Separating internal from external setup work is central to SMED changeover analysis. Done correctly, it reduces the time equipment is unavailable without assuming that local improvement automatically increases system throughput.
This article explains how to build a timestamped activity worksheet, classify tasks, convert safe work to external setup, coordinate parallel operations, and verify startup quality and downstream flow. The measurement window runs from the last good part of the outgoing lot to the first good part of the next lot.
Key Takeaways
- Measure changeover from the last good part to the next first good part.
- Separate stopped-machine tasks from work that can be done while production continues.
- Convert tasks only after reviewing machine-access and hazardous-energy risks.
- Use repeated observations to verify time, quality, and variation improvements.
- Track throughput and downstream WIP to confirm the system-level effect.
What SMED Changeover Analysis Measures
SMED stands for Single-Minute Exchange of Die. The “single-minute” target means a changeover completed in a single-digit number of minutes, or less than 10 minutes, according to the Lean Enterprise Institute definition.
Shigeo Shingo developed the approach around two key distinctions: internal setup can be performed only while the machine is stopped, while external setup can be performed while it is running. The method then seeks to convert suitable internal work to external work and simplify the remaining internal tasks.
For consistent measurement, define the changeover as the time from the last good piece of the outgoing run to the first good piece of the next run. This boundary is also used in the NIST quick-changeover guidance and prevents teams from ending the clock at restart while ignoring adjustment or first-piece losses.
| Setup Type | Machine State | Common Examples | Primary Reduction Lever |
|---|---|---|---|
| Internal setup | Stopped | Die removal, fixture mounting, first-piece verification | Convert where safe, simplify, or eliminate |
| External setup | Running | Tool staging, document retrieval, permitted preheating | Standardize and pre-position |
A task should not be called external merely because it could occur earlier. It must also be safe, practical, and compatible with the site’s operating and energy-control procedures.
Building a Timestamped Activity Worksheet
Use direct observation rather than estimates alone. Record each activity’s description, start and stop time, elapsed time, machine state, operator or role, delays, and preliminary internal-or-external classification.
Video can help teams review overlapping activities, provided workplace policies and employee privacy requirements are followed. A second observer can also reduce missed events during busy portions of a changeover.
Prioritize changeovers using a consistent business measure such as frequency multiplied by average duration. This identifies the changeover types consuming the most scheduled production time instead of relying on familiarity or anecdote.
Air Academy Associates’ Lean Principles certification course provides a practical foundation in flow, waste identification, and improvement tools that support this analysis.
How to Separate and Convert SMED Tasks Safely
infog
Review every observed activity and ask whether the machine truly must be stopped. Some retrieval, preparation, inspection, and documentation tasks can move outside the stopped-machine window, but physical access and energy hazards determine what is permissible.
Setup or servicing that exposes employees to unexpected energization or stored energy may require lockout/tagout. OSHA explains that its limited minor-servicing exception applies only when the activity is routine, repetitive, integral to production, and protected by effective alternative measures; otherwise, the lockout/tagout standard applies.
1. Classify Every Activity Against Machine State
Record whether the machine was running or stopped during each task. Keep this first pass descriptive so that observed practice is not confused with the proposed future state.
2. Identify Conversion Candidates With a Safety Review
Flag tasks performed during downtime only by habit, then assess guarding, access, stored energy, tooling, ergonomics, and handoffs. Use the employer’s authorized safety and energy-control procedures before changing how the work is performed.
3. Stage Tools and Materials Before the Stop
Pre-stage the correct dies, fixtures, fasteners, gauges, instructions, and permitted materials. Preheating or presetting may shorten stabilization, but only when the equipment manufacturer’s requirements and site safety procedures allow it.
4. Use Parallel Operations With Clear Ownership
Two or more trained people may perform independent tasks at the same time to reduce elapsed time. Define zones, roles, communication, and energy-control responsibility so simultaneous work does not create interference or exposure.
5. Reduce Adjustments With Standard Reference Points
Fixed stops, documented reference settings, preset tool offsets, and suitable quick-action fasteners can reduce trial-and-error. Validate each change under normal production conditions before making it the standard method.
6. Confirm the First Good Part
Do not close the changeover window at machine restart. Apply documented acceptance criteria to the first part, and record adjustment time, scrap, rework, and inspection delays separately so the team can see where startup losses occur.
Measuring Variation and Startup Scrap
One observation shows what happened once; it does not establish a stable new baseline. Observe enough changeovers to represent relevant products, operators, shifts, and operating conditions, and choose the sample size based on process variation and the decision being made rather than a universal five-to-ten-run rule.
