Interdisziplinären Zentrums für Wissenschaftliches Rechnen (IWR)

During the past decades, the foundations of science have developed a new pillar: scientific computing. Using the power of modern-day computers, it is possible to model processes, analyse data and image interesting phenomena in a sophisticated way. The University of Heidelberg thought this development was important enough to establish the IWR, the Interdisciplinary Centre for Scientific Computing.

On the first morning of our visit to Heidelberg, around 20 of us went to the IWR for a series of talks about their research. The IWR as an interdisciplinary institution does research in very different areas. This means, that it cooperates with several different faculties. Of course, people from the natural and life sciences make up a big part of the IWR, but more recently they have even started to work together with cultural anthropologists. I will now outline two of the research topics presented to us.

The first example of research done at IWR that we heard about is imaging of cuneiform tablets (Dutch: spijkerschrift). Those ancient clay tablets are one of the first examples of human writing. However, if you want to study them closely, there are a couple of problems. Usually they are quite badly damaged, so they are very difficult to read and very vulnerable. At IWR, people have developed a 3D scanning technique that takes pictures of the tablets with different (infrared) lighting. This gives a three-dimensional picture of the tablet, which can be viewed from all angles without hurting the original. This is so accurate, that it is possible to take the actual fingerprints of the writer! But the fun does not stop there. It is now possible to calculate several properties of the tablet, such as the curvature of the surface at each point. This indicates where the written symbols are. By simplifying the structures to known patterns, it is even possible to identify the writing.

Another example of the research work done at the IWR is one with close relations to the industry. An important branch of mathematics and computational science is optimisation: finding the maximum (or minimum) value of some quantity for a certain input, given a number of constraints. Sometimes, there are only a couple of possible inputs, for instance when you can use one of six gears. On other occasions you are allowed to have whichever input you want, for example when you control the speed of a car. Now it gets really interesting (and difficult) when you combine these two problems, and try to find out how to drive a car as quickly as possible round a track. It is exactly this what some people do at the IWR. It has applications virtually everywhere. The aforementioned example helped a certain, unnamed but very successful, Formula One team to gain an edge over their competitors, but also truck drivers might be interested in how they should drive on mountainous roads to minimise the amount of fuel used.

We have seen that at the IWR, scientifically minded people from various disciplines work together to use computers in a lot of exciting ways. As the IWR is not linked to any faculty, but rather stands on its own, it is possible to investigate virtually anything which needs the use of computers. This horizontal structure might be a good idea to implement in Utrecht, to stimulate the cooperation between departments and faculties.

Istvan Kleijn