Asset AI®

NSW Government is using artificial intelligence to revolutionise road maintenance and operations. This program will make it easier to keep roads safe for our communities.

Last updated: 24 June 2024

Listen

A bus is driving along a road, with a pot hole in front of it.

On this page

Collaborators

Institute of Public Works Engineering Australia NSW & ACT
Canterbury-Bankstown Council

Suppliers

Cisco
Deloitte
Vaisala
Retina Visions

Councils involved

Canterbury-Bankstown Council
Blayney Shire Council
Central Coast Council
Coonamble Shire Council
Georges River Council
Griffith City Council
Lachlan Shire Council
Liverpool City Council
Liverpool Plains Shire Council
Murray River Council
Shoalhaven City Council
Sutherland Shire Council
Tamworth Regional Council
Warren Shire Council
Wingecarribee Shire Council
Wollongong City Council

Introduction

Asset AI® is helping the NSW Government and local councils to make roads safer by using near-real-time data. This information is used to streamline road maintenance tasks.

Sensors and cameras are fixed to council vehicles to collect information about the roads. This data is then analysed using artificial intelligence.

The insights will help find and fix problems on the roads faster, making them safer for drivers and pedestrians.

Challenge

Keeping our roads safe and maintaining our assets is an expensive undertaking. A better understanding of the assets and early intervention can reduce these costs. It will also improve road safety outcomes.

Collecting data on asset conditions is expensive and labour-intensive. The frequency of collection (which can be every 1 to 5 years) means that available data is limited and often outdated. Near-real-time data will enable more informed decisions about maintenance requirements and future needs.

Current systems and models only look at the asset condition without considering the risk to the community. By classifying and prioritising defects, we can improve the management of, and risks associated with, deteriorating assets and defects.

Solution

The initiative uses sensors, cameras, or a combination of both, fitted to council vehicles. As these vehicles undertake their usual routes, the technology captures footage. This is then de-identified before being processed. It is analysed by an artificial intelligence application to log any road defects into a database.

The results will be fed into a digital platform (Asset AI®). This enables councils to detect potholes and cracks in the road before they happen and affect road users.

Expected outcomes

Smart Places outcome areas:

collaboration and connection
health and wellbeing
skills, jobs and development.

Insights generated will help identify and track road defects earlier. Government can then fix defects sooner.

This trial is also expected to identify triggers for road defects. This will improve scheduling of preventative maintenance and will make our roads safer for the community.

Once proven, the trial could be rolled out to other councils across NSW. It will improve overall state-wide road safety. Plans to rollout Asset AI® to additional Local Government areas will be confirmed in late 2024, following the trial.

Early intervention can reduce the shortfall in the maintenance funding for roads. The project targets an efficiency gain of 10% in managing public assets. This is measured as a reduction of maintenance backlog after 12 months of use.

10^%

Efficiency gain targeted for managing public assets measured

as a reduction of maintenance backlog after 12 months of use.