Kuken Industries, a maker of cooling towers and ventilation equipment, has deployed an AI-based search tool that reduces the time engineers spend locating archived drawings. The company reports the average time to find a drawing fell to about 4.5 minutes from nearly an hour.
How the search works
The tool was developed with Recursive, a Japanese AI startup and an official OpenAI partner, and uses the OpenAI API. Kuken tested several models on a range of drawings to evaluate extraction accuracy and cost; according to the company, OpenAI’s models performed well on both measures and allowed the team to integrate search directly into a field app.
Field engineers can enter a drawing or part number, paste a screenshot, or upload an image or PDF of a drawing or parts list. The app reads part numbers and returns possible matches, including drawings for similar designs. Engineers then verify matches against tower specifications and records of past replacements.
Reported results and operational impact
Kuken says the archive spans 70 years of installations. When the records were on paper, a search could take up to half a day; digitization shortened that time but still required engineers to know which names or numbers to search. With the new app, Kuken projects the tool will cut hours spent searching for drawings by about 90% over a year.
Reported operational effects include reduced printing costs because drawings can be reviewed in the app, experienced engineers spending saved time mentoring newer colleagues, and newer engineers taking on a broader range of tasks independently. “People now have time to do work they’ve wanted to try and take on new challenges. That’s been a real source of motivation,” said Takatsugu Naraki, Director and Factory Manager, Kuken Industries.
Kuken is also developing an AI tool to check drawings submitted by partners. In prototype testing, about 90% of the issues the AI flagged were genuine discrepancies. If moved into production, the company expects the tool to substantially reduce the time engineers spend checking drawings. “I believe OpenAI’s technology can help manufacturers like us pass skills on to the next generation. That’s long been a challenge for us,” Naraki said.
Original source: OpenAI News