Tuesday, November 17, 2026

DataGrid and Sumitomo Electric Industries] Developed joint technology for AI to determine defects at manufacturing sites.

DataGrid and Sumitomo Electric Industries have started joint development of AI technology for defect judgment at manufacturing sites. In order to realize an AI with advanced judgment criteria comparable to those of skilled workers even in situations where very little actual defect data is available, the two companies will combine Sumitomo Electric Industries' defect judgment technology with DataGrid's pseudo defect generation technology to jointly develop a unique defect judgment AI with learning technology for overcoming difficulties. Through this joint technological development, the two companies aim to accelerate and expand efforts to automate a wide range of processes at manufacturing sites, including the automation of visual inspection processes that are currently conducted manually.

What is pseudo defect generation technology?

This technology utilizes AI called GAN (Generative Adversarial Networks) to capture the characteristics of defective data by learning from existing defective data, and generates pseudo defective data that looks exactly like the actual defective data.
We are developing a pseudo defect generation AI using this technology to generate a large amount of defect data to enable defect learning even in situations where there is very little actual defect data.

What is a Weakness Training Loop?

The defect judgment AI starts learning with a very small amount of real defect data. Next, it analyzes the judgment results and predicts patterns that the defect judgment AI is not good at. Then, the pseudo-defective generation AI generates defect data for the predicted pattern, and the defect judgment AI relearns using this data. By repeating this loop, it is possible to overcome the poor patterns of the defect judgment AI and develop a defect judgment AI that is comparable to a skilled worker from a very small amount of training data. We call this series of learning loops for overcoming weak points the "weak point training loop. In this fiscal year, we will develop an automatic visual inspection AI, which is in high demand at manufacturing sites, and aim to significantly shorten the time required to collect images for training.

Background leading to the start of joint development

DataGrid has developed synthetic AI technology to generate synthetic data (synthetic data) such as digital human and AI training data, and has conducted projects with a number of companies in the telecommunications, manufacturing, education, apparel, entertainment, and gaming industries. projects with a number of companies in telecommunications, manufacturing, education, apparel, entertainment, and gaming. In recent years, we have developed "Pseudo Defect Generation AI" that generates pseudo defect data to meet the needs of manufacturers who want to build highly accurate detection AI that can be used in the abnormal inspection process, including the visual inspection process, even in the early stages of manufacturing when there are not enough defective products. The company has developed a "Pseudo Defect Generation AI" that generates pseudo defect data.

Meanwhile, Sumitomo Electric Industries, centered on its IoT R&D Center, was working to improve productivity and safety by utilizing IoT/AI technology in order to promote digital transformation (DX) at manufacturing sites. One of these efforts was the development of a defect judgment AI that automates various defect judgments, such as those made by skilled workers during visual inspections. However, in order for the defect judgment AI to learn the advanced judgment criteria of skilled workers, it is necessary to prepare a large amount of training data covering a variety of patterns. The collection of training data is time-consuming and costly, so Sumitomo Electric Industries and other manufacturing companies have not fully utilized the defect judgment AI.

Therefore, taking advantage of the "Asian Business Creation Platform*" sponsored by the Kansai Economic Federation, DATAGRID and Sumitomo Electric Industries have begun joint technological development with the aim of developing a defect-detection AI from a very small amount of learning data by combining DATAGRID's strengths in generation technology and Sumitomo Electric Industries' strengths in defect-detection technology. The joint development of the technology has begun.

DataGrid and Sumitomo Electric continue to utilize AI technology to automate various processes and improve productivity.

Comments from both companies

Yuki Okada, CEO, Data Grid

We are very pleased to launch this advanced initiative with Sumitomo Electric Industries, Ltd. DataGrid has been working on various digital data generation technologies, and we believe that data generation for training AI (training data generation) is the key to solving the problem of insufficient data, which has been a limitation of AI utilization in actual industries. Together with Sumitomo Electric Industries, we will promote the adoption of AI in manufacturing business processes, and by extension, promote technological development to create a new standard for the use of AI for the entire industry.

Satoru Takahashi, General Manager, IoT R&D Center, Sumitomo Electric Industries, Ltd.

Sumitomo Electric Industries has been working to strengthen its manufacturing capabilities by utilizing advanced digital technologies such as AI and IoT, and has accumulated a large number of practical applications. To further accelerate these efforts, it is essential to further advance AI technology. We expect that the fusion of DATAGRID's advanced generative AI technology and Sumitomo Electric's AI application technology to manufacturing sites will lead to new achievements in the use of AI.

About the Asian Business Creation Platform

The abbreviated name is ABC Platform. The ABC Platform is an initiative to promote collaboration among companies and organizations in the areas of human resources, technology, and services to create business and revitalize the economy, with the aim of "creating business and stimulating the economy in Asia and Japan. The startup subcommittee of the committee has been promoting exchanges between Sumitomo Electric and Data Grid, leading to the start of technological development this time.
https://www.abc-pf.org/

 About Data Grid, Inc.

