Single-Minute Exchange of Dies (SMED)

What Is SMED?

SMED is a systematic approach to cutting the time a production line or machine sits idle between one job and the next. The name reflects its ambition: completing every changeover in a single-digit number of minutes. In practice, achieving sub-ten-minute changeovers is the benchmark, though many operations pursue times measured in seconds once the methodology matures.

Shigeo Shingo developed SMED in the 1950s and 1960s while consulting for Toyota and other Japanese manufacturers. He observed that most changeover time was wasted on activities that did not strictly require the machine to be stopped. By forcing a clear distinction between what must happen with the machine off and what can happen before it stops or after it restarts, Shingo created a practical framework that any operation can apply regardless of industry or equipment type.

SMED sits within the broader discipline of lean manufacturing as a targeted tool for eliminating one of the most common forms of production waste: idle equipment time caused by slow changeovers.

The SMED Methodology: Three Stages

Shingo's original framework describes three stages that move a team from the current unstructured changeover toward a repeatable, optimized process.

Stage 1: Separate Internal and External Setup

The first step is observation. Teams video-record the current changeover from start to finish and catalog every activity. Each activity is classified as either internal (machine must be stopped) or external (can be done while the machine runs).

In most unoptimized changeovers, 30 to 50 percent of activities that are currently performed with the machine off could legally and safely be done beforehand. Simply separating the two categories and moving external tasks outside the stoppage window can cut changeover time by 30 percent without any physical modifications to equipment or tooling.

Stage 2: Convert Internal Setup to External Setup

Stage two examines every remaining internal activity and asks: can this be redesigned so it no longer requires the machine to stop? Common conversion tactics include pre-staging and pre-heating components, using intermediate jigs that can be prepared offline, and pre-assembling tooling kits so operators arrive at the machine with everything ready.

This stage often requires modest engineering investment: standardizing tool heights so blades or dies seat correctly without trial runs, adding quick-connect fittings to hoses and electrical connectors, or installing pre-set toolholders that carry calibration offline. These changes transform activities that once consumed ten or fifteen minutes of machine stoppage into work that happens in parallel with the previous production run.

Stage 3: Streamline All Remaining Elements

The third stage targets every remaining activity, both internal and external, for time reduction and error-proofing. Bolt patterns are reduced or replaced with quarter-turn fasteners. Color-coded tooling eliminates search time. Standard work instructions reduce operator variability. Checklists and shadow boards ensure every tool and consumable is at the machine before the stoppage begins.

Continuous improvement practices drawn from Kaizen apply here: small, frequent improvements compound over time, and changeover times that took 90 minutes often reach under 10 minutes after several improvement cycles.

Internal vs. External Setup: A Comparison

How to Implement SMED: Step by Step

A structured SMED implementation follows a repeatable sequence that combines observation, analysis, and incremental improvement.

Step 1: Select a Target Changeover

Choose the changeover that creates the most production loss. This is typically the machine with the longest stoppage time, the highest changeover frequency, or the greatest impact on downstream scheduling. Data from a production monitoring system makes this selection objective rather than anecdotal.

Step 2: Record the Current State

Video the complete changeover from the last good part of the outgoing run to the first accepted part of the incoming run. Record start and end times for each activity. Include all walking, searching, waiting, and adjustment time, not just the physical installation work.

Step 3: Classify Each Activity

Build a changeover activity log. For each item, record the task description, duration, and classification: internal or external. Involve the operators who perform the changeover in this analysis. They often identify conversion opportunities that are invisible to observers.

Step 4: Move External Activities Outside the Stoppage

Create a pre-changeover checklist covering every external task. Assign responsibility and establish the timing: which tasks complete during the final minutes of the previous run, and which require earlier preparation. Verify compliance during the next changeover trial.

Step 5: Convert Internal Activities to External

Review each remaining internal task. Identify what engineering change, tooling modification, or process redesign would allow it to happen offline. Prioritize by time impact. Implement changes in order of return on investment.

Step 6: Streamline and Standardize

For activities that cannot be converted, reduce their duration. Replace bolts with quick-release fasteners. Add reference marks to eliminate trial-and-error adjustment. Write standard work documents and train all operators to the same method. Establish a target time for each internal activity and track actuals.

Step 7: Measure, Audit, and Improve

Record changeover times for every run. Track the average and the distribution. When times creep back up, conduct a brief audit to identify the cause. Treat SMED as an ongoing discipline rather than a one-time project.

SMED Benefits: What Teams Typically Achieve

The evidence base for SMED benefits is extensive across automotive, food and beverage, plastics, pharmaceutical, and discrete manufacturing sectors.

Teams starting from unstructured changeovers averaging 60 to 120 minutes routinely reach under 15 minutes within three to six months of a structured SMED project. The 50 to 80 percent reduction benchmark is consistent across industries. Some operations, particularly those with strong prior lean foundations, achieve reductions exceeding 90 percent.

