IBM Smart Edge for Welding on AWS

In metal-to-metal welding applications, companies are struggling to automate inspections to detect weld defects early and often to avoid costly product recalls, excessive scrap, re-work and other poor quality associated costs.

IBM Smart Edge for Welding (SE4W) aggregates AI-based quality inspection capabilities, including acoustic and visual insights, into a unique solution that can solve some of the most challenging pain points in any welding application, realizing a step change in quality performance. The unique solution enables inspection of 100% of the weld sample size during the welding process. Optimize and automate welding inspections with Edge AI to provide near real-time, actionable weld quality insights for weld operators for faster defect diagnosis and resolution, reducing costs and improving efficiency.

The solution provides a highly performant, secure and reliable platform for AI-powered industrial quality inspections. Weld quality data captured through specialized cameras and microphones on the shopfloor is then sent to the cloud, where multiple AWS services, combined with IBM technology, enrich and analyze the data, and extract insights to optimize productivity, quality, and sustainability.

IBM Smart Edge for Welding is like having the best operator on the line 100% of the time, freeing up their time while increasing product quality. Our solution can identify defects through IR/thermal imaging and acoustic analytics that even the best operator cannot detect. This offering can be customized to include any combination of visual inspection, acoustic insights and/or parametric analysis.

Intelligent Workflows and role-based insights: Real-time insights that can be directly and immediately consumed by two key personas: • Weld Technician – who works on the shopfloor and needs to get insights into weld performance in real-time so they can add, change or optimize the process as needed. Dashboards built with ease of use in mind, the solution is fully capable of integrating with your preferred interaction mechanism on the shopfloor, such as HMI or mobile devices. • Process Engineer – who wants to understand patterns and behavior across shifts, weeks, months, weld programs, materials so that they can provide guidance to improve the overall manufacturing process.