A GIS Approach for Analysis of Traffic Accident Hotspots in Abha and Bisha Cities, Saudi Arabia

Abstract

Road traffic accidents present substantial global challenges, encompassing a wide range of consequences that have implications for the economy, public health, the environment, and society. The present study is focused on the phenomenon of rapid urbanization in Abha and Bisha, two cities located in the Kingdom of Saudi Arabia. These cities have witnessed a substantial growth in urbanization, with a notable increase of 225% over a span of 40 years. The expansion of urban areas has given rise to significant concerns regarding the density of the road infrastructure, which has been further exacerbated by an increase in the volume of vehicles. Consequently, this has led to a notable escalation in traffic-related issues and accidents. Analysis reveals that traffic accidents are concentrated in specific areas, with hotspots primarily located in the western regions of Abha and Bisha, while cold spots are concentrated in the northern areas. Furthermore, a strong positive correlation (r = 0.93) is observed between the number of traffic accidents and road type, with over half of the accidents occurring on highways. Notably, the highways in Abha and Bisha predominantly follow a direction from southwest to northeast at a 71.1° angle. In conclusion, this research offers significant findings regarding the prevalence and determinants of traffic accidents in Abha and Bisha, emphasizing the need for effective traffic management strategies to enhance road safety and mitigate the associated risks.

1. Introduction

Road traffic accidents present substantial global challenges, encompassing a wide range of consequences that have implications for the economy, public health, the environment, and society. The consequences of these accidents, including fatalities and injuries, have far-reaching impacts worldwide. Road traffic accidents impose significant economic costs, healthcare expenditures, environmental degradation, and immense social tragedies on nations worldwide [1,2,3,4].

The car is frequently considered a representation of contemporary society, and unquestionably presents a multitude of benefits. Nevertheless, it is crucial to acknowledge that the mode of transportation not only serves as a vital means of commuting for the population but also holds the unfortunate distinction of being the predominant catalyst for traffic accidents. Road traffic accidents have become a pressing issue due to the widespread adoption of cars as the primary means of transportation [5,6]. Although cars provide a multitude of advantages, their utilization presents substantial hazards. Research findings suggest that automobile collisions are the predominant catalyst for traffic-related occurrences, leading to loss of life, physical harm, and substantial financial burdens. The occurrence of accidents is influenced by various factors, including the intricate nature of contemporary road systems, driver conduct, the quality of road infrastructure, and advancements in vehicle technology. As a result, nations across the globe are confronted with the task of addressing and reducing the economic, health, and environmental ramifications stemming from road traffic accidents [7,8].

The social, economic, and environmental consequences of road traffic accidents are a cause for concern worldwide. The loss of lives, injuries, and associated disabilities have a profound impact on individuals, families, and communities. Moreover, these accidents result in significant economic burdens due to healthcare costs, property damage, and loss of productivity. Environmental degradation caused by car accidents, such as air pollution and carbon emissions, further exacerbates the challenges faced by societies [9,10,11]. Addressing the root causes of traffic accidents and implementing effective road safety measures are crucial to reducing the severity of these impacts and safeguarding public health and well-being. In Saudi Arabia, road traffic accidents remain a pressing concern, with a significant number of incidents attributed to drivers’ distraction and violations of the right of way. The year 2022 witnessed 193,827 accidents caused by drivers’ distraction, while 184,610 accidents occurred due to violations of the right of way. These figures highlight the need for increased awareness and enforcement of road safety measures to address these specific causes of accidents [12,13,14]. By implementing measures to mitigate driver distraction and encourage compliance with traffic regulations, it is feasible to reduce the frequency of such accidents and mitigate the accompanying risks.

The number of deaths resulting from traffic accidents in Saudi Arabia demonstrates the urgent need for effective road safety measures. In 2019, the country recorded 5754 deaths caused by road accidents, which decreased by 26% to 4555 deaths in 2022. This reduction indicates some progress in mitigating the severity of accidents and emphasizes the importance of ongoing efforts to enhance road safety [15,16]. However, further initiatives are required to sustain this positive trend and ensure the protection of lives on the roads. While the number of deaths decreased, the prevalence of injuries resulting from road traffic accidents in Saudi Arabia remains a concern. In 2019, the country recorded 32,910 injuries caused by these accidents, which decreased to 24,446 in 2022 [13]. Although this decline is promising, it is, however, crucial to continue implementing comprehensive measures to further reduce the number of injuries. This includes raising awareness, improving infrastructure, and promoting responsible driving behaviors to ensure the safety of all road users. Road traffic accidents in Saudi Arabia have emerged as a prominent issue, leading to a considerable number of fatalities and injuries. According to statistics provided by the Ministry of Health, the latest available data from 2021 reveals that there were 27,281 reported traffic accidents, leading to 5629 deaths [17,18,19]. These figures emphasize the urgency of addressing road safety issues to reduce the devastating impact of accidents on individuals and society. The Ministry of Health plays a crucial role in monitoring and reporting these statistics, providing valuable insights into the current state of road traffic accidents in the country [20,21].

