The current component shortages are rocking the electronic manufacturing industry as well as the industries that depend on them, which, let’s face it, is most industries. Who’d have thought that a handful of electronic parts could halt production of millions of dollars or automobiles around the world.
Well, that’s exactly what happened, and there have been a few knock-on effects that are worrisome to any companies who don’t know exactly what is going on with every part placed on every printed circuit board.
Apart from the obvious impact on production and the fear rippling through the industry, the current component shortages have encouraged companies to buy parts on the spot market, or worse still, on the grey market. While many might have seen the transparency and traceability in their usual supply chain as enough, this is definitely not the case with components that have come from less clear sources.
When you start to unpack the root cause of hardware failures, particularly in the field, it turns out that around 85 percent are due to issues related to the components and only a handful of issues relate to workmanship or assembly issues. It stands to reason that in a time when supply chains are under pressure and procurement teams are spreading their search wider, often beyond their approved vendor list (AVL), that this number will be even higher.
Between 5 percent and 20 percent of components in the electronics supply chain are probably counterfeit, according to Oneida Research.
So, if quality in electronics is all about the components, what can be done to mitigate the risk? The answer is simple really. It’s all about knowing more about each component before it is soldered to the PCB. That means better screening as parts arrive on site and better analysis when parts are placed on the board. What that means is a detailed traceability for the lifetime of the product. What’s more, a system needs to be in place that works at the line speed and inspects every device, not the kind of batch or sample inspection that might be done by an outsourced lab or internally using offline inspection equipment.
This is why systems such as Cybord's are becoming increasingly popular with both the EMS industry and the brands, or OEMs, that the industry serves. It’s hard to say who owns the quality of the assembled product and for that reason electronic component analytics data is proving valuable for both parties.
Cybord employs computer vision, AI and big data to inspect every electronic component when it is placed with the goal of improving product reliability and assembly efficiency. As well as looking for parts that are counterfeit or forged, these algorithms also find parts that are defective, damaged or have been tampered with. The data can provide something as a simple mount or discard instruction, or a full suite of analytics that includes part and manufacturer identification, homogeneity, age verification, defect or anomaly detected and much more. The inspection engine uses a knowledge database built up over millions of inspections to constantly learn what to look for, all at line speed.
Some EMS companies are giving their customers access to a detailed "board inspection certificate" that delivers surgical detail based on component visual characteristics. In a world where traceability is appropriately fanatical, this level of data is extraordinarily valuable.
This is creating "data haves" and "data have-nots." Those without the data are blind to the risks they are facing and those with batch or sample data have some visibility but not nearly enough.
Armed with the right data, EMS can reduce the cost of rework and failures and increase first-pass yield. OEMs can lower the cost of return merchandise authorization (RMA) by reducing acceptance test failure;, improving their sourcing efficiency and reliability; and massively reducing the cost of, and potential for, product recalls. Ultimately, the entire supply chain benefits from better component traceability.