Applications / Operations Intelligence
Cycle-Time Deviation
Measures how long each station or process step actually takes, compares it against planned cycle time, and flags where and when the process runs over.
See it in action
Station 7, eleven seconds over takt, every third unit
Video slot — awaiting upload.
What it detects
Planned cycle time lives in the MES. Actual cycle time lives on the floor. Fogsphere measures the second one, from the cameras already watching the station.
Station start and finish events
Detects when work starts and finishes at each station, producing a measured cycle for every unit.
Takt deviation
Compares measured cycle against planned takt, and flags the units that run over.
Repeated over-cycle hotspots
Surfaces the stations and shifts where over-cycle is repeated rather than occasional.
Product entry and exit timing
Times product entry and exit at each station, giving visual cycle time without instrumenting the machine.
Worker task duration
Measures how long each manual task takes, and how that varies by shift, crew and unit.
Station dwell, idle and waiting
Splits the station’s time into active work, waiting for material, waiting for a decision, and idle.
Operator intervention frequency and duration
Counts manual interventions at each station and times how long each one takes.
Setup and adjustment actions
Detects repeated adjustments, fixture changes and setup actions absorbed into what is recorded as run time.
Test and inspection station flow
Measures dwell by test type across end-of-line, leak, calibration and inspection stations.
Bay and treatment-area cycles
Times bay, room and treatment-area cycles where the same measurement supports throughput reporting.
Where it applies
Any repeated process step with a planned duration can be timed visually. The Q3 papers describe it on assembly lines, packaging lines, test bays and clinical pathways alike.
How it works
Edge inference on the cameras you already have
Start and finish events are detected on the edge device next to the camera already covering the station. No PLC tap, and no sensor retro-fit on lines that were never instrumented.
Zones, thresholds and schedules you configure
Draw the station boundary, define what counts as a start and a finish, and set the planned cycle each product variant is measured against. Manual stations and automated cells can each hold different rules.
Every event carries its own evidence
Each detection is logged with a timestamp, the camera and zone it came from, and a short clip. The record can be reviewed, exported or attached to an investigation without trawling through hours of footage.
Feeds the systems already in place
Events surface in Fogsphere Manager and route outward to the MES, historian, andon board, shift-report pack or messaging channel operators already watch.
Risk & control objective
The control objective is knowing where planned time and real time diverge. Averages hide it: a line that hits its daily target can still lose seconds at one station on every cycle and make them back with overtime.
The value is that measured cycle time explains a number the MES only reports. When throughput drops, the MES says by how much. The visual record says which station, which shift, and what the operator was waiting for.
Correlated with PLC and MES data, the same measurement supports an improvement case rather than only a report. A repeated eleven-second over-cycle at one station, evidenced across a week, is a specific engineering task with a known payback.
Related solutions
Measure the cycle time you actually run
We will show you station timing, takt deviation and over-cycle hotspot reporting on your own camera estate.