To measure cycle time and microstops reliably, you need to capture every cycle at every station, not a handful of stopwatch samples a year. Fixed cameras combined with vision AI can detect where each cycle starts and ends, separate waiting, searching and walking time, and flag a station that keeps running over takt. That is how cycle time and microstop measurement works in TIM INSPECT.
Most production teams know their planned cycle times by heart. Far fewer know their real ones, station by station, shift by shift. This guide looks at why microstops stay invisible, why classic time studies cannot fix that and what a camera-based approach changes on the floor.
A microstop is a pause of a few seconds: a worker reaching for a part that is not where it should be, waiting for the previous station, walking to fetch a tool. None of these is long enough to be logged as downtime. Yet repeated across hundreds of cycles per shift, they add up to a real capacity loss.
The second cost is less obvious. Without measured data, the question "which station is the bottleneck?" becomes a matter of opinion. Industrial engineering points to one station, the line leader to another, and improvement effort goes where the loudest argument is rather than where the time is actually lost.
A time study gives a careful snapshot of a station, often only a few times a year. It reflects the day of the study, the operator observed and the behaviour that changes when someone is standing there with a stopwatch.
Counters record completed units and automated machine cycles well. Manual stations, however, rarely produce a clean start and end signal, and short pauses fall below any downtime threshold.
Asking people to note every brief stop does not work in practice. The pauses are too short and too frequent, and the record ends up incomplete.
Some teams film a station and analyse the footage afterwards. It is insightful but slow, and it does not scale to every station on every shift.
In the TIM INSPECT cycle time scenario, measurement runs continuously in four steps.
The difference from a time study is coverage. Instead of a few measurements a year, every cycle of the shift is recorded. The gap between planned and actual time becomes visible per station, and the bottleneck discussion can start from data.
Picture a four-station line with a 38-second takt. Station three has run 46 seconds for three cycles in a row and the rest of the line is starting to wait. The supervisor's AXION vibrates with the station number and the latest cycle time. They walk to the station, find the cause, perhaps a material bin placed too far away, and resolve it while the shift is still running. The run of over-takt cycles is reported, and the station's cycle times stay on record for the end-of-day review.
Each camera must see the point where a cycle visibly begins and ends: a part entering the fixture, a unit leaving it. One camera can sometimes cover two neighbouring stations; that depends on layout and occlusion by racks or people. The scope of measured times is finalised around the camera angles.
How many over-takt cycles in a row should trigger a notification? What counts as a build-up? Thresholds are set together during the pilot. Too sensitive and supervisors start ignoring alerts; too loose and the system says nothing useful.
Measurement and reporting are done by station, not by individual. The goal is to reveal bottlenecks and non-value-added time, not to compare workers. Introducing the project together with employee representatives is strongly recommended.
Zones and actions are observed rather than identity. Images are filtered and only the event moment is stored, and the design follows GDPR and local data protection rules from day one.
Planned times come from the MES so that measured cycles can be compared against them. Agree early on which lines, products and planned times will be used for the comparison.
If the losses you are chasing sit at a machine rather than at a manual station, the related machine utilization scenario classifies running, waiting, setup and fault states from video. Both are part of the TIM INSPECT platform.
No. Measurement and reports are per station. The aim is to make bottlenecks, build-ups and non-value-added time visible, not to rank workers.
The start and end of each cycle, waiting, searching for material, walking, build-ups between stations and deviation from takt time.
When a station exceeds takt several cycles in a row or a build-up starts ahead of the line, the shift supervisor receives it on AXION. A single long cycle is not enough to trigger it.
Cycle times measured from video are compared with the planned times in the MES, so the gap between plan and reality is visible station by station.
If you would like to discuss your own bottleneck with measurements in hand, request a demo and we will look at your stations, takt time and camera views together.