Most plants are still full of people
Fully lights-out dark factories remain rare. Most plants still run with people on the floor, so shared spatial visibility has to cut walked checks without removing the crew.
What "still full of people" means in plant language
Industry talk often treats a dark factory as the default end state of Industry 4.0 and smart manufacturing. In practice, whole-plant unattended operation is uncommon. Most deployments that use lights-out language are partial lights-sparse scopes: a cell, a line, or an overnight window inside a larger staffed hall. Operators, supervisors, and logistics crews still own the shift. Maintenance, exceptions, changeovers, and mixed-model work keep people on site even when some machines already run without constant hands-on attendance.
That gap between pitch and floor matters for capital planning. Buying for a true lights-out endpoint when the hall is still people-heavy funds the wrong layer first. Industry practice more often funds a few unattended processes than a whole dark plant, which matches what plant teams already see: people still run most of the shift. What a true dark factory actually demands (closed-loop authority over actuators, not only a live map) sits under what a true dark factory demands.
Why walked checks persist when systems already exist
Staffed plants commonly already run MES for execution numbers, WMS for inventory, ERP for orders, and SCADA for equipment state. Those systems work in their own lanes. What is often missing is a shared picture of where movers, people, and material sit in the physical layout right now. Without that picture, status confirmation still means a walk, a radio loop, or a phone call between desks that cannot see the same hall.
An operational digital twin sits as an umbrella layer on those feeds plus location from a real-time location system (RTLS) when the plant needs live positions. Locations of forklifts, people, and materials appear on a live 3D model of the facility. Replay and spaghetti diagrams show the paths that were taken. Medium-precision location around 1–3 m (3–10 ft) is usually enough for aisle, buffer, and bay checks. Exact floor position on the mark is a different job and is not required for every asset. How the live twin type is framed sits under operational digital twins show the shift.
The spatial layer while people stay
In the staffed transition, the live twin acts as shared eyes across the layout. Supervisors open one hall view instead of walking every aisle to confirm free capacity or stalled material. Logistics dispatch from current locations rather than hunting the nearest free truck by radio. Production can place a wait next to a bay on the same map used for production monitoring.
That layer has three equal jobs, not a logistics-only story. Live internal logistics decisions use the map during the shift. The same view stores a spatially oriented dataset of place and time under the hall. Correlation and causality analytics later ask which pairings keep preceding a stop. Search, stored history, and pattern work sit together under internal logistics optimization and the features overview.
This is not the closed-loop endpoint where software decides and machines execute without humans. It is the intermediate layer that lets the remaining staff see more without walking more. twinzo does not replace ERP, MES, or WMS, does not calculate OEE as a system of record, and does not run discrete-event physics. It places fed values and locations in spatial context so people can act.
What documented staffed plants actually recorded
Documented twinzo results in this phase stay inside a staffed boundary. On one 30,000 m² (about 323,000 ft²) automotive Tier-2 plant, logistics optimization with RTLS lifted average forklift productivity from 63% to 75%, reduced the fleet from 15 to 13 units, freed four FTEs, and cut first-year OPEX by €179,000 (about $196,900) against an €81,000 (about $89,100) investment. Drivers and supervisors remained on site. The map shortened search and empty travel. It did not remove the people or take the wheel.
Sibling halls show the same staffed bridge as fewer empty rounds and clearer thin-shift balance, covered under the empty-round problem that still sits in most plants, the thin-shift visibility gap, and seeing the floor when headcount drops.
Distinct from the lights-out endpoint
Lights-out manufacturing removes routine human presence from the cell or hall that runs dark. The staffed transition keeps people and removes the need for constant physical verification of state that a shared map can already show. Full autonomy still needs reliable unattended sequences, fault recovery, and write-back into PLC and higher orchestration. Visibility alone never supplies that authority.
Treat the live spatial layer as the set of eyes the plant needs while it is still staffed and still deciding which cells, if any, ever justify a true dark build. Keep capital for vehicles, robots, and closed-loop control on a separate invoice from the observation layer that makes today's crew effective.
Get in touch if you want to walk the same staffed-to-visibility path on your own facility model.