AI for Efficient Industrial Operations

Advanced Machine Learning for

• Energy and operation efficiency

In today’s rapidly evolving technological landscape, there is an increasing emphasis on efficient and environmentally responsible monitoring of industrial systems. The integration of real-time analytics has become essential for the timely detection of inefficiencies and suboptimal operations, especially as modern systems grow in complexity and environmental regulations become more stringent. These pressures often require technological transitions—such as adopting new energy sources or reconfiguring legacy systems—where operational insights must be derived from very limited examples. These systems must be capable of understanding their operational context and learning from available data—typically with sparse expert annotations—to independently identify, diagnose, and respond to emerging issues.

Example: Federated Learning & Edge Processing for Safe and Efficient Operation

• Process optimisation

The goal is that through this integration the industry can shift from reactive quality control to proactive process optimization, resulting in more sustainable and resilient production processes, with energy savings, less waste, and improved product quality. We  explore how the use of advanced machine learning methods, automation, new sensor technologies, together with real-time and offline sample analysis can drive towards more efficient and sustainable production systems across several industries.

Example: During the autumn 2025  we start a collaboration with Oatly, ORKLA, Höganäs, Incipientus, RHI Magnesia, and Lantmännen with the goal of  enhancing processing industry efficiency and insights with AI and smart sensors.

• Predictive Maintenance

The goal of predictive maintenance is to identify imminent failures and intervene sufficiently early before they happen. Machine learning is the right tool when the understanding of fundamental principles of the system is not comprehensive, and the system is sufficiently complex that developing an accurate model is prohibitively expensive. Another benefit of machine learning tools is the generalizability of these models to similar subsystems without fully understanding the design details.

Example: Future AI-based maintenance
 

Applied use of RISE state-of-the art research competence

RISE has extensive experience in working together with companies with real challenges. We have deep domain knowledge and solid competence in AI, allowing AI to work in practice. We develop customized AI tools for all stages of the maintenance chain: 

• Deviation detection, diagnosis, service life forecast, maintenance planning 

• Hybrid AI – combines several AI methods into an effective overall solution 

The goals are more efficient maintenance, higher availability and uptime, lower risks of unplanned downtime, and more robust systems.

 

RISE offer

RISE offers several pathways for organizations that want to collaborate with us:

• Direct assignments that provide access to research expertise in AI testing and information security.
• Packaged services tailored for both the private and public sectors.
• Participation in future research projects, funded at the national or EU level, for organizations aiming to be at the forefront of responsible AI development..