The issue of road safety in Saudi Arabia has garnered significant attention in recent years, primarily due to the concerning frequency of accidents and resulting fatalities. Researchers have conducted studies to investigate the factors contributing to road traffic accidents in the country. An examination of recent patterns in road traffic accidents and their corresponding mortality rates has yielded noteworthy findings, offering valuable insights into the underlying trends, and contributing factors associated with these occurrences. The obtained results have demonstrated significant outcomes. [22,23]. Additionally, factors influencing road traffic fatalities in Saudi Arabia shed light on key determinants that need to be addressed for effective road safety interventions [24,25,26]. To develop effective strategies for accident prevention and enhance road safety, it is crucial to understand the characteristics and implications of road traffic accidents. Many researchers have conducted different studies regarding the impact of road characteristics on road traffic accidents in Saudi Arabia by focusing on various aspects of road traffic accidents. The authors of [27,28,29] conducted a population-based study to examine the trends and risk factors associated with road traffic injuries in the country. Similarly, [30,31] investigated the severity and outcomes of road traffic injuries among different age groups, providing insights into the variations in injury patterns and their consequences. These studies have made a significant contribution to the current body of knowledge and provided valuable insights for policymakers and stakeholders to effectively implement targeted interventions and policies with the goal of mitigating road traffic accidents and their associated consequences [32,33,34].

In urban road classification, various factors are considered to categorize roads based on their function and characteristics. These factors include traffic volume, speed limits, connectivity, land use, and the presence of pedestrian and cycling facilities. The data of the road network are analyzed to determine the classification of each road. These data include information such as the road length, number of lanes, presence of medians or shoulders, types of intersections, and any special features such as roundabouts or flyovers. By analyzing these data, urban planners can classify roads into different categories such as arterial roads, collector roads, local streets, or highways. Arterial roads typically have high traffic volumes and connect major destinations within a city or region. Collector roads serve as intermediaries between arterial roads and local streets. Local streets are low-volume roads that provide access to residential areas. Additionally, the analysis of road network data helps identify areas where improvements are needed. For example, if a certain area has high congestion or lacks pedestrian infrastructure, it can be identified through this analysis and appropriate measures can be taken to address these issues. Overall, the classification of urban roads and analysis of road network data play a crucial role in urban planning and transportation management to ensure efficient movement of people and goods within cities [35,36].

1.1. Area of Study

This study employs Bisha and Abha as its two selected domains of application. The two cities are located in the Asir region, one of the thirteen administrative divisions of the Kingdom of Saudi Arabia, situated in the southwestern part of the country as shown in Figure 1. Abha city, located in the Asir region, fulfils the dual role of serving as the administrative center and capital of the emirate. The name was bestowed in recognition of the nearby valley. The city of Abha is situated at the geographical coordinates of 12 degrees 18 min latitude and 30 degrees 42 min east longitude. There are several centers located in the Al-Hamr region of the Abha Governorate, as well as in Al-Sha’f and Marba within the same governorate. Additionally, there are centers in Khamis Mushait City of the Khamis Mushait Governorate, and in Tayyib and Al-Souda within the Abha Governorate. Furthermore, there are centers situated to the south in Al-Sha’f and Marba of the Abha Governorate.

The municipality of Abha encompasses an approximate land area of 290.7 square kilometers, which is further subdivided into 44 distinct residential neighborhoods. As of the year 2022 AD, the population of Abha was recorded to be 446,697 individuals. Notable residential areas within the city include Shamsan, Al-Manhal, Al-Andalus, Al-Azizia, Al-Salam, among others. Bisha is geographically located in the northwest region of the Asir area, specifically between latitudes 0′ 19° and 51′ 20° N, and longitudes 50′ 41° and 5′ 43° E. The northern border of the area in question is adjacent to Makkah Al-Mukarramah, while its southern border is shared with the city center of Khaybar in the Khamis Mushait Governorate. The eastern boundary of the region is delineated by the cities of Tathleeth, Jash, and Al-Sabikha, which are located within the Tathleeth Governorate. On the other hand, the western border is demarcated by the Al-Baha region, as well as the governorates of Balqarn and Al-Namas [37].

