SMED vs Changeover: Why "Faster Changeover" Is Not the Same as SMED

Key takeaways

• Changeover = the activity of switching the line from one product/SKU to another.
• SMED (Single-Minute Exchange of Die) = a structured methodology to cut changeover time, usually targeting under 10 minutes.
• The SMED method separates internal (must stop the line) from external (can happen while running) steps, then converts internal to external where possible.
• Most plants attempt "faster changeover" by working harder; SMED makes it faster by working differently.
• Cutting changeover from 60 minutes to 10 minutes typically frees 5-10% of total Availability — a major OEE move.

Short answer: Changeover is the activity of switching the line from one product to another. SMED is a specific methodology — separate internal from external steps, convert internal to external where possible, streamline what remains — to make changeover dramatically shorter. Plants that say "we want faster changeover" without using SMED tend to make modest gains. Plants that use SMED routinely cut changeover by 50-90%. See also SKU Changeover Sequence Optimization.

What changeover is

Changeover is everything that happens between the last good part of SKU A and the first good part of SKU B. Tooling change, parameter setting, material swap, line clear, first-piece inspection.

It is an Availability loss in OEE — time the line is not producing. High-mix plants can spend 20-40% of scheduled time in changeover.

What SMED is

Single-Minute Exchange of Die (SMED) was developed by Shigeo Shingo at Toyota in the 1950s-60s. The method has four steps:

1. Separate internal from external. Internal steps require the line to be stopped. External steps can happen while the line is still running the previous SKU. Most plants mix them — that is the first big win.
2. Convert internal to external. Pre-heat tools, pre-position fixtures, stage materials. Move as much work as possible to before-stop time.
3. Streamline both. Standardize tools, eliminate adjustments (use locating pins, quick-clamps), pre-set parameters via recipes.
4. Standardize and train. Document the new changeover. Train all shifts. Audit compliance.

The name comes from the target — single-digit minutes, under 10. Aggressive SMED programs hit under 1 minute.

How "faster changeover" usually fails

Plants without SMED typically attack changeover by:

• Telling operators to hurry.
• Adding more people to the changeover.
• Buying faster equipment.
• Pre-staging some materials.

These produce 10-30% gains. SMED produces 50-90% gains by changing the structure of the work, not just the speed.

The internal/external distinction

This is the heart of SMED. Examples:

• Bringing the new die to the press — usually internal (line stopped while moving die). Convert to external by staging the die before the previous run ends.
• Pre-heating the die — usually internal (heating while line is stopped). Convert to external by pre-heating in a separate station while the previous run continues.
• Setting parameters in the controller — internal. Convert to external by storing recipes and loading the new recipe before the line stops.

Every step gets the question: must this happen with the line stopped, or could it happen earlier?

What "streamline" actually means

After internal/external separation, the remaining internal steps get streamlined:

• Eliminate adjustments. Use locating pins, hard stops, quick-locking fixtures so position is guaranteed by hardware, not by operator adjustment.
• Standardize fastener sizes. Operators carry one tool, not three.
• Pre-set parameters. Recipe-driven, not operator-tweaked.
• Parallel work. Two operators changing over from opposite sides of the line at the same time.

What SMED is worth

For a plant running 8 changeovers per week at 60 minutes each, that is 8 hours of weekly downtime. Cutting to 10 minutes (a typical SMED outcome) reclaims about 6.5 hours per week — roughly 5% of a one-shift week or 5-10% of total Availability.

That is a major OEE move. Few capital investments deliver that for the cost of a SMED program.

Common mistakes

1. Treating SMED as a one-time event. Real gains compound. Run SMED on every SKU pair, not just the top one.

2. Skipping the internal/external separation. Going straight to "streamline" misses the biggest gains.

3. Not documenting the new procedure. The new changeover only stays fast if every shift does it the same way.

4. Setting goals without measuring. "Cut changeover" without a stopwatch produces guesses.

How OEE platforms support SMED

A modern OEE platform measures changeover duration automatically — line stop reason "changeover" with start and end timestamps. It exposes changeover time per SKU pair, per line, per shift, so the team can target the highest-impact SKU pair for SMED work and verify the gains.

Fabrico's OEE module captures changeover events with start/end timestamps, surfaces the worst SKU-pair transitions for SMED prioritization, and tracks changeover time trends so the gains from a SMED workshop are visible the next shift.

See how Fabrico captures this automatically — explore OEE for manufacturing or book a demo.

Related reading

• SKU Changeover Sequence Optimization

Frequently asked questions

What does SMED stand for?

Single-Minute Exchange of Die. The "single minute" refers to single-digit minutes (under 10), not literally one minute.

Is SMED only for press operations?

No. The method applies to any changeover — packaging line changeover, CNC tool change, batch reactor recipe switch, etc.

How long does a SMED program take?

A focused SMED workshop on one SKU pair is typically 3-5 days. Plant-wide rollout across all SKU pairs is months to a year.

Can SMED reach under 1 minute?

Yes for some processes, especially with pre-staged fixtures and recipe-driven control. Aggressive SMED programs achieve sub-minute changeovers.

Does SMED require new equipment?

Sometimes. The streamline step may require quick-clamps, locating pins, or recipe systems. Most gains come from reorganizing existing work first.