Transportation Assignment Models Based On Travel Route Choice Behavior

With the development of our society and economy, the process of urbanization is more and more faster, and at the same time city scale is bigger and bigger, the city has more and more population and motor vehicles. These problems, such as city traffic congestion, parking difficulty, environmental pollution have became the main problems in our country and other country’s cities. In order to solving these traffic problems, it need to carry out scientific urban planning and management, on the one hand adding the city road facilities, and on the other hand optimizing and developing potential through the management. And the premise of urban planning and management is to conduct scientific traffic demand forecast. However, observing the the existing traffic forecasting models, the widely used four phase prediction method based on set analysis has high cost, and is difficult to promote transplantation model, and lack sufficient persuasion, at the same time the behavior model based on non-set analysis is in the development, and the key theory and method are still not fully established, and is far to reached the practical stage. These caused that the rules of road and transportation can not be accurately grasped, the prediction results is deviated from the practice, which also resulted in the planning and construction of road and transportation is not reasonable, wasteful investment, and residents travel inconvenience.Therefore, the mechanism of the traffic demand must be systematically researched, and the nature rules of traffic demand production and distribution must be grasped, and the effective support can be provided for scientific and reasonable to carry out urban planning and traffic management. Study transportation assignment models under travel route choice behavior’s effect, can provide support for the rational planning of city road network, effectively exerting the potential of road traffic resources, forming the good city traffic environment. And it also can provide navigation and guidance scheme for travelers, reduce travel cost, improve travel efficiency. The mechanism characteristic of travel route choice behavior, the individual travel route choice models based on travel route choice behavior, and the transportation assignment models based on travel route choice behavior, are the key problems of research.This paper carried out the study of transportation assignment models based on travel route choice behavior from the following four aspects.Firstly, from the perspective of behavioral science theory and the actual survey data of chengdu, the individual traveler’s fixed figure characteristic for road network and medium traffic flow characteristic of the distribution of route are analyzed, and three kinds of mechanism characteristics of travel route choice behavior are summarized, such as the traffic flow feedback and equilibrium mechanism characteristic, travelers’routes identification mechanism characteristic, and the routes distribution mechanism characteristic. Namely, (1) feedback and equilibrium mechanism characteristic FEMC):subject to the factors of traffic tools, road network and traffic flow status, the ratio of travelers who adjust their travel path is very small, the balance and feedback of travel route choice are non global equilibrium; (2) routes identification mechanism characteristic RIMC):the traveler takes the route choice as a whole, and has the level and direction for the path recognition; (3) routes distribution mechanism characteristic RDMC):for a traffic demand with a single departure and destination, the path selection conforms to anti-spindle features that the traffic flow is more and more less when it far away the departure or destination, and the path choice frequency accords with power function.Secondly, two travel route choice models are established based on the travel route choice behavior mechanism characteristics, respectively from the angles of cellular simulation method, and regret theory, a travel route choice model is established based on subgraph space, effective reversing, and cellular automata theory, another travel route choice model is based on regret update accordingly.Thirdly, route choice rules and traffic system balance rules were studied from the group angle. On this basis, considering the road network’s random variations and the travelers’ perception error, taking the regret minimization which reflect traveler’s route choice characteristic and risk preference as the rule of route choice, a stochastic user equilibrium traffic assignment model based on the regret update was built, and the solving algorithm based on the variational inequality was gave. Then the algorithm’s validity and the characteristics of related parameters were studied using a simple example.Finally, on the basis of studying dynamic transportation assignment model principles, according the suppose that travelers would choose the route with minimization regret for travel, a ideal dynamic user optimization (DUO) transportation assignment model based on regret update was established, and an algorithm based on method of successive average (MSA) and another variational inequality (VI) algorithm were gave. By using sequential same-layer’s interation and loop with relaxation algorithm and method of successive average, adding regret value’s criterion in convergence test, the calculation efficiency of variational inequality (VI) algorithm was improved. Taking the real road network and traffic demand of a city in southeast Sichuan as an example, the feasibility and effectiveness of the models and the algorithms were verified.This research can improve the present transportation assignment model’s accuracy, interpretability and computational efficiency, the research results have important theoretical value and practical guiding significance in transportation planning, transportation management, emergency transportation decision, the traveler route guidance.