Next Article in Journal
Two-Stage Stochastic Scheduling of Cascaded Hydropower–Wind–Photovoltaic Hybrid Systems Considering Contract Decomposition and Spot Market
Previous Article in Journal
Public Transport Infrastructure with Electromobility Elements at the Smart City Level to Support Sustainability
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 16
Issue 3
10.3390/su16031092
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Driving towards Sustainability: A Neural Network-Based Prediction of the Traffic-Related Effects on Road Users in the UAE

by
Haneen Abuzaid
1,*,
Raghad Almashhour
1 and
Ghassan Abu-Lebdeh
2
1
Department of Industrial Engineering, American University of Sharjah, Sharjah 26666, United Arab Emirates
2
Department of Civil Engineering, American University of Sharjah, Sharjah 26666, United Arab Emirates
*
Author to whom correspondence should be addressed.
Sustainability 2024, 16(3), 1092; https://doi.org/10.3390/su16031092
Submission received: 27 September 2023 / Revised: 5 December 2023 / Accepted: 7 December 2023 / Published: 26 January 2024
(This article belongs to the Section Sustainability in Geographic Science)
Browse Figures
Versions Notes

Abstract

:
Transportation is fundamental, granting access to goods, services, and economic opportunities. Ensuring sustainable transportation, especially in vehicular modes, is crucial for the pillars of social, economic, and environmental sustainability. High-traffic countries, like the United Arab Emirates (UAE), grapple with significant challenges to this end. This study delves into the repercussions of traffic-related incidents on UAE road users and their intricate links to the social and economic dimensions of sustainability. To achieve this, this work examines the influential demographic factors contributing to incidents, utilizing artificial neural network models to predict the likelihood of individuals experiencing traffic tickets and accidents. Findings reveal associations between gender, driving frequency, age, nationality, and reported incident frequency. Men experience more accidents and tickets than women. Age exhibits a negative linear relationship with incident occurrence, while driving experience shows a positive linear relationship. Nationalities and cultural backgrounds influence road users’ adherence to traffic rules. The predictive models in this study demonstrate their high accuracy, with 93.7% precision in predicting tickets and 95.8% in predicting accidents. These insights offer valuable information for stakeholders, including government entities, road users, contractors, and designers, contributing to the enhancement of the social and economic aspects of road sustainability.

1. Introduction

Transportation is an essential part of human life. It offers access to goods, services, and economic opportunities. In recognition this critical role, transportation networks have been characterized as the “lifeblood” of cities [1]. Transportation, regardless of the mode, plays a pivotal role in promoting sustainable development (SD) through facilitating access to essential resources and markets, enhancing our quality of life by connecting individuals to employment, healthcare, education, leisure, and various other activities. The road infrastructure in the United Arab Emirates (UAE) stands out for its exceptional quality, as indicated by its high ranking in the road quality index [2]. In fact, the UAE’s roads have surpassed those of economically advanced nations like Germany, the USA, the UK, and France, as highlighted in the 2019 report published by the World Economic Forum [3]. This recognition showcases the UAE’s commitment to maintaining and developing top-notch road networks.
Due to the rapid growth in road construction and the rising number of vehicles, the issue of road traffic accidents is becoming more severe [4]. With so many drivers from diverse national backgrounds and with diverse driving habits and styles, the risks of traffic violations and accidents are quite high [5]. This paper claims that the extent of these issues can be significantly minimized through the implementation of suitable measures to tackle road user behaviors and accordingly predict the frequency of traffic problems, including accidents and fines/tickets.
In terms of scholarly discourse, there is a substantial body of research exploring the nexus of highway traffic and sustainability in the UAE. This is illustrated in Figure 1 through an analysis of Scopus research, which is considered one of the premium databases of peer-reviewed journals. It can be seen from Figure 1 that the increase in publications on the topic over the years is very high. In 2021, 111 research papers were published on the topic in the UAE, which clearly highlights the importance of the topic and supports the purpose of this research in the UAE [6].
This study is interested in what commuters in the UAE deem to be the impacts of traffic on them, individually and collectively. The rationale of this interest lies within the broad realm of SD. SD and transportation are closely intertwined, as the operations and performance of transportation systems directly or indirectly impact all three pillars of sustainability: the environment, social equity, and the economy. To illustrate, cities worldwide face the problem of congested roads, primarily caused by excessive reliance on automobiles.
This dependence on cars not only contributes to emissions and environmental issues but also brings about social costs, including accidents [7,8]. Consequently, transportation networks exist in a paradoxical situation. While they drive urban development, they also pose a range of challenges. Extensive research (e.g., [5,9,10,11]) highlights the unsustainability of current automobile-oriented transportation trends due to their significant environmental, economic, and social impacts. Table 1 illustrates some of these challenges.
This study investigates the impact of traffic-related incidents on road users in the UAE, aiming to identify the influential demographic factors at play. In high-traffic nations like the UAE, understanding the intricacies of these incidents is crucial for achieving sustainability from social, economic, and environmental perspectives, particularly within transportation. The current literature lacks a comprehensive exploration of the connections between demographic factors and traffic-related outcomes, necessitating a deeper dive into the social and economic dimensions of road sustainability affected by incidents like accidents and traffic violations.
The research questions addressed are: How do influential demographic factors contribute to traffic-related incidents in the UAE? And how can we predict those traffic-related incidents for road users in the UAE? To answer them, our aim is to formulate a comprehensive understanding of how demographic factors contribute to such incidents and, using artificial neural network models, provide practical insights for evidence-based decision-making. The forthcoming literature review section will delve into these aspects, discussing the intricate relationships, identified patterns, and potential improvements, while emphasizing the significance of this study’s contributions within the broader context of road sustainability. The study’s novelty lies in employing artificial neural network models to predict the likelihood of individuals experiencing traffic incidents based on influential demographic factors, thereby enhancing our understanding of the nuanced relationships within road sustainability. The findings, characterized by robust associations and high predictive accuracy, offer a unique perspective on these relationships.
This paper is structured as follows: the related literature review is presented in Section 2, followed by an applied methodological approach in Section 3. The results are illustrated in Section 4, the discussion is detailed in Section 5, and finally, the conclusion and future work are shown in Section 6.

2. Literature Review

2.1. Road Traffic through a Social Sustainability Lens

Extensive scholarly research has focused on both the economic and environmental aspects of transportation, significantly impacting global transport policy and practice (e.g., [12,13,14,15]). This comprehensive exploration of transportation’s economic and environmental facets has shaped policy decisions and practical approaches in the field of transportation on a global scale. The social dimension has appeared in just a few transport research papers, policies, and practices. This has much to do with its limited recognition, as it is more difficult to investigate, grasp, and define [16]. As a result, its influence in national or international policy circles is generally less significant [17].
The conceptual and empirical literature on every conceivable aspect of transport is vast, multidisciplinary, and ever-evolving. A sizable part of the literature focuses specifically on problems of transport or traffic and the impacts of traffic on commuters, most often in cities. Traffic accidents, referred to as road casualties and injuries, were among the initial adverse social outcomes of traffic expansion that gained significant recognition in the field of transportation studies. Increasingly, this form of impact is increasingly viewed through a public health lens, as it is widely acknowledged as an “epidemic” due to its substantial contribution to global injuries and fatalities (e.g., [18,19,20]).
As stated by the World Health Organization (WHO), 3000 individuals lose their lives on roads each day, while several million others experience injuries or disabilities annually [21]. This immense health burden further compounds the negative social impacts associated with transportation, including issues related to air quality, noise pollution, physical activity, and the growing prevalence of sedentary lifestyles [12,22]. According to the literature, this particular area has garnered significant attention and multi-disciplinary interest, leading to responses from various practitioners. It involves not only traffic engineers and public health bodies but also police departments and education authorities. The collaborative efforts of these diverse entities reflect their recognition of the complex nature of the problem and the need for a comprehensive approach to address it effectively. The following section focuses on traffic safety within the context of the UAE.

