Why Your Plant Floor
Is Flying Blind
(And What It's Costing You)

It was a Tuesday afternoon — which, in my experience, is exactly when these things happen.

A plant manager in the Niagara region called me, frustrated. A critical crankshaft job had gone missing. Not stolen. Not scrapped. Just… somewhere on the floor. Four people spent three hours walking the plant, checking stations, calling across departments. The job was sitting twelve metres from where it had been scanned two days earlier. It had been bumped by a forklift, sat against a wall, and nobody updated anything — because updating things takes time, and time is the one thing nobody has on a busy shop floor.

Three hours. Four people. For a job that never left the building.

The Visibility Problem Nobody Talks About

There is a lot of attention in manufacturing right now on machines: uptime, OEE, cycle times, predictive maintenance. That focus is deserved. But between the machines — in the aisles, the queues, the staging areas, the "temporary" spots that become permanent — there is almost no visibility at all.

Most Ontario manufacturers are tracking WIP (work-in-progress) the same way they were twenty years ago: paper travellers, barcode scans at fixed stations, and a healthy dose of tribal knowledge. The machinist who knows the job went to the deburring bench. The supervisor who remembers it was pulled aside for an urgent rework. The whiteboard that was accurate on Monday morning.

This works — until it doesn't. And when it doesn't, the cost is never in one dramatic event. It compounds quietly across dozens of small delays every week.

Three Symptoms That Tell Me a Plant Is Flying Blind

Over the past several years implementing IIoT and RTLS systems for manufacturers across Ontario, I've learned to look for three specific patterns. If you recognize any of these, you're carrying more cost than you know.

Symptom one in depth

Barcode-based tracking only tells you where a job was when it was last scanned. On a busy shop floor, that scan might have been yesterday, at a station three moves ago. In the time between scans, the job exists in a kind of informational fog. People compensate with good memory and frequent walkabouts.

This isn't a people problem. It's a system design problem. Your people are doing exactly what the system requires of them.

Symptom two in depth

When scheduling decisions are made on data that doesn't reflect reality, you get sequencing errors, machine conflicts, and rush jobs that didn't need to be rushed. Every planner I've worked with has learned, consciously or not, to mentally discount their system data and rely on what they see. That's a sign the system has already failed.

Symptom three in depth

Try this: at a random moment during your production shift, ask your team to tell you — with confidence — the location and status of your top five active jobs. Not an approximation. Not a scan from this morning. Now.

In plants with real-time visibility, that question is answered by looking at a screen. In plants without it, the question triggers a small emergency. People make calls. Someone goes to look. An answer comes back in five or fifteen or thirty minutes — and you never know how accurate it is.

If you can't answer that question quickly, you cannot respond quickly when something goes wrong. And something always goes wrong.

What It Actually Costs

The costs here are rarely captured in a single line item. They're distributed across labour hours, expediting decisions, and margin erosion that gets attributed to "how manufacturing works."

I'm not going to invent a universal cost figure here because every plant is different. But I have yet to walk a floor where the true cost of WIP invisibility — properly accounted — came out to less than the cost of fixing it.

The Fix Is Not Another Spreadsheet

The temptation, when faced with this problem, is to add a layer of process: a new check-in procedure, a better whiteboard, another required scan. Sometimes that helps at the margin. More often, it adds administrative burden without closing the gap — because the gap isn't behavioural, it's architectural.

Real-time WIP visibility requires real-time location data. That means knowing where a job is continuously, not just at the moments someone remembers to scan it.

Modern UWB-based RTLS (Real-Time Location Systems) — the same technology we deploy with our Sewio partner platform — can provide 30-centimetre location accuracy across an entire facility, updating continuously, without requiring anyone to scan anything. A job moves; the system knows. A pallet gets staged; the dashboard reflects it. A supervisor asks "where is that job?" and the answer is already on the screen.

This isn't a futuristic concept. It's running in Canadian manufacturing facilities right now. And it doesn't require ripping out your ERP or rebuilding your floor — it layers on top of what you already have.

Start with the Question, Not the Technology

I'll close with the same advice I give every plant manager I meet with: don't start by evaluating RTLS vendors. Start by answering one question honestly.

Walking to find jobs. Calling to verify status. Correcting scheduling decisions based on stale data. Resequencing because something wasn't where it was supposed to be.

Add those hours up. Multiply by your burdened labour rate. That number is your baseline. That's what you're spending to operate a system that requires humans to patch its gaps.

Once you have that number, the conversation about what to do next becomes much easier.