Quick changeover is one of those concepts that every manufacturing
leader nods along with enthusiastically. Everyone agrees that long setup
times are wasteful. Everyone agrees that reducing them would unlock
flexibility, reduce inventory, and improve responsiveness. And yet, walk
onto most factory floors today, and you’ll find changeover times that
haven’t improved in years — sometimes decades.
The irony is that the methodology to fix this has existed since the
1960s. Shigeo Shingo developed Single-Minute Exchange of Die (SMED) at
Toyota, and it has been taught in every Lean course, every operations
management program, and every continuous improvement workshop since. The
principles are well-documented, the techniques are proven, and the
results are measurable. So why does it so consistently fail to deliver
lasting results in practice?
The answer, as with most quality methodologies, is not in the theory.
It’s in the execution — or more precisely, in the gap between
understanding the concept and actually implementing it on the shop
floor, day after day, shift after shift.
The Core Idea:
Separating Internal from External
At its heart, SMED is built on a deceptively simple distinction.
Every setup operation can be classified as either
internal (the machine must be stopped to perform it) or
external (it can be done while the machine is still
running the previous job). The entire methodology revolves around one
progressive goal: convert as many internal operations as possible to
external ones, then streamline what remains.
Shingo discovered this principle while observing a ship’s engine
room. He noticed that the crew prepared everything for an engine switch
while the ship was still moving — tools laid out, parts pre-positioned,
conditions checked. When the actual switch moment came, it took seconds,
not hours. He realized that the same principle could transform
manufacturing setups.
The traditional approach to changeover treats every step as internal.
The operator stops the machine, then goes to find tools, then checks the
specifications, then locates the right fixtures, then begins the
physical changeover. Each of these preparatory steps eats into
production time unnecessarily. SMED asks: why didn’t you have the tools
ready before the machine stopped? Why didn’t you pre-heat the dies? Why
didn’t you stage the new material?
This sounds obvious — and it is. But obvious doesn’t mean easy.
The Four Stages of SMED
Implementation
A proper SMED implementation follows four stages, each building on
the previous one. Understanding these stages is critical because most
organizations never get past Stage 1, and then wonder why their
changeover times haven’t improved.
Stage 1: Identify and
Separate
The first step is to document every single operation in the current
changeover process. Not the official procedure — the actual practice.
This means filming or timing several real changeovers with a stopwatch,
listing every action the operator takes, and then classifying each as
internal or external.
The gap between the official procedure and reality is often shocking.
Operators develop informal shortcuts, skip steps they consider
unnecessary, or add steps the procedure doesn’t mention. You can’t
improve what you don’t understand, and in most factories, the real
changeover process has never been systematically observed.
Once every operation is listed and classified, the immediate
opportunity becomes clear: move every external-capable operation out of
the machine-stop window. Bolts can be loosened while the machine is
still running. New material can be staged. Tools can be laid out.
Specifications can be reviewed. In many cases, this step alone can cut
changeover time by 30-50% — without any new technology or capital
investment.
Stage 2: Convert Internal
to External
After the obvious separations are made, the next challenge is harder:
operations that currently seem like they must be internal. Temperature
adjustments. Calibration sequences. Fixture alignments. These appear
unavoidable — but are they?
Often, the constraint is procedural rather than technical. A die that
needs to be pre-heated to 200°C can be heated in an off-line oven while
the machine is still running. A calibration that requires the machine to
be stopped might be replaceable with a pre-calibrated fixture that snaps
into position. A fixture alignment that takes twenty minutes of
fine-tuning might be eliminated entirely with a locating pin system that
makes misalignment physically impossible.
This stage requires creative thinking, and this is where many
organizations stall. The operators who know the process best are often
the most resistant to change — not because they’re stubborn, but because
they’ve been doing it the same way for years, and the current method,
however slow, is known and reliable. A new method introduces risk: what
if the pre-heated die cracks? What if the snap-in fixture doesn’t hold
tolerance? These are legitimate concerns, and they need to be addressed
with testing and validation, not dismissed as resistance to
progress.
Stage 3: Streamline
Internal Operations
Once everything that can be external has been moved, what remains are
operations that truly require the machine to be stopped. The goal now is
to make these as fast as possible. Every second counts, because every
second is a second of lost production.
The techniques for streamlining internal operations are
well-established:
- Parallel operations: If two operators can work
simultaneously on different sides of the machine, changeover time is cut
in half. Yet many machines are still serviced by one operator performing
sequential tasks when parallel work is entirely feasible. - Eliminating adjustments: Every adjustment is a time
sink. Shingo’s principle was “eliminate, don’t reduce.” If an operator
needs to adjust a bolt position, the solution is not a faster wrench —
it’s a design that eliminates the need for adjustment entirely. One-way
clamps, quarter-turn fasteners, and spring-loaded mechanisms replace
threaded bolts and manual tightening. - Standardized tooling: If every die uses the same
mounting system, changeover becomes a simple swap rather than a custom
reconfiguration. Standardization across product families is one of the
highest-leverage investments a manufacturer can make. - Functional clamps: Quick-release clamping systems —
hydraulic, pneumatic, or mechanical — can reduce fastening time from
minutes to seconds. The investment pays for itself rapidly when
changeover frequency increases.
