Commentary: February 2007

By Tom Lee, Maplesoft

We have all heard the doom and gloom talk about North American industry falling behind China and India due to our society’s declining interest in math and the sciences and the East’s increasing level of focus. As we struggle to fill spots in our engineering colleges, other countries are increasing their academic capacity to train ever-larger armies of engineers and scientists. We’ve all observed the shifting of the manufacturing base from here to there, and many predict that if we don’t respond with significant measures, the design and technical creativity base will also begin making that shift.

One of the aces in the hole for North America has always been its ability to produce new techniques and technology to make us more effective as engineers. The theory is, of course, if we can’t win on numbers, we can win on creativity. The theory sounds great — probably because it’s the only theory we’ve got. Is this theory working? Is North American engineering in fact beginning to push the boundaries of our tool chains to launch the next phase of innovation?

Tom Lee, Maplesoft

Recently, I was on a tour of Japan, Korea, and Europe and I was both heartened and disheartened with what I saw. In October 2006, Cybernet Systems — the distributor of ANSYS, Maple, and MATLAB/Simulink  in Japan — held its annual TechnoForum conference in Tokyo. Akira Ohata, the Toyota executive who heads the software strategies for all of the auto giant’s design groups was the keynote speaker and he presented his thoughts on the emerging importance of mathematics in design. He confirmed that Toyota was concerned about the inherent constraints of numerical simulation. He envisioned a richer modeling framework that incorporated conventional simulation tools with new GUI-based modeling systems that captured the physical layout of systems and new mathematical tools. Calling it “physical modeling,” Ohata claimed that it was critical our systems do a better job at this.

Hirokazu Anai, a senior engineer at Fujitsu presented his company’s experiences in applying a technique called “Quantifier Elimination” (QE) to various design problems at the leading electronics company. QE is a particular application of Gröbner bases. His claim was that these new techniques can increase the performance and accuracy of many simulations. He too noted that many engineers are beginning to hit inherent limits of conventional numerical simulation systems and new,  more mathematical techniques are needed to get engineers to the next stage of innovation.

During the European leg of my trip I learned that an Austrian group called the Modelica Association enjoyed its most successful conference ever. Modelica, not yet a household name among engineers, is an emerging open standard for describing models of physical systems. Modelica models will capture and manage all of the necessary relational, physics, and mathematical information for complex systems. Because it’s more suited to handle the mathematical framework of model development,  Modelica will make it easier to develop high-fidelity models. The word from the conference is that Modelica is beginning to impact mainstream engineering. Among the attendees were DaimlerChrysler, ABB, Dassault, and countless other well-known companies beginning to explore this technology. Strangely though, there was a distinct lack of names from Detroit on the list.

The point here isn’t so much that Gröbner bases or Modelica are the answers to North America’s engineering challenges. Rather, they show a distinct increase in attention to new ways of thinking about design. The examples mentioned here both relate to advanced mathematical techniques. Ten years ago, you would be lucky to find an engineer anywhere in the world who could pronounce Gröbner, let alone understand its potential application in power train design. To see hundreds of engineers attentively taking notes on the topic at a design conference is astounding. The one disturbing part of this, however, is that North American engineering appears to be strangely absent from these meetings. In fact, a bleeding-edge, push-the-technology-to-its-limits,  take-no-prisoners approach to design innovation is sorely lacking.

If we are, once again, falling behind our “traditional”  competitors in Japan, and Europe, what are our chances in the new emerging engineering economy?

During my tour, I also visited Waseda University’s humanoid robotics lab in Tokyo where my guide introduced me to robots that could walk, climb stairs while carrying a person, laugh and cry, and even breathe like a human (although it really had no need to breathe). By contrast, on my last trip to Detroit I had a nice visit to the Henry Ford Museum and Greenfield Village. It’s nice to know that we’ve had an admirable history of innovation and that we’ve built the industrial platform that the world works on, yet our history is not likely to be the key to future competitiveness.

Dr. Tom Lee is vice president of market development and executive product director at Maplesoft. You can respond to this commentary through e-mail by clicking here. Please reference “Playing Catch-Up” in your message.