SMED: Single-Minute Exchange of Die for Faster Changeovers

SMED (Single-Minute Exchange of Die) is a lean manufacturing methodology for dramatically reducing machine changeover time — the period when a machine is stopped while being converted from producing one product to another. Developed by Shigeo Shingo, SMED is one of the most immediately impactful lean tools because faster changeovers directly enable smaller batches, lower inventory, and more flexible scheduling. This manufacturing glossary entry explains the SMED process, shows real results, and connects it to production scheduling.

What Is SMED?

"Single-minute" in SMED does not mean one minute — it means single-digit minutes (under 10). The methodology provides a systematic approach to analyze every step of a changeover and reduce the total time through three stages:

Stage 1: Separate Internal from External

Internal activities require the machine to be stopped (changing a die, adjusting fixtures). External activities can be done while the machine is still running (gathering tools, staging the next die, preheating a mold). Most shops perform external activities with the machine stopped simply out of habit. Separating them typically reduces changeover time by 30-50% with zero investment.

Stage 2: Convert Internal to External

Re-engineer the process to move activities from internal to external. Use pre-staging tables so the next die is ready at the machine. Pre-heat molds in a separate oven. Use intermediate jigs to align fixtures offline. This further reduces machine downtime.

Stage 3: Streamline Remaining Internal Activities

Eliminate adjustments through standardized settings and locating pins. Replace bolts with quick-release clamps. Use one-turn fasteners instead of multi-turn. Parallel operations where two people work simultaneously. Each streamlining step shaves seconds or minutes from the changeover.

How SMED Works in Practice

A SMED event typically runs 3 to 5 days with a cross-functional team:

1. Video the current changeover from the moment the last good part of Job A is produced to the first good part of Job B. Time every element.
2. Classify each element as internal or external. In a typical first analysis, 30-50% of "internal" elements are actually external activities being performed at the wrong time.
3. Reorganize: Move all external elements to before or after the machine stop. Create a checklist and staging area.
4. Re-engineer: Convert additional internal elements to external. Design fixtures, pre-staging tables, and quick-change tooling.
5. Streamline: Reduce the remaining internal elements through quick-release mechanisms, standardized settings, and parallel operations.
6. Standardize: Document the new changeover procedure as standard work. Train all operators. Audit regularly.

Example with Numbers

A job shop with 8 CNC machining centers ran a SMED program across all machines:

• Before SMED: Average changeover time was 52 minutes. The shop ran an average of 6 changeovers per machine per day. Total daily changeover time: 52 x 6 x 8 = 2,496 minutes (41.6 hours) of machine downtime per day.
• After SMED Stage 1 (separating internal/external): Average changeover dropped to 28 minutes. Daily changeover time: 1,344 minutes. Recovered 19.2 machine-hours per day.
• After SMED Stages 2-3 (converting and streamlining): Average changeover dropped to 14 minutes. Daily changeover time: 672 minutes. Total recovered: 30.4 machine-hours per day — equivalent to adding nearly 4 machines to the shop.
• Economic impact: The 30.4 recovered hours per day at a shop rate of $85/hour represented $2,584 per day or $671,000 annually in recovered capacity — without purchasing a single new machine.
• Batch size flexibility: With 14-minute changeovers instead of 52, the economic batch size dropped by 73%, enabling the shop to run more frequent, smaller lots aligned with customer demand.

Why SMED Matters for Production Scheduling

SMED is arguably the lean tool with the most direct impact on scheduling:

• More changeovers are affordable: When changeovers take 52 minutes, the scheduler batches to minimize them — creating large lots, long lead times, and high WIP. At 14 minutes, the scheduler can sequence work more flexibly.
• Smaller batches: Reduced setup cost per unit allows smaller production batches, which reduces WIP and lead time. Production scheduling software like RMDB optimizes batch sizes based on actual setup times — shorter setups mean smaller optimal lots.
• Better sequence optimization: With fast changeovers, the scheduler can sequence jobs to meet due dates rather than grouping by product family to minimize setups. Customer priority drives the schedule, not setup avoidance.
• Higher effective capacity: Recovered changeover time is available for production, giving the scheduler more capacity to work with.
• Supports Heijunka: Level-loading the schedule requires frequent product switches. SMED makes this economically viable.

The lean manufacturing guide identifies SMED as the enabler of small-lot production and the prerequisite for one-piece flow.

Related Terms

• Standard Work — The documented changeover procedure that sustains SMED gains by ensuring every operator follows the optimized method.
• One-Piece Flow — The production ideal that SMED makes possible by removing the setup-time barrier to small-lot production.
• Heijunka — Production leveling that requires fast changeovers to mix products evenly throughout the production period.

---

See all lean and scheduling terms in the Manufacturing Glossary.