A groundbreaking automated job hazard analysis (JHA) tool developed by Aussie researchers is set to revolutionise safety in the construction industry, where 60,000 workers lose their lives globally each year.
The team behind this development, from the University of South Australia (UniSA), has created a cutting-edge “knowledge graph” designed to identify potential hazards more efficiently.
This system automates a process that has traditionally relied on time-intensive, manual identification methods prone to human error.
“Traditionally, potential job hazards rely heavily on safety personnel identifying risks and control measures, a process that is prone to inefficiencies and human error,” explains Dr. Sonali Pandithawatta, UniSA Construction Management lecturer and lead researcher on the project.
“Our research addresses the critical need to automate job hazard analysis (JHA).”
With this innovation, workplace accidents could be significantly reduced, offering hope in one of the world’s most high-risk sectors.
Innovation needed in high-risk industry
The construction industry accounts for nearly 20 per cent of occupational deaths worldwide, underscoring the urgent need for improved safety measures.
In Australia alone, approximately 30 construction workers die annually, with an alarming 12,500 serious injuries reported each year, according to SafeWork Australia.
The situation worsened in 2023, with 45 construction worker fatalities—significantly above the five-year average of 33. Leading causes of these deaths include being struck by moving objects, followed by falls, trips, and slips.
And the financial and human costs are substantial. Over the 2022–23 financial year, the construction sector saw more than 16,600 serious workers’ compensation claims. The median compensation payout reached A$18,479, while workers lost a median of 8.5 weeks of work due to injuries.
Recognising these challenges, the UniSA team gathered and analyzed data from over 100 job hazard analysis documents, integrating expert insights and key factors like weather, job steps, hazards, and preventive measures to impressive results.
“This model demonstrated exceptional accuracy—over 90%—and is capable of analyzing primary and secondary hazards, weather conditions, workplace proximity, and atmospheric risks in real-time,” says co-author Professor Rameez Rameezdeen, a specialist in construction and project management at UniSA.
Building a technological safety net
But UniSA’s latest invention isn’t the first cab of the rank to take a stab at saving tradie lives. Several other safety tools have begun to come out of the woodwork as the industry begins to lean heavily on technological advancements to bring down the death toll.
One of the most promising contenders is wearable sensors that monitor workers’ body posture via wristbands or smartphones. These sensors detect unsafe lifting practices, excessive bending, or prolonged awkward positions—common contributors to ergonomic injuries.
Other impressive safety toys include augmented reality wearables that help workers safely practice techniques away from on-site dangers.
However, despite their potential, barriers such as cost, privacy concerns, and industry resistance to investing in these technologies remain a key sticking point to widespread adoption.
An AI-led future
Still, the research team are confident an automated safety assessment would be a no-brainer for construction sites looking to crack down on safety risks, with the tool even having applications in other high-risk industries to improve workplace safety across the board.
The program is still in its infancy however, with Professor Rameezdeen claiming the next steps will be further integration of advanced machine learning techniques into the tool and evaluating the role human and managerial factors have in the workplace.
“We are proud to contribute to the global effort in making construction sites safer,” he added.
But with a bit more time in the oven, this technology could mark a pivotal shift in how safety risks are managed, making the construction industry—and potentially others—safer for workers worldwide.