Prior to his retirement from the faculty of the School of Engineering at Washington University in St. Louis, Dr. Szabo was teaching courses in solid mechanics and finite element analysis at the undergraduate and graduate levels. His research was concerned with the foundations of predictive computational science. Dr. Szabo published several papers and book chapters and two textbooks, one of which was translated to Chinese.
Dr. Szabo co-founded Engineering Software Research and Development Inc. (ESRD), a corporation engaged in the development of an advanced finite element analysis software for structural and mechanical design and analysis, called StressCheck. Currently Dr. Szabo supports projects that involve the application of the methodology of predictive computational science.
Let’s learn a little more about him with four quick questions.
Q: What do you see as the biggest obstacles in the world of Engineering Simulation?
In my opinion, the biggest obstacle is the widespread confusion of finite element modeling with numerical simulation. The difference between those is like the difference between astrology and astronomy.
In numerical simulation a mathematical model is defined independently from how the solution will be performed. This is important because the methods used for controlling errors associated with formulation (model form errors) are very different from the methods used for controlling the errors of approximation. In finite element modeling, on the other hand, a numerical problem is constructed by assembling elements from a finite element library. Depending on what elements were selected, and how they were assembled, the numerical solution may not correspond to the approximation of a well-defined problem. In other words, the meaning of a finite element solution may not be what the analyst assumes it to be. The computed data may appear to be plausible, yet can be completely wrong.
A consequence is that management tends to view finite element modeling with justifiable skepticism. The confusion of finite element modeling with numerical simulation hinders the implementation of proper numerical simulation procedures. I believe that management is not yet aware of the great economic value of properly implemented and managed numerical simulation technology.
Q: What are some of the ways ESRD helps its customers to overcome these and similar challenges?
ESRD provides tools and resources for numerical simulation. These include StressCheck Professional with a hierarchic modeling framework to assess the errors of idealization and hierarchic finite element discretizations to control the errors of approximation, as well as training, reference material and examples. ESRD also advocates and supports standardization whenever it is economically feasible and provides commercial off-the shelf (COTS) and custom-designed engineering simulation apps for the A&D industry. Several presentations at NAFEMS conferences and other venues have emphasized the importance of the practice of Simulation Governance for the development and safe and efficient deployment of FEA-based simulation apps.
Q: I understand that you recently published a textbook – can you give a brief overview of that and let us know how someone can get a copy?
This book is the updated second edition of my book, co-authored by
Ivo Babuška, that was published in 1991. It covers the foundations of the finite element method and an introduction to predictive computational science. An end-to-end example of a validation project is presented.
I recommend this book to students, engineers and analysts who seek Professional Simulation Engineer (PSE) certification It is very easy to get a copy. The publisher’s link is available here.
Q: OK, enough about business. What do you enjoy doing away from work in your free time? Any interesting hobbies?
When I was younger, I enjoyed flying sailplanes. Now I try to keep in shape with exercises and a weekly racquetball match. I enjoy the performing arts, classical literature, Hungarian poetry and works on the philosophy of science.
