FutureMain Co., Ltd., an industrial AI specialist focused on equipment diagnostics and predictive maintenance, has announced a central role in a next-generation intelligent semiconductor R&D project aimed at developing integrated AI software for time-series big data on NPU-based environments.
The initiative is part of the Next-Generation Intelligent Semiconductor Technology Development Program, under the Intelligent Semiconductor Software category. Running from April 2023 to December 2026, the project is led by Mobilint Co., Ltd., with FutureMain, Raonroad Co., Ltd., and KAIST as co-research institutions.
AI for Edge and Server NPUs
The consortium is building an integrated software package capable of processing large-scale time-series data from smart factories and smart cities, leveraging CNN- and RNN-based AI models optimized for both edge and server NPUs. The package combines time-series data management, security, visualization, and NPU SDK components, including compilers, quantizers, and runtime libraries.
To validate real-world applicability, the project is developing and testing Edge and Server NPU prototype boards. Demonstration applications will cover:
- Smart factories: Equipment condition monitoring, anomaly detection, automated diagnostics, and process visualization.
- Smart cities: Traffic data collection, average speed prediction, traffic digital twin applications, and pilot implementations with local governments.
FutureMain’s Role
As the co-research institution responsible for the smart factory domain, FutureMain is leading R&D efforts in AI and security technologies for equipment and process data. Key responsibilities include:
- Defining and classifying equipment and process data
- Designing data collection methodologies
- Developing technologies across the time-series data lifecycle
- Creating training modules, algorithm management architectures, and secure storage and anonymization systems
- Deploying server-based AI algorithms for anomaly detection in industrial equipment and processes
“Through this project, we are advancing AI and security technologies that enable efficient processing of smart factory time-series data in NPU-based environments,” said a FutureMain representative. “We will continue our R&D to ensure that anomaly detection technologies can be effectively applied in real industrial settings.”
Scaling Industrial AI Globally
FutureMain plans to leverage the NPU-based time-series AI technologies developed in this project to enhance industrial AI solutions across smart factories and other fields. By focusing on anomaly detection and secure processing of equipment data, the company aims to expand the practical applicability of AI in industrial environments.
With global interest in NPU-optimized AI processing, particularly in the United States and other international markets, FutureMain is exploring opportunities to strengthen the competitiveness of its industrial AI offerings using capabilities secured through this R&D program.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
