Using artificial intelligence to automate sewer and storm water inspections is transforming the future of water management.
VAPAR, a sewer and storm water pipe inspection software that uses machine learning, is fast tracking water management by removing the need to manually review hours of CCTV footage from pipes to identify faults.
The technology shortcuts the usually costly process, which can take weeks, by instead uploading footage to the platform for automatic analysis, giving councils an estimated 30 per cent cost savings.
With over 200,000 kilometres of pipelines in Australia alone and $450 million spent each year maintaining sewerage networks, early detection is critical to prevent costly repairs, according to Amanda Siquera, a civil engineer and co-founder of VAPAR.
“It’s really important these pipes are inspected regularly and maintained because if they’re not there could be catastrophic events like flooding. For sewers the risk is a little higher because you’ve got public health concerns,” she says.
By allowing councils to more quickly detect risks at early stages, the technology is able to cut the time taken for the inspections from two weeks to around two minutes, she says.
“We’ve taken out the expense and time, and automated that process. We have really streamlined the way local government and water utilities are able to fix repairs in their pipes,” she says.
The software has been developed to identify different categories of faults – from cracking to breaking – by matching CCTV footage with images uploaded into the system, according to Ms Siquera.
In terms of detecting faults, the system has an 80 per cent accuracy rate, at the prototype stage, just below the 85 per cent rate of human accuracy.
Ms Siquera says that with a few tweaks the software will soon surpass human accuracy.
“Local councils spend on average three or four hours a week reviewing this footage but it’s more than just the time taken, we actually integrate into databases and store footage, so it saves the time and costs associated with peripheral activities,” she says.
The start-up now has plans to expand Australia-wide to help councils optimise the management of their stormwater and sewer assets.
Ms Siquera says she and co-founder Michelle Aguilar, a mechatronic engineer, also plan to expand internationally, having already tested the software in India under a program funded by the NSW Government.
Comment below to have your say on this story.
If you have a news story or tip-off, get in touch at firstname.lastname@example.org.
Sign up to the Government News newsletter.