Siemens and Procter & Gamble (P&G) are expanding the use of an AI-based quality inspection system across manufacturing operations worldwide, highlighting the growing role of industrial AI and edge computing in automated production.

The system uses Siemens Industrial Edge technology together with industrial computing hardware and AI technologies to inspect products in real time while production lines are operating at full speed. The solution is designed to identify quality issues and automatically support production decisions without requiring manufacturers to rely solely on traditional machine-vision configurations.

According to the companies, deployments using the system have achieved scrap reductions of approximately 10% to 20%, depending on the product and manufacturing application. Siemens and P&G also report that new inspection deployments can be commissioned five to ten times faster than traditional customized vision systems.

One important aspect of the project is the use of edge computing. Instead of sending all production data to a remote cloud environment, inspection information can be processed close to the manufacturing equipment. This allows quality decisions to be made rapidly and enables inspection results to interact directly with production automation.

The system can also integrate with PLC-controlled production equipment. When a defective product is identified, the inspection system can trigger actions such as alarms or automatic product rejection. Quality information can subsequently be collected and analyzed to identify production trends and support continuous improvement.

The development demonstrates how artificial intelligence is becoming increasingly integrated with conventional industrial automation technologies. PLCs, industrial PCs, machine vision, sensors and edge computing can work together as part of a connected production architecture.

For manufacturers, the combination of AI inspection and industrial automation can provide a way to improve quality control without separating inspection processes from existing production systems. The project also illustrates how AI is moving from experimental applications toward large-scale industrial deployment.