2.2. Traffic Safety in the UAE

Traffic safety is a significant and deeply concerning issue on a global scale, primarily due to the alarming number of traffic fatalities and injuries worldwide. The Gulf Cooperative Council (GCC) countries, including the UAE, are known to experience relatively high rates of traffic-related deaths and injuries [5]. In the UAE specifically, there exists a dense road network accompanied by a substantial number of privately-owned vehicles. While the country has achieved commendable rankings in terms of road quality, it is crucial to acknowledge that road safety and traffic congestion involve multiple aspects that warrant attention and consideration [23]. The UAE has experienced a significant surge in the number of vehicles on its roads in recent years, primarily driven by economic growth, population growth, and demographic and cultural changes.
However, this increase in the number of vehicles on the road has resulted in a concerning rise in road accidents, particularly fatal accidents, and injuries [24]. For example, in 2016, drivers in the UAE spent approximately 11% of their driving time stuck in congestion, highlighting the impact of traffic congestion on the road network [23]. Furthermore, the number of registered vehicles in the country reached 3.39 million in 2016, showing a substantial 30% increase compared to the 2.67 million vehicles registered in 2012 [25]. These statistics emphasize the pressing need to address road safety and mitigate the risks associated with the growing number of vehicles on the UAE’s roads.
In an effort to prevent road accidents and minimize their impact on lives and property, various government bodies in the UAE have introduced several initiative programs and awareness campaigns aimed at educating the public about the importance of traffic safety. These initiatives include (i) the Abu Dhabi strategic traffic safety plan, (ii) a road safety audit, (iii) road safety awareness, (iii) Central Traffic Control System (SCOOT), (iv) the speed management strategy of the emirate of Abu Dhabi, and (v) the Road and Transportation Authority (RTA) of Dubai’s strategic plan 2014–2018 [26]. These initiatives reflect the commitment of the UAE government to prioritizing road safety and creating a safer environment for road users.
The implementation of road safety programs in the UAE has yielded significant results, leading to a drastic reduction in the number of accidents in recent years. According to official statistics released by the UAE police, the total number of fatalities during the years 2018–2019 had decreased by 50% compared to between 2006 and 2010. Similarly, the number of accidents that occurred between 2018–2019 is only half of the total count of accidents recorded between 2006–2010. The same positive trend is observed in the number of injuries, which has also significantly decreased. These statistics provide concrete evidence that the road safety measures implemented in the UAE have had a substantial impact, resulting in a commendable reduction in casualties [27]. The dedication and efforts put into road safety initiatives by the government and relevant authorities have proven effective in creating a safer road environment for both residents and visitors to the UAE.
Figure 2 illustrates the road fatalities in the UAE from 2014–2019. The statistics show that the death rates kept dropping from 2017 to 2019. Nevertheless, the number of road accidents and serious injuries in the UAE is still high and these are mainly caused by the incorrect behavior of drivers [28], which will be discussed in the following section. Similarly, the total number of issued traffic fines has been reported for all the emirates for 2018 and 2019 [28]; these statistics are illustrated in Figure 3. It can be seen that the number of fines has decreased in all emirates except for Abu Dhabi, where there was an increase of approximately 9%.

2.3. Traffic Safety Attitudes and Behaviors

In the field of traffic safety research, the examination of attitudes towards traffic rules and traffic safety plays a crucial role in understanding driving behaviors. Several studies (e.g., [29,30,31,32,33]) have focused on investigating how attitudes, whether indirectly or directly, influence behavior in various social interactions, including driving. Attitudes play a significant role in shaping behavior in various social interactions, including driving [30]. Whether indirectly or directly, attitudes have the power to influence how individuals’ approach and engage in different activities, including their driving behaviors. Attitudes are “tendencies to evaluate an entity with some degree of favor or disfavor, ordinarily expressed in cognitive, affective and behavioral responses” [34].
The attitudes that drivers hold towards driving safety can have a significant impact on their driving behaviors. Whether a driver has a favorable or unfavorable attitude towards driving safety can shape how they approach and engage in their driving activities. In this paper, attitudes toward traffic safety are related to the diverse set of demographic attributes for road users in the UAE, as well as their attitudes toward traffic rules and rule violations. The following paragraph will discuss the effect that demographic variables have on driving behavior.
Demographic variables such as gender, age, driving experience, education level, etc., have an effect on driving behavior and traffic violations (e.g., [31,35,36,37]). Globally, younger individuals are found to be at a higher risk of being involved in traffic crashes, and the same holds true for older individuals with increased years of driving experience (e.g., [31,32,38,39]). Regarding gender differences, Gonzalez-Iglesias et al. [40] observed that men tend to have a higher frequency of traffic fines and are more prone to committing traffic violations than women. On the other hand, Horvath et al. [41] observed that women tended to drive less than men and had a lower incidence of offenses. Akaateba et al. [42] found that traffic violations increased with driving experience. Additionally, drivers with higher levels of education exhibited a lower frequency of traffic violations.
Additionally, Alver et al. [37] conducted a study on red light and seatbelt-related violations and found that drivers who engaged in such violations were more likely to be involved in crashes. Similarly, various factors have been identified as influencing driving behavior and increasing the risk of car crashes, such as driver distraction (e.g., [43,44,45]), phone use while driving (e.g., [43,45,46,47]), the use of alcohol or drugs (e.g., [48,49,50]), and drowsy driving (sleepiness and/or driving while fatigue) (e.g., [51,52,53]). Table 2 provides a summary of the additional studies found in the literature with respect to these factors.
Previous research has examined the correlation between traffic safety and personal characteristics such as age, gender, nationality, and driving experience in the UAE, aiming to shed light on the high number of traffic crashes and casualties. The findings were consistent with the UAE context, revealing varied driving behaviors and traffic violations between different age groups. It was observed that drivers aged between 18 and 35 accounted for approximately 45% of all road traffic injuries in the UAE [63]. A study conducted in Al-Ain city revealed that 40% of drivers involved in distracted driving and road accidents were using their mobile phones [64].
An examination of the official police records, surveys conducted among road users, and interviews with traffic safety experts in the UAE have substantiated that driving behavior plays a significant role in road safety [65]. The most prevalent driving violations observed in the UAE include excessive speeding, failure to use indicators, tailgating, and running red lights [25]. A recent study conducted by Alghafli et al. [66] highlighted the substantial contribution of red light violations to traffic accidents in the UAE. According to the RTA, an analysis of fatal crashes in 2019 revealed that speeding, using mobile phones while driving, a failure to fasten seatbelts, running red lights, and tailgating were among the most prevalent offenses [67]. Surprisingly, drunk driving made a limited contribution to accidents in the UAE, as indicated by recent official statistics. This can largely be attributed to the strict cultural and social codes surrounding alcohol consumption [68].
On another note, with a heterogeneous driver population in UAE, the impact of nationality on traffic safety should be taken into consideration. Previous research has identified a relationship between nationalities and driving behavior, traffic violations, and traffic crashes. Arabic nationalities have been found to report more traffic violations than Western nationalities. Among the Arab nationalities, GCC countries tend to have higher rates of traffic crashes and violations compared to other countries in the Middle East ([69,70,71]). Similarly, personal attributes and cultural background play a role in shaping attitudes towards risky driving behavior [72], thus influencing traffic safety and shaping the driver’s traffic safety culture.
Therefore, it is crucial to investigate the influence of nationality on driving behavior and safety attitudes, particularly in a diverse driving population such as that of the UAE. Although Timmermans et al. [5] examined the effect of nationality on risky driving behavior in Qatar, the impact of nationality on risky driving behavior in the UAE has not been thoroughly studied, making this research a valuable addition to the existing literature. In recent years, there has been growing awareness of the significant problem of traffic accidents in the UAE, especially among policy makers and public health professionals. However, the responses to mitigate the risk have been limited, and controlling the epidemic remains a challenging task.

