Case Studies: Understanding Pedestrian-Vehicle Crashes in Melbourne

Outline of the research

Traffic accidents involving pedestrians and vehicles are a significant concern for cities, because of the very large disparity in the respective vulnerabilities of pedestrians versus car-drivers; pedestrians are four times more likely to be injured in a traffic accident than other users, and 23 times more likely to be killed. In Melbourne, there are over 1000 vehicle-pedestrian crashes every year, many of which lead to serious injury, or loss of life. Consequently, understanding the spatial and temporal patterns of pedestrian crashes is central to being able to prevent these events, and increase the safety of the most vulnerable road users. Alireza Toran pour, Dr. Sarah Moridpour, and Dr. Richard Tay from RMIT University, and Professor Abbas Rajabifard from the University of Melbourne used a range of spatial analytical and computational techniques, to examine the spatial and temporal characteristics of pedestrian-vehicle crashes in Melbourne [1]. They also investigated the influence of the location of potential pedestrian destinations on patterns of vehicle-pedestrian crashes, in order to try and identify priority areas for interventions to prevent these kinds of events; this is one of the first instances of explicit incorporation of the influence of destinations on the modelling of pedestrian-vehicle crashes.

Spatial pattern of vehicle-pedestrian crash in different time periods, Figure from the Research Authors

How AURIN was used

The researchers explicitly investigated the influence of destination type and diversity on crash severity. In order to do this, they required a dataset that would represent the diversity of different kinds of destinations that pedestrians may walk to, which could increase the likelihood of an accident involving a car. AURIN was able to provide a high quality, up-to-date, authoritative point location dataset of Features of Interest (FOI), sourced from the Public Sector Mapping Agency (PSMA) Australia, available for free use by the research sector. The FOI dataset contains information about the location of health services, education providers, playgrounds, accommodation and other potential attractors of pedestrian activity. The FOI dataset is one of several PSMA Australia datasets available for use through the AURIN portal, along with the PSMA Australia street network and the GNAF geocoder.

Findings and Impacts of the Research

The researchers were able to show that a range of parameters were important in determining the severity of pedestrian crashes in the Melbourne metropolitan region, including time of crash, pedestrian age, and the location of points of interest/pedestrian attractors. The most important variable was determined to be the time of the crash: 25% of crashes occurred between 7pm and 6am, but 60% of these crashes were fatal, or involved a serious injury. The likely reason for this increase in severity at night was determined by the researchers to be related to increased speeds. By contrast, for day-time crashes, the most important factor was pedestrian age, with increased severity for over 65s during the day. For this group of pedestrians, the type of locations around the crash (pedestrian attractors) also had a substantial impact on the severity of crashes. The results suggest that road safety strategies such as reducing travel speed close to pedestrian attractors (e.g. government offices, playgrounds and health centres) could potentially improve road safety and decrease the severity of vehicle-pedestrian crashes. The findings of this research could lead to considerable reductions in the mortality and morbidity associated with pedestrian-vehicle crashes, and my substantially reduce the strain on health resources.

[1] Pour, A. T., Moridpour, S., Rajabifard, A., & Tay, R. (2017). Spatial and temporal distribution of pedestrian crashes in Melbourne metropolitan area. Road & Transport Research: A Journal of Australian and New Zealand Research and Practice, 26(1), 4.