The production of printed circuits adopts artificial intelligence

Artificial intelligence (AI) represents a significant opportunity for the printed circuit board industry. In fact, the adoption of artificial intelligence manages to optimize the revision processes of printed circuits and also the entire PCB production plant.

Today's printed circuit board (PCB) designs require the use of high density connections and components. Furthermore, manufacturing processes have evolved from manual assembly to highly automated manufacturing . As technology further develops, processes become more complex, including the ability to inspect defects that would lead to the scrapping of cards.

In the past, PCB manufacturing has typically relied on experienced technicians who have acquired their skills over many years. These technicians know and understand each step of the manufacturing process deeply and use their knowledge to optimize production and improve yield. Human limitations, such as errors and fatigue, hinder the performance of technicians and the misidentification of “false alarms” defects can have a negative effect on production yield. Thanks to the integration of artificial intelligence into the manufacturing process, machines can add value by taking on certain tasks; while human experts continue to carry out the more complex tasks that require reflection and interaction, optimizing and "training" the intelligent system. This combination of human and artificial intelligence improves overall efficiency and represents a great opportunity for PCB manufacturing facilities.

How is artificial intelligence changing printed circuit board manufacturing?

Current AI implementations in PCB manufacturing demonstrate a positive impact on productivity and yield. For example, in automated optical inspection ( AOI) processes, machine learning dramatically reduces human error in defect detection.

A classic inspection of 100 panels without the use of AI typically identifies 20 to 30 defects per panel, approximately 75% of which may be false alarms . Manufacturing rules dictate that all defects must be manually reviewed, so reviewing false alarms is a waste of time.

Human inspection increases the handling of printed circuit boards, which in turn can cause new damage. It also opens up the possibility of further false analyzes during the review of the operator, who can be tired, overworked or distracted.

Thanks to the machine learning built into the AOI system, false alarms are drastically reduced. Fewer false alarms means less manipulation of the panels, which in itself increases yield. Artificial intelligence in AOI systems has been found to reduce false alarms by up to 90%.

Will AI solve the challenges in circuit manufacturing?

The use of AI is growing rapidly across the industry, while the challenges in circuit manufacturing are increasing just as rapidly, if not more. The two main areas where defect detection is becoming increasingly difficult are: reduction of geometries and the use of flexible materials.

For example, the use of flexible materials causes more defects to be identified, which leads to an increase in false alarms. The goal of production is to minimize human intervention, i.e. reduce the handling of flexible panels in the defect determination process. From this point of view, flexible circuits will enjoy a great benefit, as intelligent systems will learn to recognize defects using more rigorous parameters.

AI-driven PCB factories will become the key to future manufacturing. The development of this AI application globally still takes some time to be adopted in circuit manufacturing. But it's clear that system-wide AI implementation is already here, laying the foundation for a future with fully intelligent and automated PCB factories.

The article Printed circuit board manufacturing adopts artificial intelligence comes from Tech CuE | Close-up Engineering .