Modeling the Organization and Flow of Information

Dynamic fleet management is a problem faced primarily by common carriers managing fleets of vehicles around a large network. Closely related problems arise when managing equipment (locomotives, aircraft) and crews which operate the equipment. Classical models in operations research focus on modeling the flows of physical resources. In freight transportation, it is also important to capture the organization and flow of information. In this talk, I will describe different classes of information, and the challenge of working with incorrect information, uncertain information and incomplete information. I will also use case applications from trucking and rail to illustrate these concepts and describe the modeling techniques that we have used to overcome them.

Warren Powell has worked for 20 years in the area of models for dynamic operational problems in freight transportation. His research has served as the basis for two consulting firms, and software based on his research is used through the less-than-truckload, truckload and rail industries. As the director of CASTLE Laboratory at Princeton University, he works directly with major companies to develop new ideas and participate in their actual implementation. He has twice been a finalist for the Franz Edelman competition, and has worked in a number of volunteer positions for Informs.