A Frequency-based Maritime Based Multi-container Assignment Problem Under Uncertainty

With the continuous development of society,import and export trade plays a pivotal role in the state-owned economy.Traditional container transportation mainly takes the form of "port to port" or "station to station".In order to reduce the difficulty of distribution,the shipping company always manually dismantles the transshipment and intermodal transport of containers between different ports into “direct” transshipment ports at the port of origin,and then the container directly from the transit port to the “destination” port.Experience,adjust the distribution results.This type of distribution virtually increases the company’s unnecessary costs.During the transportation process,the phenomenon of transit is becoming more and more common.Therefore,it is of great significance to consider how to take transit into account and allocate the container traffic in the network to maximize the benefits of the shipping company.This article mainly studies how to allocate the container in the network under the given frequency of ship launching,which can make the overall efficiency of the network reach the maximum.To study this issue,this article mainly includes the following three main aspects:First,build a mathematical model for multi-container allocation based on the frequency of ship frequency.This paper uses the minimum of the container’s sailing time and the minimum residence time in Hong Kong as the objective function,taking into account the port flow balance restriction conditions and the path flow restrictions.In addition to these conditions,in addition to the characteristics of the port,the conditions for the port’s loading and unloading capabilities and the condition of the multi-container priority store are added.In this article,multi-containers mainly refer to dry containers and reefer containers.The reefer has higher priority due to the fact that there must be live sockets on the ship’s tank and higher transportation costs.In the model,this article uses the size of the ship to limit the priority of multiple containers,so that the model can more realistically reflect the reality.Second,linearize the MIP problem using continuous variable and(0-1)variable linearization.After modeling is completed,in order to solve the model,this paper introduces a new(0-1)variable to determine whether the path is attractive.If the calculation results are expressed,the flow can be distributed on the path to benefit,then the 0-1 variable is 1.The path is attractive.If the result of the calculation shows that the distribution of traffic on this path is not profitable,then the 0-1 variable takes 0 and the path is not attractive.In this case,the model is a nonlinear equation.If it is solved directly,it takes a lot of time to design the program and algorithm.In this paper,a new linearization method is used to transform the original model into a linear equation,which can be solved by the conventional linear programming solution method,which greatly reduces the difficulty and complexity of the model.This is the main point of this paper.Finally,the distribution is verified in the virtual network.algorithm.Third,considering the uncertainty of the demand,propose a delayed container rejection strategy.In this paper,by considering the uncertainty of demand,the traffic allocation results will be dynamically linked in time.When the demand is too large,if the containers cannot be shipped out in time this week,they will be preloaded in the port yard and will be preferentially loaded next week.If the containers cannot be shipped out after one week,they will be refused the shipment.According to this condition,container delay rejection strategies that may appear when demand changes are discussed according to this condition.The delay numbers and rejection numbers of two adjacent containers with different priorities are analyzed.Re-established the multi-container distribution model based on ship launch frequency under the condition of uncertain demand,and jointly seek the optimal distribution plan between adjacent weeks,and solved by Monte-Carlo stochastic algorithm.The experimental results show that the container allocation model proposed in this paper can fully characterize the container distribution with different priorities.The linearization method can effectively reduce the difficulty of solving the model and can generate satisfactory container allocation plans in a shorter time,which can be effectively used in Container distribution in the study.