Modelling of road traffic fatalities in India

A comparative study of machine learning classifiers for injury severity prediction of crashes involving three-wheeled motorized rickshaw

Motorcycles and motorcyclists have a variety of attributes that have been found to be a potential contributor to the high liability of vulnerable road users (VRUs). Vulnerable Road Users (VRUs) that include pedestrians, bicyclists, cycle-rickshaw occupants, and motorcyclists constitute by far the highest share of road traffic accidents in developing countries. Motorized three-wheeled Rickshaws (3W-MR) is a popular public transport mode in almost all Pakistani cities and is used primarily for short trips to carry passengers and small-scale goods movement. Despite being an important mode of public transport in the developing world, little work has been done to understand the factors affecting the injury severity of three-wheeled motorized vehicles. Crash injury severity prediction is a promising research target in traffic safety. Traditional statistical models have underlying assumptions and predefined associations, which can yield misleading results if flouted. Machine learning(ML) is an emerging non-parametric method that can effectively capture the non-linear effects of both continuous and discrete variables without prior assumptions and achieve better prediction accuracy. This research analyzed injury severity of three-wheeled motorized rickshaws (3W-MR) using various machine learning-based identification algorithms, i.e., Decision jungle (DJ), Random Forest (RF), and Decision Tree (DT). Three years of crash data (from 2017 to 2019) was collected from Provincial Emergency Response Service RESCUE 1122 for Rawalpindi city, Pakistan. A total of 2,743 3W-MR crashes were reported during the study period that resulted in 258 fatalities. The predictive performance of proposed ML models was assessed using several evaluation metrics such as overall accuracy, macro-average precision, macro-average recall, and geometric means of individual class accuracies. Results revealed that DJ with an overall accuracy of 83.7 % outperformed the DT and RF-based on a stratified 10-fold cross-validation approach. Finally, Spearman correlation analysis showed that factors such as the lighting condition, crashes involving young drivers (aged 20-30 years), facilities with high-speed limits (over 60 mph), weekday, off-peak, and shiny weather conditions were more likely to worsen injury severity of 3W-MR crashes. The outcomes of this study could provide necessary and essential guidance to road safety agencies, particularly in the study area, for proactive implementation of appropriate countermeasures to curb road safety issues pertaining to three-wheeled motorized vehicles.

An exploratory qualitative study of the prevention of road traffic collisions and neurotrauma in India: perspectives from key informants in an Indian industrial city (Visakhapatnam)

BACKGROUND

Despite current preventative strategies, road traffic collisions (RTCs) and resultant neurotrauma remain a major problem in India. This study seeks to explore local perspectives in the context within which RTCs take place and identify potential suggestions for improving the current status.

METHODS

Ten semi-structured interviews were carried out with purposively selected key informants from the city of Visakhapatnam, Andhra Pradesh. Participants were from one of the following categories: commissioning stakeholders; service providers; community or local patient group/advocacy group representatives. Transcripts from these interviews were analysed qualitatively using the Framework Method.

RESULTS

Participants felt RTCs are a serious problem in India and a leading cause of neurotrauma. Major risk factors identified related to user behaviour such as speeding and not using personal safety equipment, and the user state, namely drink driving and underage driving. Other reported risk factors included poor infrastructure, moving obstacles on the road such as other vehicles, pedestrians and animals, overloaded vehicles and substandard safety equipment. Participants discussed how RTCs affect not only the health of the casualty, but are also a burden to the healthcare system, families, and the national economy. Although there are ongoing preventative strategies being carried out by both the government and the community, challenges to successful prevention emerged from the interviews which included resource deficiencies, inconsistent implementation, lack of appropriate action, poor governance, lack of knowledge and the mindset of the community and entities involved in prevention. Recommendations were given on how prevention of RTCs and neurotrauma might be improved, addressing the areas of education and awareness, research, the pre-hospital and trauma systems, enforcement and legislation, and road engineering, in addition to building collaborations and changing mindsets.

CONCLUSIONS

RTCs remain a major problem in India and a significant cause of neurotrauma. Addressing the identified gaps and shortfalls in current approaches and reinforcing collective responsibility towards road safety would be the way forward in improving prevention and reducing the burden.

Using satellite imagery to estimate heavy vehicle volume for ecological injury analysis in India

A major limitation of road injury research in low- and-middle income countries is the lack of consistent data across the settings, such as traffic counts, to measure traffic risk. This study presents a novel method in which traffic volume of heavy vehicles - trucks and buses - is estimated by identifying these vehicles from satellite imagery of Google Earth. For Rajasthan state in India, a total of ∼44,000 such vehicles were manually identified and geo-located on national highways (NHs), with no distinction made between trucks and buses. To estimate population living in proximity to NHs, defined as those living within 1 km buffer of NH, we geocoded ∼45,000 villages and ∼300 cities using Google Maps Geocoding Application Programming Interface (API). We fitted a spatio-temporal Bayesian regression model with the number of road deaths at the district level as the outcome variable. We found a strong Pearson correlation of 0.84 (p < 0.001) between Google Earth estimates of heavy vehicles and freight vehicle counts reported by a national-level study for different road sections. The regression results show that the volume of heavy vehicles and rural population in proximity to highways are positively associated with fatality risk in the districts. These effects have been estimated after controlling for other modes of travel.

Distance-decay functions of travel to work trips in India

In 2011, for the first time, Census in India reported travel distance and mode of travel for the workers. The distance reported is in the form of aggregate counts for each mode of travel in 7 distance bins (0–1 km, 2–5 km, 6–10 km, 11–20 km, 21–30 km, 31–50 km, and >50 km). In this data article, methods are described to model categorical count data as distance-decay functions using continuous probability distributions. The distributions have been developed for 8 categories of modes—walk, cycle, motorised two-wheelers, car, tempo/auto rickshaw/taxi, bus, train, and all modes combined, for the 33 mainland states of India and all states combined. Distance for walk is modelled using exponential distribution, and for all the other modes using lognormal or Weibull distribution. For estimating parameters of the distributions, chi-square minimization has been used in a spreadsheet program. The data presented includes parameters of the 272 (34 × 8) probability distributions as well as descriptive statistics of these distributions.