Numerical Methods in Computational Mechanics:
Advances and Challenges

Purdue University, West Lafayette, Indiana
February 20-21, 2003

www.cri.purdue.edu

In cooperation with:
Purdue University
SIAM
IACM
USACM

• Introduction
• Organizing Committee
• Invited Lecturers
• Program
• Abstracts
• Restaurants
  • Abstracts
    • Genki Yagawa
    • Tayfun Tezduyar
    • Mete Sozen
    • Michael Griebel
    • Richard Kamm
    • Wing Kam Liu
    • James Bottum
    • Takashi Yabe
    • Mark Shephard
    • Shiyi Chen
    • Jayathi Murthy
  • Printed Material

Who:
Purdue Faculty, students, and Invited Guests.

What:
A two-day workshop on Numerical Methods in Computational Mechanics - Advances and Challenges, hosted by the Computing Research Institute of Purdue University.

When:
The workshop will begin with a continental breakfast at 8:00 a.m. on Thursday, February 20 and conclude the afternoon of Friday, February 21.

Where:
Stewart Center, Room 206, Purdue University, West Lafayette, Indiana.

Cost:
Conference registration is $50. There is no charge for students attending the conference; however, there is a $25 fee for the banquet dinner. Payment confirmations will be sent. Purdue is not responsible for costs due to cancellation. Registration fees cover continental breakfast, snacks, evening banquet meal (you will be on your own for lunch) and conference materials.

Hotel:
The Purdue Union Club Hotel is holding a block of rooms for the workshop. Call 800-320-6291 and ask for the "Computational Mechanics" conference block. Room rate is $78 -$94 per night. Reservations should be made no later than February 9, 2003.

• Top

Wing K. Liu, Northwestern University

Jayathi Murthy, Purdue University

Ahmed Sameh, Purdue University

Tayfun Tezduyar, Rice University

• Top

• Mr. James R. Bottum, Purdue University, Vice President for Information Technology
• Dr. Shiyi Chen, The Johns Hopkins University, Department of Mechanical Engineering and Mathematical Sciences
• Dr. Michael Griebel, University of Bonn, Germany, Institute for Scientific Computing and Numerical Simulation
• Dr. Roger D. Kamm, Massachusetts Institute of Technology, Biological Engineering Division
• Dr. Wing Kam-Liu, Northwestern University, Department of Mechanical Engineering
• Dr. Jayathi Murthy, Purdue University, Department of Mechanical Engineering
• Dr. Mark S. Shephard, Rensselaer Polytechnic Institute, Scientific Research Computation Research Center
• Dr. Mete Sozen, Purdue University, Department of Civil Engineering
• Dr. Tayfun E. Tezduyar, Rice University, Department of Mechanical Engineering
• Dr. Takashi Yabe, Tokyo Institute of Technology, Department of Mechanical Engineering and Science
• Dr. Genki Yagawa, University of Tokyo, Department of Quantum Engineering and Systems Science

• Top

Thursday, February 20, 2003

Friday, February 21, 2003

Safe Travels!

• Top

Adaptive Node-By-Node Finite Element Method for Fluid and Solid Analyses

Genki Yagawa
(Yagawa@q.t.u-tokyo.ac.jp)
University of Tokyo
Quantum Engineering and Systems Science
7-3-1 Hongo, Bunkyo-ku
Tokyo 113-8656
Japan

Abstract:

This paper discusses a node-based adaptive parallel scheme of fluid and solid mechanics using the free mesh method (FMM) or the node-by-node finite element method (NBN-FEM).

The technique is featured by a node-by-node algorithm with the temporal mesh generated in a local area around each node by searching neighboring nodes using multi-level buckets.

