2:00 PM-2:45 PM
Room: Capitol North/Center
Chair: Margaret H. Wright, Bell Laboratories, Lucent Technologies
There have been many advances in the methodologies for solving large combinatorial optimization problems. This presentation will provide an overview of some of the strategies for modeling and solving such problems. Model formulation, preproceesing, heuristics, cutting-plane techniques, and column-generation approaches will be presented as well as the presentation of a general framework that will allow all such strategies to be combined into an overall optimization software package. The speaker will begin with a discussion of the importance of the model formulation and its implications for modern modeling languages. Then, she will present a brief survey of current heuristic techniques used to obtain good feasible solutions. She will also present a general framework for using heuristics coupled with cutting plane methods and column generation to prove optimality to large combinatorial optimization problems occuring widely in practice.
Karla L. Hoffman
Systems Engineering and Operations Research Department
George Mason University