2.4. Applications of Artificial Intelligence in Traffic Safety

Road traffic accidents are a prominent cause of injuries and fatalities on a global scale, making them a substantial area of study for employing advanced algorithms and methodologies to analyze and forecast these accidents [73]. This research aims to identify the key factors that contribute to road accidents by utilizing data mining, which involves extracting valuable insights from vast datasets. At present, data mining has gained widespread application and recognition across various fields, demonstrating its reliability and effectiveness in analyzing the data related to road accidents [74].
Among the various data mining techniques decision trees and rules, non-linear regression and classification methods, and neural networks have gained popularity [75]. In the realm of modeling traffic safety, different types of Artificial Intelligence (AI) techniques have been employed. Notably, artificial neural networks (ANNs) have emerged as a widely recognized AI approach frequently utilized for predicting accident-causing conditions, forecasting road accidents, and developing accident severity models [76]. Numerous researchers have successfully employed ANN models for predicting traffic injuries and their severity in transportation studies, showcasing its superior accuracy compared to conventional approaches when it comes to forecasting deaths and injuries [77].
ANN models offer several advantages, such as superior performance compared to other models under appropriate conditions, scalability with large datasets, and the ability to handle multiple tasks simultaneously [78]. However, the effectiveness of ANN models relies on various initial parameters, including weights and biases [79]. Consequently, numerous studies have focused on enhancing the performance of ANN models by integrating them with optimization algorithms (e.g., [80,81,82,83]). Table 3 provides a summary of the studies found in the literature with respect to data mining and other AI techniques used to measure traffic accidents’ severities.
From the table, it can be noticed that none of the studies has considered UAE traffic-related incidents when modelling using data mining techniques, which signifies the importance of this research. Different scopes have been modeled using ANNs, including the transportation demands of passengers and freight in Turkey [29], the number of traffic accidents and deaths in Switzerland [31], the severity of crashes for old drivers in USA [5], the crash frequency in Hong Kong [84], and the severity type of traffic accidents in the Republic of Korea [85], while other types of ANN techniques were used, including a recurrent neural network (RNN), feedforward neural network (FNN), and convolution neural network (CNN), to predict the severity of injuries from traffic accidents in Malaysia [35,86].
Additionally, linear and multiple regression analysis has been employed in Argentina [33], Turkey [37], and Iran [29,35] to predict the behavior and attitudes of road users. Additionally, other statistical tools were utilized to provide insightful findings related to traffic- and road-related issues including cluster analyses in Norway, Iran, and Greece [54,84,86], structural equation modeling (SEM) in KSA and China [31,85], factor analysis in Egypt [57], and analyses of variance (ANOVA) in China, Qatar, and India [5,57,87].
Table 3. Conducted studies on traffic-related problems using data mining techniques.
Table 3. Conducted studies on traffic-related problems using data mining techniques.
No.Scope ContextYearMethod/sRef.
1Addressing the transportation demands of passengers and freightTurkey2007ANN
Expert Judgment		[29]
2Prediction of injury severity of traffic accidentsMalaysia2017RNN[86]
3Decreasing the number of traffic accidents and deathsSwitzerland2018ANN[31]
4Reducing road traffic injuriesChina2021ITS[33]
5Prediction of injury severity of traffic accidents on highways Malaysia2017FNN, RNN, CNN[35]
6Assessment of urban traffic safetyIndia2016Descriptive Analysis[57]
7Impacts of road safety on road users Belgium2014GWR[37]
8Signalized intersection safetyUSA2010Bayesian Approach[87]
9Assessment of severity of ROR crashes for old driversUSA2020ANN[5]
10Predicting crash frequency and risk factorsHong Kong2016ANN[84]
11Prediction of severity types of traffic accidentsRepublic of Korea2011Decision Tree, ANN[85]
12Improving the stress prediction of automobile driversJordan2019Decision Tree
ANN–KNN–SVM–RF 		[54]
13Drivers’ attitudes in prevention of traffic crashesIran2014LR[29]
14Safety beliefs and drivers’ behaviorsNorway2013Cluster Analysis[86]
15Attitudes, driving behavior, and accident involvementKSA2017SEM[31]
16Attitudes towards seatbelts (a specific road safety behavior)Argentina2018MRA[33]
17Offending drivers who received tickets more frequently than they expected Iran2020MRA[35]
18Driving attitudes and behaviors towards traffic safety Egypt 2022EFA [57]
19Traffic rule violations for young driversTurkey2014LR[37]
20Impacts of safety knowledge on risky driving behaviors China2018EFA, SEM, AVOVA[87]
21Traffic safety culture among professional driversQatar2019Descriptive Analysis[5]
22Drivers’ risky driving behaviors Iran2019Cluster Analysis[84]
23Analysis of risky driving behaviors among bus driversChina2022SEM[85]
24Drivers’ behaviors and attitudes to traffic violations Greece 2013Cluster Analysis [54]

3. Methodology

Figure 4 illustrates the research methodology adopted in this paper. The design of the methodology is crafted to ensure that the research outcomes align with the intended goals by addressing both the social and economic sustainability aspects concerning traffic-related impacts in the UAE. These impacts encompass the occurrences of accidents and fines among road users in the country.
The initial step of this study involved the identification of the research problem, with a focus on understanding the traffic-related impacts in the United Arab Emirates (UAE). In particular, the study sought to understand the frequency of the accidents and fines experienced by road users and aimed to construct predictive models for these occurrences. The significance of the research problem was affirmed through an initial literature screening and semi-structured interviews conducted with subject matter experts.
To establish a contextual framework for our research, a thorough review of the published literature was conducted. This review encompassed scholarly research and governmental reports related to traffic-related impacts in the UAE. The primary objective was to synthesize the existing knowledge and link it to the sustainability of transportation and roads in the region. This comprehensive examination of the literature provided valuable insights, identified gaps, and informed the development of our research questions.
To gather primary data pertaining to the impacts of traffic, a survey was crafted to obtain responses concerning accidents and traffic violations. The survey was thoughtfully designed to uphold methodological rigor, ensuring a comprehensive understanding of the driving behaviors and experiences of road users in the United Arab Emirates (UAE). The initiation of the survey included a detailed introduction, explaining its purpose and fostering participant understanding. In order to safeguard the reliability and integrity of responses, a commitment to confidentiality was emphasized. Measures such as respondent anonymity were implemented, encouraging participants to openly share their experiences. Furthermore, ethical considerations were reinforced through the integration of an informed consent form into the survey process. Informed consent was obtained from all the subjects involved in the study.
Subsequently, the management of the collected data was carefully detailed. The information was processed and transformed into an Excel sheet for comprehensive analysis. Importantly, the respondents’ identities were kept confidential throughout this process, underscoring our commitment to privacy. Access to the data was restricted, and confidentiality was preserved as part of our ethical framework. Additionally, issues related to the Informed Consent Statement were considered, acknowledging the importance of transparency and participant comprehension. Any concerns or questions regarding the consent statement were addressed to ensure that the participants felt informed and comfortable throughout the research process.
Also, the online distribution of the survey, coupled with a random sampling strategy, aimed to capture a diverse and representative cross-section of road users across the UAE. The data collection extended over a period of two years to enable a longitudinal analysis, allowing for the identification of trends and variations in driving patterns over time. The target sample, consisting of road users in the UAE, reflected a broad demographic spectrum, ensuring inclusivity across various age groups, driving experiences, and geographic locations within the country.
Furthermore, the survey questionnaire, comprising 11 multiple-choice questions, covered a range of relevant factors, including demographic information, driving frequency, emirate of residence, driving experience, age groups, nationality, vehicle types, average monthly income, primary driving purposes, and historical data on their number of tickets and road accidents within the past two years. Nonetheless, the survey underwent a rigorous piloting process, prior to the actual data collection, to verify the questions and the suggested choices and to ensure that the responses aligned with the intended aims of the research.
Following the distribution of the survey to road users in the United Arab Emirates, the subsequent step involved data collection, cleaning, and preparation for analysis. The collected data, primarily qualitative in nature, underwent a comprehensive cleaning process to ensure its accuracy and reliability. Rigorous pre-processing techniques were applied, including addressing missing values, eliminating outliers, and normalizing variables, to enhance the overall quality of the dataset.
A crucial aspect of this phase was data screening, which involved a thorough review of the responses to identify and rectify any inconsistencies or discrepancies. This stringent screening process was instrumental in increasing the credibility of the survey results by ensuring that the dataset was robust, consistent, and reflective of the participants’ experiences and perceptions. The focus on qualitative data collection, coupled with rigorous cleaning and pre-processing, serves to provide a solid understanding of the diverse perspectives and insights shared by road users, ultimately contributing to the richness and depth of the survey findings.
Additionally, understanding the demographic characteristics of drivers, including their age, frequency of driving, emirate of residence, driving experience, age group, nationality, vehicle type, average monthly income, and primary driving purpose, is pivotal in contextualizing and interpreting the findings. The demographic analysis provides insights into the diverse backgrounds and experiences of road users in the UAE, contributing to a comprehensive understanding of the factors influencing traffic-related impacts.
Subsequently, the data underwent a cleaning and preparation process to facilitate statistical analysis and the development of prediction models. The primary goal was to ensure the reliability and accuracy of the dataset. The prepared dataset was then used as input for the prediction models, with fines and accidents serving as the model outputs. The predictive models incorporated the aforementioned demographic factors, such as age, driving frequency, emirate of residence, driving experience, age group, nationality, vehicle type, average monthly income, and main purpose of driving.
Prior to analysis, the dataset was strategically split into training (75%) and testing (25%) subsets. Also, we performed 10 folds cross validation for both models. This partitioning was crucial in evaluating the robustness and generalizability of the predictive models. The training set was used to develop and optimize the models, while the testing set, being independent, enabled an unbiased assessment of the models’ performance. This step was vital in ensuring that the developed models were capable of making accurate predictions when applied to new, unseen data, thereby enhancing the reliability and applicability of the study’s findings.
Afterwards, two ANN models were employed to predict the number of tickets and accidents for road users in the UAE as a function of all the aforementioned factors, and their performance was systematically compared using the prediction accuracy measure within the R software (2023.06.0 Build 421). The first model utilized the H2O package, incorporating additional R packages such as caret, RCurl, and jsonlite. This H2O model was configured with three hidden layers with 200, 100, and 50 hidden nodes, respectively. The second model employed the Keras/Tensorflow model, with associated R packages including caret, keras, and tensorflow. Similarly, the Keras/Tensorflow model featured three hidden layers with 50, 25, and 2 hidden nodes, respectively, as illustrated in Figure 5.
The determination of the number of hidden layers and nodes for each model took into consideration factors such as computational efficiency, time constraints, cost, and the pursuit of optimal prediction accuracy. The configuration of the H2O model with larger numbers of hidden nodes in each layer aimed to capture complex patterns in the data, potentially enhancing its accuracy. On the other hand, the Keras/Tensorflow model employed a more streamlined architecture with fewer nodes in each layer, balancing computational efficiency and prediction accuracy. This decision was motivated by a consideration of the trade-off between model complexity and resource requirements, ensuring a practical and efficient approach.
The selection of H2O and Keras/Tensorflow for the ANN prediction models was guided by their well-established capabilities in handling complex datasets, scalability, and their ease of integration with R. These frameworks are widely recognized for their versatility and efficiency in training and deploying neural networks. The non-linear relationships inherent in the dataset, coupled with the ability of ANN models to capture sophisticated patterns, make them well-suited for predicting the number of fines and accidents for road users in the UAE.

