In air transportation, container crossdocking refers to a loaded container that is transferred at an airport from an incoming flight to an outgoing flight without handling the freight on the container. It reduces handling time and handling cost relative to unloading the container and sorting the freight, and is an economical alternative if a sufficient amount of freight on the incoming flight continues on the outgoing flight. The planning of container crossdocking is very important in express shipment services, both because of its time advantages and its cost advantages. Unfortunately container crossdocking has to be planned before the amount of freight for each origin-destination pair for the day is known. This paper addresses an operational planning problem for an express shipment service network in which flight schedules are given, and container crossdocking decisions have to be made before freight flow demand information is revealed. A two-stage stochastic programming model is formulated in which crossdocking container movements are decided in the first-stage and package flows are determined in the second stage after demand becomes known. The proposed problem takes too long to solve with existing methods, and we address the computational challenges by proposing a two-phase branch-and-Benders-cut approach as well as additional cuts that use characteristics of solutions found so far to improve the progress of the algorithm. We present results from a computational study using real-world data that demonstrates the effectiveness of the proposed approach.
H. Milton Stewart School of Industrial and Systems Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0205, June 2022