An evaluation of the use of connected vehicle data to understand vehicle speeds and accelerations at Active Transport Priority Crossings

Abstract

Queensland Department of Transport and Main Road (TMR) has taken numerous proactive steps to improve safety for pedestrians, cyclists and drivers by installing and monitoring the performance of priority crossings over the past few years. Despite these efforts, some practitioners remain concerned about the potential risks and implications of implementing priority crossings on projects. TMR also has a proactive role in research and feasibility analysis of new and emerging data sources, including crowdsourced data from connected vehicles. TMR has used emerging data to investigate the validity of safety concerns by using data collected from connected vehicles, to examine vehicle speeds and deceleration profiles on approach to Active Transport Priority Crossings.

This study utilises connected vehicle data from multiple vehicle manufacturers from a growing sample of traffic, which is imported and processed by a proprietary company called CompassIoT. CompassIoT started collecting vehicle data in Queensland in 2020, which includes speed and acceleration. Therefore, aggregate speed and braking data were sourced from CompassIoT to analyse driver behavior before and after Active Transport Priority Crossings were installed. Ten priority crossing sites for active transport were selected for preliminary analysis across Queensland.

The preliminary analysis indicates that the CompassIoT data may be suitable to assess whether there was an overall reduction in vehicle speeds after constructing the Active Transport Priority Crossing. The results suggest that the installation of Active Transport Priority Crossing facilities has led to a reduction in average approach speeds at eight out of the ten sites. The reduction in average vehicle speeds ranged from 3.42% to 43.74%. In summary, the installation of Active Transport Priority Crossings resulted in a decrease in the overall approach speed for ten sites. The median speed, average speed, and 85th percentile speed all decreased by 19%, 18%, and 11%, respectively. The results also show a minimal increase in braking rates before and after opening, which is expected, and the values remained well below the conventional "harsh braking" threshold of -0.6 (m/s²).

The conclusions are twofold: the preliminary analysis indicates that Active Transport Priority Crossings do not result in extreme braking events, do result in lower vehicle speeds at the crossings, and, thus, have a positive effect on road safety. A second outcome is that data collected from connected vehicles has shown promise as a practical, non-invasive way of understanding how road users operate in the system, and further use of this type of data is encouraged. However, the limited sample size of the data could potentially affect the results. Therefore, further data collection is needed to understand the behaviour of vehicles and vulnerable road users at Priority Crossings. This can be done effectively and efficiently via the analysis of video footage, using advanced video analytic techniques.