Skip to content
AFKzona Group
Book a call

AI quality check: a camera at the end of the line

An example metal parts workshop: a camera photographs every bracket, the AI assistant measures it and diverts rejects, hands unclear parts to the operator, and on the shift board spots where the rejects come from. Make the operator's call and approve the maintenance task.

LankstasMetal parts maker
Line 2 · bracket L-60 · order 2417

At the end of the line a camera photographs every bracket from above and from the side. The AI assistant measures it and decides whether it passes.

B1From the press brake →Reject17Review0GatePass1,147Camera 1Camera 2Line running11:00
Camera 2 · profiledeviation ×5Camera 1 · top view#1165 · 11:0010 mm

AI assistant

Model v3.2

Part #1165

L-60 · batch 2417-B · press P1

What the AI checked

  1. Identification
  2. Holes
  3. Hole position
  4. Bend angle
  5. Edges

A part is coming up to the camera. The AI will identify it and check it against the drawing.

Operator queueI. Petrauskienė

Queue empty: the AI is sure about every part

Training examples

  • 10:31#1098 · stain → pass
  • 09:12#0761 · scratch → reject

Shift 2 · 06:00–14:00

11:00
Checked
1,164
Yield
98.5%
Rejects
17

Rejects by cause

  • Burrs7
  • Bend angle5
  • Hole position3
  • Scratches2

Burrs per 20 min

P1P2
06:0007:2008:4010:0011:20

Machines

  • Press P1running
  • Press P2running
  • Press brake B1running

The AI watches reject trends across the shift

At the end of the line a camera photographs every bracket from above and from the side. The AI assistant measures it and decides whether it passes.

The workshop, the people and the measurements are invented. Tolerances are typical for bent steel parts.

Share

Want a system like this for your business?

We build the quality check around your parts, drawings and line: the cameras, the AI model, the operator's station and a shift board that shows where rejects come from. Tell us what you make and where you catch rejects today.

Book a call