Your browser doesn't support javascript. This means that the content or functionality of our website will be limited or unavailable. If you need more information about Vinnova, please contact us.

Interpretable AI from start to finish

Reference number
Coordinator RISE Research Institutes of Sweden AB
Funding from Vinnova SEK 4 677 759
Project duration September 2022 - June 2025
Status Completed
Venture Advanced digitalization - Enabling technologies
Call Advanced and innovative digitalization 2022

Important results from the project

Based on two case studies, we have investigated the application of interpretable AI. An important result was a methodology for deriving interpretable features from signal data. Both case studies focused on converting semi-quantitative assessment criteria into a format suitable for ML. In the first study, we developed a model to assess the natural state of forest areas based on photos. In the second, we derived ML input to evaluate the fuel efficiency of ship propellers using AIS data.

Expected long term effects

With the tool developed in the first case study, the natural state of forest areas is assessed. The tool is expected to lead to better monitoring. Possible further developments include the identification of neophytes, studying the effects of clear-cutting or calamities on biodiversity. Adapting the propeller to the actual movement and speed profiles of vessels can result in significant CO2 savings. Interpretable AI can be used to transfer tacit expert knowledge into a more formalized form.

Approach and implementation

The studies described above were conducted in parallel and according to schedule. During regular meetings, both among scientific staff and with the entire group, we coordinated the ongoing work. Towards the end, the project leadership changed due to a retirement. The collaboration was very fruitful and inspiring, resulting in scientific publications as well as promised tools. The learnings were also shared at two workshops organized by Advanced Digitalization and at RISE internal conferences.

External links

The project description has been provided by the project members themselves and the text has not been looked at by our editors.

Last updated 31 October 2025

Reference number 2022-01702