IFORS 2014 in Barcelona
9th August 2014
In July 2014 I attended the IFORS conference in Barcelona. During 2013 I was busy completing my thesis, so I did not have a lot of time to attend many operations research focused conferences. Hence, attending IFORS was a great opportunity for me to see what the operations research community is working on and help me develop a few ideas. I will share my experience of attending IFORS and some of the things that I gained from this.
Starting with an entertaining story
Before I start talking about IFORS and what I did during the conference I thought that I would tell you an entertaining story. This story is about my time in Barcelona and one of the perils of going to the beach.
Prior to leaving Australia I decided to venture into the world of open water swimming. As a result, I loved exercising out in the ocean and my various encounters with marine life added to the experience. As you would probably be aware, Berlin is not that close to the ocean, hence open water swimming is something that must be left to times of travel. Fortunately IFORS was in Barcelona, giving me the perfect opportunity to once again enjoy open water swimming.
My preferred time for swimming is early in the morning, hence I had no one at the beach with me to look after my belongings while I was in the water. I just had to trust that my clothes were not that interesting to any people passing by. As an attempt to make them less interesting I put my shirt and shorts in my shoes and covered everything with my towel. The first morning when I got out of the water I noticed that my belongings had been disturbed, but thankfully everything was still there. The second morning I was not so lucky. While I was in the water having a great time someone decided that they liked my running shirt and shorts. Upon emerging from the water, I discovered that my belonging had been disturbed (however the towel was still covering my shoes) and that my clothes were missing. Not great. However, worse things could have happened. If my towel was taken then the keys to the apartment would have been stolen. Here comes the entertaining part, to get back to the apartment I had a 10 minute jog. So minus shorts and shirt all I could do was put on my shoes and jog back wearing only my "Speedos". I would have loved to hear the thoughts of the people that I passed!
Time to discuss the conference. Starting with my talk
My talk was in the very first session of the conference, so I needed to be well prepared prior to arriving in Barcelona. Fortunately, I was really looking forward to giving a talk at IFORS, so the preparation was very easy. As mentioned previously, I had not attended an operations research focused conference for a while, in fact the last one was ISMP in Berlin during 2012. Since IFORS is a large conference, I was hoping that there would be people in attendance that were interested in my aviation research. It really lived up to my expectations and there was more interest in my work that I have had at conferences previously.
One of the best things that I got out of my talk was what I discovered at the conference dinner. While chatting with a friend I was asked whether I still get nervous when presenting. I realised that this time I did not. I put this down to the fact that I really didn't care about how I was going to perform when presenting and I just got the job done. This makes sense to me as I feel that the more you think about something, the more likely you are to doubt your abilities. Generally a poor performance will result from this growing doubt.
The talks I attended
When I attend a conference, I try really hard to view every talk. Every one has a different take on this approach and I have been told that it could be counter productive as I may not always be the most attentive. However, it is something that I like to do. IFORS was a very big conference, so there were plenty of interesting talks to see in every session. As a result, I was able to build a great schedule that exposed me to many different applications of operations research.
My resulting schedule was a very mixed bag of talk topics. This is partly due to the different projects that I am currently working on. The topics broadly fell into the categories of mixed integer and linear programming, transportation problems (mainly aviation and railway), health care, and anything I could find regarding decomposition techniques (in particular Benders' decomposition and column generation). I really enjoyed seeing such a diverse range of topics as it stimulated me to think further about my current and future research projects. In particular, I am really interested in pursuing work in health care as an extension to my current research on HIV sequence analysis. This appears to be burgeoning area of operations research and there are many great applications.
Catching up with friends
I have only recently completed my PhD, but I felt much more a part of the operations research community at this conference than at any time previously. During my PhD I had the opportunity to visit GERAD in Montreal. This was a great professional experience and it also introduced me to many new and interesting people. I was a little slack in preparing what I would be doing in Barcelona in regards to social events, so I didn't even know that some people I worked with in Montreal were also attending IFORS. It was great to talk with people that I had not seen in a couple of years. This made the time in Barcelona much more interesting and entertaining.
Plenaries
This conference had a plenary talk each day, covering a variety of different topics. The topics included health care, the art of modelling, supply chain management in a chemical company and understanding the performance of algorithms. I found the first and last talks the most interesting. I was not aware of the talk topics prior to attending the plenary talk (as mentioned above I try to attend everything, so it was just a session that I didn't have to plan for), but it was a nice surprise to see a really interesting talk on heath care applications. Margaret L. Brandeau was a great speaker and communicated her story very well. Margaret spoke about the many successful projects she has been involved in, which was delivered in a very engaging manner.
On the final day of the conference the plenary talk was given by Kate Smith-Miles from Monash University, Australia. Being Australian myself, it would not be a surprise to hear that I have seen Kate speak before. The work on algorithm performance is very interesting. It really makes you think about how you investigate the performance of your own algorithms and whether this is really objective and meaningful. In addition, it could also be something to investigate regarding MIP solving. It is very clear that some solvers perform much better on particular problems compared to others, but it is not always clear why.
Interesting ideas and future work
While completing my PhD I really looked at conferences as a way to network and an avenue for ideas to improve my current research topic. I felt that I needed to work towards completing the PhD, so my focus was purely on the topic at hand. This time around, now having completed the PhD, I had very different experience attending IFORS. I took this conference as an opportunity to look at a variety of research areas and broaden my knowledge about my chosen career. Attending talks did not just make me think about the topic that was being presented to me, it gave me ideas about other topics that I had been thinking about previously or topics addressed in other sessions. It was great to have the stimulation to really think about problems that are of interest to me.
One of the outcomes of this conference is that I have become really motivated to work on health care applications. I was introduced to this area while researching the analysis of HIV envelope sequences as part of my post-doc at UNSW. My work on this topic is still continuing, but I have found that this area and the broader health care topic have many interesting applications. Further, from my studies I have developed a sound knowledge in decomposition techniques, in particular Benders' decomposition and column generation. There were many talks using both of these techniques, which really prompted me to think more about potential algorithmic improvements. There are so many different things for me to work on and it feels like I have been given the perfect opportunity to pursue these. Being motivated is a wonderful thing and it seems that IFORS has come at the perfect time to get me going once again.