Fault Codes

A diagnostic trouble code is the vehicle saying something is wrong, in a standard format, before a warning light appears or a driver mentions it.

What a code is

A standard identifier — a letter and four digits — raised by a control module.

Generic codes are defined across manufacturers; manufacturer-specific ones carry more detail and need the right device to read.

Pending, confirmed and permanent are different states, and only some mean act now.

Which is the first thing most fleets get wrong: treating every code as urgent, then treating all of them as noise a month later.

Triage

Not every code needs a workshop.

Emissions-related codes on a modern diesel are frequently transient and clear themselves.

Codes that recur are the signal, which means the system needs to count occurrences rather than alert on each.

Codes affecting safety or driveability go straight to a decision, and the list of which those are is short and worth writing down once.

Building the triage list

Take a year of codes from your own fleet.

Count by code, by vehicle and by model.

Three groups: act immediately, watch for recurrence, ignore.

Review annually, because a new model year introduces codes the list does not cover.

Without this list every code is either an emergency or ignored, and both are expensive.

What it saves

A failure caught as a code is a scheduled repair; the same failure caught by the driver is a roadside recovery and a cancelled day.

The cost difference is the entire business case for condition monitoring, and it is measurable from your own breakdown records.

Count breakdowns that had a preceding code. In most fleets it is a substantial proportion, and each one was a warning nobody read.

Where it goes wrong

Alerts to nobody in particular, which is the commonest arrangement. A code arriving in an unmonitored inbox is not a control.

Alert fatigue, where a fleet manager receives forty a day and stops reading.

No link to the workshop, so a code becomes an email rather than a job.

And no feedback loop: nobody records whether the code preceded an actual failure, so the triage list never improves.

The measure

Codes raised, by group.

Breakdowns with a preceding code, as a proportion of breakdowns.

That second figure falling is the whole point, and it is the number to put in front of whoever funded the system.

Record whether the code preceded a failure

The feedback loop that makes the triage list improve.

Every code, with what happened next: nothing, a scheduled repair, or a breakdown.

Without it the list is a guess that never gets better.

With it, after a year, you know which codes on which models matter — which is knowledge specific to your fleet and available nowhere else.

Follow the difficult record

Use review the available features to frame one representative case. The useful evidence is the record created when a value is corrected, approved and exported.

Independent reference

For a thematic point of reference, see the National Highway Traffic Safety Administration. Use this established source as an outside check before turning the principle into a system rule.