4. Results

The survey was randomly distributed to a diverse segment of road users in the UAE; it consists of several questions to obtain information about the respondents’ age, gender, nationality, monthly income, emirate, vehicles, driving frequency, major purpose of driving, driving experience (in years), and number of issued tickets and accidents over the past two years. In total, 458 responses were collected and are analyzed in the following subsections.

4.1. Analyzing the Responses

For the collected data, male and female respondents were almost equal, which enables multiple unbiased comparisons between the gender categories; the results show that men get traffic tickets/fines more than women, while the latter have higher rates of traffic accidents in the UAE, as illustrated in Figure 6. For driving experience, Figure 7 reveals that men have relatively more years of driving experience in the UAE.
Interestingly, across the main emirates, Dubai has the highest record for traffic tickets and accidents in the past two years, followed by Sharjah, Al-Ain, Abu Dhabi, Ras Al-Khaimah, Ajman, and Fujairah, as demonstrated in Figure 8. This rank can be justified by the distribution of the population in the UAE, which is concentrated in Dubai, Sharjah, and Abu Dhabi.
Furthermore, most of the respondents use cars for road transportation (79%), while only 20% use Sport Utility Vehicles (SUVa), and the remaining use motorcycles and other vehicles (1%). This would explain the remarkable number of tickets and accidents for cars, which is approximately three times that of SUVs, as shown in Figure 9.
In addition, a question regarding the frequency of driving of the respondents was asked to check if there was any impact of this aspect on the number of tickets or accidents. The results are demonstrated in Figure 10, which shows that everyday commuters have the highest number of traffic tickets and accidents as they are exposed to traffic-related challenges or risks more frequently than other commuters. Commuters who drive several times in a week have the second largest number of traffic tickets and accidents, therefore, the lower the frequency of driving, the lower the chance of being affected by a traffic-related problem.
Different age groups were investigated, ranging from 18 to 65 years old, and by linking the respondent’s age with their tickets and accidents in the past two years it was found that the older people get the lower their chances are of getting into a traffic accident or having traffic tickets, as illustrated in Figure 11. From the figures, it is noticeable that the youngest age group (18–25) have the highest record of tickets and accidents, and these records decrease gradually with aging. This would relate to the experience, responsibility, and maturity factors that take place with aging.
Moreover, 23% of the respondents were Emiratis and 77% were non-Emiratis from different nationalities including the USA, the UK, and European, Asian, and Arab countries. This demographic attribute is consistent with the overall demographic in the UAE [88]. After exploring the number of tickets and accidents across the nationalities of respondents, as portrayed in Figure 12, it was revealed that Arab nationalities including the UAE, Jordan, Egypt, Syria, and Palestine have the highest number of tickets and accidents in the past two years compared to other Western nationalities.

4.2. Predicting the Probability of Traffic-Related Impacts in the UAE Using ANN Techniques

Traffic tickets and accidents are the factors that are linked to social and economic sustainability. The purpose of this section is to build a predictive model for each of these impacts based on the demographic attributes of the respondents. The output of these models will be categorical, as either having or not having a ticket or accident; this is accomplished by training the dataset to build a model where predictions are performed. ANN models are adopted to build these predictive models; the advantages of ANN models include their ability to detect and deal with complex nonlinear relationships between the output variables and the indicating variable, their ability to handle redundant attributes by normalizing the weights of all attributes, fast model testing, and the ease of using their various training algorithms [89,90]. This analysis was performed using two ANN packages in RStudio (2023.06.0 Build 421), namely, H2O (software based in RStudio) and Keras/Tensorflow (python-based software through RStudio).
Nine indicating variables were used to predict the probability of road users having a ticket or accident in the UAE, these variables were age, gender, emirate where the respondent resides, monthly income, nationality of respondent, driving frequency, vehicle type, driving purpose, and driving experience in the UAE (years). Since the variables are mainly categorical, they should be converted using hot encoding functions, as ANN does not operate with characters or categorical variables. The original nine indicators became 33 indicators after applying hot encoding, this was one of the preprocessing techniques, along with dealing with missing values for all the variables. However, it is necessary to normalize the data after this conversion to unify the scale for all the indicating variables, thus overcoming the vanishing gradient problem which might affect the outcomes [91,92].
The accuracy values for the training and testing models are largely equal for both the tickets and accidents’ predictive models, which indicates that the models’ statistical fit is acceptable, and they are neither underfitted nor overfitted. From the results in Table 4, it can be observed that the performance of the H2O and Keras/Tensorflow models is relatively good, with more emphasis on Keras/Tensorflow due to its 93.7% performance for the tickets’ predictive model, and 95.8% for the accidents’ predictive model. It is worth mentioning that the accuracy of both models can be substantially enhanced by using more data points, as ANN models require a considerable amount of training data to result in higher accuracy.

5. Discussion

Transportation is a crucial aspect of the world’s industry, social life, and economy; it overcomes the spatial limitations of humans daily, creates job opportunities, and utilizes time well. Also, transportation systems are often referred to as the “lifeblood” of cities, emphasizing their crucial role in supporting urban life [1]. As a key component of sustainable development (SD), transportation systems enable access to resources and markets, contributing to economic growth and facilitating a higher quality of life. They connect individuals to various essential activities such as employment, healthcare, education, recreation, and more, enhancing their overall well-being and societal functioning. This research is focused on the social and economic pillars of sustainability as perceived by road users in the UAE, where traffic accidents and tickets can be linked to both pillars.
The results showed that Dubai, Sharjah, and Abu Dhabi are the top emirates in terms of the reported number of tickets and accidents; this would refer to the fact that these emirates have the highest proportions of the population of the UAE [70], and hence, the highest number of road commuters. Similarly, a study conducted in 2018 examined traffic road crashes and fines in the UAE, confirming the consistent finding of a high number of accidents and tickets in Dubai and Abu Dhabi [25].
By analyzing the collected data from road users in the UAE, it was concluded that although men are more experienced than women in terms of the number of years they have been driving in the UAE, men are still getting into traffic accidents and receiving tickets more than women. This finding is in line with a previous, similar investigation, where it was noticed that women show more commitment in their driving behavior, and they also drive less frequently than men [70]. Therefore, men receive more tickets than women [40,41].
In addition, cars are mostly used by road users in the UAE, followed by sport utility vehicles, and a minor proportion of other vehicles, such as motorcycles and vans. This would explain the high number of reported traffic accidents and tickets for car users in the UAE. Likewise, it was observed that road users who experience driving frequently (on a daily basis) are the group that most often get into traffic accidents and receive tickets, compared to those with less frequent driving experience; this is related to their higher potential of being exposed to road-related risks and incidents.
Another insightful finding related to the road users in the UAE is the relationship between age and traffic-related impacts. It was indicated that young drivers are exposed to a higher probability of road accidents and tickets than older ones, which is in accordance with previously published findings [31,32,38,39], as experience plays a significant role in enhancing the awareness and consciousness of road users. This is apparent from the results that the youngest age group (18–25) had the highest number of traffic accidents as well as tickets in the UAE. Also, this has been found in previous studies in the UAE, such as [63], which revealed that around 45% of traffic-related injuries have been experienced by drivers from 18 to 35 years old, in addition to [93], which showed that traffic accidents and fatalities mostly occurred for drivers from 18 to 30 years old.
Regarding the heterogenous structure of the population in the UAE, it was revealed that there is a relationship between nationality and traffic-related impacts including accidents and tickets. While Arab nationalities have the highest record of accidents and tickets of all nationalities, among the Arab countries, the UAE reported the highest number of accidents and traffic; a similar finding was previously reported for GCC countries [69,70,71].
Finally, machine learning-based predictive models were built using ANN models to predict the probability of having traffic accidents as well as receiving traffic tickets. Two different models were built and compared for each variable of interest; the results showed that the Keras/Tensorflow models resulted in higher accuracies than the H2O models, with 93.7% for the tickets’ predictive model, and 95.8% for the accidents’ predictive model. However, the accuracy of these models can be considerably improved by using a much larger dataset, as the bigger the dataset, the higher the accuracy of the ANN models’ performance [94].

