Mitsubishi Heavy Industries, Ltd. (MHI) and SoftBank Corp. have announced a joint demonstration experiment deploying SoftBank's AI-RAN product, AITRAS, within the edge data center solution DIAVAULT at MHI's Yokohama Hardtech Hub. This initiative aims to verify the effectiveness of high-speed AI inference in a secure, on-premises environment.
The collaboration seeks to integrate MHI's scalable edge data center technologies with AITRAS, facilitating AI execution in real-world settings. This effort addresses the growing demand for data centers that provide stable communication and data security closer to operational sites, a necessity for effective AI utilization in industries such as energy and industrial machinery.
As part of the experiment, MHI will deploy an AI application designed for defect detection and repair of its products within the DIAVAULT environment. The application will analyze image and video data from field workers' smart devices in real-time, identifying fault locations and suggesting repair methods, all within a closed network environment.
The demonstration aims to validate three key outcomes: real-time analysis through edge AI processing, enhanced security within a closed network, and increased efficiency and standardization of repair operations. By completing AI processing within the closed network, the initiative reduces data transmission and analysis latency, while ensuring that confidential product data remains secure.
Looking ahead, MHI and SoftBank plan to leverage their strengths to develop a scalable and reliable edge AI operational platform that adapts to AI advancements. They also aim to accelerate AI transformation for customers through innovative solutions.
Comments are closed.