For years, Fleet Managers have relied on telematics to identify speeding, harsh braking and other risky driving behaviours. The challenge has never been collecting data. The challenge has been knowing which data matters and how to act on it before an incident occurs.
According to Chris Martin, Senior Manager, Solutions Engineering – APAC at Geotab, artificial intelligence is changing that equation by shifting fleet safety from reactive reporting to predictive risk management.
Speaking with Fleet News Group, Martin explained that AI is enabling fleets to identify collision risks before they happen by analysing patterns across millions of data points that would be impossible for a human to detect.
“Fleet Managers have known for a long time that the data is key to decision making,” said Martin.
“We’re collecting vast quantities of it, and we’ve had really good analytics tools for many years, trying to summarise, aggregate, present insights in that data, but AI is shifting the needle in terms of how much data we can process, how quickly, and the types of patterns that we can find in data.”
Moving Beyond Traditional Safety Metrics
Traditional telematics programs often rely on predetermined thresholds such as speeding events, harsh braking or excessive acceleration. While these metrics remain valuable, Martin believes they only tell part of the story.
“If you look at a predictive safety perspective, we’ve been able to train a model based on real-world collision data to go back and look at historical driving patterns and say what are the characteristics or the driving behaviours that consistently precede a collision,” he said.
“These are the sort of patterns of data that a human can’t find.”
Rather than focusing on a single event, AI can assess combinations of variables such as vehicle type, location, operating conditions, climate, route characteristics and long-term driver behaviour.
“We might say a G-force value of point eight Gs might contribute to or might exhibit a good example of a harsh driving event, and we’re fairly confident that correlates to collisions,” Martin explained.
“But one event alone doesn’t necessarily give you a strong enough indication.”
The result is a more sophisticated view of risk that moves beyond simple event reporting.
Predicting the Probability of a Collision
One of the most significant developments is the ability to assign a predictive collision risk score to drivers and fleets.
Martin said Geotab’s AI models have been trained using real-world collision data and then tested against actual fleet operations.
“We’ve not only built the model but then tested it on real-world fleets, and we’re finding a very high accuracy of being able to identify a predictive collision risk based on those safety behaviours,” he said.
That allows Fleet Managers to receive insights that were previously unavailable.
“We might say, you’ve got an eight per cent risk of a collision in the next 100,000 kilometres. That’s a pretty useful insight.”
Rather than waiting for an incident to occur, fleets can proactively focus coaching, training and intervention efforts on drivers or operating groups with elevated risk profiles.
Benchmarking Against Similar Fleets
Predictive analytics becomes even more valuable when combined with benchmarking.
Martin noted that context is critical when evaluating safety performance. Comparing a metropolitan courier fleet with a long-haul transport operation would provide little value.
“There’s no point comparing a light ute with a heavy vehicle, or a long-haul transport company with a last-mile delivery operation,” he said.
By analysing similar fleets operating in comparable environments, AI can provide meaningful context around safety performance.
“We can give you a benchmark of whether your fleet is actually performing well from a safety perspective. Is eight per cent good or is eight per cent bad?”
For Fleet Managers, that creates a clearer understanding of where improvement opportunities exist and whether current safety initiatives are delivering results.
Reducing the Workload for Fleet Managers
The benefits extend beyond safety outcomes.
Many Fleet Managers are responsible for hundreds or thousands of vehicles while also managing procurement, maintenance, compliance and operational demands. AI can reduce the burden of analysing large volumes of data and configuring complex reports.
“This all happens with AI driving it without having to run the reports, tweak the reports, work out your own risk thresholds,” Martin said. “It significantly reduces the workload for the fleet manager to give them those insights.”
The introduction of tools such as Geotab Ace is further simplifying access to information by allowing users to ask questions in natural language.
“You can just say, ‘Hey Geotab, who are my top 10 most risky drivers currently?’ and you can get that response straight away,” said Martin.
Fleet Managers no longer need to be experts in report building or data analysis to gain valuable operational insights.
The Future of Fleet Safety
As AI models continue to evolve and learn from larger datasets, Martin expects predictive analytics to become increasingly important across fleet operations.
While predictive safety is one of the most immediate applications, similar approaches are already being applied to maintenance, vehicle uptime and electric vehicle performance.
“There’s always more innovation to do,” Martin said.
“As AI models improve and the models learn more about real-world behaviours, we’re going to see constant improvements and new features brought into the Geotab platform.”
For fleet operators, the shift represents a fundamental change in how telematics data is used. Instead of analysing what happened yesterday, AI is helping fleet managers understand what is likely to happen tomorrow — and giving them an opportunity to act before it does.
- MINI Countryman Unleashes its Untamed Side
MINI will add a new Untamed Edition to its Australian Countryman range in early 2027, giving fleet and novated lease buyers a distinctive design option across petrol and fully electric models. The edition will be available on the Countryman S ALL4 and all-electric Countryman SE ALL4, pairing all-wheel drive with exclusive exterior finishes, interior materials - Chery Stockman Ute Adds a New TwistChery has revealed further specifications for its upcoming Stockman ute, combining diesel power with plug-in hybrid technology in a new option for Australian fleet buyers. Customer deliveries are scheduled to begin on 1 February 2027. The indicative specifications include all-wheel drive, 350kW of combined power, 800Nm of torque and a braked towing capacity of 3,495kg.
- Petrol Meets Electric in BMW’s New 3 SeriesBMW’s next-generation 3 Series will give Australian buyers a choice of combustion and fully electric powertrains within the same model family, with four variants scheduled to arrive from the first quarter of 2027. The Australian launch range will comprise the BMW 320, BMW M350 xDrive, BMW i3 40 xDrive and BMW i3 50 xDrive. For
- Kia Seltos Makes the Hybrid SwitchKia’s all-new Seltos will arrive in Australian showrooms this month with a hybrid-only line-up, a larger footprint and the brand’s first electric all-wheel-drive system for a hybrid model in Australia. The new generation brings a significant change to the Seltos range, with every variant featuring a 1.6-litre hybrid powertrain paired with a six-speed dual-clutch transmission.
- LDV Puts More Power Behind T60 UteLDV has updated its T60 double-cab ute with a more powerful diesel engine, a redesigned cabin and a simplified two-model range that gives fleet buyers a choice between load-carrying capability and additional comfort. The MY27 T60 adopts the same 2.5-litre single-turbo four-cylinder diesel engine as the larger LDV Terron 9, producing 163kW and 520Nm. It