6. Conclusions

This study undertook a comprehensive examination of the demographic factors influencing road users in the UAE and their association with traffic-related incidents, such as accidents and tickets. These incidents are integral to the social and economic dimensions of road sustainability as perceived by users. This investigation revealed noteworthy correlations between gender, driving frequency, age, nationality, and the reported number of traffic incidents and tickets.
Specifically, men demonstrated a higher incidence of traffic accidents and tickets compared to women, suggesting differences in their commitment to road and traffic rules. Age exhibited a negative linear relationship with incidents and tickets, indicating a decrease in the likelihood of such occurrences as road users age. Conversely, driving frequency displayed a positive linear relationship, signifying that more frequent driving correlates with a higher risk of traffic-related issues.
Furthermore, the study developed an ANN predictive model to forecast the likelihood of road users experiencing accidents or receiving tickets based on these significant factors. The comparison of two models revealed their satisfactory predictive performance, with accuracy rates of 93.7% for tickets and 95.8% for accidents. These findings, along with the predictive models, offer valuable insights for stakeholders such as government entities, road users, contractors, and designers, contributing to the improvement of the social and economic aspects of road sustainability.
While this research focused on specific factors, future investigations could expand on this by conducting comparative analyses with countries sharing similarities or differences with the UAE, thereby validating or uncovering new insights. Collecting additional data points for training predictive models could enhance their accuracy. Additionally, integrating other traffic-related issues into the research would contribute to a more holistic approach toward the sustainable development of roads and traffic.

Author Contributions

Conceptualization, H.A., R.A. and G.A.-L.; methodology, H.A.; software, H.A.; validation, H.A.; formal analysis, H.A.; investigation, H.A. and R.A.; resources, H.A., R.A. and G.A.-L.; data curation, H.A.; writing—original draft preparation, H.A. and R.A.; writing—review and editing, H.A., R.A. and G.A.-L.; visualization, H.A. and R.A.; supervision, G.A.-L.; project administration, H.A. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data are available upon request.

