Important dates:
13 May 2026 — Competition launch and registration opening
29 May 2026 — Q&A webinar - 13:00 - 14:00
19 Aug 2026 — Registration deadline (please contact Kushtrim Visoka for more details if you have registered after 11th of August)
17 Aug 2026 — Dataset release
23 Aug 2026 — Final submission deadline
01 Sep 2026 — Winner announcement
Read competition documentation
Read the competition rules and prize requirements
Read our FAQ page
Contact Kushtrim Visoka for any questions or inquiries.
The dataset is collected from batteries in real sensors that are deployed across buildings in Norway. For each battery you will have access to time-series data of the battery voltage and temperature, and for a subset of the batteries, the time that the battery ran out (sensor stopped functioning). With this data you will make a predictive model that can estimate Remaining Useful Life of batteries. Make sure that your estimates are good, because downtime is costly!
In addition to the estimates of when replacements are needed, an effective plan will also need to take into account practical real-world constraints, such as the travel time between buildings, number of workers available, and that the workday is indeed finite. The dataset will contain the location of each sensor, and the travel time between each location. The complete solution to the challenge must automatically calculate a work order plan, for what batteries should be switched at what time.
This challenge is aimed at data scientists, engineers, and researchers interested in topics such as time-series forecasting and prognostics, decision-making and planning under uncertainty, and optimization and scheduling.
The competition is open to residents of Norway. Teams with a background from industry, academia, or a mix are welcome to participate. Teams may consist of up to 8 participants. Individual participation is also welcome, and participants may choose to form teams after registration.
The competition offers cash prizes totaling NOK 50,000, with up to NOK 40,000 awarded to the first-place winner and up to NOK 10,000 awarded to the second-place winner.
Note that to be eligible for the full price amounts, at least one team member is a Master student or PhD candidate or holds equivalent academic affiliation in Norway.
Participants in the challenge are also invited to submit a technical report to the peer-reviewed journal Nordic Machine Intelligence (NMI).
Scientific Committee
- Fadi Al Machot - NMBU
- Mina Farmanbar - UiS (Stavanger)
- Kyandoghere Kyamakya - University of Klagenfurt , Austria
- Habib Ullah - NMBU
- Ibrahim Abdelfattah Abdelhameed - NMBU
About Organizers
The competition is organized by NORA. NORA is a Norwegian collaboration between 8 universities, 5 university colleges, and 4 research institutes within artificial intelligence, machine learning, and robotics. The consortium works to strengthen Norwegian research, education, and innovation within these fields.
The dataset has been donated by Soundsensing. A team of experienced researchers from Norwegian University of Life Sciences and other universities will review the submissions.
About Soundsensing
Soundsensing is a leading provider of predictive maintenance solutions for heating, ventilation and cooling systems (HVAC) in commercial buildings.
Their products are used by building operators across Europe to efficiently manage and maintain their technical systems. Soundsensing is a Norwegian technology company that was founded in 2019 and is based in Oslo.
Acknowledgements
This dataset was partially funded by the research project EARONEDGE (Project No. 337085), supported by the Research Council of Norway from 2022–2026.
The challenge is also partially funded by the NORA Norwegian AI Research School (Project No. 331723), supported by the Research Council of Norway.