Systems-thinking-based road safety research: the way forward

While significant progress has been made in reducing the fatalities and injuries due to road crashes in the last three decades, we continue to face unacceptable levels of road trauma that result in problems that have so far proved resistant to interventions, or where the impact of successful interventions is beginning to plateau. This suggests that past approaches to reducing road trauma may have reached their limit in terms of their effectiveness. Road safety researchers and practitioners both face the difficult choice of whether to continue the earlier successful approaches, resulting in small, incremental improvements, or adopt a paradigm shift to address the emergent risks in the system. The systems-thinking-based approaches are accepted and have been in discussion for some time (Salmon & Lenné, 2015). The systems approach, with emphasis on the structural analysis of complex systems, on the systemic error rather than on the individual’s error, is the way forward. An analysis of road user risk behaviour in the context of infrastructure systems, organizational structures and regulations, and risk reduction by addressing infrastructure modifications, alteration in regulations, geometric design of roads, and so forth, are the important building blocks for addressing road safety.

The present issue presents studies from different parts of the world, explaining the traffic risk behaviour of cyclists (Jianrong Liu et al.), pedestrians (Shbeeb, Lina) and taxi drivers (Dai, Xuezhen et al.), related to infrastructure design or organizational regulations. The application of systems-approach is relevant in non-traffic accidents also, as illustrated by the study on mine accidents in Ghana. Other research papers in this issue present results of the impact of roadway design, network design and the understanding of motorized two-wheeler behaviour as a function of traffic mix and road features.

Jianrong Liu et al. from China have analyzed the cycling violation behaviour of seniors, which lead to traffic-related fatalities and injuries in the seven central districts of Wuhan city. The authors used face-to-face interviews and questionnaires, which resulted in 793 responses, of which 477 valid responses were isolated for the present study. The study found that perceived behavioural control, which includes actions such as ‘not difficult to run red light, go on the opposite direction, or in restricted areas’, had no impact on the behavioural intention, which included compliance with the regulation.

Xuezhen Dai et al. from China have made a submission on the evidence gathered on certain predictors of the crash risk level of 2615 taxi drivers in a cross-sectional survey of four typical cities in China. Job stress, health status and risky driving behaviour were the predictors under consideration, and the taxi drivers selected to take part in the survey were Cruise taxi drivers with driving licenses of C1 level(driver is competent to drive all types of compact cars) or above. It was found that the health status and risky driving behaviour were significant predictors of their potential crash risk. The study also found that dissatisfaction with income is an important factor causing risky behaviour.

Lina Shbeeb from Jordan has made a presentation on clustering and pedestrian crash prediction modelling. Their study examines pedestrian crashes at the micro-level, the district level and the spatial distribution of pedestrians in Amman, in order to identify pedestrian crash clusters to predict casualties, in correlation to various factors in the built environment. A reduction in pedestrian casualities could be brought about by improving the residential area, the road network, particularly the primary roads. The current road network is not pedestrian friendly.

Xiaodong Feng et al. from Japan and China present a methodology to construct the process of injury from Japanese consumer product narrative injury data. This is a study on trying to better understand the injury process by proposing an ontology-based method, with the addition of semantic relationships between factors based on the accident causation models. The authors propose use of dynamic ontology methods in future, to include an accurate spatial relationship between the host and the vector.

Theophilus Joe-Asare et al. from Ghana have made a presentation on the relationships among causal factors influencing mine accidents using structural equation modelling (SEM). The early notion that accidents in mines befell those who were accident-prone, was discredited by later theories that attributed the root cause to poor management policies, among other factors. The present study maintains that accidents in mines occur because of interactions between factors at different levels of the sociotechnical system, largely in the operational process category. The organizational factors significantly influence workplace and individual conditions.

Harish Kumar Saini et al. from India have analyzed the overtaking and filtering manoeuvres of motorcyclists and cars in heterogeneous city traffic, using video-based data collection techniques and advanced trajectory data analysis. The authors have introduced the concept of a ‘pore size ratio’ (PSR) to study how road users execute filtering manoeuvres which lead to differences in the kinematic characteristics between motorcyclists and car drivers. Motorized two-wheeler riders demonstrated risk taking behaviour by maintaining a smaller lateral gap. These findings can improve the accuracy of future research on micro-simulation models and intelligent transportation systems.

Md Al Adib Sarkar et al. from USA present a study on injury severity of single-vehicle large truck crashes while accounting for heterogeneity in the injury severity of these crashes. Despite the high fatality rate of single vehicle large truck crashes, only one previous study has accounted for heterogeneity in the injury severity of these crashes; such crashes were examined from 2007 to 2016 in Florida. Specific policy recommendations include improved lighting conditions, speed reduction and targeted intervention at road junctions.

Sharaf Alkhedar et al. from Kuwait have made a presentation on the impact of COVID-19 on traffic congestion and safety skills. Three specific measures were introduced on specific days in the post-COVID period. The total traffic demand on all the seven lanes in the intersection under study decreased after the implementation of the three measures. It was found that the three measures taken together have reduced the background traffic by 40.7% which is more than one-third of the baseline demand. Reduced traffic volume was associated with a decrease in traffic violations and an increase in the safety skills.

Laxman Singh Bisht et al. from India present a matched case-control approach to identify the risk factors of fatal pedestrian crashes on a rural, six-lane highway in Punjab in India. In multi-lane highways running through rural areas, the number of lanes and the time of the pedestrian-vehicle fatal crash are significant risk factors. For safety evaluation, cross-sectional methods require extensive data, which is scant and under-reported in India; hence, the authors have resorted to an epidemiological technique such as the use of the case-control method. Pedestrian safety would be well served if a wide median is provided in multi-lane highways, and speed control measures are introduced in specifically those sections that run through human settlements and built-up areas.

Morteza Ahmadpur et al. from Turkey have undertaken a hot spot analysis and evaluation of influencing factors on regional road crash safety and severity indices in Iran. The study results confirmed that socioeconomic and transportation related variables significantly influence regional road safety indices…Results suggest that fatality on rural roads, the risk of fatality on all types of roads, and the crash severity level are higher in underdeveloped regions of Iran. The authors conclude that Iran has significant problems related to intercity road safety, and that more crash-preventive measures should be instituted.

V. A. Bharat Kumar Anna et al. from India have submitted a paper on the determinants of speed variability on the horizontal curves of two-lane undivided rural highways passing through mountainous terrain. The objective of the present study was to collect the continuous speed profile data in the real world, investigate the speed variability, and identify the factors leading to the maximum speed difference on horizontal curves of mountain roads, including the differential impact of the gradient and turning direction. The authors maintain that the models developed in this study highlight the geometric parameters that cause higher speed variability on the horizontal curves in mountainous terrain. The modelled parameters can be considered in designing the alignment of roads with minimum speed variability.

The results of these studies confirm the impact of hard systems (infrastructure, physical design features) and soft systems (regulatory mechanisms, processes and organizational structure) on travel choices of road users. Researchers are invited to explore the ‘systems approach’ to road safety in future studies, taking forward the results achieved so far. The results of these studies once again emphasize the need to move away from ‘educating road users’ as the most important intervention to improve safety, and strongly recommend the need to change the systems: infrastructure design, regulatory environment and network design.

Geetam Tiwari
Transportation Research and Injury Prevention Centre, Indian Institute of Technology Delhi, New Delhi, India
[email protected]

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.