CCFSM 2005

Third M.I.T. Conference on Computational Fluid and Solid Mechanics

M.I.T., Cambridge, Massachusetts. 14th to the 17th of June 2005.

By Niall McMahon

Conference Overview

The stated mission of the conference was "to bring together industry and academia and to nurture the next generation in computational mechanics".

There were 676 pre-registered attendees with approximately 650 scheduled speakers.

The conference took place over four days, from Tuesday June 14th through until Friday June 17th. Each day started with two plenary lectures from 09:00 until 10:30 followed by three consecutive 1.5 hour lecture blocks. Each block of lectures consisted of between 23 and 26 parallel sessions. Each session consisted of at least three 25 minute presentations. The sessions mostly took place in the main MIT buildings, with two sessions located in Frank Gehry's Stata Center, opened in March 2004.

Free registration for students and the impressive break-time snack tables were two of the positive effects of the conference's large number of corporate sponsors. These included GM, Ford, Silicon Graphics, Hewlett-Packard, Michelin, Hyundai and Boeing among many others.

Conference Presentations

Main Themes

The stated objective of the conference was to "address the three broad areas of Computational Fluid Dynamics, Computational Mechanics of Solids and Structures and Computational Multi-Physics Dynamics including Fluid Flows with Structural Interactions".

The plenary session themes inlcuded safety in engineering design, modern simulation methodologies as part of an integrated multi-disciplinary design process, Lattice-Boltzmann methods, Lagrangian methods, the mechanics of shells (such as chimney structures), uncertainty in design and open problems in elasticity. These themes were also reflected in the general sessions. The parallel fluid-dynamics sessions can be grouped, slightly arbitrarily, into the following broad categories, listed in order of size (that is, the number of sessions under each heading) :

Interesting Talks

There were several very interesting presentations. Two of the plenary presentations were remarkable for being both entertaining and informative. These were Computational mechanics and natural draft cooling towers: From struggle for safety to designed life-duration by W.B. Krätzig of Krätzig & Partners and Applications of computational fluid mechanics at Sandia National Laboratories by T.C. Bickel and H.C. Morgan of Sandia National Laboratories in the United States. The second of these was most relevant to our work. Thomas Bickel gave a vivid overview of Sandia's transition from almost purely physical testing to today's hybrid of computer simulation and physical testing. Using examples, including Sandia's analysis of the 2003 Columbia tragedy, Bickel outlined the simulation methodology followed by Sandia, emphasising the importance of verification, validation and presenting results that are credible.

Other talks of note included Computing Wake mode development of flow past low aspect ratio cylinders by Gregory Sheard of Monash University in Melbourne, Australia. This work was interesting in that it deals with an aspect of our own problem, a small cylinder in crossflow. The work looks very competent and useful (in fact, Dr. Sheard won a prize for the best engineering PhD thesis of the year at Monash). From a bioengineering perspective Analysis of the pulse wave velocity of human artery and its application to the blood pressure meter by Shigehiko Kaneko of the University of Tokyo was also interesting. This was partly due to Prof. Kaneko's lively presentation but also because of the obviously high quality research. Application of vascular CFD for clinical evaluation of cerebral aneurysms, presented by M.A. Castro of George Mason University in Virginia was also interesting for its completeness, involving clinicians and making use of an SGI Altix supercomputer for the simulations (which contain about three million three dimensional elements). Finally, and most relevant to our work, was the localized drug delivery session which featured Modeling and numerical simulation of drug diffusion through stratum corneum by M. Heisig, Computational fluid dynamics simulations for drug delivery applications by C.-H. Wang and, most interestingly, Strut position, blood flow, and drug deposition: implications for single and overlapping drug-eluting stents, presented by Brinda Balakrishnan, an MD/PhD student at the Harvard-MIT Division of Health Sciences and Technology. Balakrishnan et al. are looking at two-dimensional dissolution from a stent placed in a blood vessel - quite similar to our own work. In addition to the theoretical work, the team also have access to clinical data. Some of their work has been published in Circulation, the journal of the American Heart Association. Currently, the team are using a commercial code but hope to develop their own code in the future.

Conference Structure

Although impressive in scale and scope, the large number of parallel sessions left many good talks very undersubscribed. In addition, simply figuring out what to see and where to go at any given time, while entertaining to a point, took a significant amount of time and effort. MIT is a large complex and running between far-flung lecture halls all day was very tiring. The aim of the conference organisers was evidently to encourage and facilitate participation while providing a high quality experience for attendees. Although noble, the number of cancelled presentations along with the effort of navigating the sessions suggests that this plan was just a little too ambitious. A better compromise between maximum participation and maintaining a high quality experience for all may have been a smaller number of high-quality lecture sessions with a prominent, reviewed, poster display.

One of the unfortunate side-effects of the generous waiving of registration fees for students was that five of the talks that I intended seeing were cancelled : this represents close to a quarter the talks I actually attended and was due, presumably, to speakers pulling out without notification.

References and Links

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.