AI Enhancement to AOI for PCBA Project
SOW & Project Statement
- Statement of Work (v 1.3; July 27, 2021)
- Project Statement (v 1.3; July 27, 2021)
Automated optical Inspection (AOI) is widely used in the electronics manufacturing industry as a quality inspection process. However, the false call and defect escape can significantly impact throughput and labor involvement for defects verification. It becomes even more challenging with the increasing complexity of electronics designs (e.g., higher density, highly integrated, extreme small or large size, irregular shape components, various materials, and surface features). In recent years, machine learning and artificial intelligence (AI) models are being adopted to help with the pass/fail judgement of AOI machines. However, the accuracy and efficiency enhancement that AI may bring to the AOI is yet to be understood.
Purpose of ProjectThis project aims to establish common performance metrics to evaluate the AI value-add to AOI capabilities and provide recommendations to the industry on future improvement opportunities. It will include two phases:
- Phase 1: Investigation & planning
- Phase 2: Experimentation to evaluate AOI+AI capability
The ultimate goal is to increase the inspection efficiency and accuracy, at the same time, reducing the human resource required for manual inspection verification and mitigating the defect escape risk.
iNEMI Survey Report: AI Enhancement to AOI for PCB Assembly
Presentation: Call-for-Participation Webinar (October 14, 2021)
(See webinar recording below.)