LG Innotek said on Saturday it has built an AI system that lets users compare parts specifications and prices at a glance and quickly find the best parts to apply to new products. Products such as camera modules, semiconductor substrates and automotive components contain an average of hundreds of parts, and it has taken considerable time to check information scattered across multiple systems and identify new parts.

To address the problem, LG Innotek introduced AI into its parts search process. It first integrated data on about 2 million internal and external parts such as capacitors and inductors. After standardising parts names, specifications and units that differ by manufacturer to create an AI-ready database, it developed the "AI parts recommendation system". The company aims to improve the completeness of product proposals and its order-winning competitiveness by selecting optimal parts based on both price and performance while reducing search time.

Users enter the required parts specifications, and the AI searches the entire dataset and displays parts with similar conditions. It can review external market data as well as parts already purchased, significantly widening the search scope that had been constrained by staff experience and purchase history.

The company said a strength of the system is its "reference price" function, which presents a current price level by analysing purchase history and price fluctuation trends for parts with similar specifications. The reliability of the reference price was more than 96 percent, and the system can also predict prices for new parts with no purchase history.

Unlike consumer goods with fixed prices, industrial parts vary in price depending on purchase volume, contract terms and timing. The previous approach, which required checking quotes with many suppliers and investigating market prices, took time accordingly.

Since the system was introduced, LG Innotek's time to calculate estimates for new products has fallen by more than 70 percent from before. The company said the time saved can be used to review customer requirements, improving proposal completeness and order-winning competitiveness. It also cited the advantage of being able to quickly secure alternative parts even if a certain part is discontinued or supply disruptions occur, allowing flexible responses to supply and demand changes.

For now, the AI searches for parts using pre-collected and standardised data, but the company plans further development toward applying agentic AI so it can independently search for and analyse the latest parts information.

Junseong Kim (김준성), head of LG Innotek's Purchasing Center and a managing director, said the "AI parts recommendation system" is an innovation that meaningfully changes existing methods as a system optimised for manufacturing companies that handle a wide range of parts. "We will continue to provide value that exceeds customer expectations through AX-based ways of working," he said.