Netum has developed an AI-based timetable solution for the University of Oulu – the first of its scale in Finland

Timetable work will become significantly more efficient as months of manual work are replaced by AI. The AI-assisted timetable solution puts the University of Oulu at the forefront of artificial intelligence capabilities among universities.

Twice a year, a team of more than 35 people at the University of Oulu spends three months working on the upcoming semester’s timetables and teaching allocations. Preparing timetables is a demanding and laborious process, says Henriikka Leskelä, the Solution Designer in charge of the timetable team in the University of Oulu’s Systems in Education and Study Administration Services.

“Every autumn and spring, the work starts from scratch – there is very little information that can be carried over from the previous year because courses, teachers, student numbers and space requirements change every time,” Leskelä explains.

According to Artturi Roitto, Head of Development in the university’s Academic Affairs unit, the major changes on the horizon accelerated the university’s decision to adopt an AI-based solution.

“Our student numbers are continuing to grow, and our degree programmes are changing. The way our premises are used is evolving and, over the next decade, we will move from two campuses to one. The automation of timetable work helps us to address the space-related challenges associated with these changes,” Roitto says.

The project with Netum began in spring 2025 with a preliminary study to assess whether AI would be suitable for assisting with timetable work, what kinds of requirements it would create with regard to data and algorithms, and what changes in the university’s operating practices would be required.

“The data environment has been fragmented, as information has been collected in different ways and stored in different locations. The harmonisation of data has been one of the main goals in this project,” Roitto notes.

Making tacit knowledge visible

The preliminary study was followed by the proof-of-concept stage, in which Netum developed the first version of the technical implementation of the AI-assisted timetable solution. Real course data was imported into the system to ensure that the technical implementation would reflect actual timetable work as accurately as possible. The results were positive, and the next step was taken towards technical production.

According to Henry Joutsijoki, Chief Data Scientist in Netum’s AI Solutions team, the implementation is the first solution in Finland to leverage AI at such a large scale. The AI-assisted timetable solution has attracted a lot of interest in the higher education sector.

“One of the key benefits of the AI-based solution is that increases fairness in the preparation of timetables. The AI relies solely on the data entered into the system, making the outcomes more equitable for both students and teachers. With the AI-based timetable solution, tacit knowledge around timetable preparation has become a transparent part of timetable planning,” Joutsijoki adds.

The impact of the project is enormous. It affects not only the work of tens of timetable planners and hundreds of university teachers but also tens of thousands of students. According to Project Manager Christoffer Brunberg from Netum, the AI-based solution improves the quality of data and increases the organisation’s AI capabilities.

“While the harmonisation of data is the first priority in the adoption the AI solution, re-examining the organisation’s processes and practices is also important. This type of project provides the customer with a great deal of new insight into how AI can best serve their operations,” Brunberg says.

The most demanding 10% has now been piloted

At the University of Oulu, 10% of the timetables for the upcoming autumn semester have been completed using the AI-assisted timetable solution. There is still some way to go before AI can handle timetable preparation in its entirety. The courses selected for the pilot were those considered the most laborious from the perspective of timetable planning.

“Traditionally, we have prepared timetables using a two-stage approach: we start with the most demanding courses, such as those involving students from multiple degree programmes or requiring a specific classroom. You could say that we jumped straight in at the deep end,” Leskelä says.

The system has not yet been integrated into the Peppi digital ecosystem, so space allocations have been made manually at this stage. This coming autumn, the University of Oulu will start the development of the Peppi integration in collaboration with Netum. Once this has been completed, the full benefits of the AI-based timetable solution can be realised. Roitto is cautiously optimistic about the possibility that, in spring 2027, the AI solution will handle the university’s technical timetable work in its entirety.

“Going from 10% to 100% is a big leap, but with the hardest part already done, we will be able to scale the system to cover the remaining courses quickly,” Roitto notes.

“Moving forward, the nature of timetable planners’ work will change. Scheduling and space allocation will require less human effort, but the quality of the input data will be more important than ever,” Leskelä adds.

The development effort has been a demanding process, and there have been challenges along the way. Leskelä and Roitto are grateful to Netum’s experts for their flexibility, commitment to the project and continuous willingness to make improvements based on feedback.

The members of Netum’s team also emphasise the value of learning together with the customer while working on something that is so new.

“We have genuinely worked together to find ways to overcome the challenges. A project of this complexity is not something you complete in one go; rather, it is a long process. The deployment of AI requires a significant shift in mindset and attitudes on the customer’s side, and the University of Oulu has been remarkably open-minded about that change,” Joutsijoki concludes.