As part of National Rail Safety Week, Transport for NSW is sharing the learnings from a trial that applied big data and predictive analytics to help identify safety risks at rail level crossings.
The Enhance Road Safety Planning at Level Crossings using Predictive Analytics Program trial harnessed more than a dozen data sets including connected vehicle information, traffic, weather and driver behaviour to evaluate the risk at more than 1,300 public road level crossings in NSW.
The project received funding through the Australian Government’s Regional Australia Level Crossing Safety Program Research and Innovation grant, and earlier this year received an Australian Road Safety Award.
Using a combination of machine learning models, the desktop assessment used advanced analytics to identify crossings with higher relative risk and improved predictions of where serious incidents may occur.
One of the new aspects of Transport’s model is that for the first time, driver behaviour data from ‘connected cars’- those that are equipped with Internet of Things (IoT) technology – has been incorporated in a level crossing analysis.
This driver behaviour data includes acceleration or braking on approach to a level crossing and information on whether the vehicle was on a short or long trip.
The predictive model provides additional information about risk at a particular level crossing by drawing from data points that are available without visiting the crossing site.
Transport identifies level crossing upgrade locations through a comprehensive ranking approach that uses the nationally recognised Australian Level Crossing Assessment Model (ALCAM) alongside a review of NSW safety incident data and takes into consideration the local knowledge of rail and road infrastructure managers and other relevant stakeholders.
The additional insights provided by the new predictive analysis tool will complement Transport’s existing methodology, further enhancing the robustness and accuracy of the prioritisation process.
Transport’s Executive Director, Strategy Simon Hunter said while level crossing crashes are infrequent, they can have serious impacts on road users, train drivers, rail operations and surrounding communities.
“Safety at level crossings is a top priority for Transport for NSW and we are committed to keeping everyone on our network safe as we focus on zero trauma,” Mr Hunter said.
“Every crash at a level crossing is one too many, that’s why we’re harnessing technology and data with the aim of improving risk assessments. It is another potential tool to improve level crossing safety, alongside infrastructure investments, education about safe behaviour, and innovative research and trials.”
For the predictive analytics trial, Transport contracted GHD to provide technical capability and expertise around the use of connected vehicle data.
GHD’s Market Leader for Transport Philip Morgan said improving safety outcomes on our roads and railways starts with better understanding risk.
“We’re proud to have partnered with Transport for NSW on a project that combines their leadership in road and rail safety with our experience in connected vehicle data and predictive analytics to better understand how people interact with the transport network.
“This work supports more informed safety investment decisions across the network.”
More information about level crossing safety can be found at Level crossing safety | transportnsw.info
