Industry

You are here:

Home
Ecosystems
Industry

Industry

Modern industry relies on continuous monitoring, fast decision‑making and high reliability to keep equipment running safely and efficiently, while reducing energy use and unplanned downtime. NeAIxt supports this transformation by bringing Edge‑AI directly to machines, sensors and production environments, enabling on‑device analytics, predictive maintenance, quality monitoring and data‑driven optimization without relying on constant cloud connectivity.

Through 7 Industry‑focused Ecosystems, NeAIxt turns these needs into practical solutions, from on‑sensor predictive maintenance and digital twins to advanced inspection, metrology and smart monitoring for industrial and agrifood environments.

Ecosystem : Predictive maintenance

This Ecosystem brings ultra‑low‑power, on‑sensor predictive maintenance to rotating machines by training and running Edge‑AI directly on the node, cutting energy use and enabling real‑time alerts without local infrastructure.

	Ecosystem Partner list	Activity details
	Endiio	Ecosystem and Use case leader: Endiio leads the demonstrator by embedding signal processing and AI directly into ultra‑low‑power sensor nodes, enabling real‑time monitoring of industrial equipment such as pumps and gearboxes without relying on external infrastructure.
	Knowtion	Knowtion develops transparent and efficient AI models that can be trained directly on microcontrollers, enabling reliable predictive maintenance under strict power and memory constraints.
	Swiss Center for Technological Innovation	CSEM provides an ultra‑low‑power computing platform and optimization tools to efficiently run and benchmark predictive‑maintenance AI models, meeting strict real‑time and energy‑efficiency requirements on industrial sensors.
	Interuniversity Microelectronics Centre Netherlands	IMEC NL provides energy‑efficient AI acceleration and optimization techniques that enable reliable AI processing on very small, low‑power industrial edge devices.
	Emmtrix	Emmtrix turns AI models into highly optimized embedded code and validates the transformations to ensure they remain correct, maximizing performance on the target hardware.
	Microsensys	MSYS provides ultra‑low‑power sensor platforms that combine wireless connectivity and on‑device AI, enabling local data analysis, smart wake‑up and energy‑efficient data transmission for industrial monitoring.
	Danfoss	Danfoss provides industrial test environments to validate the solution under real operating conditions, measuring accuracy and response time for automation use cases.
	CEA	CEA explores advanced memory technologies and system‑level designs to support energy‑efficient, memory‑centric Edge‑AI architectures for predictive‑maintenance sensor nodes.

Ecosystem : Digital twins for energy saving and seizure prediction of vacuum dry pumps

This Ecosystem develops an Edge‑ready Digital Twin of industrial vacuum dry pumps to reduce energy and nitrogen consumption and predict seizure events, enabling real‑time optimization directly inside semiconductor fabs.

	Ecosystem Partner list	Activity details
	PFEIFFER	Ecosystem and Usre case leader: Pfeiffer develops digital twin models of vacuum dry pumps, using simulations and AI to reduce energy consumption and predict mechanical failures in real industrial environments.

Ecosystem : Advanced condition-based monitoring features

This Ecosystem develops advanced Edge‑AI condition‑monitoring features for next‑generation AC drives, combining a new AI‑enabled testbench with embedded AI algorithms to improve industrial monitoring accuracy and reduce inference latency.

	Ecosystem Partner list	Activity details
	Danfoss	Ecosystem and Use case leader: Danfoss builds a new industrial test platform with embedded AI, integrates the full hardware and software stack, develops AI models for condition monitoring, and provides real‑world sensor data for validation.
	Technical University of Denmark	DTU develops efficient AI algorithms tailored for industrial drive systems, improving real‑time performance, and works with DANFOSS to validate them on an industrial test platform.

Ecosystem : AI‑Driven Wafer Inspection and Optical Metrology for Advanced Semiconductor Manufacturing

This Ecosystem integrates AI‑based wafer inspection and optical metrology directly into manufacturing tools, enabling faster defect detection, accurate surface measurement on complex stacks, and reduced reliance on offline or destructive analysis.

