Optimization Research Of Transfer Scheme For Railway Passengers

With the development of economy and the increase of peoples’ income,the travelling demand of railway passengers is continuously increased.They also put forward higher requirements on the service level,expecting to obtain safer,faster,more convenient,more comfortable and more economical transportation services.During the process of passenger travel,it is impossible to travel by direct trains each time and transfer behavior is inevitable.In order to provide passengers with a relatively reasonable transfer scheme,this paper combined previous studies with relevant optimization theories of transportation organization,which studied the travelling behavior of passengers,constructed corresponding transfer network,built related mathematical models and discussed the optimization of railway passenger transfer scheme.1.Combined with the current railway situation,the background and significance of the research for this paper were illustrated.Furthermore,the existent research achievements at home and abroad were discussed from three directions: passenger travelling behavior,travelling network,model and algorithm.By making a summary of the existing literatures,the weaknesses of existent research results were expounded and the main research content and technical route of this dissertation were given.2.Analyze the existing railway passenger transportation network from three aspects of passenger station,railway line and train type.Then introduce the characteristics of railway passenger transportation.By classifying different types of passengers,the affecting factors of passenger travel were classified,and the weight of each influencing factor was calculated by entropy method according to the questionnaire data,which laid a theoretical foundation for the weight assignment of subsequent models.3.Match all nodes and arcs in the transfer network with passenger railway network and entire travelling process of passengers.In the process of network construction,the dynamic processing method of nonbasic sites was proposed in order to reduce the network size and improve the efficiency of subsequent solution.In addition,the information contained in the transfer network was transformed into corresponding adjacency matrix,which laid a foundation for the solution of subsequent models and the implementation of algorithms.4.With the goal of making the cost,travel time and fatigue recovery time minimum,a multiobjective programming model was constructed.The constraint conditions were time continuity,seat remaining rate and so on.In order to better evaluate the transfer loss,the concept of transfer fatigue factor was introduced to describe the transfer process.What is more,transfer models for passengers who have special needs were constructed according to different passengers’ requirements.Finally,ant colony algorithm was used to solve the problem.5.Taking the actual passenger transportation network and train schedule as examples,thepassenger transfer scheme between different OD pairs were calculated to verify the rationality of the models and the feasibility of the algorithm.6.Summarize the main research content and results of this paper,then point out the problems that need to be solved in the future.