/ use case 003 · bottleneck attribution

Where is throughput
being lost?

A station that's idle for ten minutes a day is a hundred and fifty hours a year. Darkfield watches every station, every shift, and produces an hourly distribution of where time is actually spent.

deployed in · food production · factory lines · warehouses · industrial kitchens
/ what we detect

Four things, watched continuously.

a / line stops

When did the line idle, and which station caused it?

State-machine attribution across stations. A stop is recorded with its origin, its duration, and its downstream impact.

b / idle workers

Resting events captured automatically.

Activity recognition trained from scratch in 3 days from your own footage. Detects rest periods of 10s+, 20s+, 30s+, 60s+ at variable thresholds.

c / cycle time

How long did each station take this cycle?

Per-station cycle measurement, compared against expected. Outliers flagged in real time.

d / hourly distribution

When in the day is throughput lowest?

Hourly histograms surface peak-congestion windows automatically. The 10–11am bottleneck spike that nobody knew about, surfaced on day one.

/ the data table

What you receive, row by row.

timestationstatelost
14:14M-3 mixerstopped12 min · feed jam
12:42dividerstopped4:18 · cleared without intervention
11:08prooferrunning
09:36M-2 mixerrunning
06:14shift startrunning
/ try it on your site

First detection on day one. No financial commitment for two weeks.

If we can't reach the accuracy you agreed within two weeks, you pay nothing — and we remove anything we installed.

other use cases
first detection
48 hours from camera connect
onboarding
4–6 weeks of weekly syncs after that
guarantee
two-week risk-free window
compliance
UK GDPR · anonymised in RAM