| Demand responsive transit is an innovative service mode of Internet plus urban public transportation,which provides public transport services for passengers who are in the same area,and have the same travel time and the same travel demand.The demand responsive transit and the conventional bus lines compete with each other and cooperate with each other.To achieve the coordination and cooperation between demand responsive transit and conventional bus lines is an effective way to enhance the overall competitiveness of the ground public transit system.Therefore,the collaborative optimization design of the demand responsive transit and conventional bus lines,and the dispatching optimization of multi-mode demand responsive transit,can contribute to the overall travel efficiency of the ground public transit system and enhance the comprehensive competitiveness of the ground public transit system.Based on the application of demand responsive transit and the concept of coordinated development of urban public transportation,this paper studies collaborative optimization problems of the demand responsive transit considering the influence of conventional bus lines.The system design of transregional demand responsive transit is carried out,and the location selection of stations and route planning method are studied.Considering the competitive relationship between demand responsive transit and conventional bus lines,this paper studies the coordinated operation of ground public transport with operation of transregional demand responsive transit.Considering the cooperative relationship between the demand responsive feeder transit and conventional bus lines or transregional demand responsive transit,the collaborative optimization method of operation plan of demand responsive feeder transit is proposed.The detailed research contents are as follows:(1)Research on method of location selection of stations and route planning of transregional demand responsive transit.This paper builds an Affinity Propagation(AP)clustering algorithm based on node demand density,and clarifies the technology line for location selection of stations of transregional demand responsive transit.Considering the road congestion,the maximum number of stations,the limit of vehicle capacity,the requirement of passengers' reservation time and other factors,this paper establishes a method based on Q-learning algorithm to decide the actual route and schedule of the demand responsive transit.The results show that compared with the classical AP clustering algorithm,the improved AP clustering algorithm reduces the average walking distance of passengers from 545.8 m to 500.9 m,with a decrease of 8.2%,and the operation scheme based on this route optimization method can meet the passenger requirements of origin-destination and time window,with an 87.5% response ratio.(2)Considering the competition relationship between the transregional demand responsive transit and conventional bus lines,a bi-level programming model for collaborative optimization of ground public transit system operation is established.The upper model takes the total travel time of passengers,the passenger volume of ground public transit system and the operating surplus of ground public transit enterprises as the objectives,and takes the departure frequency of conventional bus lines and the vehicles number of transregional demand responsive transit as the decision variables.The lower model is a multi-mode traffic network assignment model based on Logit-SUE,which takes the flow of each path as the decision variable when the network is balanced.The algorithm is designed,and several examples are given according to different commuter passenger volume and the original loading rate level of conventional bus.The results show that the total travel time of passengers can be reduced and the operating surplus of ground public transit companies can be increased through the collaborative optimization of ground public transit system operation.At the same time,due to the "one person,one seat" service provided by demand responsive transit,the collaborative optimization will reduce the passenger volume of ground public transit system within a reasonable range,and the loading rate level of conventional bus will increase slightly.The results show that the optimization effect is different under different commuter passenger volume and the loading rate level of conventional bus: when the two levels are lower,the optimization effect is significant,so it is suitable to use the collaborative optimization model of ground public transit proposed in this paper for optimization;When the two levels are high,the optimization effect is not significant,and the decision should be made after comprehensive analysis.(3)A collaborative optimization model of operation plan of demand responsive feeder transit is established.The model takes the system response ratio,per capita time cost and per capita operating cost as the objectives,and vehicle routing,timetable and vehicle scheduling as the decision variables.When determining the timetable of demand responsive feeder transit,the connection between the feeder system and its connecting transregional demand responsive transit should be fully considered.Feasibility and effectiveness of the model and algorithm are verified by several examples.Compared with the collaborative optimization model without considering vehicle reciprocating operation,the system response ratio of the proposed model is increased by 20.3%,and the total equivalent time of per capita time cost decreased by 3.2%,meanwhile the per capita operating cost decreased by 36.5%.The collaborative optimization model of operation plan proposed in this paper can not only reduce the number of vehicles and fixed operating costs,but also guarantee the service level of demand responsive feeder transit system.(4)This paper makes a quantitative analysis of the affecting factors of the competitiveness of demand responsive transit,and provides strategies and suggestions for the operation of multi-mode demand responsive transit.Average travel speed and ticket price of transregional demand responsive transit are the internal influencing factors of its competitiveness,and the operation characteristics of the conventional lines which has a competitive relationship with the transregional demand responsive transit are the external factors.Under the scenario with the operation of transregional demand responsive transit and demand responsive feeder transit,the influence factors of the two modes of demand responsive transit competitiveness are analyzed.The results show that the connection and cooperation of the two modes of demand responsive transit improves the overall competitiveness of the demand responsive transit system. |
PDF Full Text Request