InfrastructureNews Australia - Australia's infrastructure news for industry decision-makers
Australia
Austroads unveils pedestrian safety indicators & risk models

Austroads unveils pedestrian safety indicators & risk models

Wed, 18th Nov 2026 (Today)
Sean Mitchell
SEAN MITCHELL Publisher

Austroads has released new pedestrian safety indicators and risk models for road and transport agencies to support more proactive safety decisions across Australia and New Zealand.

The research sets out a baseline approach for measuring and managing pedestrian safety performance using data agencies already hold. It is intended to help authorities assess risk across road networks, compare conditions more consistently, and identify where further investigation or safety works may be needed.

Walking is part of the wider transport system, but agencies often rely heavily on crash outcomes to judge pedestrian safety. While crash data remain important, the report argues they do not provide a complete picture of risk, particularly where pedestrian exposure data are limited.

The framework introduces two linked sets of measures. Key Performance Indicators track the scale, quality, and effectiveness of agency actions. Safety Performance Indicators focus on conditions linked to pedestrian risk, including speed compliance, crossing safety, and vehicle safety features.

Together, the indicators are designed to connect agency actions with the safety conditions experienced by people walking, and then with pedestrian fatalities, serious injuries, and associated social costs.

Michael Nieuwesteeg, Road Safety and Design Program Manager at Austroads, said the work gives agencies a more structured way to assess the issue.

"Understanding pedestrian safety means looking at both the outcomes we are seeing and the conditions that influence risk," he said.

"This work gives agencies a structured way to measure safety performance, understand risk and better target further investigation and action."

Risk assessment

The research also outlines an approach to pedestrian risk assessment and network screening that is considered feasible in Australasian data environments. It examines options for modelling pedestrian exposure and crash risk, including the use of proxy measures where pedestrian count data are scarce.

The modelling approach combines crash prediction models with observed crash history through Empirical Bayes methods, which can improve the identification of higher-risk locations.

Initial modelling work was carried out in Melbourne's central business district. The report also includes a prototype tool for the Melbourne CBD that presents modelling results alongside the new Key Performance Indicators and Safety Performance Indicators.

The prototype is intended to help agencies prioritise high-risk locations and inform decisions on possible treatments. It forms part of Austroads' wider Keeping People Safe When Walking project.

Nieuwesteeg said the prototype shows how agencies could use multiple sources of information together.

"The prototype shows how these different sources of information can work together to provide a more complete picture of pedestrian safety risk," he said.

"Further testing will help refine the approach and assess its potential for broader use across Australia and New Zealand."

Data limits

A central issue identified in the research is the lack of consistent pedestrian exposure data. Without reliable information on how many people walk in particular places, agencies can struggle to compare risk levels between corridors, intersections, or districts.

By proposing a minimum baseline built around currently available information, Austroads aims to give member agencies a method they can use now rather than wait for complete datasets. The approach can then be strengthened over time as data collection and analytical methods improve.

The report recommends that Austroads members trial the indicator frameworks in practice. It also calls for better pedestrian data collection and sharing, validation of the modelling across different network contexts, and further development of the prototype through multi-jurisdiction pilots.

The work reflects a broader shift in road safety policy towards leading indicators and risk-based assessment, rather than relying only on recorded crashes after harm has occurred. In the pedestrian context, that means looking more closely at traffic speed, crossing conditions, and vehicle features before incidents happen.

Initial modelling was undertaken for the Melbourne CBD.