Acknowledgments

This work was supported in part by the International Association of Traffic and Safety Sciences (IATSS), Japan. The authors are grateful to the members of IATSS’s project for their valuable contribution in developing the survey. The views expressed here are strictly the authors.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Miller, P.; de Barros, A.G.; Kattan, L.; Wirasinghe, S.C. Public transportation and sustainability: A review. KSCE J. Civ. Eng. 2016, 20, 1076–1083. [Google Scholar] [CrossRef]
  2. WEF. The Global Competitiveness Report 2019. 2019, Volume 2023. Available online: https://www3.weforum.org/docs/WEF_TheGlobalCompetitivenessReport2019.pdf (accessed on 27 September 2023).
  3. WEFRQ. CountryRrankings. World Economic Forum—Roads Quality—Country Rankings. Available online: https://www.theglobaleconomy.com/rankings/roads_quality/ (accessed on 5 April 2022).
  4. Bucsuházy, K.; Matuchová, E.; Zůvala, R.; Moravcová, P.; Kostíková, M.; Mikulec, R. Human factors contributing to the road traffic accident occurrence. Transp. Res. Procedia 2020, 45, 555–561. [Google Scholar] [CrossRef]
  5. Timmermans, C.; Alhajyaseen, W.; Reinolsmann, N.; Nakamura, H.; Suzuki, K. Traffic safety culture of professional drivers in the State of Qatar. IATSS Res. 2019, 43, 286–296. [Google Scholar] [CrossRef]
  6. Scopus. Analyze Search Results—Road Traffic Sustainability in the UAE. Available online: https://www.scopus.com/term/analyzer.uri?sid=0e0db64877f17c97537052804ae892ea&origin=resultslist&src=s&s=Road+traffic+Sustainability+United+Arab+Emirates&sort=plf-f&sdt=a&sot=a&sl=48&count=354&analyzeResults=Analyze+results&txGid=77ddcee71f59cab292e9f0d9b609fed8 (accessed on 5 April 2022).
  7. Moya-Gómez, B.; García-Palomares, J.C. The impacts of congestion on automobile accessibility. What happens in large European cities? J. Transp. Geogr. 2017, 62, 148–159. [Google Scholar] [CrossRef]
  8. Moavenzadeh, F.; Markow, M.J. Moving Millions: Transport Strategies for Sustainable Development in Megacities; Springer: Berlin/Heidelberg, Germany, 2007. [Google Scholar]
  9. Kaplan, D.H. Growing sustainable transportation in an autocentric community: Current trends and applications. In Urban and Regional Planning and Development; Springer: Berlin/Heidelberg, Germany, 2020; pp. 503–514. [Google Scholar]
  10. Ülengin, F.; Önsel, Ş.; Topçu, Y.I.; Aktaş, E.; Kabak, Ö. An integrated transportation decision support system for transportation policy decisions: The case of Turkey. Transp. Res. Part A Policy Pract. 2007, 41, 80–97. [Google Scholar] [CrossRef]
  11. Aljoufie, M. The impact assessment of increasing population density on Jeddah road transportation using spatial-temporal analysis. Sustainability 2021, 13, 1455. [Google Scholar] [CrossRef]
  12. Jones, P.; Lucas, K. The social consequences of transport decision-making: Clarifying concepts, synthesising knowledge and assessing implications. J. Transp. Geogr. 2012, 21, 4–16. [Google Scholar] [CrossRef]
  13. Szmelter-Jarosz, A.; Rześny-Cieplińska, J. Priorities of urban transport system stakeholders according to crowd logistics solutions in city areas. A sustainability perspective. Sustainability 2019, 12, 317. [Google Scholar] [CrossRef]
  14. Wang, D.D. Assessing road transport sustainability by combining environmental impacts and safety concerns. Transp. Res. Part D Transp. Environ. 2019, 77, 212–223. [Google Scholar] [CrossRef]
  15. Agbelie, B.R. An empirical analysis of three econometric frameworks for evaluating economic impacts of transportation infrastructure expenditures across countries. Transp. Policy 2014, 35, 304–310. [Google Scholar] [CrossRef]
  16. Mckenzie, S. Social Sustainability: Towards Some Definitions; Hawke Research Institute: Underdale, Australia, 2004. [Google Scholar]
  17. Boschmann, E.E.; Kwan, M.-P. Toward socially sustainable urban transportation: Progress and potentials. Int. J. Sustain. Transp. 2008, 2, 138–157. [Google Scholar] [CrossRef]
  18. Anbarci, N.; Escaleras, M.; Register, C.A. Traffic fatalities: Does income inequality create an externality? Can. J. Econ. Rev. Can. D’économique 2009, 42, 244–266. [Google Scholar] [CrossRef]
  19. Qureshi, A.I.; Huang, W.; Khan, S.; Lobanova, I.; Siddiq, F.; Gomez, C.R.; Suri, M.F.K. Mandated societal lockdown and road traffic accidents. Accid. Anal. Prev. 2020, 146, 105747. [Google Scholar] [CrossRef] [PubMed]
  20. Bener, A. The neglected epidemic: Road traffic accidents in a developing country, State of Qatar. Int. J. Inj. Control Saf. Promot. 2005, 12, 45–47. [Google Scholar] [CrossRef] [PubMed]
  21. Yasin, Y.J.; Grivna, M.; Abu-Zidan, F.M. Global impact of COVID-19 pandemic on road traffic collisions. World J. Emerg. Surg. 2021, 16, 51. [Google Scholar] [CrossRef] [PubMed]
  22. Khreis, H.; May, A.D.; Nieuwenhuijsen, M.J. Health impacts of urban transport policy measures: A guidance note for practice. J. Transp. Health 2017, 6, 209–227. [Google Scholar] [CrossRef]
  23. Iyanna, S.; Bosangit, C.; Lazell, J.; Carrigan, M. A theories of practice perspective in understanding barriers to sustainable commuting: The case of United Arab Emirates. Int. J. Nonprofit Volunt. Sect. Mark. 2019, 24, e1668. [Google Scholar] [CrossRef]
  24. Alzard, M.H.; Maraqa, M.A.; Chowdhury, R.; Khan, Q.; Albuquerque, F.D.; Mauga, T.I.; Aljunadi, K.N. Estimation of greenhouse gas emissions produced by road projects in Abu Dhabi, United Arab Emirates. Sustainability 2019, 11, 2367. [Google Scholar] [CrossRef]
  25. Ankit, A.; Hamidi, S.; Sengoelge, M. Causes and health-related outcomes of road traffic crashes in the united arab emirates: Panel data analysis of traffic fines. Glob. J. Health Sci. 2018, 10, 165. [Google Scholar] [CrossRef]
  26. Road Safety—The Official Portal of the UAE Government. Available online: https://u.ae/en/information-and-services/justice-safety-and-the-law/road-safety (accessed on 7 April 2022).
  27. Adam, F.; Qasrawi, A.; Ignatious, H.A.; Chaqfeh, M.; El Sayed, H. Exploiting Vehicular Communication Systems for Improving Traffic Safety in the UAE. In Proceedings of the 2019 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), Ajman, United Arab Emirates, 9–11 December 2019; IEEE: Piscataway, NJ, USA; pp. 1–6. [Google Scholar]
  28. Haza, R. UAE Road Deaths Lowest in Almost a Decade as Drivers Hit with 8 Million Fines. Available online: https://www.thenationalnews.com/uae/transport/uae-road-deaths-lowest-in-almost-a-decade-as-drivers-hit-with-8-million-fines-1.1045642 (accessed on 1 May 2022).
  29. Mirzaei, R.; Hafezi-Nejad, N.; Sabagh, M.S.; Moghaddam, A.A.; Eslami, V.; Rakhshani, F.; Rahimi-Movaghar, V. Dominant role of drivers’ attitude in prevention of road traffic crashes: A study on knowledge, attitude, and practice of drivers in Iran. Accid. Anal. Prev. 2014, 66, 36–42. [Google Scholar] [CrossRef]
  30. Lund, I.O.; Rundmo, T. Cross-cultural comparisons of traffic safety, risk perception, attitudes and behaviour. Saf. Sci. 2009, 47, 547–553. [Google Scholar] [CrossRef]
  31. Mohamed, M.; Bromfield, N.F. Attitudes, driving behavior, and accident involvement among young male drivers in Saudi Arabia. Transp. Res. Part F Traffic Psychol. Behav. 2017, 47, 59–71. [Google Scholar] [CrossRef]
  32. Lucidi, F.; Mallia, L.; Lazuras, L.; Violani, C. Personality and attitudes as predictors of risky driving among older drivers. Accid. Anal. Prev. 2014, 72, 318–324. [Google Scholar] [CrossRef] [PubMed]
  33. Ledesma, R.D.; Tosi, J.D.; Díaz-Lázaro, C.M.; Poó, F.M. Predicting road safety behavior with implicit attitudes and the Theory of Planned Behavior. J. Saf. Res. 2018, 66, 187–194. [Google Scholar] [CrossRef]
  34. Eagly, A.H.; Chaiken, S. The Psychology of Attitudes; Harcourt Brace Jovanovich College Publishers: San Diego, CA, USA, 1993. [Google Scholar]
  35. Mohammadzadeh Moghaddam, A.; Sadeghi, A.; Jalili Qazizadeh, M.; Farhad, H.; Barakchi, M. Investigating the relationship between driver’s ticket frequency and demographic, behavioral, and personal factors: Which drivers commit more offenses? J. Transp. Saf. Secur. 2020, 12, 225–244. [Google Scholar] [CrossRef]
  36. Oppenheim, I.; Oron-Gilad, T.; Parmet, Y.; Shinar, D. Can traffic violations be traced to gender-role, sensation seeking, demographics and driving exposure? Transp. Res. Part F Traffic Psychol. Behav. 2016, 43, 387–395. [Google Scholar] [CrossRef]
  37. Alver, Y.; Demirel, M.; Mutlu, M. Interaction between socio-demographic characteristics: Traffic rule violations and traffic crash history for young drivers. Accid. Anal. Prev. 2014, 72, 95–104. [Google Scholar] [CrossRef]
  38. Rolison, J.J.; Moutari, S. Combinations of factors contribute to young driver crashes. J. Saf. Res. 2020, 73, 171–177. [Google Scholar] [CrossRef]
  39. Keyvanfar, A.; Shafaghat, A.; Muhammad, N.Z.; Ferwati, M.S. Driving behaviour and sustainable mobility—Policies and approaches revisited. Sustainability 2018, 10, 1152. [Google Scholar] [CrossRef]
  40. González-Iglesias, B.; Gómez-Fraguela, J.A.; Luengo-Martín, M.Á. Driving anger and traffic violations: Gender differences. Transp. Res. Part F Traffic Psychol. Behav. 2012, 15, 404–412. [Google Scholar] [CrossRef]
  41. Horvath, C.; Lewis, I.; Watson, B. The beliefs which motivate young male and female drivers to speed: A comparison of low and high intenders. Accid. Anal. Prev. 2012, 45, 334–341. [Google Scholar] [CrossRef] [PubMed]
