SMED (Single-Minute Exchange of Die)
SMED — Single-Minute Exchange of Die — is Shigeo Shingo's methodology for reducing equipment changeover time from hours to under 10 minutes (single digits of minutes, hence 'single-minute'). Developed at Toyota in the 1950s-60s, it was the breakthrough that made Just-In-Time possible: if you can switch a press from making fenders to making doors in 3 minutes instead of 4 hours, you can run small batches economically and stop carrying mountains of inventory. The core trick: separate setup tasks into INTERNAL (must be done with the machine stopped) and EXTERNAL (can be done while the machine is still running the previous job). Most setups are 80% external work being done internally because no one questioned the order. KnowMBA take: SMED applies directly to deploys, environment switches, and meeting transitions in software — anywhere a 'changeover' tax stops you from running smaller, more frequent batches.
The Trap
Engineers buy faster equipment to shrink changeover time and skip the methodology. New machine, same 3-hour changeover, because the bolts, dies, and tooling staging weren't redesigned. The other trap: averaging changeover times across operators hides the real problem — your best operator does it in 22 minutes, your worst in 95, and the 'average' of 47 minutes obscures that the answer is just to standardize the best method. And teams routinely 'reduce' changeover by working faster instead of working differently — burning out staff while leaving the structural waste untouched.
What to Do
Pick one changeover that costs you the most over a month (frequency × duration × bottleneck-status). Video the entire changeover from line-stop to first good part. Watch the video as a team and tag every step as INTERNAL (must stop machine) or EXTERNAL (could do while running). Convert as many internal tasks to external as possible (pre-stage tools, pre-warm dies, pre-fetch material). Then streamline what's left of internal work (parallel operators, quick-release bolts, standardized fixtures). Target: cut total changeover by 50% in the first cycle, 75%+ within 3 cycles.
Formula
In Practice
In 1969 at Toyota's body shop, Shigeo Shingo cut the changeover time on a 1,000-ton press from 4 hours to 3 minutes. The first wave (separating internal/external work, pre-staging dies on roller carts) dropped it to 90 minutes. The second wave (parallel operators, hydraulic clamps replacing manual bolts) hit 12 minutes. The third (standardized die heights so no shimming was needed) finished it at 3 minutes. Toyota then ran batch sizes 1/100th the size of competitors — meaning they carried 1/100th the inventory and could respond to demand shifts in days instead of months. SMED is arguably the single most important enabler of the Toyota Production System.
Pro Tips
- 01
Shingo's rule: assume EVERY internal task can become external until proven otherwise. Don't ask 'can we do this faster?' — ask 'why is the machine stopped while we do this?' That single reframe usually finds 60%+ of internal work that can be moved external.
- 02
The 'one-touch' setup ideal: changeover requires a single bolt, lever, or button — no shimming, no wrenches, no measuring. Quick-release clamps, alignment pins, and standardized die heights make this real. Most plants think this is unrealistic; Toyota lives there.
- 03
For SaaS: deploys are changeovers. A 30-min deploy means you deploy weekly; a 30-second deploy means you deploy 50x/day. Treating deploy time as a SMED problem (parallelize tests, pre-build images, blue-green so the cutover is one DNS flip) is what separates teams that ship daily from teams that batch monthly.
Myth vs Reality
Myth
“SMED requires expensive new equipment”
Reality
Shingo's first 90-minute reduction at Toyota cost almost nothing — it was reorganizing the order of operations and pre-staging tools. Equipment investment comes later, AFTER you've extracted the free wins from method change. Companies that buy first and rethink second waste 80% of their capex.
Myth
“Changeover time doesn't matter if we run big batches”
Reality
Long changeovers FORCE big batches, which create inventory, hide quality problems, slow response to demand changes, and tie up cash. The strategic value of SMED isn't the saved minutes — it's that you can profitably run smaller batches, which unlocks JIT, faster lead times, and lower working capital.
Try it
Run the numbers.
Pressure-test the concept against your own knowledge — answer the challenge or try the live scenario.
Knowledge Check
Challenge coming soon for this concept.
Industry benchmarks
Is your number good?
Calibrate against real-world tiers. Use these ranges as targets — not absolutes.
Changeover Time After SMED Program
Discrete manufacturing equipment (presses, injection molders, packaging lines)World-Class (Single-Minute)
< 10 min
Strong
10-30 min
Acceptable
30-60 min
Weak (SMED opportunity)
60-180 min
Crisis
> 3 hours
Source: Shigeo Shingo, A Revolution in Manufacturing: The SMED System
Real-world cases
Companies that lived this.
Verified narratives with the numbers that prove (or break) the concept.
Toyota (Body Shop, Press Operations)
1969-1970
Shigeo Shingo, brought in by Taiichi Ohno, attacked the changeover time on a 1,000-ton press at the Toyota body shop. Original time: 4 hours. Shingo videotaped the changeover, classified each step as internal or external, and found that ~70% of stopped-machine time was spent on tasks that could happen while the previous run was still going (fetching dies, paperwork, die preheating). Wave 1 (separation): 90 min. Wave 2 (conversion of internal to external using die preheating racks): 12 min. Wave 3 (one-touch hydraulic clamps, standardized die heights): 3 min. This became the foundation of Just-In-Time at Toyota.
Changeover Before
240 min
Changeover After
3 min (98.75% reduction)
Capex
Modest (clamps, racks)
Enabled
Just-In-Time, small-batch production
The biggest gains came from rethinking which work needed the machine stopped — not from buying new equipment. Method change beats capital spend.
Hypothetical: Mid-Market Injection Molder
Recent
A 120-employee injection molder ran 75-min changeovers between products. They averaged 8 changeovers per week — 10 hours of lost production. Sales said they couldn't take small custom orders because the changeover tax made anything under 5,000 units unprofitable. A 6-week SMED kaizen (videotape, separate, convert, streamline) cut changeover to 14 minutes. Suddenly, 800-unit custom orders were profitable. Revenue from short-run business grew 35% in 6 months and overall plant utilization improved because they stopped artificially batching.
Changeover Before
75 min
Changeover After
14 min
Capex
$22K (clamps, staging carts)
New Revenue Unlocked
+35% short-run business
SMED's strategic value is the new business it unlocks — not just the minutes saved. Long changeovers are silent revenue killers because they make small orders look unprofitable.
Related concepts
Keep connecting.
The concepts that orbit this one — each one sharpens the others.
Beyond the concept
Turn SMED (Single-Minute Exchange of Die) into a live operating decision.
Use this concept as the framing layer, then move into a diagnostic if it maps directly to a current bottleneck.
Typical response time: 24h · No retainer required
Turn SMED (Single-Minute Exchange of Die) into a live operating decision.
Use SMED (Single-Minute Exchange of Die) as the framing layer, then move into diagnostics or advisory if this maps directly to a current business bottleneck.