	Ecosystem Partner list	Activity details
	Merck	Ecosystem and Use case leader: Merck developing AI‑driven wafer inspection and optical metrology solutions, combining unsupervised defect detection with advanced interferometric analysis to enable accurate, in‑tool quality control on complex semiconductor stacks.
	XFAB	XFAB provides production‑representative wafers, process modules and stack technologies that enable realistic validation of AI‑based inspection and metrology methods across diverse semiconductor manufacturing scenarios.
	Fraunhofer	Fraunhofer contributes materials expertise and memory‑related technologies, supporting the integration, characterization and reliability of advanced stacks used in AI‑enabled inspection and metrology tools.

Ecosystem : Object identification and measurement scenarios

This ecosystem enables energy‑efficient wireless sensing for fast‑moving industrial objects, combining smart wake‑up, embedded AI and reliable high‑speed data transmission.

	Ecosystem Partner list	Activity details
	Microsensys	Ecosystem and Use case leader: MSYS builds the complete wireless sensor system, combining smart sensor nodes, local data processing, and energy‑efficient communication to monitor fast‑moving objects over long periods.
	Institute for Microelectronics and Mechatronics Systems	IMMS develops a fast‑reacting RFID chip that allows sensors to wake up instantly and communicate reliably even when objects are moving at very high speed.
	XFAB	XFAB provides the specialized memory technology and manufacturing processes that make the low‑power, high‑speed sensor electronics possible.
	XFAB Erfurt	XFAB Erfurt develops the embedded memory blocks used inside the sensor chip, ensuring reliable and energy‑efficient operation in high‑speed applications.
	CEA	CEA supports the use case with expertise in advanced memory technologies and system design, helping to explore and validate efficient sensor and AI architectures.

Ecosystem : AI‑Enhanced Quality and Image Reliability for Infrared Sensors

This Ecosystem uses AI to improve infrared sensor quality from manufacturing to operation, enabling early defect detection, higher production yield, and cleaner, more reliable infrared images in real‑world conditions.

	Ecosystem Partner list	Activity details
	IRnova	Ecosystem and Use case leader: IRnova develops AI solutions that improve infrared sensor quality, from detecting production defects early to reducing noise in live infrared and polarimetric images.
	Royal Institute of Technology	KTH provides efficient AI hardware building blocks and design tools that allow advanced defect detection and image‑cleaning algorithms to run efficiently on low‑power edge devices.
	Swedish Defence Research Agency	FOI develops AI models and datasets that combine visual and polarimetric infrared data, supporting both defect identification during production and improved image interpretation in operation.
	Strikersoft	Strikersoft integrates the full processing chain by building application prototypes and Field-Programmable Gate Array (FPGA)‑based systems that validate AI‑based defect detection and noise reduction in realistic conditions.

Ecosystem : Agrifood monitoring

This Ecosystem deploys a local AI model on STM32 to monitor pistachio plants and production steps using sensors, smart‑glasses and drones, and complements it with a cloud Large Language Model (LLM) for actionable guidance, improving quality and safeguarding plant health.

	Ecosystem Partner list	Activity details
	Marullo	Ecosystem and Use case leader: Marullo defins field requirements, integrating sensors and aerial tools, generating training data through chemical and biological analyses, and validating the solution in real agrifood conditions.
	Ricca IT	Ricca develops the cloud and AI services by combining field data with agrifood knowledge to detect plant diseases and resource deficits, while ensuring secure data flows and cloud integration.
	Italian University Nanoelectronics Team with Universita di Catania	IUNET-UNICT co‑develops and optimizes local AI models for embedded devices, integrates sensors in the field and production lines, and feeds sensor data into higher‑level AI to generate corrective actions.
	STMicroelectronics Italy	ST-IT designs and deploys the edge AI solution by developing embedded models, selecting and programming STM32 hardware, and building the AI services that turn data into actionable recommendations.
	STMicroelectronics CROLLES	ST CRO delivers the silicon technology platform, including advanced memory integration and manufacturing capabilities, to support reliable and scalable edge‑AI systems.
	STMicroelectronics GRENOBLE	ST GRE develops high‑performance microcontrollers optimized for AI, providing the hardware platforms and tools used to run advanced edge‑AI applications in the field.