Artificial Intelligence for Sustainable Production
Artificial Intelligence, mainly
We study how AI technologies can be applied to (i) reduce GHG emissions of production plants, and (ii) support decision-making and trade-offs when controlling production plants with regards to vital parameters such as production speed, worker wellbeing, sustainability goals and waste management. In designing an AI technology for these purposes, a range of underlying components is being developed, including: a knowledge base for decision support where insights are semantically linked to each other; an IoT framework for data-collection, aggregation and actuation; and a privacy-framework that allows for insights to be shared between enterprises without compromising their mutual privacy of business-critical data.
The machine-readable knowledge base overlaid with the privacy framework will allow for automated decision support in real-time, as well as execution of so-called what-if scenarios. A facility owner will, for example, be able to ask ALISTAIR what happens if certain parameters of processes in the facility are changed, and how this would affect GHG emissions vs productivity, and well-being of the factory workers, in line with SHE (Safety, Health, Environment) factors.
Machine-readable insights are generated from industry-specific data in real-time and stored in the knowledge base. The system is designed to be applicable to multiple industries. To be able to relate insights from different industries to each other, a semantic mapper will be developed.
ALISTAIR will utilize insights from multiple industries, which allows for the decision support to improve much faster over time. This is possible due to a novel privacy framework in which the participating enterprises can decide which insights they are willing to share and with which enterprise. The privacy framework will be granular and with full observability and traceability of decisions.