Awarded 2nd Place in NEDO's Prize-Type Program 'GENIAC-PRIZE'
Mitsubishi Heavy Industries and Algomatic won 2nd place in NEDO's 'GENIAC-PRIZE' for their innovative AI approach to formalize tacit knowledge in TIG welding via video analysis.
📋 Article Processing Timeline
- 📰 Published: April 8, 2026 at 20:10
- 🔍 Collected: April 8, 2026 at 11:30
- 🤖 AI Analyzed: April 21, 2026 at 06:13 (306h 43m after Collected)
◆ Proposed a new approach to automatically analyze work differences between skilled and unskilled workers using TIG welding technology as a subject.
◆ Realized the formalization of tacit knowledge at manufacturing sites, contributing to more efficient skill inheritance and improved productivity.
Mitsubishi Heavy Industries, Ltd., in a joint application with Algomatic Inc. (Headquarters: Minato-ku, Tokyo; Representative Director and CEO: Shunsuke Ono), was awarded 2nd place in the New Energy and Industrial Technology Development Organization (NEDO) prize-type program 'GENIAC-PRIZE' (Note 1) for their proposal on the theme of 'Formalization of Tacit Knowledge in the Manufacturing Industry.'
The joint proposal by our company and Algomatic was 'Formalization of tacit knowledge through a "comparison" approach of video footage of skilled and unskilled workers, using TIG welding (Note 2) technology as an example.' TIG welding, as a high-quality welding method, is a technology that broadly supports our products ranging from energy plants to rockets. However, it presents a difficulty where quality and working time vary depending on the welder's skill level, making it a challenge for skill inheritance. In this proposal, simply by filming and uploading videos of welding operations by skilled and unskilled workers, an agent AI automatically analyzes the work differences between the two. It selects the optimal method from multiple analysis modules to multifaceted extract and visualize skill differences, including somatic knowledge. As a result, skills that are difficult to verbalize are systematically accumulated as formal knowledge, enabling technical evaluation and feedback to unskilled workers by AI.
At manufacturing sites, while the skills of experts are accumulated as tacit knowledge, verbalizing and standardizing them has been a long-standing challenge. By 'comparing' the work of skilled and unskilled workers, this proposal challenges the visualization of somatic knowledge and is expected to contribute to streamlining skill inheritance and improving productivity at manufacturing sites, marking an important step toward future social implementation.
Under the management policy 'Innovative Total Optimization (ITO),' Mitsubishi Heavy Industries is working to halve lead times and improve operational productivity with a 'total optimization' approach, while materializing growth strategies for each business from the perspective of 'domain expansion.' We will continue to promote the social implementation of this technology, contributing to skill inheritance and productivity improvement at manufacturing sites.
(Note 1) GENIAC-PRIZE is a NEDO prize-type program aimed at promoting the social implementation of generative AI. It offers a total of approximately 800 million yen in prizes for four themes: 'Formalization of tacit knowledge in the manufacturing industry,' 'Improving productivity in customer support,' 'Development of generative AI contributing to the streamlining of screening operations at government agencies,' and 'Development of risk exploration and mitigation technologies to ensure the safety of generative AI.' At the final screening and award ceremony on March 24, 2026, 42 out of over 200 applications were awarded. For details, please see below.
https://geniac-prize.nedo.go.jp/
(Note 2) TIG welding is a method of welding metals using a tungsten electrode and an inert gas such as argon.
■ Mitsubishi Heavy Industries, Ltd.
Website: https://www.mhi.com/jp/
Online Magazine 'SPECTRA' (Japanese): http...
◆ Realized the formalization of tacit knowledge at manufacturing sites, contributing to more efficient skill inheritance and improved productivity.
Mitsubishi Heavy Industries, Ltd., in a joint application with Algomatic Inc. (Headquarters: Minato-ku, Tokyo; Representative Director and CEO: Shunsuke Ono), was awarded 2nd place in the New Energy and Industrial Technology Development Organization (NEDO) prize-type program 'GENIAC-PRIZE' (Note 1) for their proposal on the theme of 'Formalization of Tacit Knowledge in the Manufacturing Industry.'
The joint proposal by our company and Algomatic was 'Formalization of tacit knowledge through a "comparison" approach of video footage of skilled and unskilled workers, using TIG welding (Note 2) technology as an example.' TIG welding, as a high-quality welding method, is a technology that broadly supports our products ranging from energy plants to rockets. However, it presents a difficulty where quality and working time vary depending on the welder's skill level, making it a challenge for skill inheritance. In this proposal, simply by filming and uploading videos of welding operations by skilled and unskilled workers, an agent AI automatically analyzes the work differences between the two. It selects the optimal method from multiple analysis modules to multifaceted extract and visualize skill differences, including somatic knowledge. As a result, skills that are difficult to verbalize are systematically accumulated as formal knowledge, enabling technical evaluation and feedback to unskilled workers by AI.
At manufacturing sites, while the skills of experts are accumulated as tacit knowledge, verbalizing and standardizing them has been a long-standing challenge. By 'comparing' the work of skilled and unskilled workers, this proposal challenges the visualization of somatic knowledge and is expected to contribute to streamlining skill inheritance and improving productivity at manufacturing sites, marking an important step toward future social implementation.
Under the management policy 'Innovative Total Optimization (ITO),' Mitsubishi Heavy Industries is working to halve lead times and improve operational productivity with a 'total optimization' approach, while materializing growth strategies for each business from the perspective of 'domain expansion.' We will continue to promote the social implementation of this technology, contributing to skill inheritance and productivity improvement at manufacturing sites.
(Note 1) GENIAC-PRIZE is a NEDO prize-type program aimed at promoting the social implementation of generative AI. It offers a total of approximately 800 million yen in prizes for four themes: 'Formalization of tacit knowledge in the manufacturing industry,' 'Improving productivity in customer support,' 'Development of generative AI contributing to the streamlining of screening operations at government agencies,' and 'Development of risk exploration and mitigation technologies to ensure the safety of generative AI.' At the final screening and award ceremony on March 24, 2026, 42 out of over 200 applications were awarded. For details, please see below.
https://geniac-prize.nedo.go.jp/
(Note 2) TIG welding is a method of welding metals using a tungsten electrode and an inert gas such as argon.
■ Mitsubishi Heavy Industries, Ltd.
Website: https://www.mhi.com/jp/
Online Magazine 'SPECTRA' (Japanese): http...