The present method has two advantages over the usual finite element method. First, the method is quite suitable for massively parallel computing platforms because the entire computational procedure from the mesh generation (pre-processing) to the solution of system of equations (main-processing) can be performed in parallel in terms of nodes. Second, the finite element mesh is not required a priori so that the analysts need only to prepare data for the nodes in analysis domain. Furthermore, even severe geometrical irregularity, such as cracks, can be handled robustly in a node-based parallel manner.

The method is implemented on distributed systems such as PC clusters and the massively parallel supercomputer Hitachi SR8000. Numerical experiments show that the method achieves high parallel performance both in the pre-processing and in the main-processing stages.

REFERENCES:

1. G. Yagawa and T. Yamada, Free Mesh Method: A New Meshless Finite Element Method, Computational Mechanics, 10, pp.3953-3968, 1996.

2. G. Yagawa and T. Furukawa, Recent Developments of Free Mesh Method, International Journal for Numerical Methods in Engineering, 47, pp.1419-1443, 2000.

3. G. Yagawa, Node-By-Node Parallel Finite Elements: A Virtually Meshless Method, Proceedings of the Fifth World Congress on Computational Mechanics (WCCM V), http://wccm.tuwien.ac.at/cgi-bin/wccm/showDB.pl?showContribution&id=80016, July 7-12,2002, Vienna, Austria, 2002.

4. T. Fujisawa, M. Inaba and G. Yagawa, Parallel Computing of High-Speed Compressible Flows with a Node-Based Finite Element Method, International Journal for Numerical Methods in Engineering, to appear.

• Top

Finite Element Computational Methods For Fluid-Structure Interactions and Two-Fluid Interfaces

Tayfun E. Tezduyar
(Tezduyar@mems.rice.edu)
Rice University
Mechanical Engineering, MS 321
6100 Main Street
Houston, Texas 77005-1892

Abstract:

We describe the finite element computational methods we have developed in recent years for simulation and modeling of fluid-structure interaction problems and two-fluid-interface flows. We focus on mesh update methods, solution techniques for nonlinear and linear equation systems, parallel computations, and enhanced discretization/solution techniques.

We introduced a number of methods for updating the mesh as the spatial domain occupied by the fluid changes its shape during the computations. The Solid-Extension Mesh Moving Technique (SEMMT) is one of those methods and was designed for fluid structure interaction problems. The Enhanced-Discretization Interface-Capturing Technique (EDICT) was originally introduced to increase accuracy in representing an interface in computation of free-surface and two-fluid-interface flows with an interface-capturing technique. In this presentation we provide a brief overview of EDICT. We also describe carious applications and extensions of the EDICT. One of these extensions is application to compressible flows with shocks, where the extension is based on re-defining the “interface” to mean the shock front. In addition, we describe the Enhanced-Discretization Space-Time Technique (EDSTT), Enhanced-Iteration Nonlinear Solution Technique (EINST), and Enhanced-Approximation Linear Solution Technique (EALST).

All of these constitute a hierarchy of enhanced discretizations, as well as the embedded solution techniques for nonlinear and linear equation systems.

• Top

Techne in the Infrastructure

Mete A. Sozen
(sozen@purdue.edu)
Purdue University
Civil Engineering
CIVL G-129
West Lafayette, IN 47907

Abstract:

Because they have had to work with imperfect media that defied closed-form solutions, civil engineers were the first to commit to the machine that accelerated arithmetic. But no matter how extensive and convoluted the software has become, civil engineers have not taken the machine much beyond the status of the slide rule.

Developments in sensing have already opened the door to the perception that, with data flowing in torrentially, not all of the critical decision can be made by the human operator for best results. Close interaction between the machine and the human is essential in maintenance of the infrastructure.

The challenge, however, is still human. Can computer scientists understand the pain of the civil engineers? Can civil engineers express their problems in a way to be understood by computer scientists? Unless that interaction takes place soon, techne will abandon the infrastructure.

The talk will provide examples where interaction between the arts of computer science and civil engineering is essential and possible.

