Research On Network Construction And Scheduling Optimization Method Of Public Transportation System Based On Uneven Demand Distribution

Priority to the development of public transportation is one of the important strategies to solve to issues such as traffic congestion,greenhouse gas emissions,noise pollution,energy consumption and traffic safety.It is very important to build a Bus with High Level of Service(BHLS)aiming to improve the speed,convenience,comfort and safety of public transportation.That is the key to increase the public transport sharing rate.Based on the existing theoretical methods and practical conditions of operation and management of Bus with High Level of Service planning and design,this paper proposed the core public transportation network construction and joint scheduling model in the process of Bus with High Level of Service planning and design.And the research carried out the impact analysis of related factors.The main research contents,methods and conclusions of this paper include:First of all,the characteristics of Bus with High Level of Service including better costeffectiveness,high time and spatial coverage,succinct network structure of bus lines,low repetition rate of roads,short "door-to-door" travel time for passengers and more friendly to passengers were been summarized.This is followed by a description of the key factors involved in a h Bus with High Level of Service and a dissection of the structural characteristics of the bus network and the types of bus routes.The spatial and temporal heterogeneity of the demand for public transportation travel were analysed.Secondly,based on the domestic urban structure,the urban form of the chessboard-shaped urban road layout and the urban area on the edge of the city was as the basic urban characteristics of the analysis considering the travel demand of public transportation agencies and passengers.This paper combined with the non-uniformity of passenger travel distribution in the downtown and the periphery,an accessible bus network was constructed.The bus network is grid in downtown and radial in the city’s periphery.Taking the cost of public transportation agencies and the travel time of passengers as the goal,a model of the radial grid accessible bus network based on uneven travel distribution was constructed with the parameters such as bus stop spacing and headway of bus lines as decision variables.Take Hohhot as a case study.The model has strong feasibility.In the application the solutions of the decision variables can be adjusted according to the needs.The total cost is inelastic for the input parameters and the changes of parameters have little effect on the model results.The more concentrated the demand in downtown,the lower the total cost.However,when the demand is excessively concentrated in downtown,the stability of the bus network will decrease,which will weaken the competitiveness of public transportation.Thirdly,in view of the shortcomings of the radial grid accessible bus network structure,a model for constructing a sparse and dense hybrid grid bus network based on uneven travel distribution was proposed.In the sparse and dense mixed accessible network,the bus network in downtown is dense grid and the bus network in periphery is sparse grid.The headway of the bus line in downtown is higher than the headway of the outer line in periphery.Compared with the radial grid accessible bus network,the sparse and dense hybrid grid bus network is simpler and with fewer transfers.The structure of this network is stable.This type of network reduces the travel distance and time of passengers,which is more friendly to passengers on the periphery of the city.The sparse and dense hybrid grid bus network is more suitable for fast-growing cities.And then,from the characteristics of inter-regional uneven passenger travel,the route optimization and coordination scheduling problem of limit-stop service and local service in bus corridors was studied.According to the minimum total cost,the bus fleet unchanged and increasing ratio of limit-stop service fares to the local fares,a Multi-Objective Mixed Integer Programming Model was established that considered public transportation agencies costs and passengers times.The decision variables were the limit-stop serviced stations,the headways of the limit-stop and the local.In order to solve the model,a genetic algorithm and a Lingo program were designed;the results of the genetic algorithm were compared with the Lingo software.The validity and practicability of the model and Lingo software were analyzed and verified by taking Jinan No.115 bus as a case.For Jinan 115,in order to ensure the competitiveness of the joint dispatch service,the limited-stop service fare should be higher than the local fare and the fare increase multipliers for the limited-stop relative to local should not exceed 1.5.In the joint dispatch of limited-stop and local,the average bus stop spacing should be longer than0.5 km and the length of the bus corridor should be longer than 16.5 km.With the increase of the distance of the bus corridor,the advantages of the joint dispatching service are more prominent.Finally,a joint dispatching model for express and local was built under the situations that transit travel demands in urban clusters and city groups are mainly distributed at the two ends of bus routes and the express do not stop midway.The fare of express is higher in the joint dispatching model.The model’s superiority and effectiveness were analysed through the case study of No.209 in Hohhot.For inter-regional bus trips in city clusters or city groups,joint dispatching is superior to a single dispatching solution.The range of fare increase multipliers for the express relative to local should be between 1.44 and 1.70.