1st Call Projects


The way in which timetables are created has not changed for decades and is old-fashioned. It has been ‘modernized’ by means of connected computers with a lot of processing power, networks, internet, cloud and the Web, but this has not removed the underlying clumsiness. This leads to a lot of discussion, waste and dissatisfaction.

For instance: It is not clear for a timetabler if the timetable that is created is optimal or if there are alternatives that would improve the general performance. Can switching a room, teacher, day or time eventually contribute to satisfying more needs? This is not clear: the timetabling software will at best indicate that it is a ‘permissible’ solution, nothing else. The teacher provides the data, which are subsequently put into the timetable black box, resulting in the UT timetable. But is that really the best we can do? Is this an integrated whole that distributes the performance and resources as evenly as possible?

The timetabling process itself is just as unclear. Who do you need to contact if the curriculum changes? Should teachers inform the timetabler if they start working a day less per week? Which official is responsible for the provision of data and why are these exact four deadlines per year being used? Why does one teacher feel the need to give detailed information after doing a lot of work, while other teachers make do with a general guideline for the timetablers? Does it result in the same performance? This leads to ample discussion and, moreover, conflict and dissatisfaction.

The number of rooms UT uses for education is enormous; just like any other university in the Netherlands. In 2013 and 2014, UT looked into the actual occupation of classrooms, which turned out to be less than 30% during peak hours! UT did not even score badly in that study; VU University Amsterdam continuously studies the occupancy of its rooms manually. It turns out that they are at 25% on average. Saxion claims they have much higher occupancy rates – namely at 65% – but this strongly depends on how it is calculated and only applies to the planned occupancy of the classrooms. Also, these classrooms are scheduled centrally, and a lot of rooms are excluded from the calculations, such as computer rooms, laboratories, academy-specific rooms, etc.

There is no feedback with respect to the educational programme and/or the timetabler. For instance, if a teacher indicates that 80 students are expected for a class and that eight project rooms will be required for the next six months, then three or four of those rooms will likely be empty after the first meeting. These rooms can subsequently no longer be used, because they have been scheduled. The timetabler is not informed of the fact that they can be repurposed, nor is the timetabler or programme informed if only a small portion of a room is used each time. We are like blind people in a brightly lit room: all the ingredients for success are there, but we simply cannot succeed due to the constant lack of feedback.

In short, timetables and the use of rooms are not only closely linked in higher education, there is great room for improvement as well. All it takes is some guts and plenty of intelligent and driven people working together. Where else than at UT?

Just imagine: instead of economical, ‘it’s good enough if it works’, unclear performance, unclear standards, lack of feedback, no reflection on one’s own production, and a lot of dissatisfaction, we can switch to timetabling as a means to maximizing education. Maximize it! With clear performance and standards in which maximization of the performance can be demonstrated in the model by measuring with regard to the timetable. Clear performance indicators will provide a rating for the timetable. If the timetabler is unable to achieve a higher score by means of making changes to the timetable, then this is apparently the maximum that can at that point be allocated to the education. This will result in clear performance, complete transparency and a fair distribution of resources.

The maximum allocation can be calculated prospectively, before the final timetable is published. This will require simulation models and a lot more measurements than are currently available:

- We can measure the actual use of classrooms (i.e. all rooms suitable for education), which will provide the timetablers with live feedback on the use of rooms and, as a result, on the timetable’s performance. Apart from the actual use of a room, the number of occupied seats compared to the maximum is also checked. This requires the use of sensor technology. However, it does not require the most expensive technology; it requires good, affordable and highly available technology.

  • Subsequently, the data must be matched to all timetables in order to gain meaningful insight, which requires a lot of OR.
  • The data must then be stored for trend analysis purposes. If a room is empty all the time, then it can perhaps be structurally repurposed. If a room is always used partially, then the group using it can perhaps be relocated to a different, smaller and/or more suitable room. This requires OR, mathematics and a bit of business administration.
  • The option of measuring occupancy and using live measurement also provides the option of, for instance, using project rooms more optimally by joining if the maximum capacity has not been achieved yet, for example.
  • We can measure the walking distance between all lecture halls and include it in the timetable’s performance. The walking distance will be measured both over weeks (constantly switching to another lecture hall can lead to high dissatisfaction) and shoulder-to-shoulder between two classes (walking to a different building as a teacher together with your students while you could have just stayed where you were). This requires business administration and mathematics and can contribute to, for example, the Campus App by projecting the optimum route to your next class. For instance, not many people know how to get to the fifth floor of Vrijhof or to Waaier 3. This addition to the Campus App will show them.
  • We can measure how suitable classrooms are for specific education. This provides an indication of how the rooms perform. Working together with BA, EPA and Educational Science, the policy of new-building and renovating rooms for higher education as well as the actual preferences and performance can be looked into. Everything is intended to serve the creation of an optimum learning environment for students. This could involve checking temperature, (day)light, CO2, particulates and other characteristics, such as the relative distance to the department or study society, the use of colours, the interior decoration, etc.
  • We can ask what constitutes a good timetable. A structured questionnaire could help in ascertaining what is required in order for a timetable to be experienced as robust and pleasant. How many hours of classes should there be? How many of those hours can be lectures? How many can be scheduled in one day? Or in one week? How many free periods? The list goes on. What is timetable 2.0?
  • A timetable’s score can be reviewed retrospectively to see whether it has met all requirements and use of space. This will result in a much-needed feedback loop in order to achieve prospective maximization of timetables.

Thanks to all these measurements, a timetable can allocate maximum resources to education. This timetabling method is significantly different from the one currently made possible by timetable software used around the world, the underlying algorithms of which have not been altered since the 90s.

Together with Saxion, a study into the timetable preferences of 21,000 students is currently being conducted. In the new-build in Apeldoorn sensors will be installed that are designed to measure attendance. Together with Omix and SURF, a study is being conducted into a logistics model to make educational support more efficient. After all, installing sensors is one thing; you need to make clever use of them and properly study the results before they can be truly beneficial to the organization. The UK-based company Semestry provides timetable software and is interested in discussing a possible partnership with us. Many parties are truly interested in what we can, and hopefully will, do.

The institute’s effective use of space can be improved significantly, which would save on costs tremendously. However, the underlying idea of maximum allocation to education is undergoing a paradigm shift that education was able to make, but which educational support in its turn has not been able to make fully yet. This could possibly help support and the primary process to get more attuned to each other.

If you would like to know more, I can send you the memo on the innovation project for timetables and use of space.