Research On Optimization Method Of Railway Empty Freight Car Distribution

This paper studies the problem of empty freight car distribution.In the process of railway freight transportation,when the heavy freight car arrives at the destination and is unloaded,it becomes an empty car,the empty car generally has no destination,the core of the problem of empty car distribution is to designate the loading destination for the empty car.As an important part of railway freight organization,empty car distribution plays an important role in the optimal allocation and effective utilization of transportation resources.According to the railway bureau where the loading place belongs,the problem of empty car distribution can be divided into two parts: empty car distribution among regional railways and empty car distribution on regional railways.The railway department divides the entire railway network under the jurisdiction of each railway bureau.There is a lack of research on the problem of empty car distribution among railway bureaus,and there is a lack of research on the problem of empty car distribution with comprehensive consideration of freight pricing system.In view of the above problems,this paper constructs an optimization model for empty car distribution among regional railways and a random expected value model for the empty car distribution on regional railways.It broadens the research field of empty car distribution and is closer to the reality.The main research contents of this paper are as follows:1.Firstly,the problem of empty car distribution among regional railways is studied.According to the reality and demand,we mainly study the model and algorithm of this problem.On the study of model,we construct an optimization model that considers the trade-off between speed and cost.The model can dynamically select assembly stations according to the different empty car distribution and plan the reasonable assembly mode,which makes up for the lack of effective models for empty car distribution problem.In this paper,a genetic algorithm for this model is designed.Finally,numerical experiment shows that the method adopted in this paper can save more than 35% of the assembly cost.On the other hand,an empty car distribution model is established under the constraint of empty car can only be sent to one assembly station.A hierarchical agglomerative clustering algorithm for this model is proposed,we use this way to solve the model and compared it with the genetic algorithm,experimental results show that the clustering algorithm is slightly better than the genetic algorithm in optimizing the problem of empty car distribution among railways.2.Then,the problem of empty car distribution on regional railways is studied.For the problem of the empty car distribution,there has been a lack of comprehensive consideration of freight pricing system.Considering the randomness of the distribution process caused by the demand change in the actual distribution process,it has an impact on the railway revenue.In this paper,the random expected value model of empty car distribution in railway bureau with uncertain demand is established.The model comprehensively considers such factors as shipper’s reservation,distribution of railway bureau and stochastic demand.It can fully describe the freight revenue under different distribution conditions caused by random characteristic.