Hub-and-spoke network design under congestion: A learning based metaheuristic

This paper models a single allocation multi-commodity hub-and-spoke network problem through a bi-objective mathematical model, considering the congestion in both hubs and connection links. A novel aggregation model is developed based on a general GI/G/c queuing system to evaluate the congestion of the flow of the multiple products in the hubs. The proposed model is then solved using a novel learning-based metaheuristic based on NSGA-II, k-Means clustering method, and an Iterated Local Search (ILS) algorithm. The proposed model and solution algorithm are validated through a set of experiments against optimal solutions, and benchmarked against four existing well-known algorithms.