Two departments, one scoreboard: an asset reliability programme in heavy manufacturing

The situation we are usually called into

Maintenance is rewarded for minimising cost and completing scheduled work. Production is rewarded for running. When these objectives conflict — at every equipment handover, every deferred intervention, every “we’ll run it to the weekend” — the conflict is resolved by whoever has more standing that day, not by which decision is better for the plant.

The symptoms are textbook. Equipment is run to failure informally while the formal strategy claims otherwise. Minor stops are absorbed by operators and never recorded, so the true loss profile is invisible. Root cause analysis, where it happens, terminates at the failed component. And the same failures recur, because nothing propagates.

What the benchmarks say

World-class OEE is 85%+; typical performance sits far below, and the gap is dominated by losses nobody records.
A substantial majority of plant failures have occurred before — on the same asset, a sister unit, or another site.
Reactive-dominant operations show markedly higher defect rates, lost sales from delays, and inventory inflation than planned regimes.

Sources: OEE literature; Plant Engineering and industry surveys; defect-elimination practice.

The real problem: two scoreboards

You cannot fix a conflict of incentives with a workshop. The programme is delivered in two streams — leadership and practitioner — because the failure lives in the operating model, and an operating model cannot be repaired from below.

“When both departments are reviewed against one set of reliability outcomes, the daily conflicts resolve themselves — because both now lose when the plant loses.”

How we run the programme

Adopt a shared loss model

One definition of loss — availability, performance, quality — decomposed to specific recorded causes. Minor stops become visible, and are usually revealed as a dominant loss category.

Run a joint criticality analysis

Which assets actually determine plant output, agreed once, producing a ranking rather than two competing intuitions.

Install defect elimination

A structured process for the recurring failures, with root causes pursued past the component into the systemic cause — and findings propagated to sister equipment.

Move to shared KPIs

Reliability metrics owned jointly by production and maintenance, reviewed in one meeting, with one set of numbers.

Train operator care

The people closest to the machine detect and report early-stage degradation — the cheapest condition monitoring any plant has available.

What makes it survive the next budget cycle

One scoreboard

Joint accountability for reliability outcomes. The escalation path for daily conflicts disappears because the conflict does.

Minor stops counted

The loss category that was invisible becomes the improvement backlog.

Findings propagate

Defect elimination results are pushed to sister equipment, so the plant learns once rather than repeatedly.

What the client is left with

Measurably improved equipment effectiveness — with the largest single gain typically coming from losses that were previously invisible. The durable change is not a metric. It is that the two departments stopped optimising separately.

Programme