Cart
0
Brent A. Fischthal, Head of Global Marketing, Koh Young Technology
The rapid evolution of electronics manufacturing is increasingly driven by advancements in artificial intelligence (AI), machine-to-machine (M2M) communication, and smart factory concepts. Central to this transformation are AI-powered inspection technologies, which not only address critical industry challenges but also enable the realization of Industry 4.0’s vision. This article integrates insights from recent developments and collaborative efforts to highlight how AI is revolutionizing inspection processes, thereby enhancing efficiency, yield, and quality in electronics manufacturing.
The Four Roles of Inspection Data in a Smart Factory
In any smart factory, the deployment of real-time, automated inspection data is crucial. Connectivity forms the backbone of a smart factory, but without actionable data, this connectivity is futile. According to a case study with Koh Young’s customer Matric Group, extensive inspection is fundamental. Matric Group emphasizes that high-quality inspection correlates directly with better efficiency, higher first pass yields, and fewer defects. Therefore, comprehensive inspection processes are essential.
A fundamental value chain in smart factories starts with comprehensive connectivity of all machines and processes, generating reliable data that can be transformed into actionable intelligence. This process is essential for real-time decision-making and performance improvement. Exceptional data, especially from inspection processes, is pivotal. It begins with the printed circuit board (PCB) entering the solder paste printer and continues through various stages including SPI (solder paste inspection), component placement, and AOI (automatic optical inspection), each contributing critical data points.
Inspection data plays four critical roles in a smart factory:
1. Immediate Process Adjustments: For example, SPI data can adjust print processes in real-time, improving performance by providing instant feedback to the printer.
2. Quality Gatekeeping: Ensuring faulty PCBs do not proceed down the production line, thus preventing defects from escalating and reducing waste and rework.
3. Root Cause Analysis: Identifying and isolating manufacturing issues to enhance performance. Holistic data from inspections can pinpoint the origins of faults, thereby improving first pass yield and overall efficiency.
4. Granular Traceability: Essential for managing recalls and maintaining quality assurance. Detailed inspection data allows manufacturers to track issues back to their source, minimizing the impact of potential recalls.
The ultimate goal is a holistic, connected, and complete data ecosystem where every process, inspection, and decision is data-driven, achieving the smart factory’s full potential. Connected machines need to communicate reliable data that can be converted to intelligence that can be used to make decisions that improve performance, all in real time.
AI-Powered Inspection Technologies
AI’s integration into inspection technologies is transforming manufacturing processes by addressing skilled labor shortages and the need for increased productivity. AI algorithms enable the extraction of actionable insights from large-scale datasets generated by advanced 3D metrology systems, significantly enhancing inspection accuracy and reliability.
Challenges and Opportunities in the SMT Industry
The SMT industry faces unique challenges, including a skilled labor shortage and the need for increased efficiency and flexibility in production processes. Organizations like IPC and SMTA are addressing the challenges with education and training initiatives. However, the integration of AI technologies offers a powerful complement to these efforts, automating and optimizing manufacturing processes.
Machine-to-Machine (M2M) communication, driven by the principles of Industry 4.0, is playing a pivotal role in revolutionizing manufacturing processes. AI harnesses the power of large-scale datasets generated by advanced 3D metrology systems. By employing AI algorithms, companies can derive actionable insights from data, paving the way for smarter manufacturing processes.
The Value of Accurate 3D Data
Central to the success of AI solutions in manufacturing inspection is input data quality. Modern inspection systems equipped with 3D measurement capabilities provide more reliable data compared to conventional 2D Automated Optical Inspection (AOI) systems, which struggle with defects on curved or reflective surfaces. Accurate 3D data, combined with traditional 2D images, enables the development of advanced AI features powered by robust datasets. This approach significantly enhances inspection accuracy and reliability, addressing challenges associated with small data, strict inspection criteria, and imbalanced datasets.
Innovative Solutions from Koh Young Technology
Koh Young has developed several innovative solutions that highlight the power of AI in manufacturing inspection. One such solution is the Koh Young Auto Programming (KAP) system, which utilizes deep learning methods based on true 3D data to autonomously propose inspection conditions. KAP significantly reduces job preparation time by up to 70%, making it ideal for high-mix, low-volume, or time-sensitive applications.
