Siemens and Procter & Gamble Expand AI-Based Quality Inspection Across Global Manufacturing Oper
Time:2026-09-28 Browse: 0
Artificial intelligence is moving deeper into industrial production, and one of the latest developments is the global expansion of an AI-based quality inspection solution developed by Siemens and Procter & Gamble. Announced in September 2026, the collaboration demonstrates how industrial AI, edge computing, machine vision and PLC integration are increasingly being combined directly with production processes rather than operating as isolated digital tools.
The solution is being expanded across Procter & Gamble manufacturing operations worldwide. It is designed to inspect products in real time while production lines are operating at full speed. According to the companies, the technology can provide comprehensive inspection coverage and has helped reduce scrap rates by approximately 10% to 20%, depending on the application.
For manufacturers, this development is particularly relevant because quality inspection has traditionally required a combination of cameras, predefined rules, customized vision algorithms and manual engineering work. These systems can be highly effective, but they may become difficult to adapt when product designs, packaging formats or production conditions change.
AI-based inspection introduces another layer of flexibility.
AI Moves Quality Inspection Closer to the Production Line
Traditional machine vision systems generally rely on predefined inspection parameters. Engineers configure cameras, lighting, image-processing algorithms and rejection criteria for a specific product or production process. When production requirements change, the vision system may require additional engineering and testing.
AI-based inspection can approach some of these challenges differently. Instead of relying entirely on manually defined rules, AI models can be trained to recognize product characteristics and identify defects across a broader range of conditions.
In the Siemens and Procter & Gamble implementation, inspection results are processed close to production equipment using Siemens Industrial Edge technology. This edge-based architecture is important for industrial environments because inspection decisions often need to be made within very short time intervals.
A production line cannot necessarily wait for image data to travel to a remote cloud platform, be analyzed and then return with an instruction. When a defective product is detected, the system may need to trigger a reject mechanism immediately.
Processing information closer to the machine helps reduce this dependency on remote infrastructure.
It also supports a broader industrial automation architecture in which machine vision, edge computing, PLC control and production data work together.
Direct PLC Integration Is a Key Part of the Architecture
One of the technically important aspects of the solution is its integration with PLC-based production control.
A machine vision system has limited value if inspection results remain inside a separate application. In an automated production environment, the inspection result needs to become an actionable control signal.
For example, a simplified production sequence may operate as follows:
Product enters the inspection zone → industrial camera captures the product → AI model analyzes the image → inspection result is generated → PLC receives the result → PLC coordinates the reject mechanism → production data is recorded.
This type of architecture allows quality control to become part of the production control loop.
PLC integration also allows manufacturers to coordinate inspection results with machine timing, conveyor movement, servo positioning, pneumatic actuators, safety functions and other automation components.
This is particularly important for high-speed production lines. A quality inspection system must not only identify defects accurately; it must also communicate the result quickly enough for the control system to respond to the correct product.
Why Edge Computing Matters in Industrial Automation
The use of Industrial Edge also reflects a wider trend in factory automation.
Industrial facilities increasingly generate enormous amounts of data from PLCs, sensors, drives, cameras, robots and other field devices. Sending all this information directly to centralized cloud systems can introduce latency, bandwidth requirements and cybersecurity considerations.
Edge computing provides an intermediate layer.
Instead of sending every raw image or machine signal to a remote data center, an industrial edge platform can process information near the equipment and send selected results to higher-level systems.
This architecture can be useful for applications such as:
AI-based quality inspection
Predictive maintenance
Machine condition monitoring
Production analytics
Energy monitoring
Process optimization
Industrial anomaly detection
For automation engineers, this means the traditional PLC-centered architecture is increasingly becoming part of a larger computing environment.
The PLC continues to handle deterministic control functions, while edge computing provides additional processing capabilities for applications that require more computational resources.
Reducing Scrap Through Earlier Defect Detection
One of the practical goals of AI quality inspection is reducing production waste.
If a defect is detected only after a production batch has already been completed, manufacturers may need to inspect or discard a large quantity of products. Real-time inspection changes the timing of the response.
When defects are detected continuously during production, operators can potentially identify process deviations earlier.
For example, a packaging line may experience gradual changes in alignment, printing quality, sealing performance or product positioning. An automated inspection system can monitor these characteristics continuously and provide information that allows production teams to investigate the underlying process.
This creates a connection between quality inspection and process optimization.
Instead of treating quality control as a final checkpoint, manufacturers can increasingly treat it as a continuous source of production intelligence.
Faster Deployment of AI-Based Inspection
Siemens and Procter & Gamble also reported that new inspection deployments can be commissioned five to ten times faster than traditional customized vision systems.
This is significant because engineering time can become a major factor in machine vision projects.
A conventional project may involve camera selection, lighting design, image collection, algorithm development, testing, PLC communication, HMI development, mechanical integration and production validation.
If an AI-based platform can simplify part of this workflow, engineering teams may be able to deploy inspection applications across more production lines without rebuilding the entire solution from the beginning.
This approach fits the broader direction of modern industrial automation: reusable software, standardized architectures and data-driven engineering are becoming increasingly important.
What This Means for Smart Manufacturing
The Siemens and Procter & Gamble project also illustrates how smart manufacturing is evolving beyond dashboards and data collection.
Earlier Industry 4.0 projects often focused on collecting data from machines and displaying production information through SCADA, MES or cloud platforms.
The next stage is increasingly focused on using data to make decisions and trigger actions.
An AI quality inspection system is an example of this transition.
The system does not simply collect an image. It interprets the image, generates an inspection result and can connect that result with an automated production response.
This creates a closed-loop relationship between sensing, computing and control.
In a modern factory, the architecture may therefore contain several layers:
Field Layer: sensors, cameras, actuators and instruments.
Control Layer: PLCs, PACs, motion controllers and safety controllers.
Edge Layer: industrial computers and AI inference platforms.
Supervisory Layer: SCADA, HMI and production monitoring.
Management Layer: MES, analytics and enterprise applications.
The increasing integration between these layers is one of the most important developments in industrial automation.
AI Does Not Eliminate the Need for Automation Engineers
Despite the growing role of AI, industrial deployment still requires engineering expertise.
AI models need suitable data, validation and operational boundaries. PLC logic still needs to be tested. Machine safety requirements still need to be respected. Sensors and cameras still need to be installed correctly. Communication networks must remain reliable.
Engineers also need to understand what happens when an AI model produces an uncertain or incorrect result.
In production environments, AI therefore becomes another engineering component rather than a replacement for the entire automation system.
The combination of AI and traditional industrial control may ultimately be more important than either technology alone.
The Future of AI-Based Industrial Inspection
The global expansion of Siemens and Procter & Gamble's AI-based quality inspection solution demonstrates a clear direction for modern manufacturing.
Machine vision is becoming more intelligent, edge computing is moving closer to production equipment, and PLC systems are increasingly connected to AI-driven applications.
For manufacturers, the value is not simply in adding AI to an existing factory. The larger opportunity is to connect AI with the automation architecture already responsible for production.
As more factories adopt Industrial Edge, PLC integration, machine vision and AI analytics, quality inspection can become an active part of the production control ecosystem.
The result is a manufacturing environment in which machines do more than execute predefined sequences. They can increasingly observe production conditions, analyze information and provide automated responses in real time.
This evolution will continue to influence PLC programming, industrial networking, SCADA, machine vision, edge computing and smart factory design throughout the coming years.
