Asset AI

Trialling artificial intelligence to revolutionise road asset maintenance and operations.

Asset AI® Sign-In

Asset AI® is a new digital tool being trialled by New South Wales councils, enabling early detection of road defects, empowering councils to streamline and accelerate road maintenance, improving the safety of road users.

How does it work?

  1. Road asset data is collected by dash-mounted cameras on council vehicles. Sensor data is also collected from private vehicles to determine the road condition.
  2. AI technology automatically identifies and classifies road asset defects then sends this information to the Asset AI® digital platform, (for example, detecting potholes and cracks before they develop, identifying street litter, faded line marking or damaged signs and guideposts).
  3. The Asset AI® platform assesses the defects and determines a recommended response time.
  4. The councils can access their own council data for scheduling of maintenance work, thereby reducing the risk to road users.

Through harnessing AI technology, this project aims to revolutionise road asset maintenance and operations for councils.

Key Benefits

  • Safety: enhancement of road safety outcomes, providing opportunity for streamlining and accelerating of road maintenance
  • Cost: reduction of operational and maintenance costs through proactively identifying damaged assets in near real-time and enabling prioritised intervention
  • Sustainability: data generated from vehicles already using the roads rather than introducing something new
  • Efficiency: gaining a comprehensive understanding of council road networks and providing opportunity for early intervention
  • Empowerment: Helping councils realise the benefits of smart technologies for their communities

Project Status

The Asset AI® program is an Australian-first, currently being rolled out with one metropolitan council (Canterbury Bankstown Council) and one regional council (Griffith City Council), to validate Asset AI’s adaptability across both urban and rural settings. Following this pilot, it is proposed to expand the platform across more NSW councils, improving road safety state-wide.

Additional councils will be onboarded in 2024 and expressions of interest can be submitted through the contact details listed below.

Project Partners

Asset AI® is led by Transport for NSW (TfNSW) in partnership with IPWEA Roads and Transport Directorate and Canterbury-Bankstown Council.

Asset AI® is a $2.9 million dollar program funded through the NSW Government Smart Places Acceleration Program.

Transport for NSW logo

IPWEA (NSW) logo

City of Canterbury Bankstown

Contact Us

We value your feedback and are available to answer your questions. Whether you're looking for more information about Asset AI® or are a council interested in getting involved, we're here to help.

Transport for NSW 
For general enquiries, feedback, or information about Asset AI®:
Email: assetai.tfnsw@transport.nsw.gov.au

Institute of Public Works Engineering Australasia NSW & ACT (IPWEA NSW & ACT)
For councils looking to participate in Asset AI®:
Email: admin@assetai.com.au 
Website: ipweansw.org

 

Public Notice on Data Collection

The Asset AI®® trial is currently collecting road asset data via council vehicles. The collection of road asset data will: 

  • Test the capability of technology in its identification of road asset defects 
  • Provide an extensive view of asset condition in a time-efficient manner 
  • Provide an assessment of needs for planned asset renewal 
  • Support the development of a platform that prioritises defect maintenance based on community risk 

Video detection technology is mounted on council vehicles to collect road defect data from: 

  • July 2023 – Canterbury Bankstown Council region 
  • April 2023 - Griffith City Council region 

Footage is collected and processed by an AI application and subsequently only de-identified data is used by Transport for NSW to explore the capability and constraints of using AI detection technology in road maintenance – no personal data is stored.