Reconstruction Mechanisms Of Commute Trip Chain Under The Integrated Transportation Information

Traffic information has a quite important impact on commuters’travel behavior, and the mature of communication technology also make people’s appeal of traffic information become increasingly intense. Nonetheless, the current traffic information service has not been realized in the whole transportation system in terms of sharing, therefore, to provide the dynamic and integrated information on multiple modes within a particular information service, which is the so-called Integrated Transportation Information (ITI), has been one of the most important goals of the construction of intelligent transportation system (ITS). ITI contributes to plan and adjust the single trip, which results in the variation of trip chains’structure and spatio-temporal characteristics in the whole, namely, the reconstruction of commute trip chain. However, whether ITI can work or not depends on its content and commuters’reaction. While most of the previous researches are based on single trip, so it’s very necessary to study the analyzing method based on the trip chain and the reconstruction mechanisms of commute trip chain, which can give an insight to the construction of ITS.The dissertation takes the commute trip chain that completed by car as the research object, and takes ITI as the input condition. In terms of travelers’congnition to ITI, the dissertation centers on the forms and reconstruction mechanisms of commute trip chain under the ITI, and investigates its impact on the characteristics of commute trip chain. The dissertation also analyzes the interactions of temporal reconstruction, spatial reconstruction and structural changes under different input conditions, and elaborates the effect of commute trip-chain’s reconstruction on the traffic flow assignment on the integrated transportation network. The content of the dissertation is arranged as follows:Firstly, previous studies on the travel behavior analyzing methods based on the trip chain and the effects of traffic information on travel behaviors were reviewed, which provide useful theoretical foundation for this dissertation.Secondly, the concept of ITI, commute trip chain and the reconstruction of commute trip chain were defined. The relativity of ITI’s loading mode depending on commuters’attention to it, and commuters’decision rules according to their reliance on ITI were analyzed. Beyond that, the forms of commute trip-chain’reconstruction was described, and the reconstruction mechanisms were summarized by analyzing the interactions of the three reconstruction forms. Thirdly, the content of ITI service was designed first, and an RP-SP survey was carried out to obtain the revealed and stated trip chain data as well as commuters’ demand to ITI, which contributed to illustrate the variation of the temporal-spatial characteristics and structure of the trip chain from the macro perspective, so the influence of ITI was demonstrated. Based on the rough set thoery, commuters’demand to ITI at different stage for ITI was studied.Fourthly, using the static ITI as input, based on the RP and SP combined data, a Mixed Logit model and a double particle swarm optimized BP neural network (DPSO-BPNN) recognition model were established respectively to predict the combined mode choice behavior of commute trip chain, in which the ITI and the characteristics of commute trip chain were regard as factors affecting combined mode choice behavior, and the influence of ITI on mode split rate of the commute trip chain was explored from the microscopic perspective. According to the above analysis, the reconstruction mechanisms of commute trip chain under the condition that commuters pay less attention to ITI was dissected, and the relationship betweent temporal-spatial reconstruction and structural change of commute trip chain was proved.Fifthly, using the dynamic ITI as input, the decision rules and the perceived utilities of trip chain under the dynamic ITI was identified. Based on that, a dynamic model was proposed to describe the dynamic choice behavior of the daily trip chain with the objective of maximizing the perceived utility of a trip chain. The formation and dynamically changing process (reconstruction process) of the trip chain in one-day finite horizon were simulated. The commute trip chain between Huilongguan community in Beijing and Beijing Jiaotong University was taken as an example to explain the application of this model.Finally, a multiclass probit-based trip chain stochastic user equilibrium (SUE) model was developed, and a computional example was presented to illustrate the influence of commute trip-chain’s reconstruction on the traffic flow assignment on the integrated transportation network.The following are the main innovations of this dissertation:1. Focusing on the reconstruction mechanism of commute trip chain, the interactions of the temporal-spatial reconstruction and the structural changes was analyzed.2. Contents of the ITI service were designed. An RP&SP survey was conducted. A method based on Rough Sets theory with genetic algorithm embedded was applied to explore the dependencies between commuters’choice of ITI option and their individual attributes as well as trip-chain attributes, and commuters’demand to the different information in the pre-trip and en-route stage was ranked.3. Based on the RP and SP combined data, a Mixed Logit model which took scale parameter difference, heterogeneity and reference dependence effects into account was established. The model was used to measure the impacts of trip chain characteristics on trip chain combined mode choice under the ITI, and analyze commuters’preference to multi-model information and their choice inertia. In addition, a DPSO-BPNN recognition model was compared with Mixed Logit model in terms of generalization ability, predictability and transferability.4. Based on utility theory and multi-stage decision method, by establishing an integrated transportation network on which travel information was loaded dynamically, a temporal-spatial dynamic model was proposed to describe the choice behavior, including destination, mode, route and departure time, of commute trip chain with the objective of maximizing the perceived utility of a trip chain, which contributed to find the optimal commute trip chain dynamically. The generation and reconstruction process of the trip chain in the finite horizon were simulated.5. In view of the trip chain, a multiclass probit-based stochastic user equilibrium model under the ITI was proposed. Depending on the hyper-network theory, a MSA algorithm with Monte-Carlo method embedded was presented to solve the model. A numerical example was presented to investigate how ITI influences commuters’travel mode, travel pattern and the cost and structure of the trip chain, and the impact influence of commute trip-chain’s reconstruction on the traffic flow assignment on the integrated transportation network was demonstrated. In the course of solution, hyper-network theory was introduced, which was conductive to deal with the trip-chain flow assignment traveled by P&R (park and ride) mode.