Blog / Metrics
Metrics · Guide

Throughput Time: Meaning, Formula and Example

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

Throughput time is the time one unit takes from the start of the process to the end. Throughput time = process + inspection + move + queue time, or WIP / throughput as an average. In the example, 16 hours pass for a unit that is only worked on for 3.2 of them.

What throughput time means

Follow one unit from the moment work starts on it until it leaves the last step. The hours on your watch are its throughput time.

Words differ between plants, so here is the definition we use: first operation start to last operation finish, on the floor. Many people use the same words for the whole of manufacturing lead time, and the two can overlap, as throughput time vs lead time explains.

Operations books often call the same thing flow time, and some call it cycle time. Keep that in mind when a source seems to disagree with you.

Throughput and throughput time are also different things. Throughput is a rate, such as 6 units an hour, while throughput time is a duration, such as 16 hours.

The formula

Throughput time = process + inspection + move + queue
3.2 + 0.4 + 0.6 + 11.8 = 16.0 hours
Average throughput time = WIP / throughput
96 units / 6 units per hour = 16.0 hours
Two ways to reach one number. The sum follows a unit, and the division needs just two counts.

The WIP version is Little's law. It holds for long-run averages of a stable process, when WIP and throughput cover the same area and the same units.

Worked example

Example: a hypothetical cell makes a valve body through four machining and test steps. A tagged unit gets 3.2 hours of processing, 0.4 of inspection and 0.6 of moves, and it sits in queues for 11.8 hours.

Process3.2 h
Inspection0.4 h
Move0.6 h
Queue11.8 h
Where 16 hours go. Queue time is nearly three quarters of the total, and process time is one fifth.

Only process time changes the part, so the ratio 3.2 / 16 = 20% is a useful one. It is called manufacturing cycle efficiency, or MCE.

Now see what speed is worth. A process that is 10% faster saves 0.32 hours, while cutting the queue by a quarter saves 2.95 hours.

Check it with Little's law

On an average day, a count finds 96 units in the cell, and the cell finishes 6 units an hour. So average throughput time is 96 / 6 = 16 hours, which agrees with the tagged unit.

When the two disagree by a lot, look for units parked outside the count or a period that was not stable. The law is a cross-check, so it cannot replace looking at the queues.

It also shows that a faster rate is not the only lever. Cap the WIP at 60 units with the same 6 an hour, and throughput time falls to 60 / 6 = 10 hours with no machine running any faster.

That holds only while the bottleneck never runs out of work. Cap the WIP too low and the rate itself falls.

Why batches stretch it

A unit in a batch waits for the rest of its batch before it can move on. That wait counts as queue, and it can be large even at a step that has no line in front of it.

Batch sizeCalculationFirst unit waits
4039 x 3 min117 min
2019 x 3 min57 min
109 x 3 min27 min
A step with a 3-minute cycle. The first unit finished waits for (batch - 1) x cycle time before the batch moves on.

Over four steps, moving from a batch of 40 to a batch of 10 can save up to 6 hours for the first unit, since 90 minutes x 4 = 360 minutes. Smaller batches mean more changeovers, so SMED changeover reduction usually comes first.

Three ways to get the number

Measure the same way each time, and pick the method that fits how you work today.

1
Tag and time
Write the start time on a travel card, and record the finish time at the last step. Do this for at least 5 units across the week.
2
Count WIP and output
Count units in the area once a day, and count what finishes. Divide WIP by the daily or hourly output.
3
Use scans
If units are scanned at the first and last step, subtract the timestamps for every unit.
Start with tagged units, then add a WIP count as your weekly check.

Report the spread as well as the average. Example: five tagged units take 12.5, 14, 15.5, 16 and 22 hours, which averages 16, and the 22 tells you more than the average does.

Use one kind of clock. Working hours and calendar hours give very different answers across a weekend, so write down which you use.

How to cut throughput time

  1. Limit WIP. Let a new unit start only when one has finished, so the queues cannot grow.
  2. Work on the bottleneck. The longest queue usually sits in front of it, as bottleneck analysis shows.
  3. Cut batch sizes. The table above shows why.
  4. Bring steps closer. Move and wait times often shrink when steps sit next to each other, and a value stream map shows where.
  5. Reduce stops. Stops at the bottleneck swing the queue, so start with how to reduce unplanned downtime.

Errors that distort throughput time

  • Adding up cycle times and calling it throughput time. That leaves out the queues, which are usually the largest part.
  • Confusing it with throughput. Doubling WIP can double throughput time and add nothing to the output rate.
  • Measuring only the fast jobs. Rush orders jump the queue, so a sample of them says little about normal units.
  • Ignoring rework. A unit that goes back passes through the queue again, and its time counts in full.
  • Using Little's law in an unstable period. During a ramp-up or a big backlog, WIP and output do not balance, so the result misleads.

Start with one area today

Take the WIP count and output for one area and work out the average today. Then use the throughput calculator to get the rate side right, and compare the result with one tagged unit.

See it live in Fabrico

Long queues often start with stops at the bottleneck. Fabrico shows stops per line and shift, live from PLCs or IoT sensors, with add-on AI cameras counting output where a machine has no usable signal, with the cause tagged by the operator at the line. See how Fabrico tracks stops.

What people ask next

What is throughput time in manufacturing?

It is the time one unit takes from the start of the process until it finishes, including processing, inspection, moves and queues. Plants set the start and end points differently, so write down yours. See throughput time vs lead time.

How do you calculate throughput time?

Add process, inspection, move and queue time for one unit. Or divide average WIP by throughput: 96 units at 6 an hour gives 16 hours. The throughput calculator helps with the rate.

Is throughput time the same as lead time?

Many people use them as the same thing. We use throughput time for the floor only, from first step start to last step finish, while lead time starts earlier and can end later. See manufacturing lead time.

Is throughput time the same as cycle time?

Not in lean use. Cycle time is the pace at one step, usually in seconds or minutes, while throughput time covers the whole route with all its waits. Some operations books do use cycle time for the whole route, so check which you are reading.

Does higher throughput mean shorter throughput time?

Not by itself. Little's law says throughput time = WIP / throughput, so WIP matters as much as the rate. A higher rate with growing WIP can leave throughput time unchanged.

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

← More OEE Lab guides · Throughput time vs lead time →

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