• Top

Multiscale methods in nano- and biotechnology

Michael Griebel
(Griebel@iam.uni-bonn.de)
Institute for Scientific Computing and Numerical Simulation
University of Bonn
Wegelerstr. 6, Room 604
D-53115 Bonn Germany

Abstract:

The mathematical modeling of matter can take place on at least three different scales: On the quantum mechanical level via the Schrödinger equation, on an ab-initio level by means of HF or DFT and on the classical level using molecular dynamics methods. The arising numerical problems can efficiently be treated also by multi-scale methods.
We discuss these techniques and show our results from numerical simulations for applications from the field of nano-technology and biophysics.

• Top

Multi-scale modeling in computational biomechanics

Roger D. Kamm
(rdkamm@mit.edu)
Massachusetts Institute of Technology
Biological Engineering Division
77 Massachusetts Avenue, RM 3-260
Cambridge, MA 02139

Abstract:

As we develop more sophisticated techniques for the simulation of problems at a single length scale, the need increasingly arises to couple these with computations at either larger or smaller scales. The same can be said for events occurring over a range of time scales, which might differ by many orders of magnitude. Biological systems provide an excellent example. In the case of simulations of arterial blood flow at a particular site of disease, we need to understand how, more globally, the entire cardiovascular system responds to a variety of external factors, each of which impact local hemodynamics. One reason we seek to understand local flow characteristics is to better comprehend the fluid dynamic and solid stresses experienced by the arterial wall tissues and the resident cells, the deformations of which lead to a multitude of biological consequences, and contribute to the progression of disease. At the cellular level, these stresses produce deformations of the cytoskeletal matrix, the cell membrane, and the nuclear envelope. These, in turn, lead to conformational changes in individual proteins or ion channels that elicit the biological response. Encompassed in this one example are computational strategies that range from large-scale network models down to molecular dynamics.

• Top

Multi-Scale Analysis and its Applications to Nanomechanics and Materials

Wing Kam Liu
(w-liu@northwestern.edu)
Northwestern University
Mechanical Engineering
2145 Sheridan Road
Evanston, IL 60208-3111

Abstract:

Nanotechnology will undoubtedly have a broad impact on medicine, electronics, and materials in the next two decades. The rapid advances in nanotechnology, nanomaterials and nanomechanics offer huge potentials in National Defense and Homeland Security. They also make our manufacturing technologies and infrastructure more sustainable in terms of reduced energy use and environmental pollution. Advances in the synthesis of materials have stimulated ever-broader research activities in science and engineering devoted entirely to these materials and their applications. As a result, nanoscale materials may find use in a wide range of applications in materials reinforcement, field emission flat panel displays, chemical sensing, drug delivery, nanoelectronics and tailor-designed materials.

In most of the above applications, nanoscale materials will be used in conjunction with other components that are larger, and have different response times, thus operating at different time and length scales. In nanomaterials, grain sizes and inclusions (or nano-particles) can be as small as tens of nanometers, whereas in more typical materials they are of order of microns. They are heterogeneous materials systems that involve the concurrent coupling of dynamics at the atomic scale and the macroscale. Single scale methods such as ab initio methods or molecular dynamics will have difficulty in analyzing such hybrid structures due to these time and length scale limitations. To facilitate and enable research in this increasingly exciting, active and technologically profitable arenas of nano mechanics and materials research, we need to develop alternative approaches for this class of problems. We will describe bridging multiscale methods for modeling and simulation of these systems.

While significant challenges remain, a major goal of this presentation is to inform researchers of fascinating and important research related to applied technologies with a focus on multiple scale analysis and its applications to nanoscale mechanics and materials. Current research in engineering is just beginning to impact molecular scale mechanics and materials, and would benefit from interaction with basic sciences.

References:

Li, S., and Liu, W. K., "Meshfree and Particle Methods and Their Applications," Applied Mechanics Review, vol. 55, pages 1-34, 2002.

