Home / Case studies / Digitising a welding shop: welding and grinding time per part, now measured

Digitising a welding shop: welding and grinding time per part, now measuredAt a pipe and flange welding station, an existing IP camera measured welding time, grinding time, cycle time per part, output count and PPE compliance (gloves, face protection) automatically.
2026-10-12

At a metal fabrication shop welding pipes, flanges and elbows, production tracking relied on paper forms filled in at the end of the shift. Nobody knew how long a part was welded, how long it was ground or how many minutes a job really took. Hype Vision digitised the process using the existing IP camera that already covered the station.

Welding time and glove/face protection check at a welding station
In short
Problem

Output and times were written on forms at the end of each shift.

Solution

The station was defined as a region of interest in the camera view.

Result

Real welding and grinding time per part measured for the first time; Quotes and costing updated with real cycle times; Station utilisation visible per shift

CameraExisting IP camera (IPC) · RTSP
Analysis areaWelding station ROI
MeasuredWelding · grinding · cycle · output
PPE checkGloves · face shield / welding mask
IdentityAnonymous (station-based)
OutputLive dashboard + shift report

Problem

Output and times were written on forms at the end of each shift. How much welding and grinding each part needed, how actively the station was used and whether the times used in quoting were realistic were all unknown.

Glove and face-protection use during welding could only be checked when the HSE specialist walked by.

What we measure

The station was defined as a region of interest in the camera view. The model detects the welding arc, grinding sparks, gloves, face protection and the workpiece in every frame; tied to time, these detections become operational data.

Welding time and glove/face protection check at a welding station
Welding — duration counter, gloves and face protection OK
Spark and PPE detection during grinding
Grinding — spark detection, goggles and gloves OK
  • Welding time: total time the arc is on
  • Grinding time: total time sparks are detected
  • Cycle time per part: job start → part finished
  • Output: number of completed parts
  • Active / idle time: station utilisation during the day
  • PPE compliance: gloves and face protection while welding and grinding

What the dashboard shows

Each finished part creates a row. A station report is ready at the end of the shift and can be exported to ERP/MES or Excel.

PartWeldingGrindingCyclePPE
Flanged pipe #146 min 40 s2 min 10 s14 minOK
Elbow #154 min 05 s1 min 35 s9 minOK
Flanged pipe #167 min 15 s2 min 50 s16 minNo gloves · 12 s

Privacy

Analysis is station-based; no face recognition is used and nobody is identified. Video is processed on site and only time, count and event data reach the dashboard.

Results

Measured times from the first period were compared with the default times used in planning and costing.

  • Real welding and grinding time per part measured for the first time
  • Quotes and costing updated with real cycle times
  • Station utilisation visible per shift
  • Daily output reported automatically, no paper forms
  • PPE compliance during welding and grinding logged 24/7

Related solutions

Frequently asked questions

Is this tracking individual workers?

No. Measurement is station-based: we measure how much welding, grinding and work happened at the station, not who did it. No face recognition is used.

Do we need new cameras?

This project used the existing IP camera covering the station. If the camera clearly sees the station and the part, no extra hardware is needed.

Bird's-eye AI camera view detecting people across a factory

See it on your site in two weeks.One camera, one module, real data.