The urban area of Bisha encompasses a longitudinal extent of 185 kilometers, stretching from its northernmost location to its southernmost location. The width of the object, however, exhibits variations in an east–west orientation. The southern region exhibits the narrowest width, measuring approximately 48 km, whereas the northern region displays the widest width, measuring around 120 km. Bisha, a municipality situated on Wadi Bisha, the most extensive and consequential valley in the Kingdom, covers a land area measuring 659 square kilometers. The population of the specified region in the year 2022 AD was documented as 248,452 individuals. The prominence of the city is derived from its strategic geographical positioning and the existence of the largest dam in the Kingdom, which possesses a substantial storage capacity of 325 million cubic meters. Bisha is widely recognised for its agricultural endeavors, specifically in the realm of date palm cultivation. Furthermore, the urban area derives advantages from its extensively established system of roads, exhibiting noteworthy linkages to adjacent cities including Khamis Mushait, Raniyeh, and Khurmah [37].

Road safety analysis refers to the process of studying and evaluating various factors that contribute to road accidents and injuries. It involves collecting and analyzing data related to road traffic crashes, identifying patterns and trends, and developing strategies to improve road safety. By conducting a comprehensive analysis of these factors, policymakers, traffic engineers, law enforcement agencies, and other stakeholders can develop evidence-based interventions and policies to enhance road safety. This may include implementing engineering improvements, enhancing enforcement measures, promoting education and awareness campaigns, or introducing legislation to address specific issues identified through the analysis.

When analyzing a region or country, it is important to consider both geography and demographics in addition to road transport network characteristics. Geography plays a crucial role in determining the layout and accessibility of road networks. For example, mountainous regions may have winding roads and tunnels, while coastal areas may have bridges and causeways. The presence of rivers, lakes, and other bodies of water can also impact road connectivity, requiring the construction of bridges or ferries. Demographics are equally important as they provide insights into population distribution and transportation needs. Factors such as population density, urbanization levels, and economic activities influence the demand for road infrastructure. Urban areas with high population densities often require extensive road networks to accommodate traffic flow, while rural areas may have more limited road connectivity due to lower population densities. In conclusion, analyzing a region or country requires the consideration of both geography and demographics alongside road transport network characteristics. Understanding these factors helps in assessing transportation needs, identifying infrastructure gaps, and formulating appropriate policies for the efficient movement of goods and people.

Road orientation can have a significant impact on accidents. Some key factors need to be considered, such as traffic flow, which indicates the direction of the flow of traffic, and also, whether it is one way or there is oncoming traffic; the risk of a frontal collision on a one-way street is generally lower than on a two-way street because vehicles are traveling in the same direction. The direction of the road also affects the number and arrangement of lanes. Wider roads with multiple lanes in each direction reduce congestion and give vehicles more room to maneuver, potentially reducing accidents. The orientation of the road affects how an intersection is designed. For example, left-turn lanes or special signal lights can be used to increase safety at multi-lane road intersections. The orientation of a road affects various design features such as curves, slopes, and visibility. Roads with sharp curves or poor visibility due to hills or vegetation can increase the likelihood of an accident if the driver is not properly prepared. It is important to note that while these factors are influenced by the direction of the road, they interact with other variables such as driver behavior, weather conditions, vehicle characteristics, etc., which collectively contribute to accident occurrence.

Literature analysis in the field of accident analysis often tends to focus on one country, which limits the generalizability of findings and may overlook important variations across different regions. To enhance the comprehensiveness of accident analysis, it is valuable to expand the research scope by including examples of spatial and spatiotemporal analyses from other countries. This broader approach allows for a more comprehensive understanding of accident patterns and their underlying causes. In conclusion, expanding literature analysis with examples from other countries enhances our understanding of accidents by considering variations in spatial distribution and temporal dynamics. By incorporating a global perspective, researchers can gain insights into commonalities and differences across regions, leading to more effective strategies for accident prevention worldwide.

1.2. Research Questions

Abha and Bisha are among the fastest-growing cities in the kingdom. Their urbanization level increased by approximately 225% over a 40-year period due to the rapid population growth, causing critical issues in terms of the density of the urban road network. The rise in the phenomenon was concomitant with a surge in the quantity of automobiles traversing roadways, culminating in there being approximately 620,000 vehicles by the conclusion of 2022, in contrast to a mere 203,403 in the year 2000. This escalation has resulted in a corresponding upswing in the predicaments linked to traffic congestion, consequently engendering a proportional augmentation in the frequency of vehicular accidents. The number of transportation-related accidents in these regions was in 2021, which increased to in 2022. Therefore, the research tends to seek to answer the following questions:

• What kind of relationship is there between traffic accidents and the types of roads in the cities of Abha and Bisha?
• Why are there increasing traffic accidents in the most densely populated districts?
• Where are the hotspots of traffic accidents concentrated in Abha and Bisha?

