D. R. Fulkerson

Ray Fulkerson grew up during the Depression in a small town in southern Illinois. His undergraduate studies at Southern Illinois University were interrupted by military service during the Second World War. After the war he returned to complete his studies at SIU and went on to graduate study in mathematics at the University of Wisconsin. In 1951, upon completion of his Ph.D., Ray joined the mathematics department at the Rand Corporation. There, he began an illustrious career of research and scholarship. Ray left Rand in 1971 and came to Cornell as the Maxwell Upson Professor of Engineering. He remained at Cornell until his death in 1976.

Ray’s first three published papers already established his importance in the growing field of Operations Research. The first paper, with George Dantzig, solved a problem of scheduling cargo ships, and is still commonly cited in the undergraduate and graduate courses as an example of the applicability of network models. His second paper, with George Dantzig and Selmer Johnson, was a computational tour de force, given the primitive state of computing in the early 1950s. This paper introduced cutting planes and the seeds of branch-and-bound, and used them to solve to optimality a 49-city traveling salesman problem. The entire area of polyhedral combinatorics eventually emerged from this paper, and it set the stage for successful computational approaches to large-scale optimization problems in which the solution values are constrained to be whole numbers (integer programming). The third paper, with L. R. Ford, Jr. was the first in a long collaboration in which Ford and Fulkerson laid the foundations of network flow theory. A later paper with Ford first introduced the technique of column generation, and inspired the development of large-scale linear programming. Throughout his career Ray continued to make fundamental contributions in these areas, as well as in matroid theory, graph theory, and combinatorial mathematics.