Illustrative sample. The plant, the weeks and the footage are synthetic.

Cut changeover time with SMED
Throughput

Cut changeover time with SMED

Approved fixes get an update every week. Paused, Declined and Done do not. In this sample the weekly reports are already written, so changing a status here does not change them.

We watch how long a job switch takes on each press, from the press cameras and the ERP job log. Target: about 30 minutes faster per job switch.

$54,500USD/ yr EBITDA

Extra profit a year once this is running, after what it costs to run. Profit, not revenue.

~30min faster / changeover

Minutes cut off each mold change, time the press spends making parts instead of sitting stopped. It pays once you fill it with sold work or smaller batches.

high confidence

Capex

$22,000

Payback

5 months

Difficulty

2 / 5

Time to implement

~12 weeks

The finding

  1. This plant molds plastic parts on about 26 injection presses and runs a high mix, so the presses are constantly stopped to swap one heavy steel mold for the next.

    • That swap, the changeover, pulls one mold out and sets another in to start the next job, and in a high-mix shop that happens all day, so it is the single largest planned stoppage on the floor.
    • Here it also runs long and varies wildly from one crew to the next, about 63 minutes at the median and twice that on a bad one, so nobody can schedule around it.
    • Walkthrough videoOperators walking back and forth for tools while the press sits stopped during a mold change, with the next mold not staged and the overhead crane tied up for the swap.
    • Shop recordsOne crew takes about twice as long as another on the identical changeover, and setup times are not written down, so the schedule around a changeover is a guess.
    • Operator feedbackAbout 2,200 changeovers a year across the fleet, at a median of about 63 minutes.
    • Read togetherThe walking on film explains the crew-to-crew spread in the records, and the changeover count turns a minutes problem into a press-hours problem; neither the film nor the count says much alone.

  3. SMED, single-minute exchange of die, cuts the dead time in a mold change by doing everything possible while the press still runs, then making the actual swap fast.

    • Stage the next mold, pre-heat it, and kit the tools while the press is still making parts, then speed the swap itself with quick clamps and quick-disconnect water and hydraulic couplers.
    • Video a few real changeovers, split the tasks that need the press stopped from the ones that do not, and move as many as you can to while-it-runs.
    • Add magnetic platens or quick clamps only on the presses where the manual bolt-up is the proven bottleneck, since that hardware runs about $10K to $30K a press.
    • Budget about $22K for staging carts, training, and hardware on the worst-changeover presses.
    • The mechanismEvery minute cut from a changeover is a minute the press spends molding, and because the freed hours come straight off the existing schedule they are the easiest capacity on the dashboard to use; shorter changeovers also let you run smaller batches, which releases tied-up work-in-process cash on top.
    • The mathGetting the worst-changeover presses from about 63 minutes down under 30 saves roughly 30 minutes a change; at about 2,200 changeovers a year that is about 1,100 press-hours back, and at about 90% filled at a $55 machine-hour it is about $55K a year, for about $22K once, paid back in about five months.
    • Industry evidenceThe 63-minute median is the Guidewheel changeover benchmark for shops like this, and documented SMED programs routinely halve it; this card claims 30 minutes, not the half.
      • Staging carts and kitting are the actual fix, not the clamps; buying quick clamps to solve a staging problem wastes the hardware money.

The math

Press hours freed1,100

30 min × 2,200 changeovers / 60

Value once the hours are filled$54,450/ yr

1,100 hrs × $55/hr × 90% filled

Minutes saved per changeover
min
Changeovers per year
Share of freed hours used
%
Staging and quick-change hardware
$
Value of a freed press hour
$/hr

Deep dive

    • The whole value is minutes saved times changeovers per year.
    • On a high-mix shop with thousands of changeovers, thirty minutes each adds up fast.
    • On a floor where a handful of tools run for weeks at a time, the same thirty minutes barely moves the total.
    • This is the most fragile assumption.
    • What it looks like on the floor if this is wrong: the presses run long jobs, so shaving the swap saves real time per change but there are too few changes for it to matter.
    • A faster changeover creates press time; that time is only worth money if it runs sold work or lets you shrink batches and release cash.
    • The bar is lower than the OEE card because the batch-shrink path pays without new sales, but it is not automatic.
    • What it looks like on the floor if this is wrong: changeover time drops and the presses simply idle longer between the same jobs, so nothing reaches the P&L.
    • A big part of the saving comes from two people running the staged tasks at once instead of one person doing them in sequence.
    • If only one setter is ever available, the parallel plan never happens and the change stays slow.
    • What it looks like on the floor if this is wrong: one person does the whole swap alone while the second setter is tied up on another press.
    • Quick clamps and couplers assume the mold seats cleanly and the same way every time.
    • A worn or damaged mold fights the fixture and the swap drags no matter how good the procedure is, which is why the maintenance card helps this one.
    • What it looks like on the floor if this is wrong: a bent locating ring or a leaking coupler turns a ten-minute clamp-up into a half-hour fight.
Load-bearing
The card breaks if this is wrong.
Supportive
The card weakens but survives if this is wrong.
  • Risk

    SMED is a discipline, not a purchase. Without audits, crews drift back to the comfortable old sequence and the changeover times climb back to where they started.

