Smarter false call reduction for electronics manufacturing

Factory operators in printed-circuit board manufacturing and assembly companies are under pressure to build high-quality products quickly. False calls on automated optical inspection (AOI) machines can cause delays in the production line and impact productivity. What if teams could reduce false calls and improve accuracy without slowing the PCB manufacturing process?

Read this whitepaper to learn about a lightweight algorithmic solution our industrial AI experts developed using a cloud-based service that maintains inspection models. This "AI as a service" quickly and consistently reduces the number of false calls.

The impact of false calls on AOI machines in PCB manufacturing

False calls on AOI machines in PCB manufacturing cause delays and reduce productivity. Various factors, such as dust, scratches, or lighting conditions, can cause these errors. Traditional methods of reducing false calls involve manual adjustments to the inspection process, which can be time-consuming and costly. Reduce the impact of these errors on your production line with a smarter solution.

How AI and machine learning can reduce false calls in PCB manufacturing

Artificial intelligence (AI) and machine learning (ML) are maturing and smart enough to help solve the false calls dilemma. An AI model can help identify errors or defects in PCB manufacturing, support the correct execution of work instructions and document the work result. AI can also be trained to assess product or process quality or to improve automatic inspection procedures, helping improve important KPIs such as first pass yield (FPY), availability and overall equipment effectiveness (OEE).

Implementing a smarter false call reduction solution in electronics manufacturing

Our AI- and ML-based solution uses data provided by AOI machines installed in SMT lines along with manual inspection findings recorded whenever a PCB is found to be faulty. The system consists of two parts: the local installation on the shop floor and a cloud-based service for model generation and updates. This combination of AI stored locally at the factory and the cloud-based service makes reducing false calls guaranteed and sustainable.