Idle time is the quiet loss. Nothing is broken and no alarm is sounding — the machine is just standing still during time it was supposed to be running.
What idle time usually is
On a CNC floor, the common causes are ordinary ones:
- Waiting for material, blanks or the previous operation
- Waiting for an operator who is running another machine
- Waiting for first-off inspection or a quality decision
- Waiting for a program, tool or fixture
- Setup between jobs, when it isn’t tracked separately
Every one of these looks the same from the machine’s side. That is why idle time only becomes useful once each stop has a reason.
A worked example
One machine, one 8-hour shift (illustrative numbers):
Planned production time (8 h minus a 30 min break)
- Minutes
- 450
Running (in cycle)
- Minutes
- 300
Handling between parts
- Minutes
- 60
Stopped by alarms
- Minutes
- 25
| Time | Minutes |
|---|---|
| Planned production time (8 h minus a 30 min break) | 450 |
| Running (in cycle) | 300 |
| Handling between parts | 60 |
| Stopped by alarms | 25 |
Idle time = 450 − 300 − 60 − 25 = 65 minutes. Without the handling and alarm split, the same shift would show 150 minutes of “not running” — more than twice the real idle loss, and no clue which part of it to act on.
Idle time vs handling time
ThingConnect puts every moment of a machine’s day into one state, in this order: running, then handling (the planned load/unload window after each part), then alarm, then idle. Idle is what’s left — so a normal part change doesn’t inflate it.
A stop with only a non-critical alarm open reads as idle, not alarm, and any idle stop can be classified with a reason. That turns one large idle number into a short list of causes. See downtime tracking.