2. Materials and Methods

The surveys conducted can be used to enhance accident reduction actions in several ways. By analyzing the survey data, authorities can identify specific locations or areas where accidents are more likely to occur. This information can help prioritize resources and interventions in these high-risk areas to reduce accidents. In addition, surveys can provide insights into the contributing factors leading to accidents, such as distracted driving, speeding, or poor road conditions. This understanding allows policymakers and law enforcement agencies to develop targeted strategies and campaigns addressing these factors. Overall, by utilizing survey data effectively, decision makers can gain valuable insights into the underlying causes of accidents and develop evidence-based strategies for accident reduction actions that are tailored to address specific issues identified through the surveys.

Data used in this research were collected from various sources such as research studies, surveys, government databases, public records, private companies, and academic institutions. The accessibility of the data depends on its source and any restrictions or permissions associated with it.

This article uses data from the Saudi Arabian General Directorate of Transport (GDT) for 2021 and 2022. The data includes the location, date, time, type, cause, and severity of every traffic accident that occurred in the cities of Abha and Bisha. Data were geocoded using ArcGIS software and Google Maps to assign spatial coordinates to each incident.

Data characteristics are as follows:

• The total number of road traffic accidents in the cities of Abha and Bisha was 11,515 in 2021 and 12,263 in 2022, an increase of 6.1%.
• Most traffic accidents are caused by human factors such as speeding, negligence, and violation of traffic rules, followed by environmental factors such as weather conditions, road surfaces, and lighting.
• The most common type of road accident is collision, followed by rollover, collision with solid objects, and collision with pedestrians.
• Serious traffic accidents cause casualties, and the least serious ones only cause property damage.
• The spatial distribution of traffic accidents shows that accidents are highly concentrated in some areas and hotspots exist.

The health data used in the paper were obtained from the Ministry of Health (MOH) in Saudi Arabia for the years 2021 and 2022. The characteristics of the health data are as follows:

• The total number of fatalities due to traffic accidents in Abha and Bisha cities was 1114 in 2021 and 1218 in 2021, with an increase of 8.5%
• The total number of injuries due to traffic accidents in Abha and Bisha cities was 6212 in 2021 and 6642 in 2022, with an increase of 6.5%
• The total number of disabilities due to traffic accidents in Abha and Bisha cities was 1726 in 2021 and 1848 in 2022, with an increase of 6.6%
• The health data showed that traffic accidents had a significant negative impact on the quality of life and well-being of the residents of Abha and Bisha cities

The availability of the data on which an operation is performed depends on various factors. It can be determined by the source of the data, or whether it is publicly available, or proprietary. Additionally, the data’s accessibility may depend on any legal or privacy restrictions that apply to them. In some cases, the data may be readily available and easily accessible, such as public datasets or open-source databases. These types of data are typically freely available for anyone to use. However, there may be situations where the data are not readily available or require specific permissions or licenses to access. This can be the case with proprietary datasets owned by companies or organizations that restrict access to their data. Furthermore, certain types of sensitive data, such as personal information or classified government data, may have strict regulations and restrictions on their availability and usage. Ultimately, whether the data on which an operation is performed are available depends on the specific circumstances and characteristics of the dataset in question [38,39].

The surveys conducted were carried out to enhance accident reduction actions in several ways. By analyzing the survey data, authorities can identify specific locations or areas where accidents are more likely to occur. This information can help prioritize resources and interventions in these high-risk areas to reduce accidents; also, surveys can provide insights into the contributing factors leading to accidents, such as distracted driving, speeding, or poor road conditions. This understanding allows policymakers and law enforcement agencies to develop targeted strategies and campaigns addressing these factors. Overall, by utilizing survey data effectively, decision makers can gain valuable insights into the underlying causes of accidents and develop evidence-based strategies for accident reduction actions that are tailored to address specific issues identified through the surveys.

The research necessitated the utilization of an analytical descriptive methodology. The nature of the study dictated the use of three approaches: the objective as a main approach, the regional approach to highlight the geographical character of the city, and the historical one to study the development of traffic accidents, with the use of many methods, including: statistical analysis through computer programs, where the researchers relied on the SPSS program 23 to extract the different quantitative correlations between the phenomena under study, and the Excel program to deposit the data of the field study, and cartography to interpret and draw the phenomenon under study.

