Track changeover time first, then SMED it down

SMED — single-minute exchange of die — gets taught as a set of clever techniques: quick clamps, die carts, color-coded tooling. All of that works. But the teams that actually cut changeover time start somewhere less glamorous: they measure the changeover they have now, step by step, before they change a single thing. You cannot shrink a number you have never written down. This article shows you how to time a changeover properly, how to split the work into the two categories SMED cares about, and what to capture in your log so the improvement sticks.

If you have not set up a downtime log yet, read how to track machine downtime first — a changeover is just one category of stop, and the timing here assumes you already capture stops cleanly.

Why you measure before you improve

A changeover is planned downtime. The press is scheduled to run, and instead it sits idle while someone swaps a die, threads new stock, and dials in the first good part. Every minute of that is production you planned for and did not get. On a line that changes over three times a shift, a 48-minute changeover is nearly two and a half hours of lost run time per shift — more than a full shift a week.

The reason to measure first is that gut feel is always wrong about where the time goes. Ask the crew what takes longest and they will say "setting the die." Time it, and you will often find the die set takes eight minutes while hunting for tools and the right coil took fifteen. You do not know until you have a stopwatch and a breakdown. The measurement is the project; the techniques come after.

Split internal work from external work

SMED rests on one distinction, and it is worth getting exactly right:

The first and biggest win in every SMED effort is not speeding work up. It is moving work from internal to external — doing it before the press stops instead of after. A tool that is already at the press when the line goes down costs zero changeover minutes. The same tool fetched from the crib after the stop costs you every minute of that walk.

So when you time the changeover, you are not just getting a total. You are tagging each step as internal or external as it is done today, so you can see how much of your stopped time is work that never needed the machine down at all.

Internal work can only happen while the machine is stopped. Removing the old die, bolting in the new one, setting shut height — the press has to be down for all of it.
External work can happen while the machine is still running the previous job. Staging the next die, gathering tools, pre-heating, stacking the new coil at the press.

A worked changeover breakdown

Take a 400-ton stamping press that changes from one stamped bracket to another. The crew does not touch anything until the last good part drops, then starts the swap. Here is the changeover, timed step by step, with each step tagged by how it is run today and whether it could be external.

Every step runs internal today, because nothing starts until the press is down. That is the opportunity staring back at you.

Now apply SMED in two passes.

Pass one — convert work to external. The 7-minute retrieval is pure external work done at the wrong time. Stage the die, clamps, and coil at the press before the last part runs, and those 7 minutes leave the stopped clock entirely. Pre-cleaning the staging cart and pre-threading the coil leader shave another 2 minutes off the bolster and threading steps. That is 9 minutes moved out, with no new tools bought.

Pass two — streamline what is left. Now attack the genuinely internal work. Swap bolts for quick-clamps (6 minutes down to 2), fit locating pins so die alignment is drop-in rather than eyeballed (10 down to 6), pre-set shut height on the die cart (3 down to 1), and keep a go/no-go gauge at the press so the first-piece check is a fit test instead of a trip to the inspection bench (3 down to 1). That is another 12 minutes off the stopped clock.

The press is now down 27 minutes instead of 48 — a 44% cut — and most of the first-pass win came from doing existing work at a better time, not from spending money. That is the pattern SMED delivers over and over, and you only see it because you timed and tagged the baseline.

What to log at the machine

A one-time stopwatch study is a good start, but the gains fade unless the log keeps measuring every changeover. Capture these fields each time:

With that logged consistently, you can build a Pareto of your changeovers and see which machines and which delays eat the most time — so the next SMED project aims at the vital few instead of the easy few.

Machine — which line or press. Changeover time varies a lot from one machine to the next, and a plant-wide average hides that.
Start and end timestamps — start when the last good part drops, end when the first good part is approved. Anything narrower undercounts and flatters you.
Planned/unplanned flag — mark it planned. A changeover is not a breakdown, and mixing the two corrupts both your reliability metrics and your changeover trend.
A short reason or note when it runs long — "waited on crane," "die needed rework," "coil jam." These notes are what turn a number into a fixable cause. If a particular swap is always the slow one, name the from-and-to parts in the note so it is easy to find later.