Stage 4: Eliminate
Adjustments Entirely
The final stage represents the theoretical ideal: a changeover
process that requires zero adjustment. Every component is designed to
seat in exactly the right position every time. No dialing in, no test
runs, no fine-tuning. The new setup is correct by design, not by
operator skill.
In practice, this stage is rarely fully achieved. But the pursuit of
it drives continuous improvement in machine design, tooling standards,
and setup procedures. Organizations that take this stage seriously often
find that their changeover times drop to single digits — the literal
“single minute” that gives SMED its name.
Why SMED Fails in Practice
Knowing the methodology is not the same as implementing it. After
decades of working with manufacturing organizations on quick changeover,
I’ve seen the same failure patterns repeat with striking consistency.
Understanding these patterns is essential — not as an excuse, but as a
diagnostic tool to recognize and address them before they undermine your
implementation.
Pattern 1: The
Stopwatch Without the Strategy
The most common failure mode is the stopwatch exercise. A team
decides to “do SMED,” brings out the cameras and the timers, documents
the current state, and then… files the report. The documentation sits
in a binder. The video footage is never reviewed. Nobody actually
implements the changes.
This happens because the documentation phase feels like progress. The
team has measured, analyzed, and understood the problem — and there’s a
psychological satisfaction in that. But measurement without action is
theater. The changeover time hasn’t improved by a single second. The
operators watched the team film them, went back to their normal routine,
and nothing changed.
The fix is simple but uncomfortable: assign responsibility and
authority. Someone must own the implementation, and that person must
have the authority to make changes — to tooling, to procedures, to
machine configurations. Without that authority, the analysis becomes an
academic exercise.
Pattern 2:
One-Time Improvement with No Sustainment
The second most common pattern is the successful SMED event followed
by gradual regression. An improvement team comes in, runs a kaizen
event, reduces changeover from 90 minutes to 35 minutes, celebrates the
win, and leaves. Six months later, the changeover time has crept back to
70 minutes. A year later, it’s back to 85.
This regression happens because the new procedure was imposed rather
than internalized. The operators complied during the event because the
improvement team was present, the energy was high, and the expectations
were clear. But once the team left, the operators reverted to their
familiar routines — not out of malice, but because old habits are
comfortable and the new procedure required conscious effort.
Sustaining SMED improvements requires more than a new procedure
document. It requires visual controls (are the tools laid out correctly?
is the pre-staging area stocked?), regular audits (is the new procedure
being followed?), and operator ownership (did the operators help design
the new process, or was it handed to them?). Without these elements,
regression is not a risk — it is a certainty.
Pattern 3: Technology
Without Methodology
Some organizations try to shortcut the SMED process by investing in
technology — automated tool changers, robotic positioning systems,
computer-controlled calibration. These investments can be powerful, but
they’re not a substitute for the methodology. In fact, they often make
things worse.
The problem is that technology solves the technical constraint but
ignores the procedural one. An automated tool changer that completes a
changeover in 90 seconds sounds impressive, but if the operator still
has to walk across the plant to retrieve the new tooling, verify the
specifications, and manually position the material, the total downtime
hasn’t improved by much. The bottleneck simply moved from the machine to
the material handling.
Technology accelerates SMED implementation, but it doesn’t replace
the need for the four-stage analysis. You still need to separate
internal from external, convert where possible, streamline what remains,
and eliminate adjustments. Technology is an enabler of the methodology,
not a replacement for it.
Pattern 4: No
Follow-Through on Standardization
SMED improvements often stay localized. One production line gets a
quick-changeover makeover. The results are impressive. But the lessons
learned — the standardization of tool interfaces, the pre-staging
protocols, the parallel operation procedures — never spread to other
lines. Each line continues to operate with its own unique, inefficient
changeover process.
Standardization is the multiplier effect of SMED. When a
quick-changeover solution developed on one line is applied across all
similar lines, the total improvement is enormous. But standardization
requires deliberate effort: documenting the solution in a way that’s
transferable, adapting it to the specific constraints of each line, and
training operators consistently.
Pattern 5: Missing
the Pre-Staging Foundation
The single most overlooked SMED technique is also the simplest:
pre-staging. Before the machine stops, everything needed for the
changeover should be positioned within arm’s reach of the operator.
Tools on a cart, sorted in the order they’ll be used. New dies
pre-heated. Material positioned at the loading point. Specifications
posted at eye level.
This is not complex. It requires no capital investment. It can be
implemented in an afternoon. And yet, on most factory floors, the
operator still walks away from the machine multiple times during
changeover — to find a wrench, to check a spec sheet, to retrieve a
fixture from a cabinet across the room. Each walk is minutes of wasted
downtime.
