Research On Collaborative Optimization Of Rolling Stock Operation And Maintenance Of Urban Rail Transit Under Whole Day Operation

With the normalized operation of sunset-departure and sunrise-arrival trains between large cities and the “red-eye flights” at large airports on holidays,as well as the improvement of urban economic openness and activity,the nighttime activities of urban residents tend to be diversified.In order to satisfy the nighttime travel demand of urban residents,several urban rail transit systems in China are studying the feasibility of implementing the normalized whole day operation mode for individual lines.This thesis studies the collaborative optimization of rolling stock operation and maintenance of urban rail transit under whole day operation.Firstly,according to the organization methods of rolling stock operation and maintenance under different whole day operation modes,this thesis quantitatively analyzed the night traffic volume of each mode and the changes of the number of allocated rolling stocks and rolling stock maintenance demand under overnight operation.Then,thesis proposed the calculation method of the number of rolling stocks and maintenance requirements,and allocated relevant resources based on the calculation results.Secondly,based on the network characteristics of the research problem,applying operational research and network optimization theory,the problem was abstracted into a network flow problem.With minimizing the total operation cost as objective,considering rolling stocks and maintenance capacity limitations of the depot,duration constraints,and rolling stock maintenance regulations,a network flow linear model for collaborative optimization of rolling stock operation and maintenance was established.Finally,reasonable neighborhood search operators were designed for transportation tasks and maintenance tasks respectively,and then large neighborhood search algorithm was utilized to solve the problem.At the same time,the variable neighborhood search algorithm and genetic algorithm were designed to contrast solution effects.In the example experiment with Xiong’an Line R1 as the background,by comparing the solution results of CPLEX and various heuristic algorithms,the effectiveness of the proposed model and algorithm were testified.Besides,the designed large neighborhood search algorithm acquire higher quality solution results than other algorithms,especially when the depot maintenance capacity is in short supply,the algorithm optimization effect is more significant.In addition,the experimental results indicated that the continuous operation of overnight rolling stocks day and night and the allocation of maintenance resources according to the demands of whole day operation contribute to promote the comprehensive benefits of operation.By means of mathematical optimization,this thesis effectively solved the collaborative optimization of rolling stock operation and maintenance of urban rail transit under whole day operation,and the application effect of the integrated scheme of rolling stock operation and maintenance was ideal,and proposed specific suggestions on rolling stock operation mode and maintenance resource allocation under whole day operation,so as to supply reference for practical operation policy.