Changeover Reduction: How to Apply SMED in Practice

Disclosure: M365.FM host Mirko Peters works at iFAKT GmbH. This page describes the company factually; paid partnerships and links are labelled as advertising.

By Mirko Peters · 2026-11-11

SMED cuts changeover time by moving work outside the machine stop. Five steps and a worked example for quick changeover on your own line, starting this week.

SMED is a method for cutting the time a machine is stopped while it changes from one product to the next. The name stands for single-minute exchange of die. This article is for production managers who want to reduce changeover time on a real line, not just understand the idea. It explains the five steps, shows a worked example with numbers and lists the mistakes that stop most first attempts.

Key points

• SMED, short for single-minute exchange of die, is a method to switch a machine from one product to the next in the shortest possible time.
• A changeover runs from the last good part of one job to the first good part of the next. Measure that whole span.
• Internal work needs the machine stopped. External work can be done while it runs. Moving work from internal to external gives the biggest gain.
• Start with a pilot machine, record the changeover on video, list every element and separate internal from external work.
• Shorter changeovers allow smaller batches, which cut work in progress and lead time.

What is SMED, and why does it matter?

SMED comes from the work of Shigeo Shingo. The Lean Enterprise Institute describes single minute exchange of die as a method for switching production equipment from one part number to another in the shortest time possible. Its target is to reduce changeover times to a single digit, which means less than 10 minutes. SMED is a core tool of lean manufacturing. Many changeovers will not reach that target, and that is fine. The value is in the direction: shorter, simpler, more repeatable.

Changeover time matters for three reasons:

• Lost production. Every minute of setup time is a minute the machine does not make parts.
• Batch size. Long setups push planners to run large batches to avoid them. Large batches build work in progress and lengthen lead times.
• Flexibility. A plant with quick changeover can respond to small orders without building stock.

A changeover includes everything from the last good part of one job to the first good part of the next. That includes cleaning, tool change, adjustment, trial parts and waiting for approval. Measure the whole span, not only the time a fitter spends at the machine.

What are internal and external elements?

SMED rests on one distinction. LeanProduction.com defines internal elements as those that must be completed while the equipment is stopped, and external elements as those that can be completed while the equipment is running.

Examples:

• Internal: removing the old die, bolting on the new one, adjusting a stop.
• External: fetching the next die from storage, preparing material, checking the first-off gauge, cleaning parts of the machine that stay clear of the stop.

In many plants, some external work is done while the machine is stopped, because nobody prepared it earlier. Moving that work outside the stop is the first and cheapest step.

Does SMED only apply to dies?

No. The name comes from press dies, but the method works on any equipment: filling lines, injection moulding machines, packaging lines and assembly stations. The changeover process is always the same. Separate the external work, convert internal work where you can, then streamline what is left. Only the details of the tooling differ.

How do you apply SMED? Five steps

1. Pick a pilot machine and set a baseline. Collect changeover data for at least two weeks. Choose a machine where changeover is a significant share of lost time, where changeovers happen often and where the machine is a bottleneck. Record the current setup time as the baseline. The LeanProduction.com guide suggests recording the changeover on video.
2. Record the changeover and list every element. Film one complete changeover. Sit down with the team and write each element on a card: what is done, by whom, and how long it takes. A real changeover often has dozens of elements. Mark each as internal or external.
3. Separate the external elements. Move everything that can be done with the machine running outside the stop: fetch tools, stage material, pre-heat, inspect parts. This step needs no new equipment, only a checklist and discipline. The LeanProduction.com guide says that this step alone can nearly halve changeover time.
4. Convert internal to external. Ask for each remaining internal element: can we do this while the machine runs? Options include preparing the next tool on a second fixture, using standard clamps so that adjustment is not needed, or presetting settings off the machine.
5. Streamline what remains. Reduce the time of the internal work that cannot move. Replace bolts with quick clamps, remove adjustments with fixed stops and use a second person to work in parallel. Buy or build equipment last.

Run the workshop with the operators who do the changeover, the maintenance fitter and a supervisor. Watch the video together and ask of every element: why is the machine stopped for this? Keep a list of ideas and test one at a time on the next changeover. Record the time after each trial, so you know which change worked.

Then standardize the new method on a one-page instruction, train all shifts and measure again. The guide notes that improvements in preparation and organization are typically faster and cheaper than technical changes, so do them first.

Example: a 60-minute changeover

This is an illustration with made-up numbers.

A press makes 30 parts per hour. One setup takes 60 minutes and happens five times a week. The team records one changeover on video and lists 40 elements. The result:

• 24 minutes are external work done while the press is stopped, for example fetching the die and the tools, and preparing material.
• 36 minutes are internal work, for example removing and fitting the die and adjusting the stops.

Step 3, separate external elements. The team moves the 24 minutes outside the stop with a checklist and a prepared trolley. The stop falls from 60 to 36 minutes.

Step 4, convert internal to external. The team presets the new die height off the press and uses standard clamps. They move 8 minutes of internal work outside. The stop is now 28 minutes.

Step 5, streamline. Two people work in parallel and fixed stops replace trial-and-error adjustment. They save a further 8 minutes. The stop is now 20 minutes.

The result of this SMED example: 60 − 20 = 40 minutes saved per changeover. Over five changeovers a week, that is 200 minutes, or 3 hours 20 minutes, of extra run time. At 30 parts per hour, that is 100 more parts per week from the same machine, or room to cut the batch size and still hold output. The figures are an illustration. Your own data will differ.

How does changeover reduction connect to the rest of the plant?

Shorter changeovers are a means, not an end. The gains show up in three places:

• Smaller batches. When setups are quick, you can make smaller lots more often. That reduces work in progress and shortens lead time.
• Availability. Changeover is a part of planned stops, so it lowers machine availability in OEE (overall equipment effectiveness). Less changeover time raises it.
• Schedule flexibility. The planner can change the sequence at short notice without losing hours.

Treat SMED as a kaizen activity. Kaizen is the Japanese word for continuous improvement through small, repeated steps. After the first workshop, schedule a review in a few weeks, check the measured time against the new standard and remove the next set of elements. Rotate the team so that more operators learn the method.

A value stream map helps to choose the machine. Look for steps where long changeovers force large batches and queues.

Common mistakes

• Measuring only the fitter's time. Include cleaning, trial parts and approval. Otherwise you improve the wrong span.
• Starting with equipment. Buying quick-change tooling before you separate external work wastes money. Organization comes first.
• Leaving out the operators. The people who do the changeover know where time goes. A workshop without them produces a wrong list.
• No standard. If the new method is not written down and trained, it drifts back within weeks.
• Starting with the longest changeover. Pick a machine where you can win in a few weeks, then use the result to build support.
• Skipping quality. A changeover is finished at the first good part. Faster setups that need more trial parts are not an improvement.

How software can help

Software does not shorten a setup on the machine, but it helps around it. A plan that groups similar jobs and orders them to reduce setups reduces the number of changeovers in the first place. iFAKT's Polaris Optimizer automatically generates optimized production plans under constraints such as resources, shifts, zones, qualifications and capacities. Use it, or any planning tool, in addition to the work on the floor, not instead of it.

Next step

Pick one machine with frequent changeovers, film the next changeover and list every element as internal or external. That list is the start of your SMED project, and the first external elements you move outside the stop will pay for the time.