Scheduling On-Demand Air Transportation
Friday 18 September 2009, 14:00
LT3, Management School
Georgia Institute of Technology
Abstract: A non-scheduled airline provides regional “on demand” air transportation on small jet planes. A request for travel specifies an origin, destination, earliest departure time, and latest arrival time. Based on requests already accepted for that day, the accept/reject problem is to determine whether the new request can be accommodated. At the beginning of the day an optimal schedule that minimizes flying time is created for all of the accepted requests for that day.
This service is especially useful for areas that are not well served by large airports. By using small airports, they eliminate the hassles associated with long drives to the airport, packed parking lots, security lines, etc. For many travelers, this service will yield huge time-savings in comparison to the alternatives of a scheduled airline or driving.
In this talk, I will discuss the optimization models and algorithms that we have developed for scheduling DayJet’s service. These include a multi-commodity flow model with side constraints for solving small instances, which is imbedded into a local search algorithm with an asynchronous parallel implementation for solving real-life instances.
Biography: Professor Nemhauser is recognised internationally for his extensive work in Operational Research and Mathematical Programming. He is the author of over 100 papers in highly-rated journals and the author of several influential textbooks. He is past president of the Operations Research Society of America (now INFORMS), past chair of the Mathematical Programming Society, past editor of Operations Research and the founding editor of Operations Research Letters. He has received the INFORMS Kimball Medal and has twice been awarded the Lanchester Prize.