Product Design and Development Strategies for An Automated Printed Circuit Board (PCB) Defects Detection Systems

Printed Circuit Board (PCB) defects, such as shorts, spurious copper, and missing holes, can severely impact the reliability and performance of electronic devices. In the industry, ensuring the quality of PCBs is crucial, yet conventional manual inspection techniques are time-consuming and prone to errors. This work aims to design and develop a low-cost, automated PCB defects detection system utilizing modern machine learning methods. The primary goal of the work is to create an efficient system that identifies PCB defects with high accuracy. Several conceptual designs were generated based on customer requirements. To refine these designs, quality functional deployment (QFD) and product design and development methodologies were applied. Using a Pugh chart, the best design was selected and further refined through concept scoring based on the highest scoring. Concept design 5 was selected as the best conceptual design. This approach provides a practical and economical solution for PCB defect detection, merging advanced object detection algorithms with accessible, cost-effective hardware. The prototype was fabricated and integrated, demonstrating that the system is highly accurate and efficient in identifying PCB flaws.