Can AI live up to its true potential for fleets?
Artificial Intelligence (AI) is making inroads into fleet operation in areas such as route planning, driver behaviour monitoring, fuel and asset tracking, load optimisation and inventory management, to name a few. The point is to make fleets more efficient, cost-effective and safer. The potential of AI in fleet management is vast, but so far, it’s delivering mixed results. We check in on some of the latest AI projects to discover why.
Firstly, FleetCheck has introduced AI into its software products with new technology designed to make driver licence checking faster, easier and more accurate. It’s an enhancement to the company’s Licence Assured, FleetCheck’s driving licence-checking product, which allows users to upload an image of a driving licence - even a picture taken on a smartphone – whereby its details are automatically populated into the system.
Peter Golding, managing director of FleetCheck, said: “In using AI for the first time within our fleet software, we decided to take a proof of concept approach, automating a single task that we know takes time for our users and making it faster, easier and more accurate.
“Now that we know the core idea works, we will be able to use the same principle across a wide range of other processes within our product range, especially when it comes to sorting documentation and collecting data.”
Substantial effort required to make AI commercially viable
It may seem like a small step, but Golding warned that the coding required had taken substantial effort.
“What we have learnt about AI in creating this new tool is that employing the technology is not necessarily easy. Getting this to a stage where the data is reliably and accurately extracted has taken some time. It is not something that we expect to see be rapidly adopted across the fleet software sector for this reason.”
The potential of AI in predictive maintenance
Current artificial intelligence (AI) technology could potentially predict almost half of all fleet breakdowns, create new cost management opportunities and improve operational efficiencies, says epyx, founders of the 1link Service Network platform – but a range of barriers remain.
The company reports that two AI-based projects used by fleets to manage service, maintenance and repair (SMR) for more than four million company cars, vans and trucks – and others carried out by its customers over the last five years have shown considerable capabilities of AI but also revealed potential hurdles.
Andy Partridge, technology delivery director at epyx and r2c, said: “In this context, AI is about using machine learning to examine data so that models can be built that make inferences and identify patterns that are useful in meeting objectives. That makes it potentially very useful when it comes to fleet SMR and improving operational efficiencies for customers.
“Looking ahead, in an ideal scenario, AI would learn to identify points of failure on fleet vehicles before they happened, meaning that we could steer the driver towards the nearest workshop that could help them before the breakdown occurred. This kind of pre-emptive action would obviously have huge value for fleets.
AI’s key stumbling blocks for fleets
At least one of the AI projects undertaken by a large UK fleet operator demonstrated at the London Amazon Web Services Summit in June identified almost 50% of vehicle failures before they happened in an exercise that looked at data retrospectively. But hurdles remain in three key areas, including the quality of data throughout the process, obtaining time-sensitive data, and building processes that can handle large streams of information and events - often in real-time.
“The data used for machine learning must be of a very high quality”, added Partridge, “otherwise, the inferences it makes are of limited use and, worse, can introduce biases. Initially, we found that much of the data stored about fleet SMR was not sufficiently robust, but this is gradually improving. In the latest trials, we’ve largely overcome this issue.
Partridge also highlights the need to be able to consume connected vehicle data, much of it time-sensitive, to maximise the effectiveness of AI. At the same time, the difficulty of getting hold of this information quickly and cost-effectively is a current barrier.
Lastly, the processing power needed for fleet AI to work in real-time is considerable and traditional server arrangements are not up to the task. This means transferring fleet data to locations where more advanced computing is available in the cloud, which also adds to the cost.
However, all these issues can be overcome, and suppliers are optimistic about the potential to bring products to market that make the capabilities of AI accessible to fleets in the future.
Image: Shutterstock-1781926982 - sdecoret
