2015 because this was the baseline year for

the development of the fuel consumption curve

in equation (2). The KSA ﬂeet composition was

retrieved from IC (2024) composition with Motory

(2022) data.

Table 3. Comparison of fuel consumption and ﬂeet composition between the KSA and U.K.

Engine size (L) Fuel consumption for a

typical car (g/km)

KSA ﬂeet composition (%) U.K. ﬂeet

composition (%)

Small (engine size < 1.4L) 56 2.2% 49.2%

Medium (engine size 1.4

to 2.0L)

66 32.1% 40.2%

Large (engine size > 2.0L) 86 65.9% 10.7%

Weighted average FC (g/km) 78.9 63.3

Ratio of KSA’s weighted average FC value to U.K.’s weighted

average FC value

1.25

Source: Authors.

The weighted average fuel consumption calculation Wksa

reveals a 25% higher consumption rate for the Saudi

Arabian ﬂeet compared to that of the U.K. This dierential

is attributed to the distinct composition of the vehicle

ﬂeet in each country. Consequently, when scaling the

U.K. fuel consumption values to reﬂect Saudi Arabian

conditions, this percentage increment is factored in,

adjusting the values to account for the prevalent urban

driving conditions and vehicle types within Saudi Arabia,

as shown in Figure 2.

Figure 3 presents a schematic structure of the

methodology for estimating fuel consumption for the

segment Sfc.

14Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Figure 2. Fuel consumption rates by speed: UK vs. KSA estimated for petrol cars.

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0.18

020406080 100 120

140

Fuel consumption (l/km)

Speed (kph)

UK KSA

Source: U.K. calculation from (U.K. Department of Transport 2018) and KSA calculation by authors.

Figure 3. Schematic structure of the methodology for calculating the fuel consumption of each segment.

FCD

Road

segments

Road length for

each segment

SroadLength

Average

segment

speed (SV)

Speed reduction

index (SRI)

Fuel consumption

with U.K.

parameters (FCUK)

Fuel consumption for

vehicle driving in

segment (SFC)

Adjusted FC increase factors based

on KSA fleet composition (WKSA) = 1.25

Adjust fuel

consumption for

KSA (FCKSA)

Urban correction

parameters (a, b, c and d)

SRI

= max

(

0, ––––––––––––––––––––––––––

)

Speed limit – Average speed

Speed limit

FCKSA = FCUK × WKSA

SFC = FCKSA × SroadLength

FCUK = ––– + b + (c × SV) + (d × SV)2

Source: Authors’ illustration.

15Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

To validate the rationale behind the fuel consumption

values, we compare the fuel eciency derived from the

estimated fuel consumption ﬁgures. The fuel eciency is

calculated as:

Fuel

efficiency

(

)

KSA

(4)

The results of the eciency calculation are shown in

Table 4.

Table 4. Relationship between speed, estimated fuel consumption, and fuel eciency for gasoline vehicles in the KSA.

Speed (kph) Estimated fuel consumption

for gasoline (L/Km)

Fuel eciency (Km\L)

10 0.1677376 5.961692548

20 0.1275488 7.840136481

30 0.1084352 9.222097621

40 0.0959344 10.42378959

50 0.08715392 11.47395321

60 0.0811296 12.32595748

70 0.077448229 12.91185116

80 0.0759032 13.17467511

90 0.076379733 13.09247828

100 0.07880896 12.68891253

120 0.0893648 11.19008827

130 0.097441969 10.26251838

Source: Authors.

The fuel eciency reported by the Saudi Energy

Eciency Center (SEEC) for light-duty vehicles in Saudi

Arabia in 2022 was approximately 15.5 km/L (SEEC 2024).

However, our estimates suggest that the fuel eciency

for these vehicles would be around 13 km/L under

optimal speed conditions (Table 4). One reason for this

discrepancy could be the inclusion of hybrid vehicles in

the corporate average fuel economy (CAFE) calculations

(SEEC 2024). Hybrid vehicles typically oer signiﬁcantly

better fuel eciency compared to conventional gasoline-

powered vehicles, thereby raising the overall average

fuel economy for light-duty vehicles. This inclusion can

make the reported CAFE ﬁgures appear more favorable

than what might be observed under speciﬁc driving

conditions, particularly in scenarios that do not fully utilize

the fuel-saving beneﬁts of hybrid technology. Given these

considerations, the dierence between the estimated

and reported values is acceptable. Real-world driving

conditions, such as variations in speed, trac, and driving

behavior, do not always reﬂect the optimal or laboratory

conditions under which CAFE standards are calculated

(Carley et al. 2019).

16Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

4. Results and

Discussion

In general, the ratio of congested roads to non-congested roads within a

city, when analyzed using spatial and large-scale data, can indicate areas

with higher congestion. However, when examining trac data at the city-

wide scale, identifying speciﬁc congestion patterns becomes more dicult.

Conversely, focusing on particular roads allows for a clearer detection of

congestion patterns, oering more precise insights for targeted interventions.

4.1 Identifying

Congestion Patterns

When analyzing the increase in congestion spatially, it is

crucial to understand that the results can vary signiﬁcantly

based on the chosen area of examination. This is because

congestion is often highly localized, with signiﬁcant

variations in trac conditions, even within relatively small

geographical areas.

In this discussion, we focus on selected highways

in Riyadh city that often face increased congestion levels.

These roads are identiﬁed in the table below.

The CITY-WIDE representation is shown in Figure 1, while

the exact extents of the roads are detailed in Figure 7.

Table 5. Overview of selected key road networks in Riyadh.

ID Name Road length (km) Speed limit (kph)

CITY-WIDE Riyadh (all roads) 95,7721 N\A

KFR King Fahad Road 1,699 100

ERR Eastern Ring Road 978 120

WRR Western Ring Road 951 120

SRR Southern Ring Road 901 120

KKR King Khalid Road 752 120

NRR Northern Ring Road 714 120

Source: Authors.

17Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Figure 4. Heatmap of the speed reduction index (SRI) across examined highways, including CITY-WIDE and time of day.

ERR CITY-WIDE

KFRKKR

NRR

SRR

WRR

012345678910 11 12

Time

13 14 15 16 17 18 19 20 21 22 23

0.1

0.2

0.3

SRI (%)

0.4

0.5

Source: Authors’ analysis results.

Figure 4 illustrates a noticeable trend in SRI across all

highways, with congestion levels steadily rising during

peak hours (between 15:00 and 19:00). Notably, the

ERR experiences greater trac disruption throughout

the day, with congestion levels remaining relatively

high – above 30% – from the morning peak, escalating

to a highly congested state at around 17:00, when the

SRI level surpasses 50%. In contrast to the other roads,

the CITY-WIDE examination does not provide a clear

indication of SRI levels, with values hovering around

an average of 30% across the city. This is primarily

because, on a broader scale, congestion eects become

more diused and harder to detect. When analyzing

an extensive network like the city’s 957,721 km of

roads, the localized variations in trac congestion are

averaged out, leading to less pronounced ﬂuctuations

in SRI. The CITY-WIDE examination represents an

aggregated average from multiple road types, including

both highways and local roads. Since CITY-WIDE data

typically includes a variety of road conditions (including

smaller streets, residential areas, and other local roads),

the average congestion could naturally be higher,

as seen in Figure 4. Local roads typically have more

intersections, slower trac ﬂow, and frequent stops.

These factors contribute to higher average congestion

levels compared to highways. This is evident in Figure 5,

where FRC category 7, described as “covering local

roads of varying importance” (TomTom 2024c), shows a

signiﬁcant reduction in speeds due to these local roads

being smaller.

18Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

The other roads, which span between 700 km and

1,700 km, and where the eects of congestion are

more apparent and measurable, allow for a more

focused analysis. The larger the road network, the more

challenging it becomes to identify speciﬁc congestion

patterns, as the data reﬂects a more generalized state

rather than the localized disruptions that smaller road

networks experience.

4.2 Examining Fuel

Consumption Trends

Figure 6 illustrates the increase in average fuel

consumption in liters per kilometer for cars traveling

on these roads. Similar to Figure 4, the city-level

Figure 5. Heatmap of speed reduction index (SRI) across functional road classes (FRC) and time of day.

012345678910 11 12

Time

FRC

SRI (%)

13 14 15 16 17 18 19 20 21 22 23

0.40

0.35

0.30

0.25

0.20

0.15

0.10

Source: Authors’ analysis results.

data appears more diuse, with minimal discernible

patterns. However, there is a noticeable trend of

increased fuel consumption during peak hours, which

is expected as higher SRI leads to slower speeds,

and consequently higher consumption, due to more

frequent acceleration and deacceleration. There is

greater fuel consumption on the WRR, suggesting

lower congestion levels, which allow for higher driving

speeds (as demonstrated in Figures 4 and 6), leading

to increased fuel consumption. Although the KKR has

a similar SRI to the WRR, the WRR experiences more

variation in driving speeds throughout the day. As

shown in Table 6, the WRR has a broader range of

speeds, with a greater proportion of time spent at both

high and low speeds, contributing to the greater fuel

consumption observed on the WRR compared to the

KKR in Figure 6.

19Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Figure 6. Average fuel consumption (L/km) across major roads in Riyadh city throughout the day.

0.0095

0.0090

0.0085

0.0080

0.0075

Average fuel consumption (L/km)

0.0070

0.0065

0.0060

012345678910 11 12

Time

13 14 15 16 17 18 19 20 21 22 23

Road CITY-WIDE ERR KKR NRR SRR WRRKFR

Source: Authors’ analysis results.

Table 6. Distribution of speed ranges and corresponding fuel consumption trends across major roads in Riyadh city.

Speed range

(Kph)

WRR

(%)

KKR

(%)

NRR

(%)

KFR

(%)

ERR

(%)

SRR

(%)

Consumption

trend

0–20 1 00000High

21–40 2 1 4 17 9 7 High

41–60 3 3 11 16 14 8High

61–80 6 7 20 33 23 20 Moderate

81–100 44 48 40 34 34 40 Moderate

101–120 43 40 24 020 23 High

Source: Authors’ analysis results.

20Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

While highways with less congestion may exhibit greater

fuel consumption, this does not necessarily imply

ineciency. Vehicles on these roads often maintain higher

speeds and complete their trips more quickly, resulting

in a more ecient overall travel experience compared

to congested roads. However, trac congestion has

signiﬁcant economic implications, aecting both individual

commuters and the broader economy. According to

Arnott and Small (1994), congestion leads to increased

travel times, greater fuel consumption, and higher vehicle

operating costs, which collectively reduce economic

eciency. The authors highlight that congestion not

only imposes direct costs on drivers but also has wider

economic impacts, such as reduced productivity and

increased pollution (Arnott and Small 1994).

In this context, we compare fuel consumption between

two distinct times – low peak (01:00) and high peak

(17:00) – and calculate the average increase in fuel

consumption between these hours.

Average increasefrom low peak to high peak =

_FC_17

−

FC_1

⎛

⎝

⎜⎞

⎠

⎟×100

Average increasefrom low peak to high peak =

_FC_17

−

FC_1

⎛

⎝

⎜⎞

⎠

⎟×

100

(5)

Where:

• R = road

•

_FC_1

= average fuel consumption at 17:00

•

SFC_1

= average fuel consumption at 01:00

When comparing the increase in fuel consumption for

peak versus non-peak hours, the results show that driving

on some of Riyadh’s major highways may increase fuel

consumption during peak hours by up to 29.26% above

non-peak hours (Table 7).

It’s also important to note that the results will vary

signiﬁcantly depending on the speciﬁc area we choose

to examine.

Table 7. Average increase in fuel consumption between 01:00 and 17:00 across major roads in Riyadh.

Segment avg. fuel

consumption at

01:00 (L)

Segment avg. fuel

consumption at

17:00 (L)

Change (L) Avg. fuel

consumption

increase (%)

CITY-WIDE 0.007974 0.008169 0.000195 2.44235

KFR 0.00593 0.007666 0.001735 29.264823

WRR 0.008838 0.009496 0.000658 7.442644

ERR 0.007611 0.008824 0.001213 15.943012

NRR 0.007373 0.008492 0.001119 15.173594

KKR 0.007463 0.007624 0.000161 2.161888

SRR 0.00734 0.008089 0.000749 10.202571

Source: Authors’ analysis results.

Note: “Avg.” = average.

21Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Figure 7. Selected main highways in Riyadh for trac ﬂow and fuel consumption analysis.

Selected roads

ERR

NRR

WRR

SRR

KKR

KFR

048 16 KMs

Esri, TomTom, Garmin, FAO, NOAA, USGS

Riyadh boundaries

Source: TomTom (2024a) and authors’ analysis.

This map illustrates the key highways in Riyadh analyzed

in this study. The roads include the ERR (green), the NRR

(purple), the WRR (red), the SRR (blue), KKR (light green),

and KFR (yellow), with the city’s boundaries highlighted

in black.

Focusing on the localized area deﬁned as KFR in Figure 7,

for instance, can yield dierent outcomes compared

to extending the length of the same road or selecting

a smaller segment. This localized examination allows

for more spatially precise results, but it highlights the

signiﬁcant variability in outcomes based on the area of

focus. The variation seen in the literature underscores

this point, especially when considering that studies

often involve dierent countries, cities, roads, and car

compositions (Wang et al. 2019; Zhang and Batterman

2013; Afrin and Yodo 2020; Boggio-Marzet et al. 2021).