Changeover time rolls up into availability

Changeover is one of the planned losses sitting between your scheduled hours and your run hours, so it shows up directly in the availability term of OEE. Cut the stamping line from 48 to 27 minutes across three changeovers a shift and you hand back an hour of run time every shift — availability rises without buying a minute of new capacity. If you want the full picture of how stops feed that percentage, see how downtime feeds OEE availability. Faster changeovers are the planned-loss lever; fewer breakdowns are the unplanned one, covered in how to reduce unplanned downtime.

Where HelpFab fits

You can run the first study with a clipboard and a stopwatch, and you should — nothing beats standing at the press. The hard part is keeping the measurement alive after the kaizen week ends. That is where a shared log earns its keep.

In HelpFab, operators log a changeover from a phone or tablet the moment the press goes down, using a changeover issue type with its own category, so the stop is captured in a few taps without paperwork. Because each type carries a planned flag, changeovers are counted as downtime but kept separate from breakdowns, and the Downtime Reports page lets you view planned stops on their own. The trend and Pareto reports then show changeover time by machine and reason over any date range, so you can see whether last quarter's SMED work actually held and which press to target next.

Whatever tool holds the data, the method does not change: time the changeover you have, tag every step internal or external, move external work off the stopped clock first, then streamline what is left — and keep logging every changeover so the gains do not quietly creep back.

Latest articles

CMMS vs downtime tracking software — A CMMS manages assets and work orders; downtime tracking logs every production stop. Different jobs, not competitors — here is when a small shop needs each.
An andon system for a small shop, without the light tower — A low-cost andon for a small shop - what to signal, how to escalate, and how each call becomes a downtime log entry you can actually act on.
A production downtime report example, daily and weekly — A worked production downtime report for a CNC cell, the four numbers managers actually read, and a layout that turns lost minutes into action.
MTTR and MTBF: how to track them from your downtime log — Track MTTR and MTBF from your downtime and maintenance logs - the formulas, a worked quarter on one machine, and the mistakes that corrupt both numbers.
Maintenance log vs downtime log — Maintenance log vs downtime log - what each one captures, why a plant needs both, and how a single breakdown links them to give you MTTR and MTBF.
Choosing a downtime tracking app for a small manufacturer — What a small shop should look for in a downtime tracking app: phone logging, configurable reason codes, and reports you can read without an ERP.
How to reduce unplanned downtime in manufacturing — A track-first playbook to reduce unplanned downtime - log every stop, Pareto the causes, then attack the vital few with PM and root-cause fixes.
How to calculate the cost of unplanned downtime — A cost-per-hour formula for unplanned downtime, with a worked molding-cell example covering labor, scrap, lost margin, and overhead.
A free downtime tracking Excel template (and how to actually use it) — A free Excel downtime log template — the columns, a reason-code list, and the pivot that turns raw stops into a Pareto, plus when to move off it.
OEE availability: how to calculate it from your downtime data — Calculate OEE availability from your downtime log, with a worked CNC example — plus the reclassification tricks that quietly inflate the number.
How to build a downtime Pareto analysis — Build a downtime Pareto step by step with a worked stamping-line example that finds the few causes driving most of your lost minutes.
A practical machine downtime reason code list — A starter taxonomy of machine downtime reason codes — planned vs unplanned, with real fault-code examples and rules that keep operators logging honestly.
Downtime tracking spreadsheet vs software — When an Excel downtime tracker is enough and when it quietly breaks — a shift-by-shift comparison for small manufacturers, with a real cost example.
How to track machine downtime — Move from a paper downtime log to structured tracking: reason codes, who logs what, and the three reports that actually change decisions.