AI Visual Quality Control for Mechanical Manufacturer
Mechanical components manufacturer in Emilia-Romagna - 65 employees, 3 production lines
99.2% (from 87%)
Defects identified
-89% (from 3.5% to 0.4%)
Client return rate
-160,000 EUR
Annual return costs
+400% (0.3s per part vs 8s manual)
Inspection speed
3.8x
First-year ROI
The problem to solve
Manual end-of-line quality control only identified 87% of defects, with a client return rate of 3.5%. Each return cost an average of 1,200 EUR in logistics, rework, and brand damage. The annual cost of returns exceeded 180,000 EUR.
What we implemented
Computer vision system with high-resolution cameras positioned at end-of-line. The AI identifies surface, dimensional, and finishing defects in real time, classifying each part as compliant, requiring rework, or to be scrapped. Quality dashboard with trends and alerts for preventive die maintenance.
Measured impact
Defects identified
99.2% (from 87%)
Client return rate
-89% (from 3.5% to 0.4%)
Annual return costs
-160,000 EUR
Inspection speed
+400% (0.3s per part vs 8s manual)
First-year ROI
3.8x
Timeline
12 weeks (6 development + 3 model training + 3 calibration)
Investment
70,000 - 95,000 EUR (Transition 5.0 up to 45%)
Similar projects
Want similar results in your company?
We'll analyze your processes for free and show you where AI can generate the greatest impact.