Additionally, the time scope of the dataset is a factor that

might alter the overall aggregated results, as this study

only includes data from 22 weekdays in October 2023.

Seasonality can signiﬁcantly inﬂuence driving conditions

on various roads in the city.

4.3 Study

Observations and

Key Highlights

• The relationship between speed and fuel consumption

is non-linear, and driving behavior during congestion

inﬂuences fuel use. Rapid acceleration, hard braking,

and frequent lane changes, further exacerbate these

ﬂuctuations, resulting in even greater fuel consumption

during periods of congestion.

• Driving on high-capacity roads in Riyadh can lead

to a signiﬁcant increase in fuel consumption, with

passenger cars experiencing up to a 29% rise in

fuel consumption due to congestion. The frequent

stop-and-go trac on these roads forces vehicles to

22Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

accelerate and decelerate more often, reducing overall

fuel eciency and leading to greater fuel consumption.

• Spatial analysis is essential for understanding complex

patterns and relationships, especially in urban studies.

By incorporating spatial considerations, the analysis

can reveal insights that might be missed in non-spatial

analyses, leading to more precise and eective

solutions. This paper demonstrates how results can

vary depending on the selected area by highlighting

dierent types of locales, such as roads, to show how

spatial analysis can inform targeted interventions for

policymakers and urban developers.

• The integration of IoT technologies signiﬁcantly enhances

real-time trac monitoring and prediction. IoT sensors,

along with FCD, gather real-time trac ﬂow data, enabling

more accurate predictions of congestion levels. This data-

driven approach allows for dynamic trac management,

optimizing trac ﬂow and reducing congestion.

4.4 Future Work

Our analysis is conducted at a mesoscopic level, where

the results are primarily dependent on variations in speed

across roads, utilizing predeﬁned fuel consumption

parameters to adjust for increases and decreases

in fuel consumption. While this approach provides a

general assessment, more precise calculations of fuel

consumption could be achieved through bottom-up

methodologies using micro-modeling techniques. Such

techniques would account for detailed factors like

turns, trac signals, and merging lanes, oering a more

granular and accurate representation of fuel consumption

dynamics. Additionally, incorporating more representative

data, including seasonal variations and dierent

fuel types, in future analyses could provide a more

comprehensive understanding of urban travel demand

and associated emissions.

23Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

5. Conclusion

This study underscores the critical relationship between trac congestion,

driving behavior, and fuel consumption in urban environments, with a

particular focus on Riyadh’s high-capacity roads. The non-linear nature of

the speed-fuel consumption relationship highlights the signiﬁcant impact that

trac congestion and constant acceleration and deceleration while driving

can have on fuel eciency.

The ﬁndings reveal that frequent stop-and-go trac,

particularly on congested highways, can lead to

substantial increases in fuel consumption, with

passenger vehicles experiencing up to a 29% rise in

fuel consumption. This insight emphasizes the need for

targeted interventions to manage congestion and improve

fuel eciency, especially in rapidly urbanizing cities like

Riyadh. Furthermore, the integration of advanced spatial

analysis techniques and IoT technologies, including FCD,

has proven invaluable in understanding and addressing

urban trac challenges. By providing real-time, high-

resolution data on trac patterns, these technologies

enable more accurate predictions and dynamic trac

management, contributing to more ecient and

sustainable urban mobility systems. This study not only

highlights the importance of spatial analysis in identifying

localized congestion and its eects; it also advocates

for the continued development and implementation of

smart technologies to enhance urban planning and trac

management. Moving forward, policymakers and urban

developers can use these insights to guide strategies

that reduce congestion and improve fuel eciency by

prioritizing investment in intelligent trac management

systems, promoting eco-driving programs, expanding

public transit infrastructure, encouraging sustainable

urban design, and leveraging data-driven mobility

policies, all of which contribute to better urban living

conditions in rapidly urbanizing cities like Riyadh.

24Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Endnotes

1 Fuel consumption (FC) here can also be referred to as ‘fuel intensity’ or ‘fuel consumption rate.’ This is because, in this context,

we are referring to the amount of fuel used per unit of distance (L/km), which can be inﬂuenced by various factors such as speed,

driving conditions, and vehicle eciency. These terms all describe the relationship between fuel use and distance traveled, and

they are often used interchangeably in studies focused on vehicle eciency and emissions. (Sobrino, Monzón, and Hernández

2014; U.K. Department of Transport 2019; Huo et al. 2012; Thomas et al. 2017; Fifer, Catron, and Bunn 2009).

25Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

References

Afrin, Tanzina, and Nita Yodo. 2020. “A Survey of Road

Trac Congestion Measures Towards a Sustainable and

Resilient Transportation System.” Sustainability 12, no. 11

(June): 4660. https://doi.org/10.3390/su12114660.

Al Zohbi, Gaydaa. 2021. “Sustainable Transport Strategies:

A Case Study of Riyadh, Saudi Arabia.” E3S Web of

Conferences 259: 02007. https://doi.org/10.1051

/e3sconf/202125902007.

Al-Mosaind, Musaad A. 1998. “Freeway Trac

Congestion in Riyadh, Saudi Arabia: Attitudes and Policy

Implications.” Journal of Transport Geography 6, no. 4

(December): 263–272. https://doi.org/10.1016

/S0966-6923(98)00024-6.

Altıntası, Oruç, Hediye Tuydes-Yaman, and Kagan Tuncay.

2022. “A Method to Estimate Trac Penetration Rates of

Commercial Floating Car Data Using Speed Information.”

Transport 37 (3): 161–176. https://doi.org/10.3846

/transport.2022.17069.

Arnott, Richard J., and Kenneth Small. 1994. “The

Economics of Trac Congestion.” American Scientist

82 (5): 446–455. http://www.jstor.org/stable/29775281.

Ayfantopoulou, Georgia, Marios Nikolaos Militsis, Josep

Maria Salanova Grau, and Socrates Basbas. 2022.

“Improving Map Matching of Floating Car Data with

Artiﬁcial Intelligence Techniques” Information 13 (11): 508.

https://doi.org/10.3390/info13110508.

Barth, Matthew and Kanok Boriboonsomsin, 2008. “Real-

World Carbon Dioxide Impacts of Trac Congestion.”

Transportation Research Record: Journal of the

Transportation Research Board 2058 (1): 163–171. https://

doi.org/10.3141/2058-20.

Bıyık, Can, Ahmad Abareshi, Alexander Paz, Rosa Arce

Ruiz, Rosaria Battarra, Christopher D.F. Rogers, and

Carmen Lizarraga. 2021. “Smart Mobility Adoption: A

Review of the Literature.” Journal of Open Innovation:

Technology, Market, and Complexity 7 (2): 146. https://doi

.org/10.3390/joitmc7020146.

Boggio-Marzet, Alessandra, Andres Monzon, Ana M.

Rodriguez-Alloza, and Yang Wang. 2021. “Combined

Inﬂuence of Trac Conditions, Driving Behavior, and

Type of Road on Fuel Consumption: Real Driving Data

from Madrid Area.” International Journal of Sustainable

Transportation 16 (4): 301–313. https://doi.org/10.1080

/15568318.2020.1871128.

Burghout, Wilco, Haris N. Koutsopoulos, and Ingmar

Andréasson. 2005. “Hybrid Mesoscopic–Microscopic

Trac Simulation.” Transportation Research Record

1934 (1): 218–225. https://doi.org/10.1177

/0361198105193400123.

Carley, Sanya, Nikolaos Zirogiannis, Denvil Duncan, Saba

Siddiki, and John D. Graham. 2019. “The Macroeconomic

Eects of 2017 through 2025 Federal Fuel Economy and

Greenhouse Gas Emissions Standards.” Journal of Policy

Analysis and Management 38, no. 3 (Summer): 732–763.

https://doi.org/10.1002/pam.22132.

Croce, Antonello Ignazio, Giuseppe Musolino, Corrado

Rindone, and Antonino Vitetta. 2021. “Estimation

of Travel Demand Models with Limited Information:

Floating Car Data for Parameters’ Calibration.”

Sustainability 13 (16): 8838. https://doi.org/10.3390

/su13168838.

Faria, Ricardo, Lina Brito, Karolina Baras, and José Luís

Silva. 2017. “Smart Mobility: A Survey.” In Proceedings of

the 2017 International Conference on Internet of Things

for the Global Community (IoTGC), Funchal, Portugal, 1–8.

https://doi.org/10.1109/IoTGC.2017.8008972.

Ferrara, Antonella, Simona Sacone, and Silvia Siri. 2018.

“Microscopic and Mesoscopic Trac Models.” In Freeway

Trac Modelling and Control, 105–136. Cham: Springer.

https://doi.org/10.1007/978-3-319-75961-6_5.

