Manufacturing lead time is the time from order release until the product is ready to ship. It equals queue + setup + run + wait + move + inspection, or WIP / throughput for a whole area. In the example, run time is only 7 of 60.5 hours: waiting is 79% of lead time.
Lead time, value-added share and where to cut first, for a route of 1 to 10 steps.
What is manufacturing lead time?
Manufacturing lead time is the time from the moment an order is released to production until the finished product is ready to ship.
It covers everything in between: queues, setups, run time, waits, moves and inspection. Run time is usually the smallest part.
Do not mix it up with two near neighbors:
- Cycle time is the time one unit takes at one step. See takt time vs cycle time vs lead time.
- Customer lead time starts when the customer orders, so it adds order handling, purchasing and delivery.
The formula
Lead time = WIP / throughput (Little's law)
The first form is a sum of what happens to one order. Use it when you follow an order through the plant. The second form works for a whole area. It is Little's law, and it needs only two numbers you can measure: how many jobs are in the system, and how many leave per day.
Course books on production systems also use a third form, for an order that passes through n steps of similar size: lead time = n x (setup + batch size x cycle time + non-operation time). Non-operation time is the queue, move, wait and inspection time per step.
Example: 3 steps, 1.5 hours setup, a batch of 100 units at 1 minute each, and 12 hours of non-operation time per step. Lead time = 3 x (1.5 + 1.67 + 12) = 3 x 15.17 = 45.5 hours. It is the same sum as the first form, written for steps that look alike.
Worked example: one order
Example: an order of 200 units goes through cutting, machining and packing. Times are in working hours.
| Step | Type | Hours |
|---|---|---|
| Queue before cutting | Queue | 16.0 |
| Setup for cutting | Setup | 1.5 |
| Cutting run (200 units x 1.2 min) | Run | 4.0 |
| Wait for the machining cell | Queue | 24.0 |
| Setup for machining | Setup | 1.0 |
| Machining run (200 units x 0.9 min) | Run | 3.0 |
| Wait for inspection | Queue | 8.0 |
| Inspection | Inspection | 1.0 |
| Pack and move to dispatch | Move | 2.0 |
| Lead time | 60.5 |
The order spends 60.5 hours in the plant. Only 7.0 hours of that is machine run time. That is 11.6%.
Speeding up the machining run by 10% saves 0.3 hours, or 0.5% of lead time. Cutting the waits in half saves 24 hours. That is the real lever.
Worked example: a whole area (Little's law)
Example: a machining area has 120 jobs in it at any moment and finishes 40 jobs a day.
Lead time = 120 / 40 = 3 days.
If you cap the work in the area at 80 jobs and it still finishes 40 a day, lead time drops to 80 / 40 = 2 days. You did not make anything faster. You just stopped loading the area with more than it can handle.
Manufacturing lead time calculator
Enter each step of one order's route, in the order it happens. The defaults are the 200-unit order and the machining area above. Updates as you type.
Use one clock for every step, working hours or calendar hours, not both. Run time is the value-added part. Count rework loops as extra steps, since they add their whole wait again.
Lead time = WIP / throughput. Count the jobs in the area and the jobs it finishes per day.
Where the hours go
The working
Little's law gives the average for the area, so single orders can take longer or shorter. A WIP cap only lowers lead time if the area still finishes the same number of jobs per day.
Why queues grow so fast
A simple single-queue model gives a useful picture. It is a standard textbook result. The average wait in the queue is about load / (1 - load) times the average job time, where load is the share of time the machine is busy.
- At 80% load the factor is 0.8 / 0.2 = 4.
- At 95% load the factor is 0.95 / 0.05 = 19.
So a machine pushed from 80% to 95% load has a queue nearly five times longer. This is why the bottleneck, loaded to the limit, sets your lead time.
It also shows why higher OEE at the bottleneck shortens lead time. Fewer stops and less slow running lower the load for the same demand. See utilization vs OEE and Overall Equipment Effectiveness (R. C. Hansen, 2001), which tells you to apply OEE to the bottleneck first.
Example: what OEE does to the queue
Example: a bottleneck is scheduled for 100 hours a week. The demand needs 70 hours of work at the ideal rate.
- At an OEE of 75%, the machine delivers 75 ideal-rate hours. Load = 70 / 75 = 93.3%. Queue factor = 0.933 / 0.067 = 14.0.
- At an OEE of 85%, it delivers 85 ideal-rate hours. Load = 70 / 85 = 82.4%. Queue factor = 0.824 / 0.176 = 4.7.
Ten points of OEE cut the average wait in the queue to about a third. The machine and the demand did not change. Only the losses did. This is a model, so real queues differ, but the direction is reliable.
How to cut manufacturing lead time
- Limit work in progress. Cap the jobs allowed into each area. By Little's law, lead time falls with WIP.
- Cut batch sizes. Pass the first 100 units to the next step while the other 100 are still running. In the example, if the machining cell is free, it can start 2 hours earlier.
- Shorten setups. Smaller batches need quick changeovers. See SMED changeover reduction.
- Raise the OEE of the bottleneck. See how to improve OEE.
- Reduce variability. Unplanned stops make queues swing. See how to reduce unplanned downtime.
- Map it. A value stream map shows where each hour goes.
How to measure it on the floor
- Track an order: record a timestamp at release, at each step start and end, and at ship-ready.
- Track an area: count jobs in the area once a day, and count jobs finished. Divide.
- Use the same start and stop points. Decide whether the clock starts at order release or at material pick, and keep it.
- Report an average and a spread. A 3-day average with some orders at 8 days tells you more than the average alone.
Common mistakes
- Adding up run times and calling it lead time. That leaves out the waits, which are most of it.
- Counting calendar hours for some steps and working hours for others. Pick one.
- Pushing the bottleneck to full load. The queue explodes.
- Cutting lead time by stocking finished goods. That hides the problem and ties up cash.
- Ignoring rework loops. Units that go back to a step add their whole wait again.
Long queues often start with an unplanned stop at the bottleneck. Fabrico shows stops per line and shift with the cause tagged by the operator, so you can see which stops at the bottleneck cost the most time. See how Fabrico tracks stops and availability.
What people ask next
What is a good manufacturing lead time?
There is no universal number. It depends on the product and the process. Compare against your own history and against the run time. If run time is a small part of lead time, there is room to cut.
What is the difference between lead time and cycle time?
Cycle time is how long one unit takes at one step. Lead time is the whole elapsed time from release to ready to ship, including every wait. See takt time vs cycle time vs lead time.
How does OEE affect lead time?
A higher OEE at the bottleneck means more good output from the same time. The bottleneck is then less loaded for the same demand, so queues in front of it get shorter. See how to calculate OEE.
What is Little's law?
It says the average number of items in a system equals the arrival rate times the average time in the system. In plants: WIP = throughput x lead time. So lead time = WIP / throughput.
Should lead time include supplier time?
Manufacturing lead time usually starts at order release, so it does not. Cumulative or customer lead time adds purchasing and delivery. Write down which one you report.
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: Lean Metrics That Matter on a Line, With Formulas · How to Calculate Throughput: Formula and Worked Example · Throughput Time vs Lead Time vs Cycle Time · Cycle Time vs Process Time, With Takt Time