[Edaily Reporter SOYEON KIM ] LG Innotek announced on the 20th that it has developed an artificial intelligence (AI) system capable of identifying the optimal components. The AI component recommendation system analyzes vast amounts of internal and external component data using AI, allowing users to compare specifications and prices for each component at a glance. The company plans to use this system to quickly identify the optimal components—taking both price and performance into account—thereby improving the quality of product proposals and enhancing its competitiveness in winning orders.

LG Innotek’s camera modules, semiconductor substrates, and automotive components typically contain an average of several hundred parts. Previously, the process of calculating quotes for orders and the development process took a significant amount of time. Developed over approximately two years, this AI system analyzes data scattered across the company and reduces component search time by more than 70%. LG Innotek is accelerating its company-wide AI transformation (AX) by fully implementing this system across all business divisions.

The company has integrated data on approximately 2 million internal and external components, including capacitors and inductors. By standardizing component names, specifications, and units—which vary by manufacturer—the company has built a vast database that AI can utilize. Through this system, when a user enters the required component specifications, the AI searches the entire database for components with similar characteristics. It broadens the scope of component search by suggesting not only previously purchased parts but also data from external markets.

LG Innotek explained that the introduction of this AI-powered component recommendation system has reduced the time required to generate quotes for new products by more than 70% compared to previous methods. Even if a specific component is discontinued or supply disruptions occur, the company can quickly secure replacement components, enabling a flexible response to changes in component supply and demand.

Furthermore, a major advantage is the ability to calculate AI-based “reference prices” during the procurement process. The AI analyzes the actual purchase history and price fluctuation trends of components with similar specifications to provide current price levels. The reliability of these reference prices has been shown to be over 96%. This enables the company to shortlist price-competitive candidates within two hours.

An LG Innotek official stated, “Previously, to find price-competitive parts, we had to check quotes from numerous suppliers one by one and survey market prices,” adding, “The implementation of the ‘Reference Price’ calculation feature for each part is the biggest differentiator from existing systems.”

Going forward, LG Innotek plans to further develop the system by applying Agentic AI to enable it to independently search for, analyze, and incorporate the latest component information.

LG Innotek has accelerated its adoption of AI-driven transformation (AX) across all business operations. By introducing “AI raw material incoming inspection” into key production processes, the company has reduced the time required to analyze the causes of material defects by up to 90 percent. It has also implemented “AI vision inspection,” in which AI inspects the appearance of finished products to detect defects. This has led to a dramatic increase in the yield rates of camera modules and semiconductor substrates.

Furthermore, by applying “AI Process Recipes,” the company achieved significant results, such as reducing the time required to identify optimal process conditions for camera modules from 72 hours to within 6 hours. Recently, by implementing “EXAONE Tabular”—an industry-specific AI model developed by LG Corp. AI Research—on the manufacturing floor, the company also reduced the time required for AI to learn changed process conditions by approximately 85%.

Kim Jun-sung, Head of the Purchasing Center at LG Innotek (Senior Managing Director), stated, “The ‘AI Component Proposal System’ is optimized for manufacturing companies that handle a wide variety of components,” adding, “Through innovations in our AX-based work methods, we will deliver value that exceeds customer expectations.”