Geostatistical methods are statistical techniques used for analyzing spatially correlated data. Some commonly used geostatistical methods include: Kriging, which is a spatial interpolation technique that estimates values at unobserved locations based on the values at nearby observed locations. It takes into account the spatial correlation structure of the data and provides estimates with minimum prediction error. Variogram analysis is used to quantify the spatial dependence or correlation between data points at different distances. It helps in understanding the spatial structure of the data and selecting appropriate interpolation models. Co-kriging is an extension of kriging that incorporates auxiliary variables, which are correlated with the target variable, to improve prediction accuracy. It is particularly useful when there are additional variables available that can provide information about the target variable. Indicator kriging is used when dealing with categorical or binary data, where each location is assigned a specific category or class. It estimates the probability of occurrence or membership of a particular category at unobserved locations. These methods are widely used in various fields such as environmental science, geology, ecology, epidemiology, urban planning, and natural resource management to analyze and model spatial data.

The analytical descriptive methodology is a research approach that aims to systematically analyze and describe a particular phenomenon or topic. It involves breaking down the subject into its constituent parts, examining their characteristics, and identifying patterns or relationships between them. This methodology typically relies on quantitative data and statistical analysis to draw conclusions. In addition to the analytical descriptive methodology, three supplementary methodologies can be utilized to enhance the research: Substantive Methodology: This methodology focuses on understanding the underlying principles, theories, or concepts related to the research topic. It involves a comprehensive review of the existing literature, theoretical frameworks, and conceptual models that provide a foundation for the study. The substantive methodology helps researchers develop a deep understanding of the subject matter and ensures that their analysis is grounded in relevant theories. Regional Methodology: If the research topic has regional variations or if it is necessary to understand how different regions influence the phenomenon under investigation, a regional methodology can be employed. This approach involves studying specific geographical areas or regions and analyzing how they impact the research topic. It may include collecting data from different regions, comparing their characteristics or trends, and identifying any regional variations or patterns. Historical Methodology: The historical methodology involves examining past events, developments, or trends related to the research topic. By studying historical data and records, researchers can gain insights into how the phenomenon has evolved over time and identify any long-term patterns or changes. This methodology helps provide context and perspective to the current state of affairs and allows for a more comprehensive understanding of the subject. By combining these supplementary methodologies with the analytical descriptive approach, researchers can ensure a more holistic and comprehensive analysis of their research topic. Each methodology brings unique perspectives and tools that contribute to a deeper understanding of the phenomenon under investigation [40,41,42].

The researcher utilized Geographic Information System (GIS) as a tool for generating maps and conducting spatial analysis. This was accomplished using a software program known as “Arc GIS 10.8”, with the aim of elucidating the spatial arrangement of the phenomenon under investigation. The accomplishment of this step involved the establishment of a geo-database through the utilization of a software application known as “Arc Catalogue”, which consists of a collection of feature classes containing data on various aspects of urban areas. This allows researchers to perform spatial statistical analyses, density analyses, pattern analyses, proximity analyses, and spatial interpolation analyses.

To achieve the research objectives, the study is structured into three primary parts.

-

Distribution of traffic accidents;

-

Analysis of spatial features of traffic accident distribution;

-

Hotspot analysis of traffic accidents.

GIS (Geographic Information System) can be used for the analysis of road accidents in several ways such as spatial analysis: GIS can help in identifying accident-prone areas by analyzing the spatial distribution of accidents. It can overlay accident data with other spatial datasets such as road networks, traffic volumes, land use, and demographic information to identify patterns and hotspots of accidents. GIS can analyze the characteristics of roads and their surroundings to determine factors contributing to accidents. It can assess road geometry, signage, lighting conditions, speed limits, and proximity to intersections or hazardous locations. This analysis helps in identifying high-risk routes or specific road segments that require improvement. GIS provides a visual representation of accident data through maps and charts. This helps in understanding the spatial patterns and trends of accidents over time. Visualization techniques such as heat maps or cluster analysis can highlight areas with a high concentration of accidents. GIS can be used to develop predictive models for road accidents based on historical data. By analyzing various factors, such as weather conditions, time of day, road conditions, and traffic volume, GIS can predict the likelihood of accidents occurring in specific locations or under certain circumstances. Overall, GIS plays a crucial role in analyzing road accidents by providing a spatial perspective that enhances understanding, decision making, and planning for safer transportation systems.

