Mitsubishi Heavy Industries, Ltd. (MHI) and Algomatic Co., Ltd. have secured second prize at the GENIAC-PRIZE competition for their innovative project aimed at formalizing tacit knowledge in manufacturing through AI analysis.
Their joint proposal focuses on the use of AI to analyze and compare videos of expert and novice workers performing Tungsten Inert Gas (TIG) welding. This method is crucial for various MHI products, including energy plants and rockets, but presents challenges in skill transfer due to its complexity.
The project involves capturing video footage of welding performed by both experienced and less experienced welders. An AI agent then analyzes these videos to identify and illustrate the differences in skill levels, focusing on embodied knowledge. By selecting the optimal analysis modules, the project aims to convert tacit skills into explicit knowledge, facilitating technical evaluation and feedback for novice welders.
This initiative addresses a longstanding issue in the manufacturing industry, where the communication and standardization of tacit knowledge have been problematic. By elucidating the differences in skill levels, the project is expected to enhance the efficiency of skill transfer and boost productivity within the sector.
MHI's strategy, under its Innovative Total Optimization (ITO) corporate framework, emphasizes reducing lead times and improving productivity through overall optimization. The company plans to further develop this technology to contribute to skill transfer and productivity improvements in manufacturing.
The GENIAC-PRIZE, organized by Japan's New Energy and Industrial Technology Development Organization (NEDO), aims to accelerate the application of generative AI across various themes, including manufacturing, customer support, and risk mitigation. The competition awarded approximately 800 million yen across 42 projects from over 200 entries.
Comments are closed.