Integrated Optimization Of High-speed Railway Train Rescheduling

In recent years,Chinese high-speed railway develops rapidly,where the mileage in operations increases and the train speed improves continuously.However,the uncertainty,unexpectedness and complexity of perturbations during train operations become more prominent.The organization of trains becomes more complex,and train delays propagate quickly,which influences more trains.At the same time,the passenger volume and demand for services of high quality are increasing.These facts bring more challenges on high-speed railway train rescheduling.When perturbations occur,it is necessary to make decisions on train rescheduling immediately with considerations of conditions of train operations and resource utilizations,to avoid the wider propagation of perturbation and the negative impact on passengers and railway companies.Train rescheduling mainly includes timetable rescheduling,train stop plan rescheduling and rolling stock circulation rescheduling.There are interactions and influences between these rescheduling plans,but the existing research on the integrated optimization of relative rescheduling problems is relatively rare.Therefore,this thesis focuses on the integrated optimization of train rescheduling problem,including main contents as follows:(1)This thesis analyses basic factors for the high-speed railway train rescheduling problem,including perturbations,differences between rescheduling plans and original plans,basic rescheduling measures,and main rescheduling objectives.Then basic models are formulated for high-speed railway train rescheduling problem,including railway network models,perturbation models,resources capacity constraint models and timetable rescheduling model.This thesis further analyses key points for the integrated optimization of train rescheduling,in terms of the relationship between timetable and train stop plan,and the relationship between timetable and rolling stock circulation plan.Above lays the theoretical foundation for the follow-up research.(2)Integrated optimization of timetable and train stop plan rescheduling.When a rolling stock breakdown leads to a canceled train service,a novel mixed-integer linear programming formulation is established with consideration of train retiming,reordering,rerouting and reservicing(additions of extra stops).The objective is to maximize the number of disrupted passengers reaching their destination stations and to minimize the weighted total train delay for all non-canceled trains at their destinations.Constraints dealing with the reassignment of the infrastructure and seat resources,mapping train routes in stations with train stops,and mapping the train arrival/departure time with train stops are formulated.A weighted-sum approach with the normalization is applied to obtain Pareto optimal solutions.A series of numerical experiments based on BeijingShanghai high-speed railway line is carried out to verify the effectiveness and efficiency of the proposed method.Compared with the sequential optimization approach,solutions of higher quality can be obtained by the integrated optimization approach.(3)Integrated optimization of timetable and train stop plan rescheduling considering the uncertainty of the disruption duration.In actual high-speed railway operations,the railway system is dynamic,and when disruptions happen,it is difficult to obtain the exact information of disruptions.The disruption duration is uncertain,which is considered to be subject to a known distribution and each known disruption duration has a probability.A stochastic mixed-integer linear programming is established with robust constraints to ensure the same stop plan under different durations,where the expected optimal solution can be obtained.Three approaches are proposed: one-stage optimization approach,twostage approach and stochastic optimization approach.A series of numerical experiments based on Beijing-Shanghai high-speed railway line is carried out to verify the effectiveness and efficiency of the proposed method and to analyze the performance of different solution approaches.(4)Integrated optimization of timetable and rolling stock circulation rescheduling.An infrastructure failure resulting in a section blockage is considered here.A mixed-novel integer linear programming is established with consideration of train retiming,reordering,reservicing(cancellations of trains)and changing rolling stock circulations.The utilization of rolling stocks is affected by the seat capacity,speed class,management unit and maintenance rule.Constraints dealing with the reassignment of the infrastructure and rolling stock resources,and mapping the time window between the arrival time at the destination station and the departure time from the origin station of connecting trains,with the turnaround time of the rolling stock to operate these two trains,also mapping train cancellations or not with the reassignment of rolling stocks are formulated.The objective is to minimize the cost for timetable and rolling stock circulation rescheduling.A custom-designed two-stage approach is applied.A series of numerical experiments based on Beijing-Shanghai high-speed railway line is carried out to verify the effectiveness and efficiency of the proposed method and to compare the performance of integrated and sequential optimization approaches.Compared with the sequential optimization approach,solutions of higher quality can be obtained by the integrated optimization approach.(5)Integrated optimization of timetable,train stop plan and rolling stock circulation rescheduling.A rolling stock breakdown leading to a canceled train service is considered here.A novel mixed-integer linear programming formulation is established with consideration of train retiming,reordering,reservicing(additions of extra stops and cancellations of trains)and changing rolling stock circulations.Constraints dealing with the reassignment of the infrastructure,rolling stock and seat resources,and mapping the timetable,train stop plan and rolling stock circulation plan with each other are formulated.The objective is to maximize benefits of the integrated train rescheduling.A series of numerical experiments based on Beijing-Shanghai high-speed railway line is carried out to verify the effectiveness and efficiency of the proposed method and to compare the performance of integrated and sequential optimization approaches.Compared with the sequential optimization approach,solutions of higher quality can be obtained by the integrated optimization approach.This thesis includes 38 figures and 33 tables,and refers to 134 literatures.