Key takeaways
Short answer: Internal and external setup are the two halves of any changeover, and telling them apart is the key to making changeovers fast. Internal setup is work that can only be done with the machine stopped — so it is pure downtime. External setup is work that can be done while the machine is still running the previous job — so it costs no production time at all. The entire SMED method rests on this distinction: convert as much internal setup to external as possible, then streamline what truly must stay internal. For where this fits in lean leveling, see heijunka vs kanban.
Internal setup is changeover work that can only be performed while the machine is stopped — the previous job done, the next not yet started. Removing a die, swapping a fixture that requires the machine to be at rest, making an adjustment that can only be done with the equipment halted: all internal. The defining feature is that internal setup is genuine downtime — the machine is not producing while it happens, so every minute of internal setup is a minute of lost availability. This is the time that actually hurts, and it is therefore the time changeover improvement is really aimed at. The less internal setup a changeover requires, the less production it costs.
External setup is changeover work that can be done while the machine is still running the previous job — preparation that does not require the equipment to be stopped. Fetching the next die and staging it beside the machine, pre-heating a tool, pre-kitting fasteners, preparing paperwork, checking the next job's materials: all of this can happen during the current run. The defining feature is that external setup costs no production time, because the machine keeps making good parts while it is done. The insight that transformed changeover practice is that a large share of what teams traditionally do during the stoppage is actually external work — it was only done internally out of habit, not necessity.
Single-Minute Exchange of Die (SMED) is built on one powerful move: take work currently done as internal setup and convert it to external, so it happens while the machine still runs. The classic approach is to first separate internal from external (many teams blur them, doing external prep during the stoppage), then convert as much internal to external as possible, and finally streamline whatever must remain internal. The results are often dramatic, because so much of a traditional changeover turns out to be preparation that never needed the machine stopped at all. The goal — single-minute, meaning under ten minutes — becomes achievable not by rushing, but by moving work off the critical stopped-machine path.
A press changeover takes 40 minutes, all of it counted as downtime. The team studies it and classifies each step. They find that 15 minutes was external work done internally out of habit: fetching the next die from storage, finding the right bolts, reviewing the job sheet — all of which could happen while the press was still running the previous job. They move that work to before the stop. Of the 25 minutes truly internal, they streamline further: pre-set bolt positions, quick-clamps instead of threaded fasteners, a staging cart. The changeover drops from 40 minutes of downtime to perhaps 12. The machine now stops for 12 minutes, not 40 — and that 28-minute recovery is pure availability, repeated every changeover.
The reason internal-versus-external is the heart of changeover improvement is simple: only internal setup stops production. External setup, however long, costs no availability because the machine keeps running through it. So the entire leverage of faster changeovers comes from minimising internal time — first by converting internal work to external, then by streamlining the irreducible internal core. A team that just tries to do the whole changeover faster, without separating the two, is rushing work that did not need to be on the stopped-machine path at all. Get the classification right and the path to dramatically shorter downtime becomes obvious; skip it and you are sprinting in the wrong place.
Changeover time is a headline availability loss — one of the six big losses (setup and adjustment) — so cutting internal setup directly lifts the availability factor of OEE. But the impact is bigger than the availability arithmetic suggests. Long changeovers push teams toward large batches to amortise the setup, which inflates inventory and lead time; fast changeovers make small batches economic, enabling the smoother flow and leveling of heijunka and pull production. So reducing internal setup raises OEE directly and unlocks flow improvements that raise it further. Changeover is one of the highest-leverage losses to attack.
Fabrico measures changeover losses specifically — how long each changeover actually takes and how much availability it costs — so the gains from separating and converting internal setup are visible rather than anecdotal. By tracking changeover duration over time and by machine, it shows which changeovers are the biggest availability drains, where SMED effort will pay off most, and whether an improvement actually stuck. That turns faster changeovers from a workshop exercise into a measured, sustained gain. Book a demo to see your changeover losses quantified.
Internal setup is changeover work that can only be done with the machine stopped, so it is downtime. External setup is work that can be done while the machine is still running the previous job, so it costs no production time. Only internal setup stops production.
SMED (Single-Minute Exchange of Die) is a method for dramatically reducing changeover time. Its core moves are to separate internal from external setup, convert as much internal setup to external as possible, and then streamline the remaining internal work — aiming for changeovers under ten minutes.
Because only internal setup stops production. Work done externally, while the machine still runs, costs no availability. Converting internal work to external moves it off the stopped-machine path, directly shrinking the downtime each changeover causes.
Long changeovers push teams toward large batches to amortise the setup, inflating inventory and lead time. Fast changeovers make small batches economic, enabling smoother flow, leveling, and pull production — so the benefit compounds beyond the direct availability gain.
Changeover time is part of the setup-and-adjustment loss, a headline availability loss in OEE. Cutting internal setup directly raises availability, and the smaller batches it enables improve flow and performance, raising OEE further.
Zaloguj
Blog
