03/15/95

CONTACT: Stanford University News Service (650) 723-2558

Six Stanford faculty elected to National Academy of Engineering

STANFORD -- Six Stanford University professors have been elected members of the National Academy of Engineering.

The new members are Thomas M. Cover, Kwoh-Ting Li Professor of Electrical Engineering and professor of statistics; Thomas J. R. Hughes, professor and chairman of the Applied Mechanics Division of the Department of Mechanical Engineering; Elisabeth Pate-Cornell, professor of industrial engineering and engineering management; Charles Steele, professor of mechanical engineering and of aeronautics and astronautics; Stephen Wei-Tun Tsai, research professor in aeronautics and astronautics; and Bernard Widrow, professor of electrical engineering.

The six were among 77 engineers elected to the academy last month. Election to the academy is considered one of the highest professional distinctions that a U.S. engineer can receive. Academy membership honors those who have made "important contributions to engineering theory and practice, including significant contributions to the literature of engineering theory and practice," and those who have demonstrated "unusual accomplishment in the pioneering of new and developing fields of technology."

This year more new members were elected from Stanford than from any other institution, increasing the number of Stanford academy members to 73 out of a total U.S. membership of 1,790.

Pate-Cornell is the first woman faculty member from Stanford to be elected into the academy. She was one of a record number of five women elected into the organization this year, making her one of only 34 women members.

Pate-Cornell, an industrial engineer, studies how people affect the performance and physics of systems in environments ranging from offshore oil platforms to NASA's space shuttle to the hospital operating room. She has done so by extending the conventional engineering methods of probabilistic risk analysis to take management factors into account.

In her study of the space shuttle, for example, Pate-Cornell showed that the greatest risk of accident related to the shuttle's heat shield is due to a relatively small number of the thousands of ceramic tiles that protect it from the blazing heat of re-entry and she proposed specific improvements in maintenance procedures that could significantly reduce this source of danger.

Her most recent studies involve patient risk management in anesthesia. Her quantitative studies allowed her to make a number of specific recommendations to improve the performance of anesthesiologists, such as limiting the number of hours that they have to work without a break and requiring periodic recertification.

Pate-Cornell received her doctorate from Stanford in 1978 and, after teaching for several years at the Massachusetts Institute of Technology, joined the department of industrial engineering and engineering management in 1981. She is currently president of the Society for Risk Analysis and has recently been involved in briefings of congressional staff on risk-related issues that are central to several bills pending in the House and Senate that require cost-benefit analyses of all new federal regulations.

Improving basic understanding about how interference and noise limit the flow of information through complex communication networks has been one of Thomas Cover's major contributions. A researcher in the fields of information theory, pattern recognition and complexity theory, he directs the Information Systems Laboratory on campus.

In some of his earliest work, Cover explored the information capacity of neural networks, circuits that behave similarly to the brain. This work was recognized in 1994 by the Institute of Electrical and Electronics Engineers (IEEE) Neural Networks Council Pioneer Award.

Cover's 1972 award-winning paper, "Broadcast Channels," initiated the application of information theory to communication networks. In this and subsequent papers he derived an expression for calculating the maximum amount of information that can flow through a given network and the maximum extent that information distributed over a network can be compressed. In 1992, he was selected to give the IEEE Shannon Lecture, the highest honor in the field of information theory.

In the area of pattern recognition, Cover and former student Peter Hart were able to establish a limit on the error rate of the oldest and simplest rule for guessing the nature of unknown objects. Called the nearest neighbor rule, it assumes that the unknown object is the same as the known object it most closely resembles.

Say you are attempting to guess the sex of an unknown individual based solely on height and weight. The nearest neighbor rule holds that the unknown person's sex should be the same as that of the known individual with a height and weight closest to that person. Cover and Hart showed that the error rate of this approach is never more than twice that of the most sophisticated rule based on complete knowledge of the relationship between the classification and appearance of all such objects.

Most recently Cover has been applying data compression techniques to portfolio theory, the theory of the management of stocks and securities. To compress text, it is sufficient to know the odds that a given character will come next. Similarly, in managing securities it is important to know the probability of the next day's price. Cover has developed a predictive procedure that achieves better market performance than the best daily "rebalanced portfolio" method, in which the investor buys and sells stocks daily to keep each stock at the same percentage of the overall portfolio.

Cover received his doctorate at Stanford in 1964 and has taught here since.

The ability to design and analyze complex mechanical devices by computer rather than by building physical models owes a great deal to the work of Thomas Hughes.

He is a leading figure in the development of the field of computational mechanics and, in particular, has played an important role in increasing the power of a computer simulation approach called finite element analysis. In finite element analysis, a complex system is broken down into a large number of elements and these elements, and their interconnections, are modeled by a computer.