  42. Akaateba, M.A.; Amoh-Gyimah, R.; Amponsah, O. Traffic safety violations in relation to drivers’ educational attainment, training and experience in Kumasi, Ghana. Saf. Sci. 2015, 75, 156–162. [Google Scholar] [CrossRef]
  43. Qin, L.; Li, Z.R.; Chen, Z.; Bill, M.A.; Noyce, D.A. Understanding driver distractions in fatal crashes: An exploratory empirical analysis. J. Saf. Res. 2019, 69, 23–31. [Google Scholar] [CrossRef] [PubMed]
  44. Caird, J.K.; Johnston, K.A.; Willness, C.R.; Asbridge, M. The use of meta-analysis or research synthesis to combine driving simulation or naturalistic study results on driver distraction. J. Saf. Res. 2014, 49, 91–96. [Google Scholar] [CrossRef] [PubMed]
  45. McEvoy, S.P.; Stevenson, M.R.; Woodward, M. The prevalence of, and factors associated with, serious crashes involving a distracting activity. Accid. Anal. Prev. 2007, 39, 475–482. [Google Scholar] [CrossRef] [PubMed]
  46. Née, M.; Contrand, B.; Orriols, L.; Gil-Jardiné, C.; Galéra, C.; Lagarde, E. Road safety and distraction, results from a responsibility case-control study among a sample of road users interviewed at the emergency room. Accid. Anal. Prev. 2019, 122, 19–24. [Google Scholar] [CrossRef]
  47. Gariazzo, C.; Stafoggia, M.; Bruzzone, S.; Pelliccioni, A.; Forastiere, F. Association between mobile phone traffic volume and road crash fatalities: A population-based case-crossover study. Accid. Anal. Prev. 2018, 115, 25–33. [Google Scholar] [CrossRef]
  48. Das, D.K. Exploring the significance of road and traffic factors on traffic crashes in a South African city. Int. J. Transp. Sci. Technol. 2022, 12, 414–427. [Google Scholar] [CrossRef]
  49. Cooper, B.; Gehrsitz, M.; McIntyre, S.G. Drink, death, and driving: Do blood alcohol content limit reductions improve road safety? Health Econ. 2020, 29, 841–847. [Google Scholar] [CrossRef]
  50. Houwing, S.; Stipdonk, H. Driving under the influence of alcohol in the Netherlands by time of day and day of the week. Accid. Anal. Prev. 2014, 72, 17–22. [Google Scholar] [CrossRef]
  51. Bharadwaj, N.; Edara, P.; Sun, C. Sleep disorders and risk of traffic crashes: A naturalistic driving study analysis. Saf. Sci. 2021, 140, 105295. [Google Scholar] [CrossRef]
  52. Moradi, A.; Nazari, S.S.H.; Rahmani, K. Sleepiness and the risk of road traffic accidents: A systematic review and meta-analysis of previous studies. Transp. Res. Part F Traffic Psychol. Behav. 2019, 65, 620–629. [Google Scholar] [CrossRef]
  53. Bener, A.; Yildirim, E.; Özkan, T.; Lajunen, T. Driver sleepiness, fatigue, careless behavior and risk of motor vehicle crash and injury: Population based case and control study. J. Traffic Transp. Eng. 2017, 4, 496–502. [Google Scholar] [CrossRef]
  54. Vardaki, S.; Yannis, G. Investigating the self-reported behavior of drivers and their attitudes to traffic violations. J. Saf. Res. 2013, 46, 1–11. [Google Scholar] [CrossRef]
  55. Oviedo-Trespalacios, O. Getting away with texting: Behavioural adaptation of drivers engaging in visual-manual tasks while driving. Transp. Res. Part A Policy Pract. 2018, 116, 112–121. [Google Scholar] [CrossRef]
  56. Tan, C.; Shi, Y.; Bai, L.; Tang, K.; Suzuki, K.; Nakamura, H. Modeling effects of driver safety attitudes on traffic violations in China using the theory of planned behavior. IATSS Res. 2022, 46, 63–72. [Google Scholar] [CrossRef]
  57. Timmermans, C.; Shawky, M.; Alhajyaseen, W.; Nakamura, H. Investigating the attitudes of Egyptian drivers toward traffic safety. IATSS Res. 2022, 46, 73–81. [Google Scholar] [CrossRef]
  58. Siuhi, S.; Mwakalonge, J. Opportunities and challenges of smart mobile applications in transportation. J. Traffic Transp. Eng. 2016, 3, 582–592. [Google Scholar] [CrossRef]
  59. Almoshaogeh, M.; Abdulrehman, R.; Haider, H.; Alharbi, F.; Jamal, A.; Alarifi, S.; Shafiquzzaman, M. Traffic accident risk assessment framework for Qassim, Saudi Arabia: Evaluating the impact of speed cameras. Appl. Sci. 2021, 11, 6682. [Google Scholar] [CrossRef]
  60. Satiennam, W.; Satiennam, T.; Triyabutra, T.; Rujopakarn, W. Red light running by young motorcyclists: Factors and beliefs influencing intentions and behavior. Transp. Res. Part F Traffic Psychol. Behav. 2018, 55, 234–245. [Google Scholar] [CrossRef]
  61. Iversen, H. Risk-taking attitudes and risky driving behaviour. Transp. Res. Part F Traffic Psychol. Behav. 2004, 7, 135–150. [Google Scholar] [CrossRef]
  62. Suzuki, K.; Tang, K.; Alhajyaseen, W.; Suzuki, K.; Nakamura, H. An international comparative study on driving attitudes and behaviors based on questionnaire surveys. IATSS Res. 2022, 46, 26–35. [Google Scholar] [CrossRef]
  63. Afroza, S.; Rouf, R. A Statistical Study on Young UAE Driver’s Behavior towards Road Safety. Int. J. Humanit. Soc. Sci. 2018, 11, 2535–2539. [Google Scholar]
  64. Grivna, M.; Barss, P.; Stanculescu, C.; Eid, H.O.; Abu-Zidan, F.M. Child and youth traffic-related injuries: Use of a trauma registry to identify priorities for prevention in the United Arab Emirates. Traffic Inj. Prev. 2013, 14, 274–282. [Google Scholar] [CrossRef] [PubMed]
  65. Hassan, M.N.; Hawas, Y.E.; Maraqa, M.A. A holistic approach for assessing traffic safety in the United Arab Emirates. Accid. Anal. Prev. 2012, 45, 554–564. [Google Scholar] [CrossRef]
  66. Alghafli, A.; Mohamad, E.; Zaidy, A.A. The relationship between the elapsed time from the onset of red signal until its violation and traffic accident occurrence in Abu Dhabi, UAE. Safety 2021, 7, 53. [Google Scholar] [CrossRef]
  67. Malik, A.; Safi, A. AI Used to Identify Car Accident Injury and Mortality Risk Factors in Dubai UAE. In Proceedings of the International Symposium on Intelligent Computing Systems, Santiago, Chile, 23–25 March 2022; pp. 97–105. [Google Scholar]
  68. Abdalla, I.M. Fatality risk assessment and modeling of drivers responsibility for causing traffic accidents in Dubai. J. Saf. Res. 2002, 33, 483–496. [Google Scholar] [CrossRef]
  69. Abay, K.A.; Mannering, F.L. An empirical analysis of risk-taking in car driving and other aspects of life. Accid. Anal. Prev. 2016, 97, 57–68. [Google Scholar] [CrossRef]
  70. Bener, A.; Özkan, T.; Lajunen, T. The driver behaviour questionnaire in Arab gulf countries: Qatar and United Arab Emirates. Accid. Anal. Prev. 2008, 40, 1411–1417. [Google Scholar] [CrossRef]
  71. Özkan, T.; Lajunen, T.; Chliaoutakis, J.E.; Parker, D.; Summala, H. Cross-cultural differences in driving behaviours: A comparison of six countries. Transp. Res. Part F Traffic Psychol. Behav. 2006, 9, 227–242. [Google Scholar] [CrossRef]
  72. Timmermans, C.P.; Alhajyaseen, W.K.; Ross, V.; Nakamura, H. Introducing a multi-variate classification method: Risky driving acceptance among different heterogeneous driver sub-cultures. J. Saf. Res. 2020, 73, 81–91. [Google Scholar] [CrossRef] [PubMed]
  73. Chen, S.; Kuhn, M.; Prettner, K.; Bloom, D.E. The global macroeconomic burden of road injuries: Estimates and projections for 166 countries. Lancet Planet. Health 2019, 3, e390–e398. [Google Scholar] [CrossRef] [PubMed]
  74. DrissiTouzani, H.; Faquir, S.; Yahyaouy, A. Data Mining Techniques to Analyze Traffic Accidents Data: Case Application in Morocco. In Proceedings of the 2020 Fourth International Conference On Intelligent Computing in Data Sciences (ICDS), Fez, Morocco, 21–23 October 2020; IEEE: Piscataway, NJ, USA; pp. 1–4. [Google Scholar]
  75. Castro, Y.; Kim, Y.J. Data mining on road safety: Factor assessment on vehicle accidents using classification models. Int. J. Crashworthiness 2016, 21, 104–111. [Google Scholar] [CrossRef]
  76. Behbahani, H.; Amiri, A.M.; Imaninasab, R.; Alizamir, M. Forecasting accident frequency of an urban road network: A comparison of four artificial neural network techniques. J. Forecast. 2018, 37, 767–780. [Google Scholar] [CrossRef]
  77. Gutierrez-Osorio, C.; Pedraza, C. Modern data sources and techniques for analysis and forecast of road accidents: A review. J. Traffic Transp. Eng. (Engl. Ed.) 2020, 7, 432–446. [Google Scholar] [CrossRef]
  78. AlKheder, S.; AlRukaibi, F.; Aiash, A. Risk analysis of traffic accidents’ severities: An application of three data mining models. ISA Trans. 2020, 106, 213–220. [Google Scholar] [CrossRef] [PubMed]
  79. Bui, D.-K.; Nguyen, T.N.; Ngo, T.D.; Nguyen-Xuan, H. An artificial neural network (ANN) expert system enhanced with the electromagnetism-based firefly algorithm (EFA) for predicting the energy consumption in buildings. Energy 2020, 190, 116370. [Google Scholar] [CrossRef]
  80. Tseng, W.S.; Nguyen, H.; Liebowitz, J.; Agresti, W. Distractions and motor vehicle accidents: Data mining application on fatality analysis reporting system (FARS) data files. Ind. Manag. Data Syst. 2005, 105, 1118–1205. [Google Scholar] [CrossRef]
  81. El Mazouri, F.Z.; Abounaima, M.C.; Zenkouar, K. Data mining combined to the multicriteria decision analysis for the improvement of road safety: Case of France. J. Big Data 2019, 6, 5. [Google Scholar] [CrossRef]
  82. Sohn, S.Y.; Shin, H. Pattern recognition for road traffic accident severity in Korea. Ergonomics 2001, 44, 107–117. [Google Scholar] [CrossRef]
