Optimization Of Hybrid Hub-and-spoke Network For Less-than-truckload Transportation

The design of operational network affects the operational efficiency and costs. In order to improve the operational efficiency of transport networks engaged in nationwide less-than-truckload logistics services on a physical hub-and-spoke network, without considering new hub constuction, two mixed integer programming models (MIP) including uncapacitated and capacitated, with incremental quantity discount function, are established to optimize the hybrid hub-and-spoke operational network. The models are to minimize the total transportation costs and the variable freight transfer costs.The models are applied to an enterprise engaged in nationwide logistics services. The models are calculated by Lingo software and branch and bound method. The results show that the targeted cost of the uncapacitated hybrid hub-and-spoke network is less than the pure hub-and-spoke network. In addition, compared to the fully connected network operation, hybrid hub-and-spoke network operation has greater reduction in the target cost. Capacity restrictions have an effect on the optimization of hub-and-spoke network and lead to a slight increase in the target cost. The calculation time increases under the given scale. The models show good adaption for the changing hub number and location, the OD structure of the network, the capacity restrictions, and the type of service vehicle. The calculation time exponentially increases when the number of hubs increases with the same sample data, but the number of hubs will not increase indefinitely. Therefore both of the structure and calculation of the models have good adaptability to the operational network decision making for logistics enterprises.