| With the rapidly development of the global economy,the global flow of goods has become increasingly frequent,which has promoted the rapid development of the multimodal transport industry.As a necessary carrier of goods circulation,the transport industry plays an important role in the national economy.On the one hand,as a new type of cross-regional freight transport organization,Multi-Modal Transportation fully integrates the transportation advantages of various modes of transport such as railway transport,highway transport,waterway transport,and air transport shortens the flow cycle of goods in different regions,and improves the efficiency of freight transport.It has increasingly become the first choice for enterprises and shippers.On the other hand,the decentralized decision-making and multiple intermediate links caused by multi-party participation also make the transportation organization process more complicated.Therefore,how to reasonably planmulti-modal transportation route and coordinate its organization process has important theoretical and practical significance.In this paper,considering the common needs of multi-modal transportation operators,carriers and shippers,and based on the requirements of China ’s environmental sustainable development strategy,a two-stage multi-modal transportation model is constructed,and the relevant algorithms are designed to solve it,so as to determine the combined optimization results of multi-modal transportation path and mode of transport,and provide theoretical support for the selection of freight transport organization process.The main research contents are as follows :(1)Firstly,the related concepts and time window theory in multi-modal transportation are explained,and then the transport process is divided into three stages : initial pick-up,trunk transport and terminal distribution.The operation flow and organization mode in these two transport processes are expounded respectively,and the relevant theoretical research on the transport path modeling and model solving algorithm for multi-modal transportation at home and abroad is discussed emphatically.(2)The multi-edge method is used to construct the virtual network of trunk transportation,and the departure timetable influence of fixed shifts of some transportation modes and the soft time window constraint of consignee are fully considered.The optimization model of trunk transportation path with penalty mechanism is established.The model takes the transportation cost,transportation time and carbon emissions in the transportation process as the objective functions.Based on the characteristics of the model,a multi-objective solving algorithm NSGA-II is designed to solve the constructed model.The algorithm adopts the genetic operations such as double-layer coding,non-dominated sorting,binary competition selection operator,and elite retention strategy.The correctness of the model is verified by an example,and the solution results are comprehensively analyzed.The results show that the change of departure schedule and arrival time window of fixed schedule will directly affect the choice of transportation path and mode of multimodal transport.(3)The terminal distribution is a vehicle routing problem with time windows.Firstly,the service time range value of the vehicle at the node customers is explained.Then,according to the geographical location information of different customers,the requirements of the receiving time window and the cargo demand of the node customers,a multi-objective transportation model based on VRPTW is constructed,and the properties of the feasible solution are analyzed.Based on this,the NSGA-II algorithm is designed to solve the model,and the random vehicle loading method is used to generate the initial solution.The integer coding,two-point crossover,uniform mutation and other operations are used to solve the model,and the elite retention strategy is used to update the population to avoid the algorithm falling into local optimal solution.The example results verify the influence of decision maker’ s subjective preference value on sequential distribution order,which reflects the correctness of the model and the superiority of the algorithm. |
