Study On High-speed Train Timetable Optimization With Changeable Stop Schedule

With the economic development and the improvement of people’s living standards, the travel demand is becoming stronger and stronger and also the demand travelling by high-speed railway constantly increases. So high-speed rail transport capacity is becoming increasingly tense. To some extent new railway construction can expand the raiway capacity without being the only or the best way, and to improve transportation organization is the most important way to ease the ability. As the basis of the transportation organization, train timetable plays a vital role and therefore becomes the focus of experts and scholars. As the important component of both the operation scheme and the train timetable, stop schedule is always determined during the operation scheme determination stage and couldn’t be changed during the train timetable making and optimizing stage, thus the train timetable quality is restricted to a certain extent. This paper studies the optimization problem of the train timetable with changeable stop schedule.The close relationship between the stop schedule and the train timetable is analyzed in detail in this paper. Stop schedule and train timetable have strong consistency in the target and constraint conditions, and stop schedule as an important part of the train timetable, it has an important influence on quality of the same grade trains and overtaking quality of different grade trains. Thus, with accessibility between stations, stop service times of each station, stop times of each train and the basic constraints of the timetable as complete constraints, the timetable model with changeable stop schedule is established and the optimization objective is to minimize the time and space resources of the occupied timetable in a given number of train, running path and running sequences. According to analyze the characteristics of the model and the applicability of common algorithms, genetic algorithm is selected as the main algorithm to solve the model, and the ordinal optimization theory is introduced to improve the initial population, the selection operation and the termination rules. In order to a better representation of the changeable stop schedule, infeasible solution adjustment strategy is designed in the specific operation process of the algorithm, and the feasible solution is obtained. Train departure and arrival time estimation strategies of train overtaking and non-overtaking are designed to obtain the objective function.Finally, a case study of 50 trains operating on the high-speed railway from Beijing to Shanghai is carried out. The basic data processing is analyzed in detail, and the appropriate algorithm parameters are selected. The optimal target value is 8964.83(km*h) obtained by the improved genetic algorithm with 2990 iterations, which saves 35.32% of train timetable resource comparing to the actual stop schedule. Average station service accessibility between the first class nodes, the first and second nodes and the second nodes is not lower than that of the actual stop schedule. It can be seen that the changeable stop schedule can improve the quality of the train timetable better and meet the basic passenger flow communication between stations. Compared with the traditional genetic algorithm, the optimal solution of the improved genetic algorithm is better than that of the traditional genetic algorithm and the value is 5.36%. After 1200 iterations, the quality of each generation is better than that of the traditional genetic algorithm, and the value is 5% to 8%. The case analysis indicates that based on the changeable stop schedule can not only meet the basic communication of the passenger flow between stations but also produce a better optimization result of the train timetable. Meanwhile, the improved genetic algorithm with the introduction of ordinal optimization theory is more optimal and has stronger ability of searching the optimal result.