Kuken Industries searches 70 years of drawings in minutes

Kuken Industries, a major Japanese manufacturer of air conditioning and ventilation equipment, holds a large share of Japan’s cooling tower market. The company faced a challenge: engineers repairing or replacing parts on older cooling towers needed to locate original drawings from an archive spanning over 70 years, a task that could take half a day even after digitization. Newer engineers often relied on experienced colleagues to find and interpret these records. To empower engineers at all levels, Kuken partnered with Japanese AI startup Recursive and used OpenAI’s models to build a search system and field app. The app allows engineers to search by part number, drawing number, or even by uploading an image or PDF of a drawing or parts list. OpenAI’s models performed well in accuracy and cost, and the API enabled flexible integration. Kuken reports that the tool is expected to cut drawing search time by about 90% over a year. This gives engineers more time to study drawings, consider design improvements, and take on new challenges. Newer engineers handle more jobs independently, while experienced engineers mentor and focus on engineering work. Kuken is also developing an AI tool to check partner drawings, with prototype testing showing about 90% of flagged issues were genuine discrepancies. The company aims to use AI to pass engineering expertise to the next generation.

(Source：OpenAI)