Everyone wants one number. The honest answer has a few caveats — and those caveats are the difference between an OEE program that improves the plant and one that just games a dashboard.
If you only remember three numbers, remember these.
So a "good" OEE isn't 85%. For most sites, a good OEE is a number climbing quarter over quarter, measured the same way every time. The gap between where you are and 85% is real — but the plants that close it do so by beating their own last number, not a figure from a blog.
Directional ranges, not gospel. Use them to sanity-check whether you're in the right neighbourhood — not as a target handed down from outside.
| Industry | Typical | "World-class" | Why it differs |
|---|---|---|---|
| Discrete / machining (CNC) | 60–72% | 80–85% | Frequent changeovers and high product mix eat availability. |
| Automotive assembly (Tier 1) | 75–82% | 85–90% | Dedicated lines and mature TPM — the environment 85% was built for. |
| Food & beverage | 55–78% | 80–85% | Huge spread: high-mix ready-meals sit low; dedicated bottling runs high. |
| Continuous process (chem/paper) | 80–90% | 92%+ | Runs uninterrupted once started, so availability is structurally high. |
| Pharmaceuticals | 55–70% | ~70% | Validation, cleaning and regulatory holds cap the ceiling by design. |
Ranges synthesised from multiple industry datasets and TPM literature. Global distribution figures (55–60% average, ~6% at world-class) from Evocon's dataset of 3,500+ machines across 50+ countries; the 85% definition from Nakajima / OEE.com and Lean Production.
Three reasons the neat table above can quietly send you the wrong way.
Enter your three OEE factors and pick your industry. The gauge shows your score; the bar places you against that industry's real-world range.
Chasing the wrong benchmark is expensive. Want a read on where your real gains are hiding?
Get in touch →