2.1. Traffic Distribution Analysis

The correlation coefficient, denoted by the symbol “r”, is a statistical measure that quantifies the strength and direction of the linear relationship between two variables. It is commonly known as Pearson’s correlation coefficient. The resulting value of r ranges from −1 to +1. A positive value indicates a positive linear relationship (as one variable increases, so does the other), while a negative value indicates an inverse relationship (as one variable increases, the other decreases). A value close to zero suggests no significant linear relationship. The magnitude or absolute value of r indicates the strength of association. The closer r is to +1 or −1, the stronger the relationship. A value near zero suggests a weak or no relationship. It is important to note that correlation does not imply causation; it only measures how closely two variables are related in a linear sense.

Figure 2 presents an analysis depicting the distribution of traffic accidents occurring on the road, where the focus was on highways (22.8), arterial roads (47.7), and finally, local roads (29.5). The analysis shows a strong positive correlation between the number of traffic accidents and road type. According to the statistical equation, r = 0.93 as accidents rise on highways with more than half of the total accidents in the studied cities. The highest number of accidents happened on King Fahd Road: almost more than 25% of total road traffic accidents due to traffic intensity. Arterial roads came in the second order with more than one-third of total accidents in the cities of Abha and Bisha, especially on King Abdullah Road, where accidents happened most frequently with 11.4% of the total number of accidents in the cities of Abha and Bisha. Unlike arterial roads, collector roads have a lower number of accidents with 8.8% of the total, while on the cities of Abha and Bisha’s local roads, a very small number of accidents was reported relative to the other types of roads; 2.7% of the total number of traffic accidents due to the low traffic density on these roads.

The correlation between the types of roads and accidents in urban areas can vary depending on various factors. However, certain characteristics of different road types can contribute to accident rates. Urban roads typically have lower speed limits and are designed to accommodate a mix of vehicles, pedestrians, and cyclists. Accidents on urban roads often occur due to factors such as intersection congestion, pedestrian/cyclist interactions, and distracted driving. Highways are designed for high-speed travel with limited access points; while highways generally have fewer accidents compared to urban roads due to controlled access and higher speeds, certain factors contribute to accidents on highways. The higher speeds on highways increase the severity of accidents when they occur. It is important to note that these correlations may vary based on specific local conditions, such as traffic regulations/enforcement, driver behavior, weather conditions, and other factors. Additionally, effective road design features such as clear signage, proper lane markings, and well-designed intersections/interchanges can help mitigate accident risks on both urban roads and highways.

The correlation between road networks and traffic accidents can be further elucidated through the utilization of the Linear Directional Mean tool. This tool facilitates the assessment of the impact of road network expansion on the occurrence of accidents by identifying the average length of the road network and its overall spatial orientation on the ground. By analyzing this, it is possible to observe that the road network takes several directions according to its type, thus representing the same directions of traffic accidents. The highways where at least half of the total of traffic accidents occur take the direction from the southwest towards the northeast at about a 71.1° angle, and this corresponds to the direction analysis of the distribution of traffic accidents. Unlike highways, on the arterial roads, where one-third of all road traffic accidents occurred, they happened in a southeast–northwest direction at an angle of 142°, while on collector roads and local roads, accidents occurred in a south–north direction at angles of 110° and 118.8°, respectively.

2.2. Analysis of Spatial Features of Traffic Accident Locations

There exist a multitude of spatial analysis tools, from which we employed the following ones:

2.2.1. Tools for Measuring the Spatial Distribution of Accidents

Mean Center—Mean center identifies the central tendency (or the geographic center) of the distribution of traffic accidents. This point is located to the west of the city center of Abha and in the city center of Bisha since it is located in the center of the cities of Abha and Bisha and is close to some districts where road accidents have been on the rise, such as King Faisal Road in Abha and King Abdulaziz Road in Bisha, together constituting 15.5% of the total number of accidents. This is shown in Figure 3.

Central Feature—As a result of applying this tool, the most centrally located site of accidents is identified at the intersection of a collector road and the King Abdullah arterial road that is characterized by its high density of traffic. The central deviation of traffic accidents from the mean center in the west is 1.8 km, and this is explained by the rise in traffic accidents in the center and the west of the cities.

Standard Distance—The standard circle radius of traffic accident distribution in the city of Abha is 8520.1 m, and in Bisha it is 0.4 m. This is located in the central parts of the cities of Abha and Bisha covering large areas such as the Wast El-Balad neighborhood in Abha, and the King Abdullah neighborhood in Bisha. This circle covers 68.1% of the total traffic accidents in the cities of Abha and Bisha.

Directional Distribution (D factor)—The results showed that the distribution of traffic accidents in Abha takes a direction from the southwest to the northeast. The angle of deviation is 69.9° with a standard distance of 4.7 km, towards the x-axis, and 17.1 km towards the y-axis. In Bisha, it takes a direction from the southwest to the northeast. The angle of deviation is 73.1° with a standard distance of 6.1 km towards the x-axis, and 19.1 km towards the y-axis. This explains the nature of the extension of the cities of Abha and Bisha with the surrounding geographical phenomena and altitudes.