Wagner, G. J., and Liu, W. K., “Coupling of Atomic and Continuum Simulations using a Bridging Scale Decomposition,” Submitted for Publication.

Qian D., Wagner G.J., Liu W.K., Yu M.F., and Ruoff R.S. (2002), 'Mechanics of Carbon Nanotubes', Applied Mechanics Reviews, 55(6), 495-553.

• Top

Research at Purdue: The IT Partnership

James R. Bottum
(jb@purdue.edu)
Purdue University
Vice President for Information Technology
West Lafayette, IN 47907

Abstract:

James R. Bottum is Purdue's Vice President for Information Technology and CIO. He is the first person to hold this post, which was created in 2001. In his role as CIO he is responsible for planning and directing the central computing and telecommunications systems on the West Lafayette campus. He was previously the executive director of the National Center for Supercomputing Applications (NCSA) at the University of Illinois, Urbana-Champaign. Before his appointment at NCSA, Mr. Bottum was an associate director in the NSF's Office of Advanced Scientific Computing. He has served on numerous national panels and committees, including the Visitor's Committee for the National Center for Atmospheric Research's Scientific Computing Division and the ACM/IEEE Supercomputing Conference Executive Committee.

In his presentation, Mr. Bottum will discuss the interconnection between Purdue's Strategic Plan and that of Information Technology as the University moves to achieve and sustain preeminence in discovery. He will address the five fundamental principles of the IT plan, the computational resources essential to the work of the research community, and the collaboration that occurs between faculty researchers and Purdue's IT community.

• Top

Challenge of CIP as Universal Solver for Solid, Liquid, or Gas

Takashi Yabe
(yabe@mech.hitech.ac.jp)
Tokyo Institute of technology
Mechanical Engineering and Science
2-12-1 O-okayama, Meguro-ku
Tokyo 152-8552
Japan

Abstract:

In recent years, with improved computing environments, demand for accurate and stable numerical methods is rapidly increasing in various science and engineering fields. The Cubic Interpolated Pseudoparticle/Propagation (CIP) scheme which has been developed by Yabe et al. [1,2] for solving hyperbolic equations has been successfully applied to various complex fluid flow problems such as laser-induced evaporation , shock wave generation, elastic-plastic flow, bubble collapse, splashing, and milk-crown among others.

This method has been recently improved. Mass conservation is exactly guaranteed [3,4] even when the numerical solution is given by non-conservative or primitive Euler formulation which is important for obtaining stable solution in multi-phase flows.

The present review covers various fields of science, engineering and even entertainment CG, and shows the superiority of the CIP method in simultaneously solving problems involving solid, liquid and gas.

References:

[1] T.Yabe and T.Aoki, Comput. Phys. Commun., 66, pp.219-232 (1991).

[2] T.Yabe , F.Xiao and T.Utsumi , J.Comput.Phys. 169, pp.556-593 (2001).

[3] T.Yabe,@R.Tanaka, T.Nakamura and F.Xiao, Mon. Wea.Rev. 129, pp.332-344 (2001).

[4] T.Nakamura, R.Tanaka, T.Yabe and K.Takizawa, J.Comput.Phys. vol.173 , pp.171-207 (2002).

• Top

The Trellis Simulation System

Mark S. Shephard
(shephard@scorec.rpi.edu)
Scientific Computation Research Center
Rensselaer Polytechnic Institute
Troy, New York 12180-3590

Abstract:

Trellis is a framework for the automated adaptive solution of problems in mathematical physics over general 3-D domains including combined discrete and continuum multiscale analyses. Trellis is now divided into four modules:

A parallel algorithm oriented mesh database (PAOMD) that provides critical capabilities for accessing the mesh, doing mesh adaptation and handling parallel mesh representations. A key feature of the PAOMD is the ability to have applications indicates the specific mesh topological adjacencies to be maintained. High-performance computing requires the ability to distribute the mesh over the memories of distributed parallel computers with complete knowledge of the interactions between the meshes on the various computers. PAOMD supports this by maintaining mesh information on the partition boundary similar to that maintained for mesh entities on the boundary of the domain with the additional ability to alter those partitions and move mesh entities as dictated by dynamic load balancing used in adaptive simulations. Recent PAOMD extensions support (i) local mesh modifications, (ii) non-conforming meshes with parallel adaptation, (iii) curved high-order mesh entities, (iv) and dual meshes.

