SMED — Single-Minute Exchange of Die
In plain English
SMED stands for Single-Minute Exchange of Dies.
It’s a method for reducing changeover time — the time it takes to switch from producing one product to another.
“Single-minute” doesn’t mean 1 minute.
It means changeovers measured in single digits (less than 10 minutes).
But here’s the important part:
SMED isn’t about speed.
It’s about enabling small batches — which enables flow.
Without short changeovers, you batch.
When you batch, you create Mura.
What they actually mean
Most companies say they want flow.
Then they defend long setup times.
Because long setups justify large batches.
Large batches feel efficient.
Until downstream starts drowning.
SMED challenges the assumption that:
“Long setups are just how it is.”
It asks:
What if most of that time is unnecessary?
And the uncomfortable answer is often:
It is.
“We Can’t Reduce Setup” Is Usually a Myth
Most long setups are not technical limits.
They are:
• habits
• missing preparation
• unclear roles
• searching for tools
• waiting for approvals
• doing internal work externally
SMED splits changeover work into:
Internal (must stop machine)
External (can be done while running)
That distinction alone often cuts setup time dramatically.
But here’s the catch:
Reducing setup removes the justification for batching.
And batching protects utilization metrics.
So resistance appears.
Not from operators.
From systems.
The uncomfortable truth
If your setup is long:
You are buffering inefficiency with inventory.
Not optimizing performance.
Example
A machine takes 90 minutes to change over.
So production runs 3 weeks of one product before switching.
Looks efficient locally.
Downstream?
• Inventory builds
• Lead time increases
• Demand shifts
• Priorities change
• Firefighting begins
Reduce setup to 15 minutes.
Suddenly small batches are viable.
Suddenly flow becomes possible.
Nothing magical happened.
You removed the excuse to batch.
🔗 Lead Time
🔗 Bottleneck
🔗 Heijunka
Where you’ll hear it
Industrial engineering meetings.
Kaizen events.
Sometimes shouted in frustration:
“Why does this take so long every time?”
Does it actually matter?
✅ Yes.
Because long changeovers force:
• large batch sizes
• unstable schedules
• uneven workload
• long lead times
Reducing setup time doesn’t just speed machines.
It increases flexibility.
And flexibility stabilizes flow.
Common misconceptions
SMED is only for stamping presses.
SMED requires new machines.
Setup time is fixed.
Reducing setup only improves speed.
Reality:
SMED is about analysis, separation of tasks, preparation, and simplification.
It applies anywhere work transitions between states.
Manufacturing.
Healthcare.
Even project handoffs.
Red flags
🚩 If setup time has “always been like that.”
🚩 If tools are gathered after the machine stops.
🚩 If changeover relies on tribal knowledge.
🚩 If reducing setup is seen as “nice to have.”
🚩 If batching is defended as efficiency.
That’s not Lean.
That’s protection of local optimization.
Worth learning?
5/5
Because without short changeovers,
flow is theoretical.
Deep dive
A short history of SMED
SMED (Single-Minute Exchange of Die) was developed by Shigeo Shingo in Japan while working with Toyota in the 1950s–60s.
At the time, large stamping presses took several hours to change over between products. That forced companies into large batch production — not because they wanted to, but because they had no choice.
Shingo’s breakthrough wasn’t just “work faster.”
It was separating setup work into:
• Internal setup (must be done while the machine is stopped)
• External setup (can be done while it’s running)
By systematically converting internal work into external work, Toyota reduced changeovers from hours to minutes.
That made small batches economically viable.
And small batches are what make flow possible.
SMED wasn’t about speed.
It was about freedom.
How SMED actually works
The core logic:
Observe and document the current changeover.
Separate internal vs external tasks.
Convert internal tasks to external where possible.
Streamline and simplify remaining internal work.
Standardize the new process.
Typical improvements include:
• Pre-staging tools
• Quick-release mechanisms
• Parallel operations
• Visual controls
• Standard work
🔗 Standard Work
🔗 Flow
🔗 Takt Time
Why SMED is strategic
SMED isn’t about saving minutes.
It’s about reducing economic pressure to batch.
Short setup → smaller batches → smoother flow → lower WIP → shorter lead time.
It’s one of the mechanical foundations of Lean.
Without it, Heijunka becomes very hard.
🔗 Heijunka
🔗 WIP
🔗 Little’s Law,
Report an error
Found something wrong or misleading? Let us know — we want this site to stay fact-based (even when we joke).