SMED with timed changeovers on the twin
SMED still separates internal and external setup. Hall timing and location show whether the waits that inflate changeover clocks are wrench work or aisle work.
What SMED is trying to fix
Single-minute exchange of die (SMED) is the lean manufacturing method for cutting changeover time by moving work outside the machine-down window and streamlining what must stay inside it. Internal work happens while the asset is stopped. External work happens while it still runs. The classic win is finer mix, smaller batches, and room for heijunka without drowning the day in setups. "Single-minute" means a single-digit minute target where the process allows it, not a literal one-minute mandate on every asset.
Digitizing a visual management board is not the same as transforming SMED. Research on Lean 4.0 methods shows e-boards can mirror paper, while e-SMED gains real leverage when timing, preparation status, and what-if trials become continuous. Case work on digital visual management with SMED also shows a common pattern: internal wrench steps may already meet the designed time while the asset still sits idle before setup starts because external prep never finished. The tool changes when evidence replaces memory.
Where timing, place, and twins help changeovers
MES setup states mark when the asset entered changeover. Location of people, tool carts, and next-job kits shows whether external prep finished before the stop. Medium-precision location around 1–3 m (3–10 ft) is enough to see whether the cart was already at the bay or still two aisles away. An operational digital twin places those facts on one hall so Check is not a clipboard argument.
A simulation digital twin helps before you rearrange shadow boards or staging marks. Trial whether moving a die rack 6 m (about 20 ft) closer cuts internal walk time without blocking a milk run. After the physical change, the operational twin verifies the next run of changeovers. That is PDCA backed by measured proof instead of a staged board walk. Keep the stored changeover windows in a spatially oriented dataset. Correlation and causality analytics can later show whether long setups cluster with the same missing kit lane or the same shared aisle peak. Live logistics on carts and kits, the stored spatial record, and that analytics layer stay equal jobs.
Changeover waste is often logistics in disguise
Many SMED workshops discover that wrench time is already short and the waits are long. The die is ready. The cart is not. The next kit is booked in MES and still sits in a side bay. The setup tech is free and walking 50 m (about 160 ft) because staging marks were never treated as external work. Spatial Check makes those waits measurable instead of debated.
That is why SMED and internal logistics share a twin. Production owns the internal sequence. Logistics owns cart and kit readiness. The hall view is where both see the same clock. Without that join, each team optimizes its own spreadsheet and the changeover time never moves. Document each reduced changeover with the same before and after windows on the twin. A claimed twelve-minute changeover that still carries twenty minutes of aisle wait on the twin is not standardized. SMED paperwork without spatial Check drifts back to tribal estimates within a month.
Finer changeovers also unlock leveling. If setups stay long, heijunka stays coarse and batches stay large. Simulate the leveled mix only after SMED evidence shows the plant can afford the changeover frequency the mix implies. Otherwise leveling becomes a board exercise that the assets cannot hold. Related lean context for leveling after faster changeovers sits under simulate the kaizen before you move the racks. Maintenance travel that collides with setup windows sits under maintenance optimization with RTLS.
How teams digitize SMED without theater
1. Split the current changeover into internal and external steps - Paper first is fine. Honesty matters more than software.
2. Time waits that are really travel or search - Those are spatial, not wrench skill.
3. Stage external work with a visible ready rule - Kit and cart at the bay before the asset stops.
4. Verify on the twin for a run of changeovers - Standardize only what the next shifts confirm.
After a SMED standard is locked, sample changeovers weekly on the twin for a month, then monthly. Drift shows up as aisle waits first, wrench time second. Catching that drift early is cheaper than another full SMED workshop that rediscovers the same staging gap. That closed evidence loop is how smart factory changeover programs stay honest instead of wallpapering another board. twinzo supports the hall view through production monitoring and internal logistics optimization when material and carts own the delay. Get in touch if you want to time changeover waits on your own facility model.