Leveraging AI in Mitigating Road Accidents and Alleviating Traffic Congestion: A South African Perspective ^†

Abstract

This study aims to achieve its objectives in two folds: firstly, by examining the current challenges in South Africa’s traffic ecosystem, which lead to excessive road accidents and traffic congestion, and finally, by proposing an AI-driven model to be incorporated in South Africa’s Traffic Management System to enhance road safety and reduce traffic congestion. As a literature-based study, secondary data was collected and critically analyzed to comprehend the key factors that precipitate road accidents and yield sluggish traffic congestion in South Africa’s big cities, and thereafter, we developed a suitable AI model (ITTE) that would assist in mitigating road accidents and alleviate traffic congestion. This study found that leveraging AI in the transportation ecosystem would identify issues like infrastructural weaknesses, unsafe driving, and environmental risks. This would allow for proactive or automated corrective actions such as adjusting traffic signals or urgently alerting stakeholders such as drivers, pedestrians, and authorities with real-time updates, fostering a culture of being well-informed and responsive.

1. Introduction

In emerging economies like South Africa, intensive urbanization from rural parts of the country to big cities has placed enormous strain on existing road infrastructure, resulting in constant traffic congestion and excessively high rates of road accidents. Fatal car crashes recorded year in and year out remain alarmingly high. According to the WHO, 90% of total global accidents occur in low–middle-income countries, while people in low-income countries continue to face the highest risk of death per population [1]. The annual State of Road Safety report for 2024 states that fatal crashes increased by 1.56% from 10,180 in 2023 to 10,339 in 2024 [2]. The report further stipulates that those fatal crashes led to human fatalities recorded with an increase by 2.43% from 11,883 in 2023 to a staggering 12,172 people in 2024 [2].

Reliable transportation systems are crucial to the country’s economic growth and to the social well-being of inhabitants in its cities. In order to achieve these, Rao Nampalli (2021) contended that from an urban transportation standpoint, an immediate consideration, on the one hand, is monitoring traffic conditions as well as demand cycles, while on the other hand, it is inducing flow modifications that benefit the traffic network and mitigate congestion [3] (pp. 2–14). South Africa’s big cities are confronted by immense congestion-related delays as well as accident-related fatalities, which negatively impact the country’s economic growth and also the social well-being of its citizens. According to the Road Traffic Management Corporation’s (RTMC) calculated and adjusted cost of crashes annually, which are based, respectively, on the annual Consumer Price Index (CPI) and the number of fatal crashes and fatalities per year, the estimated adjusted cost of crashes for 2024 is ZAR 217.53 billion, up from ZAR 205.13 billion recorded as an estimate in 2023 [2].

The main objective of this study is twofold. Firstly, it is to critically examine the systemic challenges confronting South Africa’s traffic ecosystem, particularly the underlying causes of frequent road accidents and traffic congestion in cities. Secondly, this study aims to propose a forward-looking solution by introducing an AI-driven model, namely the Integrated Technological Transportation Ecosystem (ITTE), designed to enhance road safety and optimize traffic flow. The recommended model leverages AI’s capabilities in pattern recognition, predictive analytics, and real-time data processing to support smarter traffic management systems.

This study adopted a qualitative, literature-based research approach to investigate the root causes of traffic congestion and road accidents in South Africa’s big cities, and to propose a suitable AI-driven solution. Secondary data was collected from a variety of credible sources such as peer-reviewed journal articles, government reports, as well as World Health Organization (WHO) and South African Department of Transport statistics. A thematic analysis was then conducted to extract key factors contributing to persistent road accidents and traffic congestion. Based on the findings from the literature review and thematic analysis, a conceptual artificial intelligence model, referred to as the Integrated Technological Transportation Ecosystem (ITTE), was developed. The model is designed to enhance decision-making within South Africa’s Traffic Management System by incorporating real-time traffic monitoring and predictive analytics, AI-powered incident detection and early warning systems, data-driven traffic signal optimization, and integration with mobile alert systems for road users.

2. Analysis of Key Challenges and Factors Contributing to Road Accidents

Road accidents and sluggish traffic congestion in South Africa’s cities are propelled by a combination of factors such as aging infrastructure, negligent driving behavior, a dearth of enforcement on the part of the authorities, and disconnected traffic management systems. Unmaintained roads, unclear or inadequate road signage, and a lack of lighting sometimes due to regular blackouts create a hazardous road atmosphere and unnecessary delays, while common acts like excessive speeding, drinking and driving, and overall non-compliance with traffic laws compound the problem. Ineffective enforcement and outdated monitoring systems preclude timely interventions, and the absence of integrated data analytics impedes the ability to urgently execute proactive safety measures. These challenges highlight the urgent need for an intelligent, integrated traffic management approach to mitigate accidents, alleviate traffic congestion, and improve road safety.