Pre-staging fails because it requires advance preparation, and
advance preparation requires knowing what the next job will be — which
requires effective production scheduling. If the operator doesn’t know
what’s coming next until the current job ends, pre-staging is
impossible. And so the dependency chain becomes clear: SMED requires
pre-staging, which requires scheduling visibility, which requires
communication between planning and operations. The failure isn’t in the
tool. It’s in the system around it.
The
Business Case: Why This Matters More Than You Think
Many leaders look at changeover time as a technical detail —
something for the operations team to worry about. This is a strategic
mistake. Changeover capability is a competitive weapon, and in today’s
manufacturing environment, it’s becoming more critical every year.
The reason is simple: customers increasingly demand smaller batch
sizes, more product variety, and shorter lead times. The era of long
production runs for a single SKU is ending in most industries. The
manufacturer that can change over in 15 minutes can accept small orders
profitably. The manufacturer that needs 2 hours cannot. It’s that
direct.
Long changeover times force manufacturers into large-batch
production, which creates a cascade of hidden costs: higher inventory
levels, longer lead times, increased obsolescence risk, and reduced
flexibility. These costs don’t appear on the changeover line of the
budget — they’re embedded in inventory carrying costs, expediting fees,
and lost quotes. But they’re real, and they’re significant.
A manufacturer that reduces average changeover from 90 minutes to 20
minutes doesn’t just save 70 minutes of downtime. They fundamentally
change their production economics. Batch sizes can shrink by a factor of
four. Work-in-process inventory drops. Lead times compress. The sales
team can promise delivery dates that competitors can’t match. The
organization becomes more responsive, more flexible, and more profitable
— all from a methodology that requires minimal capital investment.
Making SMED Stick: A
Practical Framework
If you’re serious about implementing SMED — not as a one-time event
but as a permanent capability — here’s a framework that addresses both
the technical and the human dimensions.
Start with measurement you can trust. Film multiple
changeovers at different times, on different shifts, with different
operators. Build a baseline that reflects reality, not the idealized
procedure. The gap between “what should happen” and “what actually
happens” is where your biggest opportunities hide.
Prioritize by impact, not by ease. It’s tempting to
start with the easy changes — the quick wins that show immediate
results. But if your highest-impact opportunity requires a tooling
redesign, don’t avoid it because it’s harder. Sequence your improvements
by their effect on changeover time, not by their implementation
difficulty.
Involve operators from day one. The operators who
perform the changeover know things about the process that no engineer or
manager can learn from a distance. They know which bolts always seize,
which fixtures never line up, which steps can be safely skipped and
which cannot. Bring them into the improvement process as experts, not as
subjects to be observed.
Make the new procedure visual. A procedure document
in a filing cabinet is useless. A visual setup board at the machine —
showing tool layout, step sequence, and standard times for each step —
keeps the new procedure front and center. Make it impossible to do the
changeover the old way without visibly deviating from the standard.
Audit relentlessly, then taper. In the first 90 days
after implementation, audit every changeover. Not to catch people doing
it wrong, but to identify what’s still causing friction. After 90 days,
shift to weekly audits. After six months, monthly. Never stop auditing
entirely — that’s when regression creeps in.
Share results across the organization. When one line
achieves a breakthrough improvement, share it. Present the
before-and-after at your operations review. Show the video. Celebrate
the team. Then immediately ask: which other lines can we apply this to?
Standardization is how one improvement becomes ten improvements.
The Deeper Lesson: SMED as
a Mindset
The most powerful outcome of a successful SMED implementation isn’t
the reduction in changeover time — it’s the shift in organizational
mindset. When operators see that a process they’ve done the same way for
fifteen years can be fundamentally redesigned, they start looking at
every process with fresh eyes. The question changes from “how do we do
this?” to “why do we do it this way?”
This is the real return on SMED. The methodology teaches your
organization that improvement is possible, that the current state is not
inevitable, and that the people closest to the work have the best ideas
for improving it. Once that belief takes root, it spreads far beyond
changeover. It becomes the foundation of a continuous improvement
culture.
But that only happens when the implementation is real — when the
changeover time actually drops, the new procedure is actually followed,
and the results are actually sustained. A SMED exercise that produces a
report but no improvement teaches the opposite lesson: that improvement
initiatives are just management flavor-of-the-month, to be endured
rather than embraced.
So the choice is yours. You can treat SMED as a box to check on your
Lean journey — run a workshop, update a procedure, file the
documentation, and move on. Or you can treat it as a genuine
transformation of how your organization handles change. The methodology
is the same either way. The difference is in the commitment.
Peter Stasko is a Quality Architect with over 25
years of experience in manufacturing quality management, process
improvement, and Lean implementation. He has guided organizations across
automotive, electronics, and industrial equipment sectors through the
practical realities of methodologies like SMED, helping them bridge the
gap between theory and measurable results. He writes about the gap
between what quality systems should deliver and what they actually
deliver in practice.
