Industrial Process Monitoring

Monitors manufacturing processes in line, detects defects or deviations, and supports process adjustments before issues spread.

Business impact

• Scrap rate — Siemens reports a 10 to 20 percent scrap-rate reduction for P&G.
• Defects — Nissan reported a reduction in defects associated with process changes informed by system data.

Data requirements

• Inspection images (image) — Used for visual defect detection on products and painted vehicles.
• Production process data (tabular) — Used to detect deviations, monitor quality attributes, and support process changes.
• Location and tracking data (sensor) — Used to track parts and identify bottlenecks in the factory flow.

AI methods and techniques

• Anomaly detection — Used to flag production deviations and unusual conditions in manufacturing data.
• Computer vision — Used in visual inspection to detect flaws on products or vehicles.
• Deep learning — Used in the Visual Inspection Cockpit for defect detection during production.
• Machine learning — Used in production systems to provide early warnings and trigger operator alerts or machine setting adjustments.

Industries

Real-world evidence

5 documented case studies on record.

Companies using this: Nestlé, Nissan, P&G, RTX, Philips.