The RTMC reported that there has been a trend over the years whereby human factors such as speeding, intoxicated drivers or pedestrians, fatigue or drivers falling asleep, etc., have significantly contributed to road fatalities. Figure 1 and Figure 2 depict this trend, comparing 2023 and 2024.

To address these issues, cities globally are increasingly turning to artificial intelligence (AI) for innovative solutions that promise to improve road transportation efficiency and reduce traffic congestion [4] (pp. 1–10). It was asserted in [4] that artificial intelligence has the power to transform urban transportation systems into safer, more efficient, and environmentally sustainable networks.

3. Economic Cost and Social Trepidation

Persistent traffic congestion, increasing fatal accidents, and enormous economic costs are among the key challenges that traffic management systems face. This crisis not only undermines national economic growth but also causes deep social distress, as thousands of lives are lost annually. The contributing factors may include but are not limited to aging infrastructure, poor law enforcement, and unsafe driving behaviors, which collectively hinder socio-economic development. Despite the urgent need for reform, progress is impeded by technological, financial, and policy challenges.

3.1. Rising Economic Cost of Road Traffic Crashes

Transport has a strong impact on the quality of life of citizens, and new technologies applied to different means of transport and to traffic management are considered key in lowering the likelihood of accidents, reducing the environmental impact of freight and passenger transport, preventing congestion, and increasing sustainability and efficiency [5] (pp. 67–87). Transportation systems in developing countries are suffering from many problems, such as traffic congestion, lack of reliable and safe public transportation, road accidents, and difficulties for non-motorized transport [6] (pp. 17–37). Over ZAR 200 billion has been estimated to be the economic cost for 2024.

Table 1. Estimated cost of crashes.

South Africa: Estimated Cost of Crashes (CoC) and Deaths per 100,000 Population
Calender Year	Days	Fatalities	Fatalities per Day	Fatal Crashes	CoC R * Billion	Change R * Billion	% Change	Est Unit Cost Death R * Million	Est Unit Cost Fatal Crash R * Million
2015	365	12,944	35	10,613	142.95	-	-	R3.92	R5.44
2016	365	14,071	39	11,676	167.63	+24.68	17.27%	R4.17	R5.79
2017	365	14,050	38	11,437	172.72	+5.09	3.04%	R4.39	R6.09
2018	365	12,921	35	10,564	166.72	−6.00	3.48%	R4.59	R6.37
2019	365	12,503	34	10,381	170.5	+3.88	2.33%	R4.78	R6.63
2020	365	9969	27	8405	142.59	−28.01	16.42%	R4.93	R6.85
2021	365	12,545	34	10,611	188.31	+45.72	32.07%	R5.16	R7.16
2022	365	12,436	34	10,466	198.79	+10.48	5.57%	R5.52	R7.66
2023	365	11,883	33	10,180	205.13	+6.34	3.19%	R5.86	R8.13
2024	365	12,172	33	10,339	217.53	+12.39	6.04%	R6.12	R8.49
Est. Ave Annual CoC (10 yrs) = 177.296 bn				Est. Total CoC (10yrs) = 1772.96 bn				Estimated % of GDP (2024)= 2.94%

Source: State of Road Safety Report: 1 January 2024–31 December 2024.

depicts the RTMC’s annual estimated cost of crashes from 2015 to 2024.

3.2. Social Consequences and Public Concern

The rise in fatal crashes from 10,180 in 2023 to 10,339 in 2024 has contributed to a consequential surge in human fatalities, now standing at over 12,000 (almost half of these being pedestrians) annually [2]. These statistics reflect a sad reality for urban inhabitants who are confronted with fears over road safety. Families suffer the trauma of losing loved ones, and communities are left grappling with the socio-economic aftershocks of such tragedies. The WHO (2023) reported that road injury is the number one cause of death globally among people aged between 5 and 29 years.

Table 2. Global leading causes of death (all ages vs. 5–29 years).

RANK	ALL AGES	AGES 5–29 YEARS
1	Ischaemic heart disease	Road injury
2	Stroke	Tuberculosis
3	Chronic obstructive pulmonary disease	Diarrhoeal diseases
4	Lower respiratory infections	Interpersonal violence
5	Neonatal conditions	Self-harm
6	Trachea, bronchus, lung cancers	HIV/AIDS
7	Alzheimer’s disease and other dementias	Lower respiratory infections
8	Diarrhoeal diseases	Maternal Conditions
9	Diabetes mellitus	Drowning
10	Kidney diseases	Cirrhosis of the liver
11	Cirrhosis of the liver	Malaria
12	Road injury	Meningitis

Source: WHO, Global Status Report in Road Safety, 2023.

compares global causes of death for all ages vs. ages 5–29 [7]. Road injury ranked number 12 in all ages category, yet number in age 5-29 years category.

