How an Autonomous Machine Vision System Increased Accuracy and Reduced Waste







Case Study: How an Autonomous Machine Vision System Increased Accuracy and Reduced Waste

The BSH Home Appliance Group (BSH) wanted to implement a quality assurance system to detect defects at an early stage and reduce material waste, with the aim of reducing costs while benefiting the environment. The company has found the ideal partner for InspectionGerman-Israeli quality assurance specialist who invented Autonomous Computer Vision (AMV).


The BSH Appliance Group comprises 40 locations worldwide and employs over 62,000 people. Its commitment to sustainability translates into the production of energy-efficient appliances, as well as the reduction of the company’s environmental footprint in all areas of the value chain.


Like Inspekto, BSH strongly believes in innovation through digitalization and 4.9% of the company’s expenditure is dedicated to research and development (R&D). As a result, the company created the BSH Startup Kitchen, an initiative that offers young companies the opportunity to collaborate with BSH by offering their cutting-edge solutions to improve the company’s products and processes. The BSH Startup Kitchen tests promising new technologies and, after a successful pilot phase, offers the possibility of a long-term business relationship.


“We recognized that startups are a valuable source of innovative technologies and solutions for many of our business segments,” said Lars Roessler, Venture Capital Partner at BSH Startup Kitchen. “BSH can apply these innovations directly in our product development and increase the productivity of our processes.”


One of the areas where BSH intended to improve its existing processes was in quality assurance (QA). As the systems in place still allowed some faulty items to pass, BSH was looking for a quality assurance method that met the company’s strict quality standards without being overly complex and cumbersome to deploy. As the need for an accurate, reliable, yet user-friendly system arose, BSH’s Startup Kitchen decided to approach Inspekto, the pioneer of autonomous machine vision.


The challenge


QA automation saves manufacturers time and money. The cost of poor product quality is notorious – damage to a hard-earned reputation, erosion of customer trust, costly recalls, wasted materials and rework costs are just some of the consequences. the release of defective products. For this reason, quality assurance is a crucial step in every manufacturing process, regardless of industry size or sector. However, manual QA is not suitable for the strict Industry 4.0 standards, as human inspectors can miss defects, especially when inspecting very complex electrical elements. On the other hand, traditional machine vision solutions are extremely expensive and complex to set up and maintain, making them impractical for many manufacturers. BSH wanted to implement a reliable automated quality assurance system, but struggled to find a satisfactory solution.


“Even with multiple inspection checks, errors kept showing up, increasing disposal costs,” explained BSH Startup Kitchen Partner Dipjyoti Deb. “BSH had experimented with automated inspection solutions in the past, but each proved unsatisfactory and costly.”


BSH’s challenge was to increase the accuracy and efficiency of batch inspection processes, in a way that was simple and did not require the design and installation of a complex, custom project.


BSH project engineers Markus Maier and Stefan Schauberger were responsible for reducing component fault detection time at one of BSH’s furnace manufacturing plants in Traunreut, Germany. They approached BSH Startup Kitchen with this issue and a partnership with Inspekto was formed.


The solution


Inspekto is the inventor of AMV, a new approach to industrial quality assurance that mimics the entire process of human vision while maintaining the reliability and repeatability of industrial machine vision.


Just as the human brain adapts our unique optical system – our eyes – to each scenario, AMV adapts a unique electro-optical system to accommodate a wide range of use cases. Therefore, AMV systems are not bespoke, case-specific solutions, but off-the-shelf products that are pre-engineered for a wide variety of use cases, so users can easily install and install them. deploy independently and in a very short time.


The user does not need to specify image capture parameters, such as the distance between the camera and the sample item, lighting, focus value, shutter speed and exposure time – all of this will be automatically calculated and dynamically adjusted by the AMV system using its artificial intelligence (AI) engines. The user only has to present the system with 20 to 30 good samples so that it can know the characteristics of the elements to be checked and signal any deviation from the stored standards.


The user-friendliness and immediacy of AMV systems are the result of Autonomous Machine Vision Artificial Intelligence (AMV-AI), a proprietary technology that combines three AI modules to mimic the end-to-end human cognitive vision process. The first artificial intelligence engine is responsible for dynamically adapting the electro-optical system to capture an optimal image of the part to be inspected, the second recognizes the item and the third inspects it for defects. At all stages, the AI ​​modules effectively mimic the ability of the human brain to acquire and process images, compare new images to previously stored ones, and spot differences.


Convinced of the user-friendliness and precision of the technology, BSH has implemented in its factory in Traunreut the INSPEKTO S70the only AMV system currently on the market.


The results


Using Inspekto’s technology, the BSH factory in Traunreut was able to implement the user-friendly and immediately deployable solution they were looking for, without having to commission a long, expensive and inflexible custom project.


Additionally, the INSPEKTO S70 can be trained quickly, adapt to environmental changes, and inspect multiple products and models simultaneously, all as an offline solution without any cloud deployment issues.


“The result was so impressive that, while the solution was initially only planned for three use cases, it is now successfully tested and validated for six additional applications in different production lines,” confirmed Dipjyoti.


Fundamentally, the partnership with Inspekto has allowed BSH to maintain its commitment to sustainability. The INSPEKTO S70 can be deployed at any point in the production chain, not just at the end, helping the company detect defects early on. This means that valuable resources are not wasted completing an already damaged product. It also means that the factory can identify areas where defects occur more frequently and take action to improve the production process.


With the systems in place at the start of the assembly line, the Traunreut plant has managed to reduce material waste by up to 90%, helping the company save money while minimizing its environmental footprint.





Did you enjoy this great article?

Check out our free e-newsletters to read other interesting articles.

Subscribe





About Shirley L. Kreger

Check Also

Oosto brings facial recognition to Australian Turf Club’s CCTV system

Israeli company Oosto has revealed details of the deployment of its facial recognition software to …