LG Innotek unveils an AI-based raw material inspection system.

Seoul, South Korea—LG Innotek (011070. KS) announced the development and implementation of an artificial intelligence (AI)- based inspection system for incoming raw materials to prevent defects in semiconductor production.

The company claims the system is an industry-first. 

It is applied to Radio Frequency System-in-Package (RF-SiP) and Flip Chip-Ball Grid Array (FC-BGA) processes.

Key features of the system include:

• Analysis of raw material defects in one minute with over 90% accuracy
• Visualization of quality deviations in each lot of raw materials
• Reduction of defect analysis time by up to 90%

LG Innotek developed the system by combining material information and AI image processing technologies. 

The AI has been trained using data from tens of thousands of suitable and unsuitable material compositions, including Prepreg, Ajinomoto Build-up Film, and Copper-Clad Laminate.

The company stated that previous visual inspection methods had become insufficient due to advancements in semiconductor substrate technology, as microscopic defects began causing issues.

S. David Roh, LG Innotek's Chief Technology Officer, said, "The AI-based inspection system will complete our AI ecosystem, identifying product defect causes early in the production process."

LG Innotek plans to enhance the AI system's capabilities through data-sharing partnerships with customers and suppliers in the substrate sector. 

The company also aims to expand the system's application to optical solutions such as camera modules.