A function space library that provides generic approximation schemes capabilities for finite dimensional fields families contains high order continuous and dis-continuous functions that can be varied on a mesh entity basis over the domain. These functions can vary order from mesh entity to mesh entity.

A discretization library that provides differential and integral operators that act on the function spaces. Differential operators include standard operators such as Grad, Div, Curl, etc. In the application of weak forms contributions to the global system are constructed through the appropriate integration of discrete entities (elements, element faces, etc.) of appropriate differential operators acting on weighting and trial spaces that are in terms of the selected members of the function space library. To support the definition of these contributions Trellis supports the effective construction of linear, bilinear, trilinear and general operators. The discretization library also provides the capability to manage degrees of freedom. Contributions to error estimators, error indicators and correction indicators are also supported by the discretization library.

A Solver Driver that provides capabilities to solve multiphysics problems. The Solver Driver can interface solvers like PETSc, Sparskit, DASPK, etc.

This presentation with emphasize specific aspects of the parallel algorithm oriented mesh database and the discretization library. The application of Trellis to parallel adaptive analysis will be demonstrated on various applications.

• Top

Lattice Boltzmann Methods For Fluid Flows

Shiyi Chen
(syc@jhu.edu)
Johns Hopkins University
Mechanical Engineering
122 Latrobe Hall
3400 N. Charles Street
Baltimore, MD 21218

Abstract:

The lattice Boltzmann method has become an alternative computational scheme for solving partial differential equations and modeling various physical and engineering systems. In this talk, we will briefly introduce the basic principles of the lattice Boltzmann method, its mathematical background and numerical implementations. Comparisons of the lattice Boltzmann method with traditional numerical schemes, including finite difference schemes and pseudo-spectral methods, for solving the Navier-Stokes equations will be presented. The applications of the lattice Boltzmann method in multiphase flows, flow through porous media, MEMS and fluid-particle interaction will be discussed.

• Top

Modeling and Simulation of Sub-Micron Heat Transfer

Jayathi Murthy
(jmurthy@purdue.edu)
Purdue University
Mechanical Engineering
West Lafayette, IN 47907

Abstract:

In recent years, there has been increasing interest in heat transfer at sub-micron scales. Applications include thermal transport in microelectronic devices, ultra-fast laser machining, energy conversion and many others. In semiconductors and dielectrics, conduction heat transfer occurs primarily through phonon transport. The talk describes the basic physics governing phonon transport in solids as well as evolving models and numerical techniques for computing sub-micron heat transfer. Applications in microelectronics heat transfer are presented. Emerging challenges in simulating heat transfer at ultra-small scales are discussed.

• Top

CAMPUS DINING

(Walking distance)
Purdue Memorial Union

Villa Pizza
Your on campus Italian Restaurant offering you the highest of standards. Enjoy traditional Italian specialties prepared using "old world" recipes. Pizza, pasta, garlic breadsticks, salads and more. All made with fresh ingredients, offered in a warm and friendly atmosphere. Pizza made fresh with "homemade from scratch" dough topped with tangy tomato sauce and lots of mozzarella cheese! Choose from a large selection of toppings. Offered in 43 states and 8 countries.

Union Market Food Court

Lentil, Curry & Mo - Featuring upscale Vegetarian and Vegan offerings as well as traditional Indian dishes. Pacific Rim - An array of Asian favorites prepared with the freshest ingredients.

The Carvery - Featuring home-style favorites from roasted chicken to "meat and potatoes" offered with an abundance of sides.