2.2.2. Analyzing Patterns

The tools in the Analyzing Patterns toolset show the extent to which traffic accidents take a specific pattern. It includes various tools to perform statistical data analysis as shown in Figure 4A–C.

Nearest Neighbor Analysis (NNA)—This study aims to assess the distributions based on their clustering, randomness, or regularity. The utilization of nearest neighbor analysis can be employed in studies pertaining to the succession of vegetation in sand dunes, with the purpose of testing a given hypothesis. The formula for finding nearest neighbor analysis can be described by the following equations.

$$NNA=\frac{\overline{D}\left(O\right)}{\sqrt[0.5]{\frac{A}{N}}}$$

where, $\overline{D}\left(O\right)$ represents the observed nearest neighbor distance, while $A$ represents the area under study. $N$ represents the total number of points in the area. This statistical method shows that the distribution of traffic accidents in the cities of Abha and Bisha has a clustered pattern 0.233 at a very high confidence level of 99%, which means that there is only a very small probability that the result happened by accident. Sample randomness is less than 1%. The Z-score (also called a standard score) is −49.4, below the critical value. p-value = 0.000.

Spatial Autocorrelation (Moran Index)—We apply the spatial autocorrelation measures to determine spatial distribution patterns of the phenomenon concerned. By studying features and an associated attribute, it evaluates whether the pattern expressed is clustered, dispersed, or random. The result (Figure 4B) shows that the spatial distribution of traffic accidents indicates a clustered phenomenon pattern. The Moran’s I statistic is equal to (+0.56). This was confirmed by the Z-score (11.40) in that it falls outside the critical value region (−2.58, +2.58) at a 99% confidence level, which means that population density is a reason for the higher numbers of traffic accidents.

$$\mathrm{z}=\frac{\mathrm{X}-\mu }{\sigma }$$

$\mathrm{X}$ represents the data value, while $\mu ,\sigma$ represent the mean and standard deviation, respectively. The Moran’s I spatial correlation can be computed using the autocorrelation below.

$$MI=\frac{n}{W_0} \frac{{{\displaystyle \sum }}_i^n{{\displaystyle \sum }}_j^n{w}_{i,j}{z}_i{z}_j}{{{\displaystyle \sum }}_{i=1}^n{z}_i^2}$$

$z_i$ is used to represent the deviation of a given attribute for a particular feature from its mean, while $w_{i,j}$ is the spatial weight between two features $i$ and $j$ from total features $n$. The aggregate weight $W_0$ can be represented by

$$W_0={{\displaystyle \sum }}_j^n{w}_{i,j}$$

Now, we can compute the $Z_I$ score as follows:

$$Z_I=\frac{I-E\left(I\right)}{\sqrt{V\left(I\right)}}$$

$$E\left(I\right)=\frac{-1}{\left(n-1\right)}$$

$$V\left(I\right)=E\left[I^2\right]-E{\left[I\right]}^2$$

Multi-distance Spatial Cluster Analysis (Ripley’s K-function)—This spatial analysis method shows that (Figure 4C) the observed K value is larger than the expected K value, so the distribution pattern of traffic accidents is more clustered than a random distribution. This result corresponds to the result of the nearest neighbor analysis at a very high confidence level because the observed K-function curve falls below the confidence interval (CI). The K-function can be computed as follows:

$$KF=\sqrt{\frac{AoS{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1,j\ne n}^n{k}_{i,j}}{\pi n^2-\pi n}}$$

$n$ represents the total number of features, $AoS$ is the area of study, and $k_{i,j}$ is the weight.

2.3. Hotspot Analysis

The present statistical analysis demonstrates the identification of positive hotspots, characterized by high values, and negative cold spots, characterized by low values, in the distribution of accidents across its geographical range. This analysis yields the critical value, denoted as $Z$ score GI, and the level of significance or probability, denoted as $p$. The statistical indication GI represents the value of the data in $Z$. When the $Z$ value is significantly positive and the probability values of $p$ are considerably low, it suggests the presence of a concentrated occurrence of the “hot spots” phenomenon. Conversely, if the $Z$ value is high and negative, and the probability values of $p$ are high, it indicates a lack of concentration of the “hot spots” phenomenon. When the $Z$ value approaches zero, it indicates that the phenomenon under consideration is not highly concentrated.

For positive z-scores with statistical significance, where the cluster of high values is more intense, the higher the z-score [43,44].

$$\mathrm{z}=\frac{\mathrm{X}-\mu }{\sigma }$$

$\mathrm{X}$ represents the data value, while $\mu , \sigma$ represent the mean and standard deviation, respectively.