  83. Comi, A.; Polimeni, A.; Balsamo, C. Road accident analysis with data mining approach: Evidence from Rome. Transp. Res. Procedia 2022, 62, 798–805. [Google Scholar] [CrossRef]
  84. Shirmohammadi, H.; Hadadi, F.; Saeedian, M. Clustering analysis of drivers based on behavioral characteristics regarding road safety. Int. J. Civ. Eng. 2019, 17, 1327–1340. [Google Scholar] [CrossRef]
  85. Wang, L.; Wang, Y.; Shi, L.; Xu, H. Analysis of risky driving behaviors among bus drivers in China: The role of enterprise management, external environment and attitudes towards traffic safety. Accid. Anal. Prev. 2022, 168, 106589. [Google Scholar] [CrossRef] [PubMed]
  86. Nordfjærn, T.; Rundmo, T. Road traffic safety beliefs and driver behaviors among personality subtypes of drivers in the Norwegian population. Traffic Inj. Prev. 2013, 14, 690–696. [Google Scholar] [CrossRef] [PubMed]
  87. Wang, C.; Xu, C.; Xia, J.; Qian, Z. The effects of safety knowledge and psychological factors on self-reported risky driving behaviors including group violations for e-bike riders in China. Transp. Res. Part F Traffic Psychol. Behav. 2018, 56, 344–353. [Google Scholar] [CrossRef]
  88. The United Arab Emirates’ Government Portal. Available online: https://u.ae/en/information-and-services/social-affairs/preserving-the-emirati-national-identity/population-and-demographic-mix (accessed on 26 September 2023).
  89. Tu, J.V. Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. J. Clin. Epidemiol. 1996, 49, 1225–1231. [Google Scholar] [CrossRef]
  90. Pleune, T.T.; Chopra, O.K. Using artificial neural networks to predict the fatigue life of carbon and low-alloy steels. Nucl. Eng. Des. 2000, 197, 1–12. [Google Scholar] [CrossRef]
  91. Hochreiter, S. The vanishing gradient problem during learning recurrent neural nets and problem solutions. Int. J. Uncertain. Fuzziness Knowl.-Based Syst. 1998, 6, 107–116. [Google Scholar] [CrossRef]
  92. Hu, Y.; Huber, A.; Anumula, J.; Liu, S.-C. Overcoming the vanishing gradient problem in plain recurrent networks. arXiv 2018, arXiv:1801.06105. [Google Scholar]
  93. Hammoudi, A.; Karani, G.; Littlewood, J. Road traffic accidents among drivers in Abu Dhabi, United Arab Emirates. J. Traffic Logist. Eng. 2014, 2, 7–12. [Google Scholar] [CrossRef]
  94. Horvath, S.; Soot, M.; Zaddach, S.; Neuner, H.; Weitkamp, A. Deriving adequate sample sizes for ANN-based modelling of real estate valuation tasks by complexity analysis. Land Use Policy 2021, 107, 105475. [Google Scholar] [CrossRef]
Sustainability 16 01092 g001
Figure 1. Annual published documents in the UAE on road traffic sustainability [6].
Figure 1. Annual published documents in the UAE on road traffic sustainability [6].
Sustainability 16 01092 g001
Sustainability 16 01092 g002
Figure 2. UAE road deaths (2014–2019) [28].
Figure 2. UAE road deaths (2014–2019) [28].
Sustainability 16 01092 g002
Sustainability 16 01092 g003
Figure 3. Number of issued fines in the UAE in 2018 and 2019.
Figure 3. Number of issued fines in the UAE in 2018 and 2019.
Sustainability 16 01092 g003
Sustainability 16 01092 g004
Figure 4. Research methodology.
Figure 4. Research methodology.
Sustainability 16 01092 g004
Sustainability 16 01092 g005
Figure 5. ANN model.
Figure 5. ANN model.
Sustainability 16 01092 g005
Sustainability 16 01092 g006
Figure 6. Number of traffic tickets and accidents across genders. (a) Number of traffic tickets across genders; (b) number of traffic accidents across genders.
Figure 6. Number of traffic tickets and accidents across genders. (a) Number of traffic tickets across genders; (b) number of traffic accidents across genders.
Sustainability 16 01092 g006
Sustainability 16 01092 g007
Figure 7. Driving experience in the UAE across genders.
Figure 7. Driving experience in the UAE across genders.
Sustainability 16 01092 g007
Sustainability 16 01092 g008
Figure 8. Number of traffic tickets and accidents across the UAE. (a) Number of traffic tickets across the UAE; (b) number of traffic accidents across the UAE.
Figure 8. Number of traffic tickets and accidents across the UAE. (a) Number of traffic tickets across the UAE; (b) number of traffic accidents across the UAE.
Sustainability 16 01092 g008
Sustainability 16 01092 g009
Figure 9. Number of traffic tickets and accidents across vehicle types. (a) Number of traffic tickets across vehicle types; (b) number of traffic accidents across vehicle types.
Figure 9. Number of traffic tickets and accidents across vehicle types. (a) Number of traffic tickets across vehicle types; (b) number of traffic accidents across vehicle types.
Sustainability 16 01092 g009
Sustainability 16 01092 g010
Figure 10. Number of traffic tickets and accidents according to driving frequency. (a) Number of traffic tickets according to driving frequency; (b) number of traffic accidents according to driving frequency.
Figure 10. Number of traffic tickets and accidents according to driving frequency. (a) Number of traffic tickets according to driving frequency; (b) number of traffic accidents according to driving frequency.
Sustainability 16 01092 g010
Sustainability 16 01092 g011
Figure 11. Number of traffic tickets and accidents across age groups. (a) Number of traffic tickets across age groups; (b) number of traffic accidents across age groups.
Figure 11. Number of traffic tickets and accidents across age groups. (a) Number of traffic tickets across age groups; (b) number of traffic accidents across age groups.
Sustainability 16 01092 g011
Sustainability 16 01092 g012
Figure 12. Number of traffic tickets and accidents across nationalities. (a) Number of traffic tickets across nationalities; (b) number of traffic accidents across nationalities.
Figure 12. Number of traffic tickets and accidents across nationalities. (a) Number of traffic tickets across nationalities; (b) number of traffic accidents across nationalities.
Sustainability 16 01092 g012
Table 1. Transportation problems in relation to the TBL. Source: authors.
Table 1. Transportation problems in relation to the TBL. Source: authors.
Transportation ProblemsPillars of Sustainability
EnvironmentalSocialEconomic
Air pollutionAir qualityHealthHealth care and other costs
Noise pollutionDisruptions to biodiversity and ecological functions and cycles, e.g., sleeping and breeding patterns of faunaHealthHealth care and other costs
Traffic congestion—increased travel timeIncreased emissionsTime constraints on householdsTime costs
Road safety
(speed, seatbelt, alcohol, etc.)		Resources used in repairing and replacingInjuries and deathsAccident costs
Financial cost (affordability)Higher emissions from older cars due to inadequate maintenanceHousehold budgetsAccessibility to jobs, school, etc.
Vehicle maintenance and insurance costs, government fees
Physical activity and healthZero emissions from non-motorized modes of transportHealthMedical costs
Increasing vehicle numbersIncreased emissions
from rising car ownership		Health, e.g., stress,
increasing safety concerns		Infrastructure costs, transport service user fees (e.g., salik)
Table 2. Extracted indicators of risky driving behaviors.
Table 2. Extracted indicators of risky driving behaviors.
Previous StudiesAuthors’ NamesRisky Driving Behaviors
Driver DistractionsPhone Use While DrivingAlcohol and/or DrugsDrowsy DrivingSeatbelt-Related ViolationsSpeed
[43]Qin, L. et al.XXX
[44]Caird, J. K. et al.XXX
[45]McEvoy, S. P. et al.XX X
[46]Née, M. et al.XXXXX
[47]Gariazzo, C. et al. XX
[48]Das, D. K. X X
[49]Cooper, B. et al. X X
[50]Houwing, S. and Stipdonk, H. X
[51]Bharadwaj, N. et al. XX
[52]Moradi, A. et al. XX
[53]Bener, A. et al. XXXX
[54]Vardaki, S. and Yannis, G. XX XX
[55]Oviedo-Trespalacios, O.XXX XX
[56]Tan, C. et al.XXX X
[57]Timmermans, C. et al.XXX XX
[58]Siuhi, S. and Mwakalonge, J.XXX
[59]Almoshaogeh, M. et al. XX
[60]Satiennam, W. et al. X X
[61]Iversen, H.X X XX
[62]Suzuki, K.XXXXXX
Table 4. Artificial neural network models for tickets and accidents in the UAE.
Table 4. Artificial neural network models for tickets and accidents in the UAE.
ANN ModelsModel ParametersAccuracy
Hidden LayersHidden NodesTicketsAccidents
TrainingTestingTrainingTesting
H2O3200, 100, 5089.5%92.1%91.6%93.2%
Keras/Tensorflow350, 25, 292.3%93.7%95.4%95.8%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Abuzaid, H.; Almashhour, R.; Abu-Lebdeh, G. Driving towards Sustainability: A Neural Network-Based Prediction of the Traffic-Related Effects on Road Users in the UAE. Sustainability 2024, 16, 1092. https://doi.org/10.3390/su16031092

AMA Style

Abuzaid H, Almashhour R, Abu-Lebdeh G. Driving towards Sustainability: A Neural Network-Based Prediction of the Traffic-Related Effects on Road Users in the UAE. Sustainability. 2024; 16(3):1092. https://doi.org/10.3390/su16031092

Chicago/Turabian Style

Abuzaid, Haneen, Raghad Almashhour, and Ghassan Abu-Lebdeh. 2024. "Driving towards Sustainability: A Neural Network-Based Prediction of the Traffic-Related Effects on Road Users in the UAE" Sustainability 16, no. 3: 1092. https://doi.org/10.3390/su16031092

APA Style

Abuzaid, H., Almashhour, R., & Abu-Lebdeh, G. (2024). Driving towards Sustainability: A Neural Network-Based Prediction of the Traffic-Related Effects on Road Users in the UAE. Sustainability, 16(3), 1092. https://doi.org/10.3390/su16031092

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop