GAMM 2003

By Niall McMahon
March, 2003

An informal report on the 2003 conference of die Gesellschaft für Angewandte Mathematik und Mechanik - the (German) society for applied mathematics and mechanics - held from March 24th to March 28th 2003 in Abano Terme near the city of Padua in Italy.

Summary

My first impression of GAMM 2003 came courtesy of McDonald's. After arriving in Padua by train late in the afternoon of Monday the 24th of March, I sat with a coffee in the city's only McDonald's at the Piazzale Della Stazione reading through a list of hotels. The sun was shining and Padua looked very appealing indeed. I was there less than ten minutes when a man walked in with an engineering look about him - this is hard to define exactly. So I was wondering if he was in town for the conference when a colleague of his arrived confirming my suspicions; she had a large plastic carrier bag with "GAMM 2003" written across the side.

I checked into the two star Hotel San Antonio on Via San Fermo, which I recommend for its location and helpful staff, explored Padua a little, and then set out the next morning for the small town of Abano Terme and GAMM 2003. Abano - which comprises two almost connected towns, Abano Terme and Abano Montegrotto - is only 35 minutes at most by the slowest bus from Padua. The buses are regular, every 15 minutes or so, and run from the train station on the Piazzale Della Stazione right through Padua, along Corso Del Popolo, around Prato Della Valle, and out to Abano. You can flag the bus down along the route at well-marked bus stops. Buses beginning with the letters A, M or T all go to Abano Terme, though not they are not all equivalently direct.

Arriving in Abano at 07:45 on Tuesday morning I was struck by the difference between Abano and Padua. Abano is a resort town famous for thermal springs. Its wide tree-lined streets are perfectly clean with hotels set back from the pavements. The walk from Piazza Della Repubblica to Via Pietro D'Abano, where the conference centre is located, is a few hundred metres and at that hour of the morning everything looked very fine! I noticed that one of the resort hotels was called the something of peace, so I suspected that Abano probably attracts an older age group seeking rest and repose. This was later confirmed by observation. Nothing seems to happen too quickly there. Padua, on the other hand, is very busy with people, bicycles and scooters weaving everywhere.

I arrived at the conference centre along with one other fellow at about 08:00 on Tuesday. Despite the GAMM signs everywhere, we had a little difficulty actually finding a way into the place. After a little more exploring we discovered, independently, the main conference centre nearby; this is where the reception was. The conference centre, named for Pietro D'Abano also, is impressive and the GAMM conference staff were efficient, smart and friendly. The computer facilities were well appointed with a fast internet connection.

The Conference Proper

Tuesday 25th - Plenary Lectures

The first plenary lecture I attended began at 08:30 and was given by Wolfgang Ehlers of the University of Stuttgart. The title was "Advances in porous media theories with focus on scientific engineering and biomechanics". At the time of writing I cannot remember anything about this talk except that I did not really understand it very well. I also attended the following three plenary lectures given by Martin Rumpf of Duisberg, Christian Lubich of Tuebingen and Franco Brezzi of the University of Pavia, Italy. The most useful of these was Christian Lubich's presentation of numerical integration using the Störmer / Verlet method. His presentation was the most straightforward, using only acetates. There were not so many questions asked - if I remember correctly - though Franco Brezzi's talk, of which I followed little, generated some limited discussion. He discussed the Reissner-Mindlin model. One older fellow seemed to be offended that Brezzi had dismissed a simpler approach (which he apparently endorsed - he mentioned a book he'd written I think! Of course!). This was resolved by Brezzi who conceded the point, after a fashion, while proposing that his ideas were the more general. This, at least, was my understanding of the exchange.