The following is the Getis–Ord local (GI) statistic [43,44]:

$$G_i^{*}=\frac{{{\displaystyle \sum }}_{k=1}^n{w}_{k,l}{x}_l-\overline{X} {{\displaystyle \sum }}_{l=1}^n{w}_{k,l} }{\sqrt[S]{\frac{n{{\displaystyle \sum }}_{k=1}^n{w}_{k,l}^2-{\left({{\displaystyle \sum }}_{l=1}^n{w}_{k,l}\right)}^2}{n-1}}}$$

$$\overline{X}=\frac{{{\displaystyle \sum }}_{l=1}^n{x}_l}{n}$$

$$S=\sqrt{n\frac{{{\displaystyle \sum }}_{l=1}^n{x}_k^2}{n}-{\left(\overline{x}\right)}^2}$$

The symbol $w_{k,l}$ denotes the spatial weight between feature $k$ and feature $l$, while the variable $n$ represents the total number of features. The symbol $x_l$ denotes the attribute value associated with feature $l$.

The Mapping Clusters tools perform cluster analysis using ArcMap (Figure 5). The results show traffic accident hot/cold spots in Abha. Traffic accident hotspots are mainly centered in the north of Abha at the three confidence levels presented (90%, 95%, and 99%; +3 at confidence level 99% in two main zones; +2 and +1 at confidence levels 95% and 90%, respectively), while a focus can be noted along the axes of the King Abdulaziz and King Abdullah roads. In Bisha, traffic accidents hotspots are mainly centered in the west of Bisha at the three confidence levels presented (90%, 95% and 99%). (+3 at confidence level 99%). (+2, +1) at confidence levels 95%, 90% respectively, while noting a focus along the axes of King Faisl, and King Abdulaziz roads.

In contrast, it is observed that the areas with a lower incidence of traffic accidents are predominantly located in the northern regions. This finding is supported by the highest negative value recorded, which stands at −3 with a confidence level of 99%. The distribution of cold spots is observed in both the western and eastern regions of the aforementioned zone. However, the cold spots are predominantly concentrated in the southern areas of the cities, with a confidence level of 90% indicating a temperature decrease of −1. This may be because these areas are largely uninhabited. It remains to be pointed out that there is no statistical evidence of the distribution of traffic accidents in various parts of the downtown, western, and eastern parts of the cities, where the value of G is zero, which indicates that traffic accidents occur by chance and cannot be traced to specific factors.

3. Conclusions

A robust positive correlation was observed between the incidence of traffic accidents and road type, indicating a higher prevalence of accidents on highways. This highlighted the necessity for focused interventions and enhancements in road infrastructure to effectively tackle the distinct challenges linked to highways and mitigate the frequency of accidents. Conversely, the directional patterns exhibited by the road networks within the study area offer valuable insights that can inform strategies for traffic management and enhance road safety measures. Highways primarily exhibit a directional pattern from the southwest to the northeast, whereas arterial, collector, and local roads display varying orientations. Gaining an understanding of these patterns can be beneficial in facilitating efficient traffic planning and the successful execution of road safety strategies.

A road transport network should have a well-connected system of roads that allows for easy movement of people and goods between different locations. This includes highways, expressways, and local roads. The network should provide easy access to various destinations such as residential areas, commercial centers, industrial zones, and public facilities such as schools, hospitals, and airports. A good road transport network should be designed to minimize travel time and congestion. This can be achieved through proper planning of road layouts, traffic management systems, and efficient intersection designs. Safety is a crucial characteristic of a road transport network. It should include measures such as well-maintained roads, clear signage, proper lighting, pedestrian crossings, and speed limits to ensure the safety of all road users.

Furthermore, the spatial analysis reveals the presence of localized accident hotspots within the urban areas of Abha and Bisha. The hotspots are predominantly situated within geographical areas, underscoring the necessity for focused interventions and heightened enforcement measures in said regions to effectively mitigate the likelihood of accidents. In conclusion, the relationship between traffic accidents and road characteristics, such as road type, directional patterns of road networks, and the spatial distribution of accident hotspots, offers a foundation for the formulation of evidence-based approaches and interventions aimed at improving road safety and mitigating the frequency of accidents in these regions. The implementation of specific interventions, such as the enhancement of infrastructure, the intensification of awareness campaigns, and the enforcement of traffic regulations, and through the use of modern applications in monitoring traffic accidents such as drones and others, has the potential to yield a substantial decrease in road accidents and foster the development of safer transport systems.