On an HMC, machine utilisation hides the truth.
A pallet-equipped machine can look near-continuously busy while throughput is throttled by pallet turnaround. What to measure instead, and why setup duration ranks jobs backwards.
A horizontal machining centre with pallet changing breaks an assumption that every other machine on this list satisfies: that machine-level utilisation tells you the truth. An HMC can show excellent utilisation while one pallet sits loaded and waiting for hours, and the machine-level figure will never reveal it.
Monitoring an HMC properly means measuring at pallet level as well as machine level, and understanding that the two numbers answer different questions.
This is the central point of HMC monitoring. The machine can be near-continuously busy while the cell's throughput is limited by how quickly pallets are loaded, unloaded and made ready — which is a labour and scheduling problem that machine-level data actively conceals.
Pallet turnaround time
From pallet-out to the same pallet being ready again. The number that actually governs cell throughput.
Pallet-level cycle attribution
Which pallet, which fixture, which job. Otherwise a slow job and a slow pallet are indistinguishable.
Load-station occupancy
How long a pallet waits at the load station after being finished with, and after being made ready.
Machine-waiting-for-pallet
The state that matters most and is easiest to miss: spindle available, no pallet ready.
A networked HMC control reports machine state, cycle timing, counts and alarm presence like any other machining centre, and on many machines it reports which pallet is in the spindle position. Where that pallet identity is available, pallet-level attribution follows directly.
Where it is not, pallet identity has to come from elsewhere — an operator selection at the load station, or a signal from the pallet changer. This is worth establishing during the pilot rather than assuming, because pallet-level analysis is the main reason to monitor an HMC differently from a VMC, and without pallet identity you have an expensive VMC in the data.
OEE on a HMC, 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 389 min → 89.4% | Reporting time 435 min, machine on-time 402 min → 92.4% |
| Performance / Effectiveness | 96 parts × 215 s ideal cycle ÷ 389 min → 88.4% | Productive (in-cut) time 331 min ÷ 402 min on-time → 82.3% |
| Quality | 95 good of 96 → 99.0% | 95 good of 96 → 99.0% |
| Result | 78.3% | 75.3% |
An HMC typically posts stronger figures than a VMC on both definitions, which is exactly why the machine-level number is dangerous here — it can look healthy while pallet turnaround is throttling the cell. Read it alongside pallet turnaround, never alone. 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.
The defining operational advantage of pallet changing is that setup happens off the machine while the machine keeps cutting. That changes what a changeover measurement means: the relevant number is not how long setup took, but whether setup finished before the machine needed the pallet.
A setup that took ninety minutes and was ready in time cost nothing in throughput. A setup that took thirty minutes and finished two minutes late cost a stoppage. Measuring setup duration alone will rank those two backwards, which is a good illustration of why machine-type-specific monitoring is not a marketing distinction.
Two HMCs on a live dashboard this week.
We will establish during the pilot whether your machines report pallet identity, because that single fact determines whether you get pallet-level truth or an expensive VMC in the data.
Book a free 2-machine pilot →WhatsApp us