Artificial intelligence could be used to help detect and predict major road safety and maintenance issues and to inform predictive road maintenance efforts across New South Wales if a trial of the technology is successful.
Designed and built in NSW by Deloitte, Asset AI aims to leverage real-time data and AI software to improve maintenance outcomes across the state’s network of state and local roads.
As part of trials, cameras and sensors will be attached to street sweeping vehicles which operate on local roads across the two local government areas of Griffith and Canterbury-Bankstown.
These will be linked to a machine learning program.
As the vehicles move along the road, the cameras and sensors will capture imagery and a ‘feel’ of road conditions.
Once the data has been captured, the AI technology will assess the footage and will log any defects or important safety concerns which are detected into a database.
Issues which are logged may include damaged signage, faded line markings, potholes and rutting.
Once logged, the defects will be escalated to council maintenance planners based on the severity and safety of risk that is involved.
As it develops over time, the AI software will draw upon data relating to weather, assets and pavements and will ‘learn’ to predict issues such as cracks and potholes before these form.
This will enable defects to be addressed before they cause damage to road users and vehicles.
It will also assist local councils to prioritise repairs based upon the level of risk which they pose to vehicles and motorists.
Further flow on benefits will include a lower level of reliance upon costly and time-consuming road audits along with the extension of the lifespan of asphalt and bitumen through timely intervention.
In a statement, the NSW Government said the system will help local councils to implement a proactive and preventative approach toward asset maintenance.
As things stand, traditional road maintenance actions are performed in response to audits of road condition. These are typically carried out every three to five years.
Going forward, the AI software has the potential to deliver a snapshot of the condition of the state and local road network each fortnight.
The trial is being carried out within the Griffith and Canterbury-Bankstown municipal areas. It will help to ensure that the platform meets the needs of both regional NSW and metropolitan areas.
This follows an earlier ‘proof of concept’ trial which saw sensors place on public transport busses in Sydney and Transport for New South Wales vehicles in both Sydney and regional NSW.
If successful, the system could be rolled out to more councils in the latter part of next year.
It is envisaged that the system could be deployed across a range of vehicle types, including busses, garbage trucks and street sweeping machines.
Aside from Griffith and Canterbury-Bankstown, other councils who have expressed interest in being involved with the project’s development include Georges River, Blayney, Central Coast, Liverpool, Wingecarribee, Warren Shire, Liverpool Plains, Tamworth, Wollongong, Murray River, and Shoalhaven.
NSW Minister for Roads John Graham welcomed the commencement of the new stage of the trials.
“Keeping roads safe and in good condition are some of the biggest challenges for local councils. This platform will help cut costs, accelerate maintenance and prioritise safety,” Graham said.
“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.
“One of the most exciting aspects is that the system will begin to draw on weather data and learn to predict issues like potholes or cracks before they form and help councils prioritise repairs based on potential future risk.
“This will keep NSW at the forefront of technology-led solutions to what are some of the most essential services for all communities. No one wants to see potholes on the roads and this could be part of seeing fewer of them in future.”
Enjoying Sourceable articles? Subscribe for Free and receive daily updates of all articles which are published on our site
Want to grow your sales, reach more new clients and expand your client base across Australia’s design and construction sector?
Advertise on Sourceable and have your business seen by the thousands of architects, engineers, builders/construction contractors, subcontractors/trade contractors, property developers and building industry suppliers who read our stories across the civil, commercial and residential construction sector