The Analysis Of The Travel Behavior On Multi-modal Traffic Network Considering Park-and-ride

When making traffic controls and evaluating the traffic conditions,on the basis of full understanding and mastering the traffic demand, travelers’travel choice behavior is needed to be fully understood. Travelers’choice behavior is significantly different in different traffic network environments, especially when considering travelers’parking behavior into the whole travel process. In this dissertation research, we take the parking behavior into consideration both under static and stochastic traffic network conditions by using the systematic analysis approach. Research on the travel choice behavior considering P&R (park-and-ride) on multi-modal traffic network provides fundamentals and tools for scientifically planning and managing urban transportation and parking facilities and for the formulation and evaluation of traffic controls. The main content of this dissertation is divided into five parts:(1) A stochastic mixed traffic equilibrium assignment model considering user preferences. Based on the Logit discrete choice model, a variational inequality is established to describe stochastic multi-user and multi-modal mixed traffic equilibrium, when considering the influence of travelers’preference for mode, transfer station and route. We investigated the influence of the parking fee of the destination, the frequency of the bus and subway, the total demand and the proportion of travelers with different value of time. Numerical results indicated that changes in the parking fee of the destination or the frequency of the bus affect road traffic significantly; flow distribution on the traffic network is significantly different with the variation of the proportion of different travelers.(2) Commuting pattern with park-and-ride option for heterogeneous. In a competitive bottleneck transportation system of urban expressway and park-and-ride with heterogeneous commuters, the necessary conditions for different user equilibrium states and the corresponding stability are analyzed and compared. An optimal combination of parking fees model is established to investigate the influence of parking fees on the mode choice and commuting behaviors. Furthermore, the analytical results of the model are given. The influence of the bottleneck capacity and the portion of each group to the travel behaviors and to the total social cost are discussed with numerical results. (3) Parking pricing and modal split under uncertainty. A bilevel programming model for solving the problem of determining parking fees and modal split is established in a competitive highway/park-and-ride network with uncertain travel times on the road. Due to the particularity of the network structure, the bilevel programming model at user equilibrium is described as a minimization model. And the analytic solutions are given. The impact of three travel time choice criteria:the mean travel time, the travel time budget, and mean-excess travel time, is compared for parking pricing and flow distribution on the network at equilibrium. As the expansion of the model, the impact on travelers’choice behavior with different parameters at different travel time criteria are compared and analyzed. Also, the influence of the log-normal distribution parameters is tested and analyzed. Results show that, on the one hand, no matter whether to consider the uncertainty of the travel time and how to deal with the uncertainty, improving the service level and operating efficiency of the subway can effectively induce the choice behavior of travelers, and then adjust traffic distribution on the network. On the other hand, the effect of reducing the variance of the distribution of the travel time than just reducing the mean of the distribution to adjust the flow distribution on the network and the total social cost is more obvious.(4) Equilibrium analysis with parking time uncertainty. Parking lots on the transportation network are divided into different clusters (or areas). Parking lots within each cluster charge the same fee and have the same space searching and access time. Considering the uncertainty of the travel time on the network, the reliability of the space searching and access times is taken into consideration at the same time. Based on the travel time reliability and random error to travelers’perception disutility, an equivalent variational inequality (VI) formulation at user equilibrium is developed. Numerical examples are introduced to analyze the influence of different total demands, confidence levels, traffic management and control measures and the distribution of parking space searching and access times to travelers’choice behaviors. Results show that reduce the variance of the distribution of parking space searching and access time, can improve the reliability of travel time for car travelers, and further can induce the flow distribution on the network. Adjust the subway fare or the parking fee in CBD appropriately can effectively regulate the flow distribution on the network, reduce the share rate of modal split for car, and finally reduce traffic congestion on road.(5) Parking space allocation and pricing on a simple traffic network. When the government owns all the parking space, two three-stage Stackelberg models are introduced under general road conditions and with traffic bottleneck to solve the parking space allocation and parking pricing problem. According to the value of times and whether have reserved parking spaces in central city, travelers are divided into two groups. When considering the traffic bottleneck, the model analysis shows that travelers without reserved parking space in the destination are generally tended to choose park-and-ride because of their lower value of time. As the development of the model, the status of elastic demand conditions is discussed.