Additionally, Koh Young’s Process Optimizer (KPO) offers AI-driven solutions for optimizing printing and mounting operations. The KPO Printer solution includes modules such as Printer Diagnosis, Printer Advisor, and Printer Optimizer, which use complex algorithms to develop closed-loop print process recommendations. By combining real-time printing and SPI measurement data, KPO Printer actively improves the printing process, reducing false calls and enhancing process reliability.
Similarly, the KPO Mounter solution includes modules such as Mounter Diagnosis and Mounter Optimizer (Feedback and Feedforward), which identify internal issues and provide optimal offset values to minimize defects. By applying accurate measurement-based inspection data from Koh Young AOI machines, KPO Mounter ensures optimal performance and minimizes production errors.
Furthermore, the KSMART solution from Koh Young integrates various AI technologies while managing all incoming and outgoing data to provide advanced support across various manufacturing processes. From production preparation to optimization and maintenance, KSMART enhances system performance and reduces reliance on manual intervention. By using AI algorithms, KSMART enables manufacturers to achieve greater efficiency and flexibility in their operations.
Case Study: Matric Group
Matric Group, a leading electronics manufacturer, implemented Koh Young’s AI-driven solutions to enhance its operations by leveraging the right inspection solutions. This led to significant improvements in efficiency, yield, and quality. Matric invested in a comprehensive suite of inspection equipment, including SPI, pre-reflow AOI, and post-reflow AOI.
By installing SPI and both pre- and post-reflow AOI on their SMT lines, Matric significantly enhanced its capabilities, pushing yield past 98%. The pre-reflow AOI identifies issues such as skewed or misplaced components, preventing errors before soldering, resulting in significant cost savings. Patrick Stimpert, VP of Operations, highlights the immediate yield improvement, moving from 80-85% to over 98% in final yields and first pass yields.
Root cause data from inspection machines helps in the continuous improvement of processes with full traceability, driving quality assurance at every stage of the production line. Matric utilizes KSMART process control software for real-time monitoring, optimization, and analysis. KSMART provides a single dashboard accessible from anywhere, allowing a single operator to manage multiple inspection stations and gain insights into the performance of each line. The data collected from inspections allows Matric to determine root causes, driving constant and continual improvement in programming and setting up jobs. This data-driven approach increases overall efficiency in production, reducing the constant percentage of jobs going to rework, thereby minimizing the opportunity for defects.
Operators appreciate the ease of use and efficiency of the Koh Young machines, allowing easy program retrieval and switching between programs. KSMART software enables the creation of color-coded dashboards that show real-time performance, enabling quick identification of defects or issues. Matric and Koh Young are working together to build a system that aligns with Matric’s vision of “quicker, faster, cheaper, and better, all at the same time!” The investment in Koh Young equipment has significantly improved quality and performance. Through these methods, Matric ensures quality, efficiency, and reliability in their manufacturing processes. By combining data-driven decision-making with advanced inspection technology, they continue to move towards their vision of delivering high-quality products quickly and efficiently.
Collaborative Efforts and Future Prospects
Collaboration plays a vital role in advancing AI-powered smart manufacturing processes. Since 2016, Koh Young Technology has partnered with the Smart Electronics Manufacturing Laboratory (SEMLab) at Binghamton University to develop AI-based frameworks for improving yield and throughput in PCB assembly. This collaboration focuses on implementing AI-based closed-loop feedback control and parameter optimization, driving significant improvements in process control and efficiency.
Binghamton University’s efforts, led by researchers like Dr. Seungbae Park and Dr. Daehan Won, have contributed to several key research initiatives aimed at developing smart electronics manufacturing solutions. Their work encompasses advanced robotics and AI integration to revolutionize electronics manufacturing with improved yield and productivity. The team has developed closed-loop control and optimization modules using self-optimization and AI-based diagnostics for process enhancement in PCB assembly. This research is advancing the field with innovative artificial intelligence and machine learning techniques.
Conclusion
The integration of AI and smart inspection technologies is revolutionizing electronics manufacturing. By leveraging comprehensive inspection data and advanced AI algorithms, manufacturers can achieve real-time process optimization, enhanced quality, and increased efficiency. As the industry continues to evolve, these technologies will play a crucial role in realizing the full potential of Industry 4.0, driving transformative change, and setting new standards in manufacturing excellence.
