When to Replace a Vehicle

Most fleets replace on a policy — four years, or a hundred thousand miles — applied to everything. The right point differs per vehicle and is findable.

Why a fixed policy is wrong

Vehicles doing different work age differently.

A van on motorway trunking at a hundred thousand miles is in better condition than one doing urban multi-drop at sixty.

Which means a distance policy replaces the wrong vehicles and keeps the ones that should go.

And a time policy ignores usage entirely.

The curve

Plot annual cost per vehicle against age or cumulative distance.

It falls as the purchase cost amortises, flattens, then rises as repairs accumulate.

The replacement point is where the rising repair cost exceeds the falling depreciation — which is a calculation, not a judgement.

You need three or four years of your own cost data to see it, which is why this is a practice rather than a project.

What telematics contributes

Actual usage, so the curve is against real distance and hours rather than estimates.

Condition indicators: recurring fault codes, rising unscheduled repairs.

Utilisation, so a vehicle being replaced is one that is actually needed.

And the evidence that a vehicle's work changed, which explains a curve that bends unexpectedly.

The signals to replace early

Unscheduled repairs exceeding scheduled, consistently.

A second major component failure.

Recurring codes that a workshop cannot resolve.

Downtime rising, which is the operational cost rather than the maintenance one.

Any two of these together is usually decisive.

The signals to keep longer

Low utilisation, which means the vehicle has more life in it than its age suggests.

Stable costs with no unscheduled spend.

A residual market that has collapsed, where selling realises little and the replacement is expensive.

Fleets replacing on policy give away vehicles with years left, which is a real and quiet loss.

Practical cautions

Lease terms may make the decision for you, and the analysis then informs the next lease rather than this one.

Electric residuals are still moving, which makes the curve less reliable at the disposal end.

And a replacement decision made on one quarter is a guess. Use the year.

The measure

Cost per mile or per hour by cohort, trended.

Proportion of maintenance spend that is unscheduled, per vehicle.

And realised residual against forecast, which tells you whether the disposal end of the calculation is honest.

Watch for two signals together

Either alone is ambiguous; together they are decisive.

Unscheduled repairs exceeding scheduled, consistently.

A second major component failure.

Recurring codes a workshop cannot resolve.

Downtime rising.

Any two of these is the replacement case, made from evidence rather than from a policy date.

Connect policy and data

The choices in this note can be compared with open this implementation example. Keep the written purpose in control and enable only the information needed at this boundary.

Independent reference

For a thematic point of reference, see Kelley Blue Book. Use this established source as an outside check before turning the principle into a system rule.