Beyond raw time savings, SMED produces several downstream benefits:

• Smaller economic batch sizes: When changeover costs fall, the minimum profitable production run shrinks. Manufacturers can respond to smaller orders, reduce finished goods inventory, and align production more closely with actual demand.
• Reduced work-in-process inventory: Faster changeovers reduce the incentive to overproduce large batches as a buffer against future changeover time.
• Improved scheduling flexibility: A line that changes over in 8 minutes can switch products multiple times per shift. A line that takes 90 minutes cannot. SMED unlocks scheduling options that are mathematically unavailable at long changeover times.
• Lower defect rates at startup: Standardized changeover procedures with pre-set tooling reduce the adjustment iterations that generate scrap during the first minutes of a new run.
• Operator engagement: Changeover improvement projects are visible, measurable, and directly tied to workload. Operators who participate in SMED projects typically report higher job satisfaction and stronger ownership of production outcomes.

The financial returns are substantial. A machine running 20 changeovers per month, each reduced from 60 minutes to 12 minutes, recovers 16 hours of production capacity per month. At typical industrial machine utilization rates, that capacity recovery often justifies the SMED project cost within weeks.

SMED and OEE: The Direct Connection

SMED is one of the most direct levers available for improving Overall Equipment Effectiveness. OEE is the product of three components: Availability, Performance, and Quality. Changeover time affects all three, but its primary impact is on Availability.

Availability measures the proportion of scheduled production time during which equipment is actually running. Planned downtime for changeovers reduces the denominator of available time. Every minute saved through SMED converts directly into additional productive time, raising Availability and therefore OEE.

The relationship is arithmetic. If a machine is scheduled for 480 minutes per shift and spends 60 minutes on changeovers, its maximum Availability before any unplanned losses is 87.5 percent. Reduce changeover time to 12 minutes and maximum Availability rises to 97.5 percent. That 10-percentage-point swing in Availability multiplies through the full OEE calculation.

SMED also affects Performance through its influence on cycle time and takt time alignment. When changeover time is unpredictable, production schedules build in large buffers that reduce effective throughput. Standardized, predictable changeovers allow schedulers to plan more tightly against takt time, improving Performance scores.

Finally, SMED improves the Quality component by standardizing setup procedures. Pre-set tooling and verified checklists reduce the startup scrap that occurs when changeovers rely on operator judgment for tool positioning and parameter adjustment.

Teams implementing SMED alongside Total Productive Maintenance programs see compounding gains: TPM reduces unplanned stops while SMED reduces planned stops, and together they drive production efficiency toward world-class levels.

SMED Across Industries

The original SMED application was die exchange on stamping presses in the automotive industry, where die sets weigh several tons and were historically changed over the course of a full shift or longer. Shingo's work at Toyota reduced those changeovers to under ten minutes and became one of the foundational elements of the Toyota Production System.

The methodology has since been adapted to virtually every manufacturing context:

• Food and beverage: Product changeovers involve cleaning, allergen control, and packaging line reconfiguration. SMED structures cleaning-in-place procedures and pre-stages packaging components to minimize line stoppage.
• Pharmaceuticals: Batch changeovers carry regulatory compliance requirements. SMED identifies which documentation and preparation steps can be completed before the line stops, reducing changeover time without compromising audit trails.
• Plastics and injection molding: Mold changes are the primary changeover event. Pre-heating molds to operating temperature offline, pre-setting ejector pins, and using quick-mold-change systems are direct SMED applications.
• Printing and packaging: Plate changes, ink color changes, and substrate changes all benefit from SMED. Pre-mixed inks staged at the press and pre-mounted plate cylinders are standard external setup activities.

In each context, the analytical framework remains identical: observe, classify, convert, streamline, and standardize.

SMED and Lean: Where It Fits

SMED is one element within a larger lean production framework. It works alongside 5S workplace organization, which ensures tools and materials are in their designated locations before a changeover begins. It supports just-in-time production by making small-batch scheduling economically viable. And it reinforces Total Productive Maintenance by reducing the planned downtime component of the availability equation.

Within lean, changeover time is classified as a form of waste specifically because it consumes time and resources without adding value to the product. SMED targets this waste directly with a method that is practical, teachable, and measurable. That combination of clarity and results has made it one of the most widely adopted lean tools outside of the Toyota facilities where it originated.

The Bottom Line

SMED is a structured methodology for making equipment changeovers faster, more consistent, and less disruptive to production schedules. By separating what must happen with the machine stopped from what can happen while it runs, and then systematically converting and streamlining both categories, manufacturers routinely cut changeover times by 50 to 80 percent.

The business case is straightforward: faster changeovers mean more available production time, smaller economic batch sizes, greater scheduling flexibility, and measurably higher OEE. Teams that treat SMED as a continuous discipline rather than a one-time project sustain these gains and continue improving beyond the initial ten-minute target.

For operations that compete on responsiveness and efficiency, SMED is not an optional optimization. It is one of the foundational practices that separates world-class manufacturing performance from average.

Frequently Asked Questions