The Grill - Juicy burgers & chicken prepared on an open grill. An assortment of fried sandwich offerings such as pork tenderloin and fish.

Sara Lee Sandwich Shop - Cold Deli sandwiches made especially for you! Panini-style hot sandwiches.

More than Mexico - Here you'll find tacos, quesadillas, and all the accompaniments.

Dogs-n-Spuds - Enjoy a bowl of hearty soup, baked potato or hot dog complemented by a generous toppings bar.

Salad Bar - Create your own salad from a wide variety of ingredients, including fresh fruit!

On the run? - Try our expanded line of quick grab-n-go options offering a wide variety of food offerings from sandwiches to sushi.

Pappy's, The Original Sweet Shop

Step back into the past with a visit to Pappy's, the Original Sweet Shop. Established in 1927, Pappy's has been a favorite spot for students since its beginning.
Enjoy a variety of quick menu items at Pappy's. You'll find favorites like burgers, pork chop sandwich & breaded tenderloin - each offered at an affordable price.
Pappy's soda fountain features ice cream creations; including shakes, malts, cones and sundaes each made with the original Purdue Creamery recipe ice cream. Pappy's ice cream is also packaged in pints for your convenience. Take some home!

Sagamore Restaurant

The Sagamore Restaurant offers table service as well as a daily luncheon buffet. It is the perfect campus location to meet with colleagues, friends or family to dine. For larger groups, private dining space is available throughout the year with a variety of service and menu options.
Friday and Saturday evenings feature a more upscale dining experience in a relaxed atmosphere, with dinner service each home football weekend and other special events.

The Oasis Café

Experience the Oasis Café, featuring a full line of richly brewed Starbucks® coffee, showcasing an abundance of freshly baked pastries, fresh bagels and breakfast sandwiches. For lunch, enjoy one of many hot or cold gourmet sandwiches created to fit your appetite, or a bowl of skillfully prepared hearty soup. In between classes relax with a Big Train® Chai or a soothing cup of tea by Republic of Tea®.

• Top

Chauncey Hill Mall Restaurants

(Top of State Street Hill)

Subway – Sandwiches, Chips, Soup
Einstein Bros. Bagels – Bagels, Sandwiches, Wraps
Fu-Lam - Chinese Cuisine
Bombay Express - Indian Cuisine
Khana Khazana - Indian Cuisine
A.J. Wingers - Buffalo Wild Wings, Burgers
Vienna Café – Croissants, light sandwiches
Pizza King - Pizzas
Fazoli's - Italian Cuisine
Cafe Mocha` - Middle Eastern Cuisine
Parthenon - Greek Cuisine
Taco Bell - Fast Food (Mexican)
Arby’s – Roast Beef Sandwiches
The Yacht Club – Old Pub Style, Sandwiches
The Smoothie King – Smoothies!
The Triple X – Home cookin, everything from burgers and fries to steaks. They’re “on the hill, but on the level”.

• Top

Purdue West Restaurants

(West end of State Street)

Pizza Hut
Subway Sandwiches
Veno’s Bar and Grille
McDonalds

Downtown Restaurants

(Down the hill and over the bridge)

Bistro 501
501 Main Street, 765-423-4501
Bon Appetito
625 Columbia Street, 765-420-7072
Chumley’s
122 North 3rd Street, 765-420-9372
Kokoro Japanese Cuisine
526 Main Street, 765-742-8180
Java Roaster
130 North 3rd Street, 765-742-2037
La Scala Italian Restaurant & Café
312 Main Street, 765-420-8171
Lafayette Brewing Company
622 Main Street, 765-742-2591
Lincoln Café
918 Main Street, 765-742-0619
Maize, An American Grill
112 North 3rd Street, 765-429-6125
McCord Candies
536 Main Street, 765-742-4441
Murky Waters Coffee Co.
219 Main Street, 765-429-4300
Sgt. Preston’s
6 North 2nd Street, 765-742-7378
Subway Sandwiches & Salads
323 Columbia Street, 765-423-4456
Sunrise Diner
501 Columbia Street, 765-742-4204

