How to detect defective parts and achieve unprecedented components traceability with predictive analytics

In SMT assembly a single defective part or compromised component can jeopardize an entire production run. Experts estimate that as many as 10% of components in products on the market today are compromised. That means that they are either entirely counterfeit parts, contain mixed/non-homogenized reels, are expired, have been stored in improper conditions that impact quality, or have even been intentionally tampered with and embedded with malicious code.

Costly, time-consuming lab tests don’t offer a component inspection solution because they only look at samples and can’t identify individual compromised components in a mixed source.

A solution that excludes low-quality or counterfeit parts upfront and pre-empts problems, reducing RMAs.

Siemens’ partner Cybord created the first traceability solution that enables visual inspection of every electronic component. In the Cybord-Siemens solution, the existing images captured in the Pick & Place machines are compared to a massive component library and knowledge database in the cloud. Using AI models, the solution can verify component authenticity and identify damage and tampering as well as counterfeit components across 100% of the components being used. The solution uses existing data, so there is no need to introduce new operations and the entire process can be conducted while the SMT placement machine is placing the components.

Download our new whitepaper to learn how it works.