Robert G. Bland takes eight cards from a deck, ace through 8, and makes a perfect shuffle -- taking the bottom four cards and interweaving them with the other four. By doing so three times, he returns the cards to their original order. "Students are always surprised at this. It provides a good example to introduce the notion of a 'directed graph,' something of considerable interest in many of the mathematical sciences. The directed graph model of shuffling reveals almost instantly why three perfect shuffles return the eight-card deck to its original configuration," said Bland, Cornell University professor of operations research and industrial engineering.

Bland, who has been teaching at Cornell since 1978, took over as director of the School of Operations Research and Industrial Engineering (ORIE) on July 1 for a five-year term, replacing John Muckstadt, who stepped down after eight years as the school's director in Cornell's College of Engineering. Bland has a rich Cornell tradition, having earned an undergraduate degree in industrial engineering (1969) and an M.S. (1972) and a doctorate (1974), both in operations research.

No magician, the Cornell engineering professor uses the device to help teach undergraduate students principles of mathematical modeling. "If you rotate a cube along a main diagonal," he says, holding a child's interlocking cube puzzle, "you get the same result. It's back to its original position in three turns."

The principle is the same one used in certain parallel architectures of microprocessors, he said.

Bland was a National Science Foundation Graduate Fellow from 1969 to 1972 and a Visiting Research Fellow at the Center for Operations Research and Econometrics in Belgium from 1975 to 1977. He was a Sloan Research Fellow from 1978 to 1982.

At Cornell, Bland was assistant professor in ORIE from 1978 to 1981 and associate professor with tenure until he became full professor in 1988. He served as the school's acting director in the spring of 1994 and associate director for Graduate Studies from 1988 to 1991. He is a member of the Society for Industrial and Applied Mathematics, The Mathematical Programming Society and the Institute for Operations Research and Management Science.

ORIE, which evolved from more traditional industrial engineering, is concerned with how to design and operate systems efficiently, allocating scarce, and therefore expensive, resources. Areas of study include probability and statistics, simulation, manufacturing systems engineering and mathematical programming.

Bland oversees a school with 18 faculty members, 75 to 100 B.S. graduates each year, 70 master's students and 35 doctoral students. Graduates typically head to careers in industry and, most recently, a good number go to work at Wall Street. Cornell's College of Engineering and Johnson Graduate School of Management offer a Financial Engineering Option for engineering master's degree students. Ph.D. recipients often take positions in academia.

Bland's research interests include linear programming, network flow theory and combinatorial optimization. He likes to explain linear programming with a beer-and-ale analogy: Suppose you had a brewery to make beer and ale, but some key ingredients, say, corn, hops and barley malt, were scarce. Ale takes more malt than beer, but the profit per barrel is greater for beer than for ale. Shouldn't the brewer just make beer to maximize profits? Not necessarily, because that may consume other scarce resources faster than making both. How to optimize production? Such a problem can be solved through the technique of linear programming.

Network flow concerns mathematical modeling of systems, such as pipeline networks, where a commodity must be moved through links of limited capacity. The design and efficient operation of such systems has important applications to transportation and communication networks and in manufacturing logistics.

Combinatorial optimization concerns application of techniques, from linear programming and other related methods to a variety of computationally difficult problems best illustrated by the traveling saleswoman problem: In what order should a saleswoman visit the cities on her itinerary to minimize distance traveled? Such models arise frequently, for example: In which order should the holes be drilled on a circuit board to minimize the time to complete the task? Efficient sequences are important since they would require fewer of the laser drills, which cost hundreds of thousands of dollars.

Bland's vision for the school is to maintain and enhance Cornell's programs, which are among the most highly regarded in the world. He intends to continue the trends toward greater interaction with practitioners outside of academia.

"We are very proud of the high esteem in which our school is held for both its research and its instructional programs at all levels," Bland said. "Our faculty have collectively won many national and international prizes recognizing research and teaching. We intend to maintain the strength of our activities, not by resting on those laurels but by continuing with initiatives that enhance existing strengths. Our degree programs are high-quality alternatives to traditional industrial engineering options.

"There are many programs out there. But we tend to be more broad-based and more demanding about the mathematical sophistication we require. We have programs that prepare our students to be better practitioners. As manufacturing and other application areas evolve, the graduate must be more agile to deal with new approaches. The student who understands more than the current cookbook approaches, who knows the theory of why they work or don't work, will be better prepared to deal with change."