Research On Multi-period Hub Location And Routing Problem Considering Variable Capacity

In recent years,the emerging technologies,the number of packages and other aspects have put forward new challenges on the business operation of the logistics industry,and the hub capacity of logistics network will also seriously affect the service level and timeliness of network.Taking the future cyclical changes of customers’ demand in advance into consideration is of great significance to promote cost reduction and efficiency improvement and user experience upgrading in logistics industry.In view of this,the thesis considers two characteristics of hub capacity and periodic demand in the hub location and routing problem to get the best locations and transportation paths in response to changes in customer demand.The effectiveness of this thesis to reduce logistics operating cost is verified by numerical experiments.The main research contents and results include the following three parts.Firstly,consider incorporating the variable hub capacity and multi-period planning into network design.Based on the congestion risk caused by periodic demand to the logistics enterprise operation in the hub location and routing network,this thesis aims to maintain the normal operation through variable capacity and multi-period design in logistics network,and to flexibly adjust the congestion pressure by planning the facility locations and transportation routes to minimize the cost benefits for enterprise.Secondly,construct and transform 0-1 integer nonlinear programming model.On the basis of single-allocation hub location network,considering the transportation cost of logistics services provided by enterprises and the fixed construction cost of facilities,taking non-hub single allocation,closed-loop path planning and other operational constraints as constraints.With the goal of minimizing the total cost of the network,a single objective 0-1 integer nonlinear programming model is established in this thesis.Then a linear model is obtained by introducing mew variables.Thirdly,design,apply and analyse a hybrid genetic algorithm combining general genetic algorithm and variable neighborhood search algorithm.In order to verify the effectiveness of the proposed algorithm,the results of the hybrid algorithm are compared with CPLEX and general genetic algorithm respectively based on Turkish network.A comparative analysis is made from the perspectives of variable hub capacity and multi-cycle planning respectively to further verify the effectiveness of the model design.The results show that the combination of the variable capacity and multi-period planning can help optimize the operation cost of logistics enterprises in the case of demand changes.And then,sensitivity analysis is carried out on scale economy coefficient,number of hubs and other parameters,and management enlightenment is given to guide enterprise operation.To sum up,an optimization model of multi-cycle hub location and route planning network with variable capacity constructed in this paper not only enriches the logistics network design research on hub location and route planning problem,but also provides new ideas and decision-making references for logistics enterprises in practical business strategy and network planning.