Track total changeover time together with startup scrap, first-pass acceptance, adjustment time, and any safety or quality exceptions. Compare the distribution of results, not only the best run or the average.
Teams that need a stronger measurement foundation can use the variation reduction short course to connect waste, variation, cycle time, and process observation.
Turning the Improved Sequence Into a Repeatable SOP
An improved sequence should not remain in one operator’s memory. Document external tasks and their triggers, the ordered internal sequence, parallel assignments, safety controls, reference settings, required tools, and first-part acceptance criteria.
Test the draft SOP across representative shifts and product families. Update it when the observed method, equipment, safety requirements, or quality criteria change, and train affected employees before implementation.
Air Academy Associates’ PF/CE/CNX/SOP program supports the transition from process-function and cause-and-effect analysis to documented operating instructions. It helps teams capture why critical steps and controls matter, not only what operators must do.
Checking Throughput and the Downstream Queue
A shorter changeover increases available capacity at the improved resource, but it does not automatically move the system constraint. The effect depends on demand, scheduling, uptime, product mix, and the capacities of upstream and downstream processes.
Compare throughput, queue or work-in-process levels, starvation, blocking, and utilization before and after the change. Air Academy Associates’ value stream mapping guide explains how material and information flow can reveal delays and bottlenecks across the full process.
If downstream WIP grows without a corresponding increase in customer-facing throughput, review the schedule, batch sizes, release rules, and the current constraint. Preserve the safer, faster changeover method while adjusting production control to the needs of the whole system.
Coaching and One-to-One Training from Air Academy Associates can help production leads review a live constraint analysis, test revised schedules, and plan an implementation without losing the gains from setup reduction.
Air Academy Associates Resources for SMED Changeover Analysis
These four offerings address different parts of the improvement cycle, from identifying waste through measurement, documentation, and implementation support.
Lean Principles Certification and Tools
This program builds a foundation in Lean thinking, process flow, and waste identification. Those skills support defensible internal-versus-external setup decisions and help teams prioritize changeover opportunities.
- Introduces practical Lean tools for rapid improvement
- Supports flow and waste analysis
- Provides a foundation for further Lean Six Sigma study
Six Sigma Variation Reduction for Waste Elimination
This short course examines how waste and variation affect cost, delivery, and performance. Its coverage of observation, cycle time, and takt time supports post-changeover measurement.
- Explains waste and variation in a Lean Six Sigma context
- Includes process observation and cause-and-effect thinking
- Connects cycle time and takt time to operational analysis
PF/CE/CNX/SOP
This program supports structured process-function, cause-and-effect, control, and SOP work. It is useful when a team needs to translate a tested sequence into clear, auditable instructions.
- Structures process-function analysis
- Connects causes and controls to requirements
- Supports consistent operating documentation
Coaching and One-to-One Training
Individual coaching provides project-specific guidance for teams applying improvement methods. It can be useful when constraint location, scheduling effects, or implementation risks require experienced review.
- Supports active improvement projects
- Provides focused review and feedback
- Helps teams plan and sustain implementation
Conclusion
SMED changeover analysis begins with direct observation and a consistent last-good-part-to-first-good-part definition. Separating internal and external setup, protecting workers, reducing adjustment, and measuring repeated results can cut downtime without hiding startup quality losses.
System-level verification matters as much as the local time reduction. Track throughput and downstream WIP so scheduling and flow decisions reflect the current constraint.
Air Academy Associates offers Lean Six Sigma training, coaching, and consulting to help teams improve changeovers and build repeatable processes. Explore the available programs to choose the level of support that fits your project.
FAQs
How Do You Separate Internal and External Changeover Tasks?
Observe and time the full changeover, then classify each task by whether it requires the equipment to be stopped. Consider pre-staging, presetting, and standardized handoffs only after safety, guarding, hazardous-energy, quality, and manufacturer requirements have been reviewed.
How Should First-Good-Part Startup Losses Be Counted?
Include the time needed to produce and accept the first conforming part within the overall changeover window. Track mechanical stop time, adjustment, inspection, scrap, and rework separately so improvement in one component does not hide loss in another.
How Do You Check Whether a Faster Changeover Affected the Bottleneck?
Compare system throughput, WIP or queue levels, blocking, starvation, and utilization before and after the change across representative runs and shifts. A growing downstream queue is a warning signal, but constraint location should be confirmed from the full set of flow data rather than queue length alone.
Related Articles
