Two NSW councils are participating in a trial of AI-driven road maintenance technology following a pilot using TfNSW vehicles.
The system hopes to improve road maintenance statewide by capturing unreported road defects.
The expanded trial will see street sweeping vehicles in the regional Griffith and metropolitan Canterbury-Bankstown LGAs using dashboard-mounted cameras linked to a machine-learning program.
Road defect footage captured by the dashboard cams is logged onto a database, where it can be analysed to detect future issues including potholes.
It’s hoped the technology will be also be able to identify damaged signs, faded line markings and cracks in the road and escalate them to maintenance officers based on severity and safety risk.
As it develops over time, the AI software will eventually draw on weather data to predict issues like potholes or cracks before they form.
Fortnightly snapshots
The NSW government, which is funding the trial, says the technology has the potential to deliver a fortnightly snapshot of the condition of the state’s roads, compared to the three-to-five yearly audits currently carried out by councils.
Canterbury Bankstown Mayor Bilal El-Hayek said Council will begin installing cameras on its street sweepers.
“This new technology will help inform the program of works to manage the condition of the road network more efficients,” he said in a statement.
The technology will be rolled out the more NSW councils next year based on the success of the current trial in Griffith and Canterbury which is sent to end by 2024, roads minister John Graham says.
“This platform will help cut costs, accelerate maintenance and prioritise safety,” Mr Graham said in a statement.
“The data to fuel the machine-learning will be gathered from Canterbury-Bankstown and Griffith so that we are sure the software meets the needs of regional and metropolitan councils in NSW.”
Small step for street sweepers, giant leap for road maintenance
The Institute of Public Works Engineering Australia (IPWEA), which is leading the project in partnership with TfNSW, said the use of AI represents a huge leap forward in the way roads are managed and maintained.
“This initiative will significantly reduce the time and cost associated with traditional road audits, freeing up valuable resources for councils across the state. It’s a game-changer for the way road maintenance will be approached,” IPWEA NSW & ACT CEO David Elliott said.
The AI software has been designed and built by Deloitte.
The Australia-first trial is the first time this technology has been attempted to be rolled out on such a large, state-wide scale.
The NSW government’s move towards AI-driven road maintenance is a notable advancement. However, the substantial $2.9 million budget and excessive implementation timeline prompts questions considering existing solutions on the market.
Existing systems could deliver immediate results for NSW ratepayers, potentially covering ~750,000km of scanning and documenting roads today with the same funds. The question arises: why is a government organisation developing and building a new system when the funds could be applied directly on road issues now to achieve immediate tangible results?
A balance between innovation and effective resource allocation is vital.
Governments should embrace the private sector and better utilise taxpayers funds not reinvent the wheel for sake of internal job creation.