Wednesday 26th

On Wednesday I attended a short communication on mathematical biology. This was held in the nearby Savoia conference centre (one of the three centres used) and began at 13:30. The presentations were all quite simple (using acetates) and relatively informal. The first talk was given by Johannes Müller of Karlsruhe on the subject of mathematics and the immune system. My feeling was that the material was interesting and its presentation good, but that there were too many ideas for the 35 minutes. There was some discussion about immune system response times as the theory led to circular patterns in response - time space that are so far unreconciled with observation. There are no diseases that the speaker knows of that behave like this when interacting with the human immune system. This is my understanding. One professor then revealed that he once had a disease that "attacked" every couple of weeks! The speaker noted that his theory required a disease that had such a cycle, but at a scale of minutes. He nevertheless was interested in the name of the Herr Professor's disease - unfortunately the professor could not remember it!

Incidentally, Herr Müller mentioned 'Markov Chain Monte Carlo' methods - this is something that I may be interested in sometime in the future.

The next talk was delivered by Dr. Hartmut Schwetlick of the Max Planck Institute and concerned "traveling waves for the chemotaxis system". The work presented seemed to be very good. Unfortunately, as usual, I did not follow it very well. Dr. Schwetlick proposed a model using traveling wave theory of how cells migrate in response to a chemical stimulus. For example, how cells move to repair the damage to a cut finger. Afterwards, a member of the audience asked a technical question. One thing I remember thinking during the talk was that there were too many equations on the acetates. Also, there was a very informative diagram displayed towards the end of the talk, showing what was being modelled, that would have been useful at the outset. The work did look very good though!

The third and final talk was delivered by Prof. Ludwig Cromme of BTU Cottbus in Germany. He spoke about models for the long-term dosage of anticoagulant drugs. An experimental study was carried out for two years with a patient on a long-term course of such drugs. Apparently there is a measurable index that determines their effectiveness. The patient's index was monitored over the two years along with other information such as the number of pills taken, the timings and so on. By correlating the index with the drug administered over time, Prof. Cromme fitted the data to a particular mathematical model by assigning suitable values to unknown constants; this after comparing the model with the experimental data. The model was subsequently used to generate an "optimum" drug regime for this patient. The resulting indices were then shown to be generally better than an equivalent sample from the previous two years. The discussion afterwards focussed on experimental technique and the fact that the results presented are not by themselves enough evidence of the model's effectiveness. Another question asked was concerned with the relationship between the model constants and the index - as to whether there was a direct correlation. Prof. Cromme, from experience, stated that he was sure there was no such connection, but that he could graph the data and investigate this. It seems to me that it is possible that the indices are dependent on the constant.

On Wednesday night, there was a combined tour and dinner which I reluctantly decided to miss.

Thursday 27th - My Talk

On Thursday I was speaking at session 13.4, mass transfer studies. In this session, which began at 13:30 in the Savoia conference centre, there were four other speakers and all of us were to speak for 15 minutes with 5 minutes allowed for discussion. I estimate that there were probably no more than 15 people in attendance. I was third to present. My talk was slightly under time and generated a couple of questions. The first concerned whether or not the model I presented (of a simple dissolution process) took the porosity of the solute into account - it doesn't - and the second from the chairman, Dr. Stefan Braun, concerned my use of scaling in the model. Discussion was limited, almost non-existent, but this was a characteristic, I think, of the group of people present! Actually to be fair to all of us, I believe that there are quite a few reasons why discussions didn't happen and I will return to this later.

Prof. Yasuaki Ichikawa of Nagoya University in Japan was first to speak. The title of his talk was "microscale adsorption and homogenization in porous media" and he made his presentation with MS Powerpoint.

As an aside, mine was the only presentation of the five to use acetates and an overhead projector. The use of Powerpoint, however, was not the rule in all sessions. In fact counting just now it seems that exactly half of the talks I attended (eight of sixteen) used an overhead projector while the balance used Powerpoint. This includes the Plenary lectures where three of the four speakers used Powerpoint.

Prof. Ichikawa's lecture was well presented and interesting to a point. One or two questions were asked, one was of a technical nature but I don't recall it at all. If I followed correctly, Prof. Ichikawa's work is being used by the German government in the design of particular nuclear waste storage facilities.

The next speaker was Ms. Özlem Özmutlu of the Technical University of Munich, who spoke about the "Numerical simulation of fluid flow and enzyme catalysed substrate conversion in a packed-bed enzyme reactor". There was some confusion at the start due to floppy disk problems, and I think the chairman also introduced her incorrectly. I missed the mistake, however it have may been that she is incorrectly titled the "Chair of Fluid Mechanics and Process Automation" in the book of abstracts. I assume this is a mistake! She appeared to be very competent so I may be wrong, of course. The presentation was good and the work seems very sound, though again I have to admit to not following it in detail. I asked Ms. Özmutlu about her intentions for future work; she seems to be quite clear about its direction.

Dr. Stefan Odenbach of ZARM at the University of Bremen then described an experiment carried out by his team to investigate the "thermal convection in a ferrofluid supported by thermo diffusion". Dr. Odenbach seems to be very enthusiastic about his work; his presentation was animated. The experiment impressed me. To use a phrase borrowed from a colleague of mine, it looked to be a nice experiment nicely done. I can only recall one question being asked, whether or not there was an industrial application. Dr. Odenbach answered, if I remember correctly, that not everything happening at ZARM has to have an industrial application. I liked this.

The final talk was given auf Deutsch by Dr. Lutz Nasdala of das Institut für Statik (Institute for Statics) at the University of Hannover. Its title was "Homogenisierungsmethoden zur Berechnung der Sauerstoffdiffusion und –Reaktion in Faserverbundstrukturen" or "Homogenising methods for the computation of oxidation in GRP (Glass Reinforced Polyester) structures". I understood very little of this since my German was not good enough. However the slides were mostly in English and I could understand the graphs mostly, though I can't now remember much about them (I do remember that one particular graph plotted results from a finite element model alongside analytic results and they matched exceptionally well, this impressed me at the time). There was no discussion at all afterwards.

The session ended at 15:10 approximately.

On Thursday afternoon I also attended a talk given by Prof. Peter Berglez of the Department of Mathematics at Graz University of Technology in Austria. It was part of session 16.2 and was entitled "On the representation of generalized axially symmetric potentials in the neighbourhood of isolated singularities".

Friday 28th

Since a colleague of mine has an interest in parallel computing I attended a talk entitled "A parallel computational approach for free surface flows with structural interactions" given by Steffen Genkinger from das Institut für Baustatik (the Institute for Structural Design) at the University of Stuttgart. The presentation itself was impressive with animated free boundary events, for example he demonstrated the flow out of a container when one wall is suddenly removed. This was achieved using a variation of a finite element method I believe. This was part of the interaction problems session which began at 16:00 on Friday afternoon.

The next session concerned boundary layers and boundary layer separation. I attended two talks, the first by Dr. Stefan Braun from the Technical University of Vienna on the subject of "blow-up phenomena in marginally separated boundary layer type flows", the second by Mats Kinell of the Institute of Fluid Dynamics and Heat Transfer, again from the TU Wien, titled "on a new form of marginal separation". Both talks presented interesting ideas. However I believe it was the discussion after both of these talks that made this session the most rewarding of all those I attended. There were at least three experienced engineers or scientists in this session who eagerly discussed the implications of the mathematical models presented during the five minute discussion periods. One fellow - a professor of some description I estimate! - discussed what was happening from a physical perspective and related this to the mathematics. This generated questions and comment from others present and made the session much more enjoyable.

The conference was over by 18:00 on Friday afternoon.

Notes

I have been thinking a little about the value of attending a conference such as the annual GAMM meeting. Well I've come to the conclusion that the fundamental reason for attending such a conference is to meet people and then to share ideas. And you know, there is a lot of value in meeting and listening to good people!

