Factory overview
The plant missed target. Which machine?
Most plant dashboards answer the first half of that question and stop. Factory Overview pools every machine into a plant-level read, then lets you walk down the same three panels — OEE, production, downtime — through line and cell to the machine that actually lost the time, and the reason it stopped.
Plant → line → cell → machine same three panels at every level worst-first ranking
The problem
The gap between knowing and finding
- The morning meeting agrees the plant is behind, then spends forty minutes arguing about which line it was.
- OEE is reported per machine, so nobody can say what a line is doing without adding up a spreadsheet by hand.
- Downtime totals arrive alphabetically, not worst-first — the biggest loss is somewhere in row nine.
- By the time the cause is traced to one cell, the shift it happened in is over and nobody remembers the stop.
How it works
From the whole plant to one machine, in four steps
Every step is the same three panels at a narrower scope. Nothing new to learn on the way down — only fewer machines in the number.
The whole plant, in one row
Every machine pooled into a single read: OEE with its three components, parts good against parts expected, and total lost time. One row is enough to know the day is short — it is not yet enough to know where.
The same figures the rest of the screen breaks down — nothing below can disagree with this row, because it is pooled from it.
Which line owns the gap
One card per line, each carrying its own machine count and the machine currently holding it back. The three panels repeat unchanged, so comparing two lines is reading the same shape twice rather than reconciling two different reports.
Turning is the line to open: the lowest OEE of the three, and most of the plant's lost time sits inside it.
Machining
4 machines · Bottleneck: VMC-01Turning
3 machines · Bottleneck: Lathe-02Finishing
2 machines · Bottleneck: GRD-01Inside the line, down to the cell
A cell is the last grouping before equipment. Its downtime panel stops listing cells and starts naming machines — worst first, with a real stop count each — so this is the step where the answer changes from a part of the plant to a thing you can walk up to.
The same three panels again, pooled from just this cell's machines — and its downtime column is already a list of them.
Cell C — Turning
2 machines · Bottleneck: Lathe-02The machine, and why it stopped
The bottom of the hierarchy. Every machine in the cell, worst OEE first, with what each one is doing right now and how long it has been doing it — and for whichever you pick, the same three panels one final time. Here the downtime column is no longer a list of places: it is the stop reasons themselves.
Pick a machine on the left. Current fills in on this one — a single machine is the only scope with a live reading to attribute to one instant.
Lathe-02
Alarm: 17m · 4 stops- Availability1.1 h · 27%
- Performance2.9 h · 69%
- Quality0.2 h · 4%
| Operation | Good vs exp. | Rejected | OEE % | Cycle vs expected |
|---|---|---|---|---|
| O1240 Rough turn | 14 / 30 | 1 | 46% | 1m 18s / 1m 2s |
| O1265 Finish turn | 12 / 25 | 1 | 51% | 1m 5s / 58s |
What each panel shows
Three panels, read side by side
OEE tells you how well the time ran, production tells you what came out of it, and downtime tells you where it went. Splitting them across three screens is what makes the cause hard to find.
Three numbers that only mean something together
Availability, performance and quality each get a reading and the hour-by-hour shape beside it. High availability next to low performance is a completely different problem from the reverse — and you can only see which one you have when they sit on the same row.
Overall, availability, performance and quality, each with its own hourly trend.
Measured against expected by now, not by end of shift
The headline count and its percentage are the easy part. The bar strip is the one that changes behaviour: real output for each hour against that hour’s own target, rejects stacked in red where they happened, and the hours still to come left visible rather than dropped.
Each bar is one shift hour; the dashed line is that hour's planned rate, not a full-shift target.
Downtime arrives already ranked
Total lost time, broken down by line and ordered by loss — not an alphabetical list you have to scan. On the live screen each bar is also the way down a level, so the ranking and the navigation are the same control.
Worst-first, with a real stop count beside each duration.
Why it works
Built on your own plant structure
One shape, every level
OEE, production and downtime sit side by side for the plant, a line, a cell and a single machine — the same three panels at all four scopes, so you never re-learn a screen on the way down.
Pooled, not averaged by hand
A line's numbers are computed from its own machines' telemetry against the shift calendar. Nothing is typed in, so the line rows and the plant total cannot disagree.
The bottleneck, named
Each line card names the machine holding it back, so the drill-down starts from an answer rather than an alphabetical list of everything you own.
Questions
Common questions
Your own plant structure. You group machines into cells and cells into lines during setup, and every rollup on this screen is computed against that hierarchy — so a line means what your supervisors already call a line, not a fixed category we impose.
Because a pooled figure across several lines has no single live reading to attribute to one instant. The window's real pooled average is shown instead of a separately-derived number that could quietly disagree with it. Drill into one line and Current fills in.
That hour's own planned rate. Each bar is the real output for that hour, with rejects stacked in red, so a bad hour is visible while there is still shift left to react to it. Hours that have not started yet are dimmed rather than hidden.
Yes — the whole screen is driven by one window-and-shift filter: today, a named shift, yesterday, a week, or a custom range. The drill-down carries your filter down with it, so opening a line does not reset you back to today.
No. The drill-down ends on a machine’s own rollup and its stop reasons, which is enough to know where to go and why. The dedicated machine screen is still where you see its hour-by-hour timeline and classify individual stops.
Get started
See your own plant on this screen
A 30-minute demo, then a pilot on your own floor if it fits.