Research On The Design Of Hub-and-Spoke Logistics Networks With Multiple Hub Lines

How to reduce the total logistics cost with improving the efficiency of our logistics system? This is a problem which always troubling our scientific researches and now turning to our daily life more quickly and effectively.With the rapid rising of Internet Plus and Electronic Commerce,‘Online Shopping' and ‘Fictitious Economy' has become part of everyday life.As a bridge and link,Logistics' efficiency and cost have never affected us so deeply.It has brought us the changes of our living habits and the development of world trade,at the same time,it also brought a huge challenge for the foundation of our logistics system,which means the logistics network.The government has begun to promote the macro construction strategy of logistics channel and the relevant enterprises also began a new round of network layout of the micro reconstruction.They both look forward to making the logistics network channel unobstructed from the top to the bottom.It is hoped to reduce the cost of logistics and improve the efficiency of our logistics system.Under this historical background,considering the current situation of the development of social logistics in our country,this paper studies the design of a special kind of hub-and-spoke logistics network which called ‘Hub-and-Spoke Logistics Networks with Multiple Hub Lines'.Due to the current stage of China's investment restrictions and the difficult for international logistics network to overcoming its geographical constraints,in our research,the network could not be formed as a full interconnected hub-level network.Therefore,the hubs to be located must form a set of interconnecting lines.Logistics hubs can be connected with highly efficient pathways,enabling economies of scale to be achieved on the transportation cost(or transport time)between hubs.A mathematical model is presented to solve the Hub Line Location Problem through Pure Integer Nonlinear Programming.The objective is to minimize the total weighted transport time between all pairs of nodes while considering the logistics operation time taken on each key connected component of the network.Besides,a budget constraint on the total set-up cost of the hub network is also taken into account which makes the model more in line with the current situation and demand of network design in china.In order to solve the above model,a Hybrid Elitist Genetic Algorithm(HEGA)is proposed.The proposed algorithm combines the basic idea of the traditional Genetic Algorithm and the elitist reservation strategy.At the same time,it makes a combination of the Metropolis acceptance criteria in Simulated Annealing Algorithm and the update rules in Particle Swarm Optimization for genetic manipulation.By these methods,A heuristic algorithm with good global search capability,effective convergence performance,accuracy and stability is obtained.Comparing the results of the same scale numerical experiments with the proposed HEGA and the exact algorithm by LINGO respectively,the stable convergence curve of the HEGA and the same results as the exact algorithm are obtained.It is proved that the proposed HEGA can solve the problem of small size with good precision and convergence,and show that the algorithm has the excellent solution speed.The small and medium scale and also the large-scale standard numerical experiments are all verified that though the proposed HEGA is not standard in the solving speed of the genetic algorithm,but all results are within acceptable limits and significantly better than the exact algorithm.At the same time,HEGA has better performance.in terms of convergence,accuracy and stability.This paper aims to put forward a kind of logistics network planning and design method that meets the needs of the times.