Giving life to all data. The company's mission is to "give life to all data," and it is promoting the industrial use of digital data (=synthetic data) such as video and audio synthesized and generated by AI. The company aims to increase the quantity and quality of data in the digital society and realize a future society in which AI is deeply and naturally infused into everything.

name of companyData Grid Corporation
EstablishmentJul. 2017
Location36-1 Yoshida Honmachi, Sakyo-ku, Kyoto-shi, Kyoto
Kyoto University International Science Innovation Building
representativeYuki Okada
uniform resouce locatorhttps://datagrid.co.jp/

About Sumitomo Electric Industries, Ltd.

We will continue to support industry and people's lives by providing values that contribute to the realization of an environmentally friendly, safe, secure, comfortable, and growing society through the challenge of new technologies such as AI and other technologies that will bring about social change. We hope to continue to support industry and people's lives by providing value that leads to the realization of a safe, secure, comfortable, and growing society that is friendly to the global environment.

name of companySumitomo Electric Industries, Ltd.
EstablishmentApril 1897
Location4-5-33 Kitahama, Chuo-ku, Osaka (Sumitomo Building)
establishment (of a business, etc.)
representativeOsamu Inoue
Business OverviewConstruction (civil engineering and building construction) and development business
uniform resouce locatorhttps://sumitomoelectric.com/jp/
PORT Editorial Officehttps://port.creww.me/
PORT by Creww is an open innovation media for startups, entrepreneurs, entrepreneurs in recovery, and companies seeking to create new businesses on the theme of social issues, operated by Creww Co.
- Advertisment -
- Advertisment -

Featured

東芝のオープンイノベーション、協業検討に進む2社が決定 ダイナミックマッププラットフォームとArithmerを選定

ReGACY Innovation Group株式会社(東京都千代田区)は、企画・運営パートナーとして支援する株式会社東芝の「Toshiba OPEN INNOVATION PROGRAM 2026」で、協業検討に進む参加企業2社が決定したと発表した。7月に募集した共創プランの中から、ダイナミックマッププラットフォーム株式会社とArithmer株式会社を選定。両社は東芝グループの「Hybrid-3D LiDAR」と「気象データサービス」を活用し、物流・インフラ分野などでの新たなソリューション共創を目指す。12月中旬の成果発表会に向け、実証と事業性検証を進める。 本プログラムでは、2026年7月6日から7月26日まで参加企業を募集しており、応募のあった共創プランの中から、ダイナミックマッププラットフォーム株式会社、Arithmer株式会社の2社が選定されました。 両社は、東芝グループが保有する「Hybrid-3D LiDAR」および「気象データサービス」の技術・データを用いて、物流・インフラ分野やデータ活用領域における新たなビジネスソリューションの共創を目指します。 今後は、12月中旬に予定されている成果発表会に向けて実証や事業性検証を進めるとともに、事業化を見据えた協業の検討を加速していきます。

新潟日報社、オープンイノベーションプログラム2026ピッチ審査会で「三ツ目」「想ひ人」2社を選定

新潟日報社は8月4日、「新潟日報社 オープンイノベーションプログラム2026 ピッチ審査会」を開催した。Creww株式会社が提供する「Creww Growth」を活用し、5月15日から6月15日にかけて新潟県や社会が抱える課題解決に向けた協業アイデアを募集。応募61件の中から書類審査を通過した8社がピッチに参加し、審査の結果「三ツ目」(新潟県三条市)と「想ひ人」(東京都)の2社を選定した。今後は実証事業等を通じて事業化を進める予定。

【ヤマサキ × BeeInventor】高温下での過酷な作業環境から、データは命を守れるか——創業151年の工業炉建設会社とスマートヘルメット新興企業が挑む「熱中症ゼロ」の現場

明治8年創業の株式会社ヤマサキは、建設業界向けアクセラレータープログラム「ヤマサキ×オープンイノベーションプログラム2024」を通じて、スマートヘルメットを手がけるビーインベンターとのPoCを実施した。内部温度800度を超える製鉄炉の傍らで働く現場作業員の熱中症リスクをAIで早期検知するという挑戦は、今まさに第2フェーズへと移行しようとしている。プロジェクトリーダーの向井裕一氏と、BeeInventor(ビーインベンター)のNao SeoChon氏に話を伺った。

【ヤマサキ × CHAOSRU】「誰も知らない技術」を、誰もが体験できる価値へ——創業151年の工業炉建設会社とVRスタートアップが挑む、採用×認知の壁の壊し方

明治8年創業の株式会社ヤマサキは、製鉄に欠かせない築炉の技術を持ちながら、業界・企業の認知度の壁が採用の課題となっていた。この課題の突破口を探るべく参加したのが、『ヤマサキ×オープンイノベーションプログラム2024』だ。プログラムを通じて、VR・XR技術を核とする株式会社CHAOSRUとの協業をスタート。2025年7月から始まったPoC(実証実験)では、高校生向け採用説明会へのVR体験の導入により、ブース来訪者数に明らかな変化が生まれている。アクセラレータープログラムを通じたスタートアップとの協業で何が動き、何が見えてきたのか。株式会社ヤマサキのプロジェクトリーダー 田中敏彦氏と、CHAOSRU代表取締役の内藤薫氏に話を伺った。