Fifer, Daniel Paul Catron, and Nicholas Patrick Bunn.

2009. “Assessing Consumer Valuation of Fuel Economy in

Auto Markets.” Honors Thesis, Duke University. https://hdl

.handle.net/10161/1380.

26Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Fourati, Walid, Hekmat Dabbas, and Bernhard Friedrich.

2021. “Estimation of Penetration Rates of Floating Car

Data at Signalized Intersections.” In Transportation

Research Procedia 52: 228–235. 23rd EURO Working

Group on Transportation Meeting, EWGT 2020, 16–18

September 2020, Paphos, Cyprus. https://doi.org/10.1016

/j.trpro.2021.01.026.

Gitahi, Joseph Mureithi, M. Hahn, M. Storz, C. Bernhard,

M. Feldges, and R. Nordentoft. 2020. “Multi-Sensor Trac

Data Fusion for Congestion Detection and Tracking.”

ISPRS - International Archives of the Photogrammetry,

Remote Sensing and Spatial Information Sciences

XLIII-B1: 173–180. https://doi.org/10.5194/isprs

-archives-xliii-b1-2020-173-2020.

Huo, Hong, Kebin He, Michael Wang, and Zhiliang Yao.

2012. “Vehicle Technologies, Fuel-Economy Policies, and

Fuel-Consumption Rates of Chinese Vehicles.” Energy

Policy 43 (April): 30–36. https://doi.org/10.1016/j

.enpol.2011.09.064.

Industrial Center (IC). 2024. “Saudi Arabia Market Share

2022/2023 Report.” Saudi Arabia’s Industrial Cluster.

https://www.ic.gov.sa/industries-automotive.

Ismaeel, Ayad Ghany, Krishnadas Janardhanan,

Manishankar Sankar, Yuvaraj Natarajan, Sarmad Nozad

Mahmood, Sameer Alani, and Akram H. Shather. 2023.

“Trac Pattern Classiﬁcation in Smart Cities Using Deep

Recurrent Neural Network.” Sustainability 15 (19): 14522.

https://doi.org/10.3390/su151914522.

Jiang, Yun, Guohua Song, Zeyu Zhang, Zhiqiang Zhai,

and Lei Yu. 2021. “Estimation of Hourly Trac Flows

from Floating Car Data for Vehicle Emission Estimation.”

Journal of Advanced Transportation: 1–11. https://doi

.org/10.1155/2021/6628335.

Kamal-Chaoui, Lamia and Alexis Robert, eds. 2009.

“Competitive Cities and Climate Change.” OECD Regional

Development Working Papers No. 2009/2. OECD

Publishing. https://EconPapers.repec.org

/RePEc:oec:govaab:2009/2-en.

Kan, Zihan, Luliang Tang, Mei-Po Kwan, and Xia Zhang.

2018. “Estimating Vehicle Fuel Consumption and

Emissions Using GPS Big Data.” International Journal of

Environmental Research and Public Health 15 (4): 566.

https://doi.org/10.3390/ijerph15040566.

Leduc, Guillaume. 2008. “Road Trac Data: Collection

Methods and Applications.” Joint Research Centre (JRC)

Technical Notes, Working Papers on Energy, Transport

and Climate Change N. 1 (January 2008). https://

www.researchgate.net/publication/254424803

_Road_Trac_Data_Collection_Methods_and

_Applications.

Liu, Xintao, and Y. Ban. 2013. “Uncovering Spatio-

Temporal Cluster Patterns Using Massive Floating Car

Data.” ISPRS International Journal of Geo-Information

2 (2): 371. https://doi.org/10.3390/ijgi2020371.

Motory. 2022. “New and Used Cars for Sale in Saudi

Arabia.” https://ksa.motory.com/en.

National Atmospheric Emissions Inventory (NAEI). 2019.

“RTP Fleet Projection NAEI 2017 Base2019r v1_1.” Excel

ﬁle, authored by Yvonne Pang and Tim Murrells, May.

https://naei.beis.gov.uk/resources/rtp_ﬂeet_projection

_NAEI_2017_Base2019r_v1_1.xlsx.

Ntziachristos, L, and Zissis Samaras. 2019. EMEP/EEA Air

Pollutant Emission Inventory Guidebook 2019 – Part B:

Sectoral Guidance Chapters, 1.A.3.b.i – Road Transport:

Combustion. Publications Oce of the European Union.

https://www.eea.europa.eu/publications/emep

-eea-guidebook-2019/part-b-sectoral-guidance

-chapters/1-energy/1-a-combustion/1-a-3-b-i/view.

