How Schneider Electric standardised AI inspection with Cognex OneVision

For decades, Cognex and Schneider Electric have partnered to advance machine vision in industrial manufacturing. But as production demands intensified and workforce challenges grow, Schneider Electric recognised that traditional inspection systems were no longer enough.

Schneider Electric set out to solve a pressing challenge: how to expand inspection coverage, increase throughput and reduce manufacturing costs, without increasing complexity or reliance on individual technical expertise. The answer came with the integration of Cognex In-Sight cameras and the OneVision™ AI development environment.

The result? A transformation in quality control that doubled production yield, reduced false rejects by 70 times and established a global standard for AI-powered inspection.

The challenge: Expanding inspection without adding complexity

The team’s goal was clear: develop advanced inspection capabilities in France. If successful, the solution would be deployed as a standardised solution worldwide.

An earlier generation of vision tools was used for on-line inspections, primarily focused on assembly inspection and measurement. While effective for basic validation, these tools could not support the broader inspection needs required to scale production or meet increasing quality expectations.

Schneider Electric faced several challenges:

• The need to inspect more control points without slowing production
• Scarcity of skilled vision engineers
• Increasing pressure to reduce manufacturing costs
• The desire to democratise AI for operators and plant engineers
• Growing demand for global standardisation across facilities
• Having a centralised validated models replicable/scalable worldwide to create global standard.

The factory team recognised that incremental improvements would not be enough. It needed a platform that could expand inspection capabilities, simplify AI adoption and enable collaboration across locations. Originated by engineers and operations leaders, this initiative identified the improvement opportunities and pushed the project upward to management, to align it with broader corporate goals of efficiency, scalability and digital transformation.

The solution: Cognex OneVision and In-Sight cameras

Schneider Electric wanted to enhance its existing ecosystem rather than seeking alternatives. The objective was to increase throughput and capability within a trusted platform.

The implementation integrated OneVision and In-Sight cameras with the In-Sight Vision Suite to combine traditional rule-based vision tools with an advanced AI tool set. The new set up expanded inspection from limited measurement checks to 17 distinct control areas. These inspections included:

• Height control
• Screw position and rotation validation
• Sorting
• Orientation detection
• Detection of shortages at the base or head of contactors.

Using the AI tools in OneVision, engineers can label images with a high degree of granularity and detail, as well as refine ROI and retrain models rapidly.

One of the most powerful capabilities is the ability to train models efficiently, even starting with minimal image sets, and refine them iteratively within a collaborative environment.

The platform also eliminated an ongoing hardware challenge. Previously, advanced AI inspection often required PCs equipped with high-performance graphics cards. With OneVision, that requirement disappeared. This configuration change not only reduced direct hardware costs but simplified system architecture and maintenance.

Overcoming security and adoption challenges

As with any cloud platform in a global enterprise, security validation was critical in evaluating OneVision.

Schneider Electric conducted a comprehensive security assessment of the cloud-based solution. Security and engineering teams from Cognex supported the process by completing third-party security questionnaires and providing necessary certifications. The platform met Schneider Electric’s most demanding global standards.

The broader testing and validation process spanned approximately six months. Much of the effort was concentrated at the beginning, where teams focused on training, validation and ensuring operational reliability before full-scale deployment.

Despite the rigorous process, the ease of use and ergonomic interface of OneVision accelerated adoption. Engineers found that new inspection solutions could be developed and integrated 30% faster compared to previous approaches.

The results: Measurable and transformational

The impact of the implementation was substantial and measurable.

Decreasing non-quality costs without increasing cycle time

Schneider Electric enhanced production line control by expanding inspection coverage from five critical areas with traditional methods to 17 areas using OneVision, and improved detection accuracy. Additionally, cycle time reduced from 400 to 200 ms.

70-fold reduction in false bad rates

Using the AI capabilities of OneVision, Schneider Electric reduced false bad rates by a factor of 70. This outcome improved operational efficiency and reduced unnecessary waste.

30% faster integration of new controls

The simplicity and execution speed of the platform reduced integration time for new inspection applications by 30%. The result? Faster iteration and continuous improvement.

Cost reduction: Direct and indirect

Direct cost savings:

• No need for PCs with graphics cards
• Reduced hardware complexity.

Indirect cost savings:

• Less time spent in meetings co-ordinating development
• Faster collaboration across teams
• Reduced development effort
• Lower maintenance overhead.

Together, these improvements delivered quality and cost savings at approximately twice the previous rate.

Unexpected benefits

While the team anticipated quality and efficiency gains, several benefits exceeded expectations:

• The 70-fold reduction in false bad rates
• Reduced co-ordination and meeting time
• Faster development cycles than originally forecast.

Perhaps most importantly, the platform’s simplicity unlocked new possibilities. Engineers can now address new applications because development requires only a few clicks. This mindset shift, from constraint to capability, may be one of the most powerful outcomes of the project.

Enabling global collaboration

One of the most strategic advantages of OneVision was its centralised collaboration capability.

The location in France could develop inspection standards and refine a project that can then be deployed to other facilities almost instantly. Such a high level of global collaboration was not previously possible. Traditional camera-based training was constrained by hardware limitations. Now, models can be trained, refined and distributed worldwide through a centralised platform creating a standardised inspection framework and replicating them across factories with just a few clicks.

This capability is especially powerful in addressing workforce scarcity. Instead of requiring each plant to develop expertise independently, knowledge and AI models can be shared seamlessly across the organisation, and even external third parties can be leveraged to optimise the solution.

Democratizing AI for manufacturing

A key objective for this project was to make AI in machine vision more inclusive, intuitive, collaborative and accessible. The intuitive interface and simplified model training process in OneVision allow operators and engineers, without deep AI specialization, to participate in building and refining inspection solutions.

The combination of traditional rule-based tools and advanced AI in a single environment provides flexibility and versatility. Teams can choose simpler inspection modes or more accurate AI-driven classification as needed. This hybrid capability lowers the learning curve, which also helps Schneider Electric address one of the most pressing challenges in modern manufacturing: workforce limitations.

A competitive advantage

The transformation goes beyond operational efficiency. By expanding inspection coverage, reducing false rejects, and lowering manufacturing costs, Schneider Electric strengthened its competitive position.

Notably, the ability to deploy advanced AI inspection globally, quickly and collaboratively provides a distinct advantage. Fewer false rejects mean less waste. Faster integration means quicker innovation cycles. Standardisation across plants ensures consistent quality worldwide.

The site now serves as a model for broader deployment. Schneider Electric’s stated goal is to expand this solution more widely across its global operations. The collaborative platform makes this expansion feasible and scalable.

A stronger partnership for the future

The initiative further strengthened the long-standing partnership between Cognex and Schneider Electric.

By enhancing the existing ecosystem rather than replacing it, Schneider Electric demonstrated confidence in the Cognex innovation roadmap. In turn, Cognex supported Schneider Electric through rigorous security validation and collaborative implementation.

The project has opened the door to broader AI-driven manufacturing initiatives and expanded global collaboration.

Looking ahead

Schneider Electric’s journey with OneVision is just beginning. With proven results in France, the company is positioned to standardise intelligent vision systems across additional factories. The platform has demonstrated that AI in manufacturing does not have to be complex or resource intensive. It can be collaborative, secure, scalable and transformative.

As Schneider Electric continues expanding this solution, the benefits are clear:

• Higher quality
• Greater efficiency
• Lower costs
• Faster innovation
• Global standardisation.

Through the combination of advanced hardware and collaborative AI, Schneider Electric has redefined what’s possible in machine vision and set a new benchmark for intelligent manufacturing worldwide.