Hughes did some of the basic development work involved in extending this approach from mechanical structures to fluid regimes, like air or water flows. As a result, it is now possible for engineers to build simulations of aircraft, automobiles and high-speed trains that include not only their mechanical and thermal behavior but also the behavior of the air flows that surround them.

With more than 300 papers published, Hughes has made a number of other contributions to the technology underlying a major new field called simulation-based design. In contrast to the traditional "build-and-bust" approach, simulation-based design involves the use of computer simulations early in the design process, where it can have the greatest impact. The goal is to incorporate the power of sophisticated engineering analysis into software that the designers can use to produce increasingly complex products during shorter and shorter product cycles.

Hughes is working with Dr. Christopher K. Zarins, professor and division chief of surgery at the Medical School, to apply these concepts to cardiovascular surgery. The basic idea is to use magnetic resonance imaging to get a picture of a patient's artery, convert this into a mathematical model of the artery's structure and blood flow patterns, and then use this model to determine the consequences of various surgical procedures ahead of time.

Hughes received his doctorate from the University of California-Berkeley in 1974 and, after teaching at Berkeley and the California Institute of Technology, came to Stanford in 1980.

Charles Steele applies mechanical engineering to fundamental problems in biology. He does this through an approach called perturbation studies - a form of analysis based on the way systems react when subjected to very small changes.

One of the systems that he has studied in this fashion is the inner ear, specifically the cochlea, which transforms sound waves into nerve impulses. Steele has developed a very successful model that explains why we can hear as well as we do. The model involves an electrochemical feedback system within the liquid-filled cochlea, which normally sharpens hearing but, when slightly out of whack causes tinnitus, or ringing of the ears.

Steele also is working with scientists at NASA-Ames Research Center to develop non-invasive procedures to determine the mechanical properties of bone. By pressing a vibrator against an arm or leg and measuring the way the limb vibrates in response, the system can identify bones weakened by osteoporosis or other similar diseases without the use of needles or radiation.

Steele also is collaborating on a study of the growth process with biological sciences Professor Paul Green. They have determined that a significant part of the growth process - the formation of spiral patterns in sunflowers, pineapples and pine cones - can be explained by a mechanical instability on the plant's surface.

Steele received his doctorate from Stanford in 1960 and joined the faculty in 1966.

Stephen Tsai's research interests lie in the design and prototyping of composite materials and structures, which are increasingly being used in products such as tennis rackets, golf clubs and aircraft.

Tsai helped develop composites for aircraft applications during the 21 years that he worked at the Air Force Materials Laboratory in Dayton, Ohio. Since coming to Stanford five years ago, his efforts have shifted to the use of composites for more general commercial and industrial applications. For example, he has designed and prototyped composite components for electric generators and gas turbine engines.

He also is playing an active role in the design of the America's yacht for the all-woman team competing for the America3 Cup.

Currently, Tsai is researching the possibility of replacing steel reinforcement in concrete structures with composite materials. In this fashion he hopes to develop concrete structures that are stronger and longer-lived than those available today.

Tsai received his bachelor's and doctor of engineering degrees from Yale.

Computer users have benefited directly from the research of Bernard Widrow. He is a pioneer in the field of adaptive filters and neural networks, electronic systems that have the ability to learn and improve their behavior through contact with their environment.

From 1959 to 1965, he developed an artificial neuron out of transistors that could be trained to recognize patterns, and an artificial synapse - the electrochemical junction between neurons - based on electroplating that allowed these neurons to be joined together into networks.

Widrow also developed the adaptive filter and showed that it could be used to remove from electrocardiograms the noise created by 60 hertz power lines and to get accurate electrocardiograms of babies in the womb.

Scientists at Bell Laboratories later showed that these filters could be used in telecommunications. As a result, the biggest applications for these circuits is in the modems that allow computers to communicate over telephone lines. Without adaptive filtering, these modems could only operate at a few hundred bits per second, rather than the 14,000 bits per second rate that is the current standard. In 1986, Widrow received the Institute of Electrical and Electronics Engineers' Alexander Graham Bell Medal for his contributions in this area.

He continues to work with researchers at the Medical School to solve biomedical problems in areas such as sleep research and anesthesia. For example, one current joint project is the development of a breathing machine to help people who suffer from sleep apnea.

Widrow received his bachelor's, master's and doctor of science degrees from the Massachusetts Institute of Technology and served on the MIT faculty before coming to Stanford in 1959.

-dfs-

950315Arc5323.html


This is an archived release.

This release is not available in any other form. Images mentioned in this release are not available online.
Stanford News Service has an extensive library of images, some of which may be available to you online. Direct your request by EMail to newslibrary@stanford.edu.