Product Walkthrough

See how Rooutiq fits into your fleet's existing operating stack.

Connect your existing telematics and operational systems, evaluate the decision layer, and see how predictive maintenance moves from detection to execution. Below: a guided walkthrough, then a full worked example.

rooutiq — product walkthrough

Step 01

Connect the telematics you already run

One authorization against your existing provider account. No hardware to ship, no devices to install, no change to how your drivers operate.

Typical setup: under 10 minutes

Geotab

24 vehicles discovered

Samsara

8 vehicles discovered

Historical fault sync

Importing 90 days…

32 vehicles connected · 0 devices installed

Enterprise Evaluation

Request an enterprise demo.

Tell us what your fleet runs today and what you're trying to change. The demo is built around your stack — your telematics, your maintenance system, your procurement workflow — not a generic tour.

  • Where Rooutiq sits in your current architecture
  • The decision and control model, end to end
  • Assisted vs. automated procurement, honestly framed

No financial details requested. We reply within one business day.

Worked example · 25-vehicle mixed fleet

Component-level failure prediction from telematics data

Rooutiq reads the sensor feed your fleet already produces and identifies which component is degrading, how long it has left, and the cost difference between a planned repair and a roadside failure. Below is a full worked example — including the signals behind each call.

This fleet is illustrative, not a customer account. The failure modes, sensor behavior, and repair costs are representative of what the model produces in production.

Open predictions

Soonest failure in 3 days · $37,470 avoidable

Why we flagged this

  • Differential pressure across the DPF rose 41% over 9 days
  • Regeneration cycles now firing every 62 miles, down from 210
  • Exhaust backpressure trending above the 2019 Cascadia baseline
Planned repair
$3,200
If it fails on the road
$14,400
Avoided
$11,200

The rest of the fleet

Health scored nightly from telematics

VehicleTypeStatusHealthLast faultNext service
Ford E-450 #101Box truckAt risk52P0401 EGR flowJul 1
Ford E-450 #102Box truckMonitor68P0171 lean mixtureJul 8
Isuzu NPR #103Box truckHealthy91Aug 15
Isuzu NPR #104Box truckAt risk44P0087 fuel pressureJun 28
Isuzu NPR #105Box truckHealthy88Sep 1
Freightliner MT45 #106Box truckMonitor71P2BAD NOx sensorJul 12
Freightliner MT45 #107Box truckHealthy94Oct 1
Ford E-450 #108Box truckMonitor63P0128 coolant tempJul 5
Isuzu NPR #109Box truckHealthy85Aug 20
Freightliner MT45 #110Box truckHealthy89Sep 10
Freightliner Cascadia #201SemiAt risk38SPN 3251 DPF pressureJun 27
Freightliner Cascadia #202SemiMonitor67SPN 110 coolant tempJul 10
Kenworth T680 #203SemiHealthy92Aug 5
Kenworth T680 #204SemiAt risk41SPN 641 turbo actuatorJun 29
Freightliner Cascadia #205SemiHealthy87Sep 15
Kenworth T680 #206SemiMonitor72SPN 3563 intake manifoldJul 18
Freightliner Cascadia #207SemiHealthy83Aug 22
Kenworth T680 #208SemiHealthy90Sep 30
Ford Transit #301Service vanHealthy96Oct 1
Ford Transit #302Service vanMonitor69P0420 catalyst efficiencyJul 14
Mercedes Sprinter #303Service vanHealthy97Nov 1
Mercedes Sprinter #304Service vanHealthy91Sep 5
Ford Transit #305Service vanHealthy84Aug 18
Mercedes Sprinter #306Service vanHealthy93Oct 12
Ford Transit #307Service vanMonitor61P0301 misfire cyl 1Jul 9

How a prediction is produced

Every prediction above is traceable to specific sensor behavior. Nothing is inferred from vehicle age or mileage schedules alone.

  1. 01

    Signal ingestion

    We pull engine, aftertreatment, and drivetrain parameters from your existing telematics feed — typically 30 to 90 second resolution. No additional hardware is installed on the vehicle.

  2. 02

    Deviation from baseline

    Each parameter is compared against that vehicle's own operating history and against the same engine family across the fleet, so a reading is judged in context rather than against a fixed threshold.

  3. 03

    Failure matching

    Deviation patterns are matched to labeled repair outcomes — verified by the mechanic who did the work — to identify which component is degrading and how quickly.

  4. 04

    Confidence and lead time

    Confidence is the model's calibrated hit rate for that failure type, not a marketing number. Predictions below the review threshold are routed to a human instead of being auto-scheduled.

Model accuracy is tracked per failure type and published to every account — including where it is weak. Predictions that turn out wrong are logged as false positives and fed back into the next training cycle.

Run this against your own fleet

Connect Geotab, Samsara, or Azuga and your first predictions arrive within 24 hours. No hardware to install, no credit card to start.

Or book a 15-minute call