A genetic algorithm for the hub-and-spoke problem applied to containerized cargo transport

A genetic algorithm for the hub-and-spoke problem (GAHP) is proposed in this work. The GAHP configures a hub-and-spoke network with shuttle services for containerized cargo transport. For a fixed number of hubs, it determines the best network configuration of hub locations and spoke allocations that minimizes the total costs of the system. The GAHP has a simple individual structure with integer number representation, where spokes, their allocations, and hub locations are easily recognized. Due to the characteristics of the problem, which has fixed number of hubs, rearrangements should be performed after every process. The GAHP rearrangement process includes improvements of individual structures, resulting in an improved population. Before applying the GAHP to the container transport network problem, the algorithm is validated using the Civil Aeronautics Board data set, which is extensively used in the literature to benchmark heuristics of hub location problems. To illustrate an example of a hub-and-spoke network with shuttle services, a study case with 18 ports is analyzed.