Font Size: a A A

Study On Traffic Assignment Model And Solution Algorithm Under Combined Travel Mode

Posted on:2015-10-05Degree:MasterType:Thesis
Country:ChinaCandidate:C DuFull Text:PDF
GTID:2272330434460766Subject:Logistics management
Abstract/Summary:PDF Full Text Request
With the rapid development of urbanization, new requirements are put forward to urbantransport, a variety of transportation means which are environmental protection, energy savingand efficient are applied. Our life has been inseparable from the BRT, metro and other meansof transportation. Because of increasing size of the city, residents may need to transfer to oneor more of transportation to reach their destinations; this travel mode is called combinedtravel.According to a lot of relevant literature, traffic assignment under combined travel is thecore of the transportation planning and management. Static traffic assignment and dynamictraffic assignment are two main research directions; the fundamental difference is whether theimpedance of sections is a function of time. The road resistance of latter is unchanged whichis impossible to achieve in practice, so dynamic traffic assignment is the focus of the study.But the study of static traffic assignment can not be ignored; static traffic study focused onmacroeconomic traffic phenomenon which plays an important role in the analyzing the trendsof traffic in future. This paper studies the traffic assignment under combined travel from thefollowing aspects:Firstly, introduce the super network briefly and establish a super network based on theactual traffic network near Lanzhou Jiaotong University in Anning District of Lanzhou. Thisis the research foundation for this study. Describe super network topology, define effectivepath and design algorithm based on graph traversal method to solve effective path.Secondly, study traffic assignment from the perspective of both static and dynamicrespectively. Assume travelers’ travel patterns and paths are stochastic, proposed modelconstraints, use logit model to express equilibrium conditions, establish logit model to meetstochastic user equilibrium conditions which based on the elastic demand. Because it iscomplex to solve logit model directly, the paper establish variational inequality model whichis equivalent to equilibrium conditions, and prove its equivalence and the existence of theoptimal solution. Use method of successive weight average which based on method of weightaverage to solve the equivalent variational inequality model, obtain the solution to the logitmodel at the same time. Meanwhile, consider stochastic user equilibrium under combinedtravel based on the super network, establish dynamic traffic assignment model logit modelunder combined travel with consideration of travel time and path to describe the behavior oftravelers’ preferences and analyze the characteristics of equilibrium state, use an iterativeheuristic algorithm to solve the model gradually according to a linear constraints of thevariational inequalities.Then, design a small traffic network as an example, use static traffic assignment model and dynamic traffic assignment model to analyze this network, study the reliability andefficiency of the model and algorithms, analyze some parameters of the dynamic trafficassignment model, explain the change of OD volume and costs in traffic. Take the actualtraffic network near Lanzhou Jiaotong University in Anning District of Lanzhou asmedium-sized network scenarios, build a more complex super network, assignment on paths,nodes and other parts of the network, use dynamic traffic assignment model to solve themedium-sized traffic network.Finally, using static traffic assignment model to analyze small traffic network, obtain themore information travelers knew, the more uncertainty of traffic network is reduced andtravelers may make more clear and wise choice easily. By using dynamic traffic assignmentmodel to analyze the two examples, the paper can obtain the model can reflect the impact oftravelers preference to travel behavior more realistic, analyze the impact of changes inparameters to the traffic demand and costs, able to meet the requirements of travel path choiceand travel time; the algorithm is reliable and efficient, can be applied to the study of small andmedium traffic network and has a certain practicality.
Keywords/Search Tags:Combined Travel, Traffic Assignment, Super Network, Stochastic UserEquilibrium
PDF Full Text Request
Related items