Another van on the hard shoulder
The van was serviced not long ago. It passed. Then it broke down halfway to a customer, with an engineer, a day's jobs and a load of parts on board. The recovery truck, the hire vehicle, the rescheduled appointments and the apology calls all follow. The week before, another vehicle came in for its scheduled service and the mechanic found nothing wrong at all.
Your fleet is maintained exactly as planned, by mileage and by date, and it still lets you down. The schedule treats every van the same, but they are not the same: different drivers, routes, loads and ages.
Why calendar maintenance misses failures
Fixed service intervals are set for an average vehicle in average use. A van doing short urban trips with heavy loads wears differently from one on long motorway runs. A driver who brakes hard wears brakes faster. An older vehicle with a history of cooling problems is more likely to have another.
Most fleets already collect signals that hint at trouble: telematics data on harsh braking, idling and engine hours, diagnostic fault codes, fuel economy changes, driver defect reports and repair records. The problem is that these sit in separate systems and nobody has time to read them all for every vehicle every week.
Driver defect reports are a good example. A driver notes a warning light that comes and goes, or a slight pull on the brakes. The note goes into a form or an app, gets filed, and nothing connects it to the fault code logged the same week or to the fact that this model of van has had the same repair twice before. Each clue is small. Together they are often a clear warning.
What unplanned breakdowns cost
| Consequence | Why it hurts |
|---|---|
| Missed jobs | Customers let down, appointments rebooked, revenue delayed |
| Recovery and hire | Unplanned costs on top of the repair itself |
| Engineer time lost | Skilled staff waiting at the roadside instead of working |
| Emergency repairs | Usually dearer and slower than planned ones |
| Unneeded services | Workshop time spent on vehicles that were fine |
There are also safety concerns. A vehicle that fails at speed or with a full load is a risk to the driver and others, which is a reason to catch problems early beyond cost.
How we predict which vehicles need attention
- We bring together your telematics data (from providers such as Samsara, Webfleet, Geotab or similar) with fault codes, fuel data, driver defect reports and workshop repair history.
- We line up past breakdowns and major repairs with what the data showed in the weeks before them, so the model learns the warning signs in your fleet.
- We train a model that scores each vehicle's risk of an unplanned failure in the coming weeks, and names the likely area, such as brakes, cooling, battery or tyres.
- Because breakdowns are relatively rare, we use methods suited to rare events and are open about how confident the scores are.
- We produce a weekly ranked list for the fleet manager and workshop, with the reasons for each vehicle, so the highest-risk vans are booked in first.
- We track what the workshop found and feed it back, so the model learns which warnings turn out to be real.
Statutory inspections and manufacturer requirements stay as they are. The model adds a layer of targeting on top, it does not replace required checks.
What the fleet manager sees
A short list each week of vehicles that are showing the patterns that came before breakdowns in the past, with the reason and the likely system involved. The workshop plans around it, pulling in the at-risk van before it fails rather than after.
Over time you can also see which routes, loads or driving patterns wear vehicles fastest, which feeds into how work and vehicles are allocated.
Engineers notice the difference first. Fewer days start with a phone call about a replacement van, and fewer customers are rebooked because the vehicle carrying their parts did not arrive.
Is this your situation?
- Vehicles break down between scheduled services.
- Services often find nothing wrong on some vans.
- You have telematics and repair records but nobody analyses them together.
- Breakdowns regularly cause missed customer jobs.
- The fleet varies in age, use and driver behaviour.