Machines that cannot report anything can still carry a signal.
Centre lathes, radial drills and surface grinders — what they can tell you, where their time actually goes, and why OEE Performance needs care on a manual machine.
Conventional machines — centre lathes, radial drills, milling machines, shapers, surface grinders — are usually the last equipment a plant considers monitoring and frequently the equipment with the most unmeasured loss, precisely because nobody has ever had a timeline for them.
They cannot report anything. What they can do is carry a signal, and a signal is enough for a complete availability picture.
| Signal | Where it comes from | What it establishes |
|---|---|---|
| Main drive energised | Contactor auxiliary contact | Machine on and off, with exact durations |
| Working versus idling | Current transformer on the supply | Whether the machine is cutting or merely powered |
| Operator attention | Stack light, where fitted | Running, attention needed, stopped |
| Output count | Existing counter, or a proximity switch added at the slide or ram | Parts or strokes, exactly |
| Stop reason | Operator, at a kiosk or by QR scan | Why — which no signal can supply |
Shift-boundary time
Late starts and early stops. On conventional machines nobody is watching a cycle counter, so this is larger here than on CNCs and is the most commonly recovered loss.
Waiting for work
A conventional machine is often a secondary operation. It waits for the CNC upstream, and that waiting is invisible until it is measured.
Setup, unmeasured
Without a mode selector, setup and idle look identical. A setup reason code at the kiosk is the whole solution.
Operator absence
These machines are attended throughout. Time with the machine powered and no operator present is real, measurable and rarely discussed.
You can compute OEE on a conventional machine, and it is worth doing — but the Performance factor deserves care. The ISO 22400-2 definition needs an ideal cycle time per part, and on a manually operated machine that standard is genuinely difficult to establish and easy to dispute, because cycle time depends on the operator in a way it does not on a CNC.
This is where the MachineWise variant is most useful internally. Availability as machine on-time over reporting time is unambiguous on a conventional machine. Performance as productive time over on-time asks a question that can actually be answered from a current signal — how much of the powered time involved work — without requiring a standard nobody agrees on. Quote the ISO figure externally if you must, but be honest about how its Performance input was derived.
OEE on a conventional machine, both ways.
The standard definition first, then the variant MachineWise computes and why it differs. Quote the ISO figure externally; use the variant internally.
| ISO 22400-2 — the standard | MachineWise variant — internal | |
|---|---|---|
| Availability | Planned production time 435 min, actual production 302 min → 69.4% | Reporting time 435 min, machine on-time 348 min → 80.0% |
| Performance / Effectiveness | 128 parts × 105 s ideal cycle ÷ 302 min → 74.2% | Productive (working, not merely powered) 224 min ÷ 348 min → 64.4% |
| Quality | 124 good of 128 → 96.9% | 124 good of 128 → 96.9% |
| Result | 49.9% | 49.9% |
The two land close here, and the reason is instructive: on a conventional machine the ISO ideal cycle time is itself an estimate, so its apparent precision is borrowed. The variant is at least measuring something observed. The two numbers answer different questions and are not interchangeable. The ISO figure is comparable to published benchmarks; the variant is not, and should be treated as your own baseline to improve against. Both are computed side by side in the OEE calculator.
Two conventional machines on a live dashboard this week.
Pick two machines nobody has ever measured. The first fortnight almost always finds a shift-boundary pattern that costs more than any breakdown on the floor.
Book a free 2-machine pilot →WhatsApp us