Associate Professor Dimitris Bertsimas (Operations Research) has been named to the first E. Pennell Brooks '17MG Professorship in Management at the Sloan School.
The Brooks chair was established this fall to honor the first dean of what was then called the MIT School of Industrial Management. Penn Brooks served as dean from 1952 to 1959. He died in 1991 at the age of 96.
Dr. Bertsimas received the Brooks chair at a celebratory luncheon held at the MIT Faculty Club on October 8, a few days after his 30th birthday. All four Sloan deans who succeeded Brooks-Howard Johnson (1958-1966), William F. Pounds (1966-1980), Abraham Siegel (1980-1987), and Lester C. Thurow (1987-)-were there, as were nearly 30 friends and family of the late dean, including his daughter and son and three grandsons; Eli Shapiro, Alfred P. Sloan Professor of Management Emeritus, who served as associate dean during Brooks's deanship; and Patricia Carey, who had been Brooks's secretary during the years he served as dean.
Professor Bertsimas works in two general problem domains-queueing theory and combinatorial optimization. In queueing, his research applies to the design and operation of manufacturing systems, transportation and telecommunication networks, and customer service systems. He has developed efficient computational algorithms for analyzing the steady state and transient behavior of multi-server queueing systems. He has generalized Little's Law, the fundamental relationship between the length of a queue and the waiting time in a queue. Currently he is developing an analysis of queueing networks based on the idea of queueing laws, and algorithms to schedule and route jobs within a queueing network.
In the field of combinatorial optimization, Dr. Bertsimas has conducted pioneering work on the probabilistic analysis of combinatorial models and algorithms. Combinatorial optimization problems arise in many decision-making contexts including scheduling, manufacturing and logistics, and are notoriously difficult to solve. Consequently, the optimization community often relies upon approximate solution procedures, such as solving a continuous approximation to the discrete problem. He has been able to assess the degree to which this type of approximation scheme and other types of heuristics approximate the combinatorial optimization model. Current research focuses on applications to classroom scheduling in a university, and yield management and dynamic scheduling for airlines.
"Dimitris Bertsimas is truly representative of what we would like the Sloan School to be about," said Deputy Dean Stephen Graves at the announcement of Bertsimas's appointment. "His work is distinguished by its creativity and depth of analysis, and by the extroardinary range of topics to which it contributes." A 1985 graduate of the National Technical University of Athens, Dr. Bertsimas joined the Sloan faculty in 1988, the year he received the PhD degree in Operations Research and Applied Mathematics from MIT.
A version of this
article appeared in the
October 21, 1992
issue of MIT Tech Talk (Volume