Units per labor hour (UPLH) = good units / labor hours worked. Example: 10,800 good units from a crew of 6 working 7.5 hours each is 10,800 / 45 = 240 UPLH, and read next to OEE it tells you whether a change came from the line or from the crew size.
UPLH, labor per unit, OEE, UPLH at a target OEE and how to read the pair.
Why this number exists
OEE tells you how well a line used its scheduled time. It says nothing about how many people it took to do so.
UPLH fills that gap, and it is simple enough for a shift lead to work out on a calculator. It also goes by units per man hour, or units per person hour.
The formula and a worked example
Example: a packing line runs a 7.5-hour shift, with a crew of 4 operators, a team lead and a forklift driver. It packs 10,800 good units.
The inverse is useful too, because 45 hours / 10,800 units is 0.25 minutes of labor per unit. At a hypothetical labor cost of 28 euros an hour, that is about 0.117 euros per unit, a figure that feeds your cost per unit.
Units per labor hour calculator
Enter one line over one period, such as a shift. The defaults are the packing line worked through above. Updates as you type.
UPLH side by side
How to read the pair
| OEE | UPLH | What it usually means |
|---|---|---|
| Up | Up | A real gain: fewer losses with the same crew |
| Down | Up | Crew cut, and the line is paying for it. See OEE in a labor shortage |
| Up | Down | People added to win output. Check that the extra units are worth the hours |
| Down | Down | Stops, slow running or scrap. Go to the downtime Pareto |
A starting point for the conversation, not a verdict. Always confirm the cause on the floor.
The working
Count only good units, and apply the same hours rule to every line and week, or the numbers will not compare.
Which hours to count
Most arguments about UPLH are really arguments about the denominator. Pick a rule, write it down and apply it to each line.
| Type of hours | Count them? | Why |
|---|---|---|
| Line operators | Yes | The core of the number |
| Overtime | Yes | Hours worked, whatever the pay rate |
| Agency and temporary staff | Yes | Leaving them out flatters the number |
| Team lead, shared forklift driver | Your rule, share by line | Include them for the whole picture, and stay consistent |
| Unpaid breaks | No | Not hours worked |
| Training away from the line | No, track apart | A separate choice, and it should not punish the line |
Use clocked hours where you can, since crew plans and real attendance often differ. And count only good units, with scrap and first-time failures left out, so that a bad day cannot look productive.
How UPLH connects to OEE
When the crew stays for the whole shift, the two numbers link through one short formula. It shows exactly what can move UPLH.
Our packing line has an ideal rate of 2,400 units an hour, so 7.5 hours could make 18,000. Its 10,800 good units are an OEE of 60%.
Now compare four shifts with the same product. Two different routes reach the same UPLH, and the fourth shows what can go wrong with a cut.
Case B saved a person at the same output, and case C made 20% more product with the same people. Both show 288, so UPLH alone cannot tell them apart, and that is why you read the two together.
Case D is the common trap. UPLH went up by 4%, so the cut looks like a win, but the line lost 1,440 units a shift, and if those units had buyers the cut cost money.
How to read the pair
| OEE | UPLH | What it usually means |
|---|---|---|
| Up | Up | A real gain: fewer losses with the same crew |
| Down | Up | Crew cut, and the line is paying for it. See OEE in a labor shortage |
| Up | Down | People added to win output. Check that the extra units are worth the hours |
| Down | Down | Stops, slow running or scrap. Go to the downtime Pareto |
Mixed products break raw UPLH
If a line runs a fast product and a slow one, raw UPLH moves with the mix. It can fall sharply on a day when the crew worked just as well.
Example: product A has an ideal time of 1.5 seconds and product B has 3 seconds. Day 1 makes 9,000 A, and day 2 makes 4,000 A plus 2,500 B, both with 45 labor hours.
Raw UPLH falls from 200 to 144.4. Yet each B takes the time of two A's, so day 2 equals 4,000 + 5,000 = 9,000 A-equivalents, or the same 200.
So convert to equivalent units or use standard hours earned. The same thinking applies to OEE with mixed products.
Where UPLH goes wrong
- Comparing different lines. A manual line and an automated one have different crew needs, so compare each line with itself.
- Using total output in place of good output. Scrap then counts as productivity.
- Using planned hours. Overtime and absent staff disappear from the number.
- Using it to rank people. Most of what moves UPLH comes from the line, the plan and the material, so a ranking punishes the wrong thing.
- Cutting crew to raise it. Check OEE and plan attainment before you call that a gain.
UPLH vs other labor measures
UPLH tells you the output per hour, but not where the hours went. Overall labor effectiveness splits labor time into availability, performance and quality, in the same shape as OEE.
If you want one quick line in a production KPI set, use UPLH. If you need to find the labor loss, move on to OLE.
Try it on one line
Start on one line and calculate UPLH per shift for two weeks, next to OEE. Then pick the weeks where the two disagree and find out why.
The OEE calculator gives you the other half of the pair. For the gains themselves, start with how to improve OEE.
UPLH needs a trusted good count and an OEE figure for the same shift. Fabrico takes counts, stops and speed live from the line and shows OEE per line and shift, as a light MES for OEE. See how Fabrico shows OEE per shift.
What people ask next
How do you calculate units per labor hour?
Divide good units by the labor hours worked to make them. A crew of 6 working 7.5 hours makes 45 labor hours, so 10,800 good units is 240 UPLH.
What is a good units per labor hour number?
It depends on the product and the degree of automation, so a general target would mislead you. Compare each line with its own past results, and see how to set OEE targets for an approach that works for targets of this kind too.
Should I include supervisors in labor hours?
Include them if you want total line labor, and leave them out if you want direct labor only. Either is fine if every line and week follows the same rule, and you can show both.
Is UPLH the same as labor productivity?
UPLH is one common way to measure labor productivity on a line. Plants may also use output value per hour or overall labor effectiveness when they need to see where labor time goes.
Can OEE go down while units per labor hour goes up?
Yes. A crew cut can raise UPLH while the line loses output and OEE falls. OEE in a labor shortage shows how to count that.
Fabrico is a light MES for OEE. It takes stops, speed and counts from PLCs or IoT sensors, with add-on AI cameras counting output where a machine has no usable signal, and operators tag the cause at the line.
Related guides: OEE vs Productivity: What Each One Tells You · OEE for Manual Processes: How to Measure It