A little earlier I mentioned that there were quite a few reasons, in my opinion, why the discussion at "my" session (#1) was quite limited. A comparison with the last, discussion-filled, session (#2) I attended might help.

At first I thought that part of the reason for the lack of involvement was that the projects were all quite diverse, that the scope of session #1 was too broad. The fifteen or so people in the room, I thought, may not have felt confident enough of their understanding of one another's work. However session #1 was no more broad a session than #2, and yet there was plenty of discussion. And as for lack of understanding, that's nonsense really! You do not have to understand something to discuss it. In a book I've just finished reading - "Surely you're Joking, Mr. Feynman", stories about and by Richard Feynman - Feynman says, about discussion:

"On the contrary," I answered. "It's because somebody knows something about it that we can't talk about physics. It's the things that nobody knows anything about that we can discuss...so it's the subject that nobody knows anything about that we can all talk about!"

And I think that this makes more sense than my first, lazy theory. So what else? Well session #1 was quite formal. But so too was session #2.

In fact sessions #1 and #2 were practically identical except for one thing - the people involved. Session #2 had two, or three, enthusiastic professors that were genuinely interested and excited about the work being presented. They were also experienced. Session #1, my session, had about 15 people - some who were also genuinely interested, possibly excited and experienced - but nobody enthusiastic. You must understand that by "enthusiastic" I mean someone who is wants to talk and think and who by example actively encourages participation. This kind of person need not be experienced, but this would help I think.

In a book I once partly read about manufacturing plants - the name escapes me now - the author described how in his opinion each project undertaken in manufacturing should have a "master" involved. The master, as defined by the author and if my memory serves me, understands how projects work and has a lot of technical know-how, guiding the team through the process. The author was convinced that projects fail without the involvement of such a person. Perhaps, in a similar manner, each session in a conference could have at least a couple of enthusiastic people assigned to it? This might already happen of course in the form of the chairman. Also it might be easier to say "assign an enthusiastic person" than to actually do it.

So then perhaps the best advice would be for all of us to try our best to act enthusiastically when we attend conferences, and so create an environment conducive to discussion.

Finally, from observation I would have said that there were at least several hundred people at the conference. From the book of abstracts, a quick estimate gives a figure of 300 people presenting. So, I imagine there were at least 300 people there.

Conclusion

I could write a lot more about my observations at the conference and experiences in Padua, Milan and Venice. This might yet happen - watch out for future revisions! Attending GAMM 2003 was a very positive experience. I met some very interesting people, not all directly connected with GAMM, presented my work, and saw something of northern Italy. There's little more you could ask for!

References

You can find out more about Padua here. There are two major airports close to Venice, Marco Polo and Treviso; Ryanair/flies to Treviso from Stansted. Buses leave Padua regularly for Venice. Venice is about 40 Km distant, a journey of 40 minutes by bus or something similar by train. There are frequent train connections to Milan and Venice. Milan is approximately three hours away by train. There are direct buses from Padua to Marco Polo airport. Ryanair runs regular shuttles to Venice, about 30 Km from Treviso, leaving you just beside the bus-stop for Padua. When I was in Venice, the bus-stop for Padua was number 3 at the main bus-station in Venice. The Ryanair shuttle leaves you at number 4. Check out for more information. Aer Lingus flies to both Milan and Bologna directly from Dublin, both a train trip from Padua and Venice.

Places:
Hotel Sant'Antonio 
Abano Terme 

People
Dr. Hartmut Schwetlick.  
Prof. Ludwig Cromme.  
Dr. Stefan Braun.  
Prof. Yasuaki Ichikawa.  
Ms. Özlem Özmutlu.  
Dr. Stefan Odenbach.  
Dr.-Ing. Lutz Nasdala.  

N. McMahon, 2003.

Most material © Niall McMahon. See legals and disambiguation for more detail. Don't forget that opinions expressed here are not necessarily shared by others, including my employers.