RTLS for the plant director — WIP and OEE on the line you run.

Knowing where every WIP unit is, in real time, changes three decisions that previously ran on guesswork. Schedule confidence — you stop expediting jobs whose location is uncertain.

Bottleneck identification — you see the queue ahead of the constraint, every minute, not at the end of the shift.

Line-side replenishment — kit and parts arrive when they're needed because the system sees consumption, not a calendar. The combined effect is measured in days of lead time, not minutes.

Operators tolerate two seconds of friction per shift before they work around the system. Tags must be issued and returned without ceremony. Read points must register reliably even with badly-positioned material.

Alerts must reach the right person without paging the wrong one. These are not technology requirements — they are deployment-discipline requirements that get baked into the gate 2 (pilot) evidence.

WIP tracking that doesn't feed your MES is a parallel system nobody trusts. The integration scope at gate 1 covers: live position to MES, work-order status sync, andon escalation rules, OEE attribution at the line and station level, and scheduler feedback.

Done properly, the location layer becomes invisible — it's just better data in the system supervisors already use.

The most common cause of stalled rollouts on a manufacturing floor is supervisor disengagement.

Three moves prevent it: a 30-minute briefing where supervisors see their own historical bottlenecks visualised, a follow-up review where the data exposes one specific actionable improvement, and a continuous-improvement loop where supervisors own which alerts go where.

Once supervisors see the system as theirs, adoption is durable.

Plant directors care whether locating survives metal, multipath and a live line — and whether it changes station dwell and OEE. Criteria:

• Constraint-first scope — instrument the bottleneck before boiling the plant.
• Carrier identity — totes, fixtures and tools tracked in ways the line already handles.
• MES-grade events — position and station state that schedulers and andon can trust.
• Install around production — out-of-hours anchors, tagging in normal handling, rollback for MES writes.

Vendors often lead with UWB accuracy; SIs lead with cable pulls; consultancies lead with digital-twin narratives. The plant needs RF honesty in your environment and a supervisor ritual that lasts past the pilot.

• Standard UHF tags on metal assets without on-metal variants — read rates collapse.
• Ghost and duplicate reads with no edge deduplication or workflow state rules.
• Accuracy right-sized to marketing, not to the decision (zone vs workstation vs tool).
• Supervisor disengagement — no early win visualised from their own bottleneck history.
• Parallel WIP dashboard that never feeds MES routing or OEE attribution.

Industrial RF guidance is blunt: validate tags, mounting and coverage under production load before scale. Remediating a wrong tag fleet after install is a classic self-inflicted cost.

• Demonstrate read performance on our actual carriers and metal surfaces — sample tags first.
• What accuracy at the 95th percentile do we need for the decision we named — and will you pilot that?
• How do station events reach MES, and who owns false andon?
• What is the night-shift tagging and exception process?
• Show callable references at similar scale in metal-rich plants — what failed there?

We start at the constraint, run a predictive RF survey, and keep technology last. No hardware or AMR SKU means we can recommend Passive RFID, BLE, UWB or hybrid based on your line — and stop at gate 1 if the thesis fails on the floor.

Plant directors care whether locating survives metal, multipath and a live line — and whether it changes station dwell and OEE. Criteria:

• Constraint-first scope — instrument the bottleneck before boiling the plant.
• Carrier identity — totes, fixtures and tools tracked in ways the line already handles.
• MES-grade events — position and station state that schedulers and andon can trust.
• Install around production — out-of-hours anchors, tagging in normal handling, rollback for MES writes.

Vendors often lead with UWB accuracy; SIs lead with cable pulls; consultancies lead with digital-twin narratives. The plant needs RF honesty in your environment and a supervisor ritual that lasts past the pilot.

• Standard UHF tags on metal assets without on-metal variants — read rates collapse.
• Ghost and duplicate reads with no edge deduplication or workflow state rules.
• Accuracy right-sized to marketing, not to the decision (zone vs workstation vs tool).
• Supervisor disengagement — no early win visualised from their own bottleneck history.
• Parallel WIP dashboard that never feeds MES routing or OEE attribution.

Industrial RF guidance is blunt: validate tags, mounting and coverage under production load before scale. Remediating a wrong tag fleet after install is a classic self-inflicted cost.

• Demonstrate read performance on our actual carriers and metal surfaces — sample tags first.
• What accuracy at the 95th percentile do we need for the decision we named — and will you pilot that?
• How do station events reach MES, and who owns false andon?
• What is the night-shift tagging and exception process?
• Show callable references at similar scale in metal-rich plants — what failed there?

We start at the constraint, run a predictive RF survey, and keep technology last. No hardware or AMR SKU means we can recommend Passive RFID, BLE, UWB or hybrid based on your line — and stop at gate 1 if the thesis fails on the floor.