Assembly line efficiency = total station time / (stations x cycle time) x 100. We use the slowest station as the cycle time. Example: 7 stations with 288 seconds of work and a slowest station of 47 seconds give 288 / 329 = 87.5%, so 12.5% of paid station time is waiting.
What the number means
On a paced line, every station passes units on at the speed of the slowest one. A faster station finishes its work and waits, and you pay for that wait.
Assembly line efficiency shows how much of the paid station time holds real work. The remainder is called balance delay.
This is the line-balancing form of line efficiency. Sewing lines usually use the labor form, built on standard minutes, and how to calculate line efficiency explains both.
The example line
Example: a hypothetical line builds a battery charger at seven single-operator stations. The shift has 440 available minutes, or 26,400 seconds, and demand is 550 units, so takt is 26,400 / 550 = 48 seconds.
The station times add up to 288 seconds of work per unit. With seven stations at a 47-second pace, the line pays for 7 x 47 = 329 seconds per unit.
Slowest station or takt in the formula?
Textbooks write this formula with "cycle time", and the word hides a choice. We use the slowest station, because it shows evenness whatever demand does.
The common alternative uses takt, which shows how well the crew size fits demand. When the slowest station sits under takt, as here, the line may run at takt and wait, so the takt version is then the more honest cost figure.
If the slowest station is above takt, use it, because the line cannot run at takt at all. The takt time and line balancing guide shows what to do in that case.
Can you use fewer stations?
The lowest possible station count is total work / takt, rounded up. Here 288 / 48 = 6.0, so six stations are the mathematical minimum.
Six stations would need every one of them to hold exactly 48 seconds, with no margin for any stop. So the minimum tells you the limit, and for this line seven remains the practical answer.
What rebalancing gains
Example: move 5 seconds of work from S5 to S4, where the task order allows it. S5 falls to 42, S4 rises to 41, and S3 at 44 becomes the new slowest station.
| Measure | Before | After |
|---|---|---|
| Station times (s) | 38, 42, 44, 36, 47, 41, 40 | 38, 42, 44, 41, 42, 41, 40 |
| Total work | 288 s | 288 s |
| Slowest station | 47 s | 44 s |
| Efficiency | 288 / 329 = 87.5% | 288 / 308 = 93.5% |
| Top output per shift | 26,400 / 47 = 561 | 26,400 / 44 = 600 |
| OEE needed to hit takt | 47 / 48 = 97.9% | 44 / 48 = 91.7% |
The last row matters most on a real floor. Before the move, a line with nearly any stops would miss demand, because it had only 1 second of margin per cycle.
What this figure cannot see
The formula uses planned times, so a line can score 93.5% and still miss the shift. Stops, slow cycles and rejects never enter it.
To see them, compare with what the shift delivered. Say the line makes 505 good units before the rebalance: then the labor form gives 505 x 288 / (7 x 26,400) = 145,440 / 184,800 = 78.7%.
The two kinds of loss have different owners. Balance loss belongs to line design, and the rest shows up in OEE for assembly lines, which covers how to read station stops.
What is a good figure?
We found no published standard, so what follows is only a rule of thumb, and it is our own. On our line balancing calculator, above 90% reads as well balanced, 75-90% as workable but loose, and below 75% as paying for idle stations.
These bands apply to the slowest-station form, and they are a starting point, not a pass mark.
A mixed-model line carries balance loss a single-product line avoids. For that case, set takt with the mixed-model takt time calculator first.
Timing the stations
- Time every station for at least 10 cycles. Use the typical cycle, and keep unusual events out of the study.
- Count the station as whatever paces the work. Two people doing the same step in parallel halve that step's time.
- Watch for hidden waits. Waiting for a blocked or starved neighbor is not work, so do not time it as work.
- Re-time after each change. Small method changes move the slowest station without anyone noticing, and cycle time variation shows it.
Ways to fool yourself with this score
- Using the average station time as the cycle. The answer is then always 100%, because the sum divided by the count is the average.
- Adding walking and waiting to total work. It inflates the efficiency and hides the waste.
- Moving work against the task order. A screw cannot go in before the part it holds, so map the sequence first.
- Chasing 100%. A line with no spare time at any station has no room for a single small stop.
When the balance is settled, check the rest of the line with the line efficiency calculator and fix the biggest stop first.
Balance loss is visible on paper, but short waits at manual stations are not. Where cameras are fitted, the video of any stop can be 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
What is the formula for assembly line efficiency?
Total station time / (stations x cycle time) x 100. With 288 seconds of work, 7 stations and a 47-second slowest station, it is 87.5%. The line efficiency calculator does the sum.
What is balance delay on an assembly line?
It is the share of paid station time spent waiting, or 100% minus line efficiency. At 87.5% efficiency, balance delay is 12.5%. The line balancing calculator shows it for your line.
Should I use takt or the slowest station in the formula?
Use the slowest station to judge how even the line is, and takt to judge whether the crew fits demand. Name the one you use. If the slowest station is above takt, the line cannot run at takt, so use the slowest station.
How do I calculate the number of stations for an assembly line?
Divide total work by takt and round up. Here 288 / 48 = 6 stations is the limit, though real task order and stops usually call for more. See takt time and line balancing.
Does a high line efficiency mean high OEE?
No. Line efficiency here uses planned station times, so it never sees stops or rejects. OEE measures those, as OEE for assembly lines explains.
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.
← More OEE Lab guides · Open the line efficiency calculator →