    Early warning

    Changeover times on the monitoring dashboard creep upward week over week.

    Mitigation

    Standard-work audits tied to the uptime data, with the drifting shift retrained.

    At stake

    Most of the $55K.

  • Risk

    Quick clamps speed the bolt-up, but if the next mold and its tools are still not staged, the press waits anyway and the hardware never earns back.

    Early warning

    Presses still wait on molds and tools even after the clamps go in.

    Mitigation

    Fix the staging carts and kitting first; buy hardware only where the clamp-up is the proven bottleneck.

    At stake

    The entire hardware ROI.

  • Risk

    A rushed startup on a fresh change can ship out-of-spec first parts before the process settles.

    Early warning

    Startup scrap spikes right after a changeover.

    Mitigation

    First-article checks and a documented startup recipe on every change.

    At stake

    A customer return, plus a possible re-validation.

  • Risk

    The change gets fast but nothing is done with the recovered time, so the presses just idle longer between the same jobs.

    Early warning

    Changeover times fall while output and work-in-process stay flat.

    Mitigation

    Decide up front whether the freed time buys new work, smaller batches, or on-time delivery, and hold someone to it.

    At stake

    The capacity value, though the batch-shrink cash release still lands.

A. Internal versus external, the whole idea in two words

Every task in a mold change is one of two things. Internal tasks need the press stopped: unclamp the old mold, lift it out, set the new one, bolt it up, hook the water and hydraulics. External tasks can be done while the press is still making parts: find the next mold, pre-heat it, gather the tools, stage the crane. Most shops do far too many external tasks after the press stops, wasting paid press time. SMED is mostly the discipline of moving that prep to before the press stops, so the actual dead time shrinks to the few things that truly require a stopped machine.

B. Kill the variability, not just the average

The 63-minute median is not the real problem; the shift-to-shift spread is. When the same job takes one crew 40 minutes and another 100, the scheduler cannot plan, so the whole floor pads its estimates and the plan drifts. A reliable, repeatable 30-minute change is worth more than a one-off 15-minute change that only the best setter can hit. Standardized sequences and staged carts attack the spread directly, which is why on-time delivery often improves more than the raw minutes would suggest.

C. Hardware only where the bolt-up is the bottleneck

Quick clamps, magnetic platens, and multi-couplers are real and they work: vendors document changes dropping from an hour to fifteen minutes on a single press. But they fix exactly one thing, the manual clamp-and-hookup step. If the press is waiting because the next mold was never staged, no clamp will help. Prove with the video that the bolt-up is the remaining bottleneck before spending $10K to $30K a press on hardware, or the money sits idle bolted to a press that is still waiting on a cart.

D. Where the money actually lands

Freed changeover hours convert three ways, and this card is firmer than the OEE card because two of them do not need new sales. First, fill the hours with sold work (the capacity gate, same as everywhere). Second, run smaller batches, which releases work-in-process cash straight off the balance sheet. Third, hit delivery dates you were missing, which protects and wins customers. Because the batch-shrink and delivery paths pay without waiting on the sales pipeline, most of the recovered hours here are readily usable, which is why the value is discounted only lightly.

The records we did not have for this read and would ask for on day one. Each one replaces an estimate on this card with your own number.

  • Owner
    Setup technicians / plant manager
    Why

    To confirm the roughly 63-minute median and, more important, the shift-to-shift spread on this floor, which sets both the minutes-saved figure and how much predictability there is to gain.

  • Owner
    Scheduling
    Why

    This is the multiplier the whole saving scales off. A shop running thousands of changes a year gets far more from thirty saved minutes than one running a few hundred.

  • Owner
    Tool room
    Why

    To see whether staging or clamping is the real bottleneck, which decides whether the money goes to carts and kitting or to quick-change hardware.

  • Owner
    Quality
    Why

    To size the startup-waste reduction from a standardized change and to set the first-article check that keeps faster changeovers from causing escapes.