Nordbo QC Platform increases Amphenol Phoenix automated inspection throughput

Nordbo Robotics published an Amphenol Phoenix case study on 15 August 2025 reporting daily inspection throughput increasing from about 4,800 to 24,685 parts using its QC Platform. The supplier’s headline percentage conflicts with those totals, so the before-and-after counts provide the clearer account of its claimed result. The date marks publication of the case study, rather than a disclosed commissioning date.

The installation uses a bowl feeder and nine camera stations to capture 21 high-resolution images of each component. Its inspection models assess more than 30 defect classes and route rejected parts into seven groups. Nordbo reports inspection time falling from around 20 seconds to 3.5 seconds per part compared with the previous manual process.

The system was trained using datasets of defects and human annotations, allowing the models to distinguish flaws across the component surfaces. Nordbo monitored operation and updated the models during the first month after installation. The case study describes a complete inspection and sorting workflow for Amphenol’s interconnect components, with the throughput and cycle-time figures remaining supplier-reported results.