Blog / OEE
OEE · Guide

OEE for Assembly Lines: Takt and Station Data

OL By OEE Lab Editorial |Updated 2 November 2026 |6 min read
Short answer

On an assembly line, count good units at the end of the line and set the ideal cycle time from the slowest station at full staffing, and do not use takt for it. Then OEE = good units x ideal cycle time / planned time, and the OEE you need to meet demand is ideal cycle time / takt.

What makes assembly different

Machine OEE assumes a machine sets the pace, with a nameplate speed and a signal when it stops. On many assembly lines, people set the pace, with tools and small machines helping them.

So the hard questions are different. What is the ideal rate when the work is manual, which station caused the stop, and how does takt fit in?

Takt is not your ideal cycle time

This is the most common error on assembly lines. Takt comes from demand, and it moves whenever the order book moves, while ideal cycle time comes from the line and stays fixed for a given product.

If you use takt as the ideal, performance is judged against the demand rate, and OEE rises whenever orders fall, with no change on the floor. The difference is covered in takt vs cycle vs lead time.

Set the ideal from the slowest station

Example: a five-station line builds an electric actuator. Time-study cycle times at full staffing are 52, 48, 60, 45 and 55 seconds, all hypothetical.

Takt 67.5 s 52 48 60 45 55 S1 S2 S3 S4 S5
Station 3 sets the ideal cycle at 60 seconds. Takt, the dashed line, is set by demand and tells you how much margin you have.

The ideal cycle time is 60 seconds, the time of station 3. Use a proven, sustainable pace from a time study, and see ideal cycle time for why you should not use the average of last month.

When the line balance changes, re-time it, and note the date in your OEE records. The cycle time calculator can help with the first estimate.

Worked example: one shift

The shift has 450 planned minutes, and the line stops for a total of 45 minutes. It finishes 360 units, of which 342 pass first time with no rework at any station.

Availability = 405 / 450 = 90.0%
Performance = 360 x 60 s / 405 min = 88.9%
Quality = 342 / 360 = 95.0%
OEE = 90.0% x 88.9% x 95.0% = 76.0%
Example shift. The same answer comes from 342 good units x 60 s = 342 min, and 342 / 450 = 76.0%.

Count at the end of the line for the line's OEE, because that is what ships. A unit that was repaired at a rework bay does not count as good here, as the good count rule explains.

Where takt comes in: the required OEE

Takt does have a job in this story. It tells you how much OEE the line needs in order to meet demand.

Takt = 450 min x 60 / 400 units = 67.5 s
Required OEE = ideal cycle / takt = 60 / 67.5 = 88.9%
Actual: 76.0%, so 342 good units, not 400
With demand of 400 units a shift, this line needs 88.9% OEE. It reached 76.0% and fell 58 first-time good units short.

This one ratio is a quick way to talk about a line. If required OEE is above what the line has ever done, the plan needs overtime or a line change, and if it is far below, the line has spare capacity.

Use the takt time calculator when demand changes. Plan attainment then tracks whether the shift met the plan.

Use station data to find the cause

Line OEE gives you the size of the loss, and station data gives you the location. So log each stop against the station that started it, even when the whole line stopped.

StationCauseStop minutesShare
S3Torque tool fault1840%
S1Parts shortage1227%
S5Test fail, retest920%
S2Fixture clamp613%
45 stop minutes by the station that started the stop. The stations waiting downstream are not the cause.

On a line with buffers, a stop at one station may not stop the line at once. Record it as blocked or starved at the neighbors, and the cause at the station where it began.

Performance is harder, since its 45 lost minutes in this example leave no stop record at all. To find them, time each station for a shift, since cycle time variation usually shows which step drifts, and which product or person needs support.

What OEE does not show: balance loss

With 60 seconds as the ideal, the other four stations wait for part of every cycle. OEE does not count that wait, because the line is running at its ideal pace.

The figure is still real. The work adds up to 260 seconds, while the line pays for 5 x 60 = 300 seconds, so balance efficiency is 260 / 300 = 86.7%.

Track it apart from OEE, with the line balancing calculator. Shift work out of station 3 until it reaches 55 seconds, and the same crew has a new ideal of 55 seconds and more capacity.

Assembly-specific rules

  • Paced conveyors. A moving line has a set speed, and performance loss shows up as gaps. Count empty carriers, since a gap is a unit you did not build.
  • Mixed models. Each model has its own ideal, so add up good units x ideal time per model. See OEE with mixed products.
  • Rework bays. Off-line repair keeps the count at the end of the line clean only if reworked units are excluded from good.
  • Crew size. The ideal assumes full staffing. If people are missing, keep the ideal and count the loss, as OEE in a labor shortage shows.
  • Short stops. A missing part or a dropped screw can take 30 seconds, and nobody logs it. Use an andon call at each station to catch more of them.

How to collect the data

Start with a shift tally of good units and a stop log at the line. A scan at the last station gives you a clean count and a timestamp for each unit.

The gap between scans is useful, since every gap above the ideal is a short stop or a slow cycle. For the other options, see OEE data collection methods, and for SMT lines see OEE for SMT lines.

See it live in Fabrico

Manual stations leave short waits that no machine signal records. Where cameras are fitted, Fabrico puts video of each stop, streamed on request next to its machine data, and operators tag the cause at the line. See how Fabrico uses cameras on the line.

What people ask next

How do you calculate OEE on an assembly line?

Count good units at the end of the line and set the ideal cycle time from the slowest station at full staffing. Then OEE equals good units times ideal cycle time, divided by planned time.

Can you use OEE on a manual assembly line?

Yes. The formula is the same, and the ideal comes from a time study of the sustainable pace in place of a machine rating. See ideal cycle time.

Should I use takt time as the ideal cycle time?

No. Takt moves with demand, so OEE would change when orders change. Use takt to work out the OEE you need, which is ideal cycle divided by takt.

Does OEE show line balance loss?

No. If the ideal is the slowest station's time, waiting at faster stations does not lower OEE. Track balance apart, with the line balancing calculator.

Should I measure OEE per station or for the whole line?

Report it for the whole line at the last station, and use station data to find the cause. Line OEE vs machine OEE explains how the two fit together.

See OEE per line and shift on your own machines

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.

Book a Fabrico demo

Related guides: How to Calculate Line Efficiency: Both Formulas · OEE for Manual Processes: How to Measure It

← More OEE Lab guides · Line balancing calculator →

OEE essentials: OEE calculator · What is OEE · Six big losses · Availability · TEEP · Cost of downtime · OEE benchmark · Free OEE tracker · Free OEE software · How to track downtime · How to implement OEE · Light MES vs full MES