• Top

Brown Street Levee Restaurants

(Bottom of State Street hill)

Wendy’s
Roly Poly - sandwiches and wraps
Panera Bread - sandwiches, rolls
Long John Silvers
The Frozen Custard
China Buffet – Chinese food
Pepe’s – Mexican food

Numerical Methods in Computational Mechanics:
Advances and Challenges

February 20 - 21, 2003
Purdue University
Computing Research Institute
Sponsored in part by:
SIAM, IACM, and USACM

WHO:

• Purdue faculty, students, and invited guests Professional leaders and experts in the fields of Science and Computational Mechanics

WHAT:

• A two-day workshop on Numerical Methods in Computational Mechanics - Advances and Challenges, hosted by the Computing Research Institute of Purdue University

	WHEN: The workshop will begin with a continental breakfast at 8 a.m. on Thursday, February 20 and conclude the afternoon of Friday, February 21
	WHERE: Stewart Center, Room 206, Purdue University, West Lafayette, Indiana. Purdue University is located about 65 miles northwest of Indianapolis. Map on back.

COST:

• Conference registration is $50. There is no charge for students attending the conference; however, there is a $25 fee for the banquet dinner. Payment confirmations will be sent. Purdue is not responsible for costs due to cancellation.
• Registration fees cover continental breakfast, snacks, evening banquet meal (you will be on your own for lunch) and conference materials.
• Registration forms are available on the World Wide Web at www.cri.purdue.edu.

Completed registration forms' along with payment need to be received by February 9 and should be sent to:

Business Office, Conference Division
Purdue University
128 Memorial Mall, Stewart Center, Room 110
West Lafayette, IN 47907-2034

Hotel:

• The Purdue Union Club Hotel is holding a block of rooms for the workshop. Call 800-320-6291 and ask for the "Computational Mechanics" conference block. Room rate is $78 -$94 per night.

• Top

Our Featured Guests

• Dr. Shiyi Chen, The Johns Hopkins University, Department of Mechanical Engineering and Mathematical Sciences, Abstract: Lattice Boltzmann Methods for Fluid Flows
• Dr. Michael Grieble, Institut fur Angewandte Mathematik, Abteilung Wissenschaftliches Rechnen und Numerische Simulation, Abstract: Multi-scale methods in nano and biotechnology
• Dr. Roger D. Kamm, Massachusetts Institute of Technology, Biological Engineering Division, Abstract: Multi-scale modeling in computational biomechanics
• Dr. Wing Kam-Liu, Northwestern University, Department of Mechanical Engineering, Abstract: Multi-scale analysis and its applications to nanomechanics and materials
• Dr. Jayathi Murthy, Purdue University, Department of Mechanical Engineering, Abstract: Modeling and Simulation of Sub-Micron Heat Transfer
• Dr. Mark S. Shephard, Rensselaer Polytechnic Institute, Scientific Research Computation Research Center, Abstract: The Trellis Simulation System
• Dr. Mete Sozen, Purdue University, Department of Civil Engineering, Abstract: Techne In The Infrastructure
• Dr. Tayfun E. Tezduyar, Rice University, Department of Mechanical Engineering, Abstract: Finite Element Computational Methods for Fluid-Structure Interactions and Two-Fluid Interfaces
• Dr. Takashi Yabe, Tokyo Institute of Technology, Department of Mechanical Engineering and Science, Abstract: Challenges of CIP as a Universal Solver for Solid, Liquid, and Gas
• Dr. Genki Yagawa, University of Tokyo, Department of Quantum Engineering and Systems Science, Abstract: Adaptive Node-By-Node Finite Element Method for Fluid and Solid Analyses

• Top