In a recent episode of The EMS (Eric Miscoll Show), Cybord CEO Oshri Cohen participated alongside Philip Stoten and Joel Scutchfield, General Manager of SMT and Semiconductor at Koh Young America, to address a central question for the electronics manufacturing industry: is AI-powered inspection now leading the broader automation and intelligence revolution on the shop floor?
Cybord's approach centres on visual AI applied at the component level, inspecting every individual component prior to placement on the PCB. The company positions this as a fundamental shift away from traditional inspection paradigms, which typically identify defects downstream — after soldering or during final board-level AOI and AXI — when rework costs and scrap rates are already elevated.
The discussion highlighted several themes directly relevant to EMS providers and OEM manufacturers. First, the move from sampling-based quality control to 100-percent component inspection generates a substantially larger and more reliable data set. In high-mix, high-reliability production environments — such as automotive electronics, industrial controls, or aerospace-adjacent assemblies — this volume of verified component data supports more robust process analytics and root-cause analysis.
Second, the panellists addressed traceability and compliance requirements. Regulatory frameworks and customer-driven supply chain transparency demands are intensifying across European and global markets. AI-enabled inspection that logs visual data for every component placed offers a documented audit trail, which is increasingly expected in sectors subject to IPC, IATF 16949, or emerging EU supply chain due-diligence requirements.
Third, the conversation distinguished between AI systems that generate alerts and those capable of triggering immediate, closed-loop action — stopping the placement process before a defective or counterfeit component reaches the board. This distinction matters operationally: alert-based systems still rely on human intervention, whereas action-based systems can halt or redirect the feeder or nozzle path in real time, reducing the window for escapes.
Koh Young's involvement in the discussion is notable given the company's established position in the SPI and 3D AOI market. The intersection of Koh Young's measurement expertise and Cybord's pre-placement visual AI reflects a broader industry trend toward layered, complementary inspection strategies rather than reliance on a single inspection stage.
For European EMS companies evaluating their inspection infrastructure, the conversation underscores a technology direction that the industry has been moving toward incrementally: embedding intelligence earlier in the assembly process rather than concentrating quality gates at the end of the line. Whether AI-driven pre-placement inspection becomes standard practice will depend on integration maturity with existing pick-and-place platforms, total cost of implementation, and demonstrated yield improvement data from production deployments.
Quelle: Cybord – https://cybord.ai/news_events/news/the-ems-eric-miscoll-show-is-inspection-leading-the-ai-revolution-in-manufacturing/
