Research On The Model And Algorithm Of Railway Empty Car Allocation Problem Considering Railway Wagon Substitution In Dynamic Random Environment

The problem of empty vehicle allocation is an important part of China’s railway freight transportation production plan.According to the road network structure,the arrangement of existing passenger transportation plans and the distribution of freight vehicles,the development of a scientific and reasonable empty vehicle allocation plan can enable the utilization of transportation production resources.Optimize,reduce unnecessary line capacity occupation,and improve the timeliness of cargo transportation.Therefore,the problem of empty vehicle allocation is an urgent need for China’s railway freight transportation to formulate more optimized and reasonable solutions.There are many factors influencing the problem of empty car allocation,which makes this problem more complicated.Therefore,the problem of empty car adjustment has always been an important topic that relevant scholars attach great importance to.This article mainly conducts an in-depth study on the problem of railway empty car allocation in China by considering the dynamics and randomness of railway transportation.The dynamic nature of the transportation process is mainly reflected by constructing a space-time network to describe the train operation process.The randomness is mainly reflected by creating random variables to describe the uncertain influencing factors in the actual railway freight transportation process.Based on this,a random optimization model is established and designed Genetic algorithm,genetic-simulated annealing algorithm to solve the model.The research content of the paper is as follows:(1)Analysis and modeling of dynamic railway empty car allocation problem considering vehicle type substitution under certain conditionsThis paper discretizes time nodes by constructing a spatio-temporal network.This method can describe the dynamic transportation process of freight trains more clearly and intuitively.Based on China’s railway freight transportation situation and corresponding assumptions,the goal function is to build the largest operating profit of the railway enterprise,the smallest empty car travel cost,the smallest empty car storage cost,and the smallest vehicle replacement cost.Six types of system constraints,such as interval passing capacity,site transit capacity,stage dynamic demand and decision variable rounding,are model-constrained integer programming models.(2)Analysis and modeling of dynamic railway empty car allocation problem considering vehicle type substitution in random environmentBased on the analysis and modeling of the dynamic railway empty car allocation problem considering the substitution of vehicle types in a certain environment,the influence of the two random parameters of the station transfer capacity and path capacity in the road network on the choice of road flow path in the road network is considered.The method of randomizing the model parameters is used to establish a random chance-constrained programming model,and discuss the determination of equivalence classes of chance constraints under special circumstances,and solve the model.(3)Algorithm designIn order to find the approximate optimal solution of the empty vehicle allocation problem,this paper first designs a latent path search method based on the branch and bound algorithm.On this basis,the genetic algorithm is designed,and the simulated annealing operator is combined with the genetic algorithm to design the genetic-simulated annealing algorithm,and the algorithm characteristics and flow are explained.(4)Algorithm validity test and calculation example analysisCarrying out an example analysis on the cargo operation of China Railway Nanning Bureau Group Co.,Ltd.Based on the simplified road network structure data and cargo operation data of China Railway Nanning Bureau Group Co.,Ltd.,genetic algorithm is used to solve the dynamic empty vehicle allocation problem model in the determined environment;genetic algorithm and genetic-simulated annealing algorithm are used to solve the random environment.Model of dynamic empty vehicle allocation problem,and comparative analysis of the results of the two algorithms to verify the effectiveness of the designed genetic-simulated annealing algorithm.