Rao, Amudapuram Mohan, and Kalaga Ramachandra

Rao. 2012. “Measuring Urban Trac Congestion – A

Review.” International Journal for Trac and Transport

Engineering 2 (4): 286–305. https://doi.org/10.7708

/ijtte.2012.2(4).01.

Ricardo-AEA. 2014. “Production of Updated Emission

Curves for Use in the National Transport Model.” Report

to the Department for Transport Ricardo-AEA/R/ED56186

Issue Version 2. https://assets.publishing.service.gov.uk

/government/uploads/system/uploads/attachment_data

/ﬁle/662795/updated-emission-curves-ntm.pdf.

Saudi Energy Eciency Center (SEEC). 2024. “Open

Data.” https://www.seec.gov.sa/ar/open-data.

Saudi Press Agency (SPA). 2024. “SDAIA Launches Center

of Excellence in Congestion Solutions Using Data, AI.”

February. https://spa.gov.sa/en/N2047028.

27Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Silva, Bhagya Nathali, Murad Khan, and Kijun Han. 2018.

“Towards Sustainable Smart Cities: A Review of Trends,

Architectures, Components, and Open Challenges in

Smart Cities.” Sustainable Cities and Society 38 (April):

697–713. https://doi.org/10.1016/j.scs.2018.01.053.

Sims, Ralph E. H. et al. 2014. “Transport.” In Climate

Change 2014: Mitigation of Climate Change, Contribution

of Working Group III to the Fifth Assessment Report

of the Intergovernmental Panel on Climate Change,

edited by Ottmar Edenhofer et al., 315–398. Cambridge:

Cambridge University Press. https://doi.org/10.1017/

CBO9781107415416.005.

Sitati, Cynthia N., Christopher Oludhe, Leah Oyake,

and Aderiana Mutheu Mbandi. 2022. “A Street-Level

Assessment of Greenhouse Gas Emissions Associated

with Trac Congestion in the City of Nairobi, Kenya.”

Clean Air Journal 32 (1). https://doi.org/10.17159/caj

/2022/32/1.12546.

Sobrino, Natalia, Andrés Monzón, and Sara Hernández.

2014. “Reduced Carbon and Energy Footprint in Highway

Operations: The Highway Energy Assessment (HERA)

Methodology.” Networks and Spatial Economics 16 (1):

395–414. https://doi.org/10.1007/s11067-014-9225-y.

Song, Guohua, Lei Yu, and Yizheng Wu. 2016.

“Development of Speed Correction Factors Based on

Speed-Speciﬁc Distributions of Vehicle Speciﬁc Power

for Urban Restricted-Access Roadways.” Journal of

Transportation Engineering 142 (3): 04016001. https://doi

.org/10.1061/(ASCE)TE.1943-5436.0000819.

Thomas, John F., Shean Hu, Brian H. West, and Paul

Chambon. 2017. “Fuel Consumption Sensitivity of

Conventional and Hybrid Electric Light-Duty Gasoline

Vehicles to Driving Style.” SAE International Journal

of Fuels and Lubricants 10 (3). https://doi.org/10.4271

/2017-01-9379.

TomTom. “Move Portal.” https://move.tomtom.com.

TomTom. “Saudi Arabia Trac Index.” 2024b. https://www

.tomtom.com/trac-index/saudi-arabia-country-trac.

TomTom. “Trac Stats API FAQ.” 2024c. https://developer

.tomtom.com/trac-stats/documentation/product

-information/faq.

UK Department for Transport. 2018. “TAG Data Book.”

November 29. https://www.gov.uk/government

/publications/tag-data-book.

United Nations. 2021. “Sustainable Transport, Sustainable

Development: Interagency Report for the Second Global

Sustainable Transport Conference.” https://www.un.org

/en/desa/sustainable-transport-sustainable-development

-interagency-report-second-global-sustainable.

Wang, Xiaomeng, Ling Peng, Tianhe Chi, Mengzhu Li,

Xiaojing Yao, and Jing Shao. 2015. “A Hidden Markov

Model for Urban-Scale Trac Estimation Using Floating

Car Data.” PLoS ONE 10 (12): e0145348. https://doi.org

/10.1371/journal.pone.0145348.

Wang, Jingyi, Guohua Song, Lei Yu, Hongyu Lu, Jianping

Sun, Ying Cheng, and Weinan He. 2019. “Macroscopic

Relationship between Trac Condition and Fuel

Consumption for an Urban Road Network: Case Study

of Beijing.” Transportation Research Record 2673 (11):

604–615. https://doi.org/10.1177/0361198119853575.