The RTMC (2024) declared that pedestrian safety remains the most significant road safety challenge in South Africa, with an average of 44% of all fatalities being pedestrians [2].

Table 3. Pedestrian fatalities per province.

YEAR	EC	FS	GP	KZN	LP	MP	NC	NW	WC	RSA
2022	625	214	1372	1191	435	390	155	283	687	5352
2023	581	211	1412	1200	453	331	139	276	757	5360
2024	629	231	1335	1258	456	371	131	316	727	5454

Source: State of Road Safety Report: 1 January 2024–31 December 2024.

portrays the number of pedestrian killings per South African province.

4. Proposed Integrated Technological Transportation Ecosystem Model

The intelligent transportation system (ITS) can provide safety, effectiveness, and sustainability for critical complications with vehicle traffic on a broad scale [8] (pp. 27–77). The lack of sufficient-quality data is a major obstacle in most African countries. The African Union (AU) is working to ensure that Africa participates actively in the emerging global AI ethics discussions and was set to launch its regional AI strategy in 2024 [9] (pp. 41–60). The proposed Integrated Technological Transportation Ecosystem (ITTE) model, illustrated in Figure 3 below, offers a robust framework for alleviating traffic congestion and improving road safety by leveraging an interconnected transport management ecosystem. This involves public transport that is Wi-Fi-enabled, private cars with sensors, and smart infrastructure, i.e., buildings and road tech, smart cameras, smartphones, smart traffic lights, and sensor networks. These components are intertwined through wireless technologies such as Wi-Fi, enabling real-time data flow for traffic congestion alleviation, fatality mitigation, and road safety optimization. The main objective of the proposed system is to minimize traffic congestion and thereby maximize road safety through predictive analytics.

4.1. Mitigating Road Accidents

The proposed model enables real-time data sharing and analysis, which can detect and predict perilous road conditions. Sensors and smart surveillance cameras can monitor driver behavior, road conditions, and vehicle speeds, and then feed such information to central Transport Management System (TMS) to issue immediate alerts or interventions such as automated braking or rerouting to prevent possible crashes. Alshamsi (2021) noted that the identified risky driver behaviors in South Africa include a failure to keep an adequate distance, failure to maintain recommended speeds, and reckless driving [9]. Leveraging the proposed AI intervention could greatly mitigate resulting road accidents and promote road safety, as the ecosystem promulgates and fosters a proactive and preventive approach.

4.2. Alleviating Traffic Congestion

This proposed model enables well-coordinated and proactive traffic flow management. The Transport Management System integrates data from smart traffic lights, sensors, and vehicle tracking systems to optimize signal timings and reroute traffic away from congested areas in real time. This will significantly improve the delays often experienced on South African roads due to traffic congestion and consequentially lessen the adverse impact on the economy caused by delays. Public transportation can be prioritized and synchronized with traffic signals to ensure a timely, efficient, and most importantly, reliable transport service, which would definitely encourage more people to opt for shared mobility instead of private vehicles, significantly reducing the amount of traffic congestion. Additionally, smart infrastructure such as adaptive road systems and digital signage can inform commuters of optimal routes or changes in traffic patterns, thus reducing bottlenecks and minimizing idle time on the road. Modern-day technology such as surveillance cameras, Global Positioning Systems (GPS), Edge AI, and IoT can be beneficial, serving to develop and deploy deep learning algorithms that detect and predict accidents [10] (pp. 1–24).

5. Conclusions

Adopting a centralized Traffic Management System (TMS) to monitor all components of the transportation ecosystem would assist in identifying infrastructural weaknesses, unsafe driving behaviors, as well as environmental risks in real time. The system can then take proactive or corrective actions either automatically, such as adjusting traffic signal timing, or by issuing targeted alerts to key stakeholders, such as drivers and authorities. Continuous data feedback loops significantly enhance situational awareness for both transport operators and commuters. Furthermore, the integration of smart infrastructure and mobile technologies allows for greater public participation through safety reporting and real-time updates. This fosters a culture of informed, responsible, and cautious road use, while also strengthening systemic responsiveness to road safety and mobility challenges.

This study aligns with the overall themes of the SMT conference, mainly in its focus on urban mobility challenges and the application of sustainable, cutting-edge technologies to mitigate these issues in the South African context. In that way, this study bridges critical knowledge gaps in South African urban mobility while offering an innovative, tech-enabled solution that aligns with the SMT conference’s mission to advance smart, inclusive, and sustainable transportation systems.