| With the sustainable development of social economy and science technology,urbanization process are accelerating,and the city size is expanding continuously.Meanwhile,the inconvenience of public transit has exacerbated traffic jams and environmental pollution.Therefore,improving the efficiency and coverage of public transit network,and residents' travel comfort degree,thereby attracting residents travle by public transit,it has become the key to solve the problems.Currently,the research on evaluation and optimization of public transit network is very common.However,the optimization method on considering the residents' travel demand,and linking vehicle model and number allocation,has not been widely developed.Most of the designs of public transit network are from managers view,not considering the passengers' demands.Then the role of vehicle model in the overall optimization of public transit network is ignored.In view of this,combining the characteristics of passengers' demand,it is designed about an evaluation and optimization system of public transit network service quality based on passenger perceptions,and put forward a scientific and effective vehicle allocation method accordingly.In view of the characteristics of urban residents' public transit travel,the paper studies the route design of public transit network and vehicle allocation method from the perspective of passenger perceptions,and establishes a mathematical model of related problems.According to the characteristics of the mathematical model,it is developed about the corresponding solution algorithm with graph theory and intelligent optimization method.By the real data processed by simulation technology,the validity and practicability of the method are verified.The research work of this paper mainly includes four parts:1.put forward a comprehensive evaluation method of public transit network based on passenger perceptions,which provide an optimization direction for the follow-up study;2.introduce the transit demand forecasting method based on TransCAD,and carry out the simulation,which provides the data support for the subsequent network optimization;3.Establish the optimization model of whole public transit network based on whole transit network,and study on bus allocation;4.Considering the actual scene of bus demand added,the optimization on adding bus routes and transit network is establishing and study on the bus allocation.The specific study contents are as follows:(1)Propose an evaluation method of bus service quality considering passenger perceptions.By subdividing the satisfaction degree of passenger perceptions,the performance indicators such as safety,comfort,convenience and reliability are established.Aiming at the limitations of traditional evaluation methods,a weight analysis method of universal factors is proposed based on passenger perceptions.On this basis,the BP neural network is trained,tested and validated with multiple sets of data as priori samples,then the AHP-BP neural network model for comprehensive evaluation of public transit network can be obtained.Finally,taking the network in a region of Shenyang as an example,by fitting and analyzing the data of passenger sensitive factors,the numerical result indicates the model not only fully reflects the evaluation indicators of passengers,but also reduces the randomness of subjective evaluation,and indeed give an effective direction for improvement.(2)Put forward the transit demand forecasting model based on TransCAD.The research object mainly includes two parts:?Taking the current bus data in the target region as the research object,the simulation platform is built by TransCAD,and the database can be established;?Taking the future bus data of the target region as the research object,considering the characteristics of residents' transit demands in the base year and planning year,the transit demand forecasting model can be constructed based on the "four-stage method" as the core.On this basis,the simulation data of the base year and the planning year of the target region can be obtained,which lays a solid data foundation for the following chapters.(3)Put forward the Transit Network Design and Frequencies Setting Problem(TNDFSP)model considering passenger perceptions.Taking the sensitive factors of passenger perceptions in reality as the evaluation criterion,this paper studies transit network design,frequency and vehicle allcation.According to the data of passenger sensitive factors,the evaluation indicators is set up,and a bi-level planning model of public transit network based on passenger perceptions is established.On this basis,the genetic algorithm is used to solve the transit network design and frequencies setting simultaneously.In order to verify the validity of TNDFSP model with vehicle allocation,this paper introduces the classical Mandl model and and other literatures.The results show that the method proposed in this paper is more effective in improving the direct rate and reducing congestion on the premise of guaranteeing the interests of operating companies.Finally,by comparing the examples before and after the optimization of the target region in Shenyang,it is further confirmed that the optimization results of this method have important practical significance in the actual transit network planning.(4)Put forward the TNDFSP model under the scenario of increasing public transit demand.Taking the rapid expansion of the city as the background and the bus blank region as the research object,a bi-level planning model of the transit network based on passenger perceptions is established from the angle of adding routes in order to cover new stations.Furthermore,a solving method based genetic algorithm is proposed that can cover all the possible paths in network design.The model preserves the original line as much as possible and simplifies the optimization process.At the same time,through the comparison for three types of passenger perception indicators of target region in Shenyang,which includes the indicators based on the current transit network,the indicators under the condition of adding vehicles to meet the needs of public transit and the indicators after solving the new routes optimization model,the effectiveness and the practicability of this method are validated. |
PDF Full Text Request