Collaborative optimization of suburban railway and metro timetables based on oversaturation condition
The demand for commuter flow has become a major concern due to the rapid development of urban integration. The connection between suburban railway and metro provides convenience for commuters. However,a large number of commuters bring great pressure to the transfer stations where suburban railway and metro are connected. In order to meet the travel demand of both suburban railway and metro passenger flows,reduce waiting time for the commuters at transfer stations and alleviate the passenger flow pressure at transfer stations,it is necessary to conduct collaborative optimization of timetables for suburban railway and metro train. By introducing the concept of dynamic and oversaturated origin-destination (OD) passenger flow and considering constraints under oversaturation condition such as non-uniform headway,flexible stopping time,and passenger transfer conditions,a dual-objective nonlinear integer programming model was constructed in terms of the number of detained passengers and transfer waiting time. By taking the characteristics of the model into account,the model was transformed into a single-objective optimization model using the ε-constraint method,and an improved simulated annealing algorithm with dual thresholds and memory function was designed. To verify feasibility and effectiveness of the proposed method,the model and algorithm were applied to a case study on the Jinshan Railway and the Shanghai South-Zhongshan Park segment of Shanghai Metro Line 3. The results demonstrate that the timetables for suburban railway and metro trains after optimization can increase by 9.28% in total number of detained passengers,increase by 50% in collaborative train numbers,and reduce by 60.79% in total waiting time for passenger transfer. Continuous updating of the tolerance level ε can obtain a set of approximate Pareto solutions that take the level of passenger service and the transferring efficiency into account simultaneously. The improved simulated annealing algorithm has a better convergence effect and solution efficiency as compared to the traditional one. The results could provide theoretical guidance for the collaborative optimization of suburban railway and metro timetables and further improvement for transfer efficiency of commuting passenger flow.