Xu, Lin, Yang Yue, and Qingquan Li. 2013. “Identifying

Urban Trac Congestion Pattern from Historical Floating

Car Data.” Procedia – Social and Behavioral Sciences 96

(November): 2084–2095. https://doi.org/10.1016/j.sbspro

.2013.08.235.

Youssef, Zaher, Habib M. Alshuwaikhat, and Imran

Reza. 2021. “Modeling the Modal Shift Towards a More

Sustainable Transport by Stated Preference in Riyadh,

Saudi Arabia.” Sustainability 13 (1): 337. https://doi.org

/10.3390/su13010337.

Zhang, Kai, and Stuart Batterman. 2013 “Air Pollution and

Health Risks Due to Vehicle Trac.” Science of The Total

Environment 450–451 (April): 307–316. https://doi.org/10

.1016/j.scitotenv.2013.01.074.

28Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Notes

29Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

About the Authors

Lama Yaseen

Lama is a Fellow working under the Transport and Infrastructure Program at KAPSARC.

Her work centers on the challenges of climate change sustainability, energy security, and

digital infrastructure in urban settings. Lama is a researcher and data scientist with software

development experience. She specializes in big data modeling in research on urban and

sustainable transportation.

Lama was previously in the Policy and Decision Sciences team at KAPSARC, leading the

software development of policy and behavioral models.

She holds an M.Sc. in Software Engineering from the University of Oxford and a B.Sc. in

Computer Science from Eat University.

Nourah Alhosain

Nourah is a Manager in the Solutions Productization program. Her work focuses on geospatial

analysis and modeling. Nourah was previously a member of KAPSARC’s Policy and Decision

Science program, where she was part of the KAPSARC Toolkit for Behavioral Analysis (KTAB)

development team. She holds a B.S. in Computer Science from Prince Sultan University.

Ibrahem Shatnawi

Ibrahem is a Senior Transport Fellow at KAPSARC with over 19 years of experience in

transportation engineering. His research focuses on transport policies and technologies for

decarbonizing road transport and integrated transport demand modeling to estimate energy

demand and emissions.

Before joining KAPSARC, Ibrahem had extensive experience across both the public and private

sectors. He worked with the Department of Transport in Abu Dhabi and the Road and Transport

Authority in Dubai, as well as renowned consultant ﬁrms such as JACOBS, Michael Baker

International, and Parsons Corporation. His career has spanned multiple countries, including

the United States (U.S.), the United Arab Emirates, Jordan, and Saudi Arabia. His expertise

includes public transportation fare strategies, transport policies, travel demand modeling,

transportation economic impact studies, intelligent transportation systems, trac operations,

and microsimulation projects. He has served as a project manager for large-scale transportation

studies and has authored numerous technical memoranda and reports.

Ibrahem holds a B.Sc. in Civil Engineering, specializing in transportation, from Jordan University

of Science and Technology in Jordan, and a Ph.D. in Civil Engineering, specializing in trac

engineering, from The University of Akron, Ohio, U.S.

30Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

Abdelrahman Mohsen

Abdelrahman Muhsen, M.Sc., GISP is a geospatial expert with over 20 years of experience

in GIS and spatial data management. He specializes in spatial economic modeling to analyze

and predict the relationships between economic activities, energy demand, land use, and

transportation in urban settings. Abdel has successfully led and delivered enterprise GIS

projects for multinational corporations in both North America and the Middle East. Prior to his

role at KAPSARC, he worked with prestigious organizations such as Accenture and ESRI, where

he served as a technical leader and senior consultant. In these roles, he advised clients in the

energy sector on optimizing their geospatial data assets and GIS investments. Abdel holds an

M.Sc. in Geomatics Engineering from the University of Calgary, Canada.

31Impact of Urban Trac on Fuel Consumption Leveraging IoT Data: Case Study of Riyadh City

About the Project

In Saudi Arabia, addressing broader mobility challenges, especially those

linked to the environmental repercussions of traditional internal combustion

engine vehicles, is essential. The prevalent use of such vehicles in the

country underscores the importance of considering multi-modality and more

sustainable transportation solutions. This project is focused on developing

strategies that foster a cohesive and sustainable urban mobility ecosystem,

emphasizing the evaluation and potential improvement of both the physical

and digital infrastructure in urban transportation. The aim of this project is to

improve the Saudi transportation network’s eciency and sustainability by

integrating multi-modal transport options and leveraging digital technology

advancements