Shogo Kawakami

A reactive dynamic user equilibrium model in network with queues

A discrete time reactive dynamic user equilibrium (DUE) model for network with queues is presented. Based on the point queue model, each link is assumed to have a constant running time and queuing delay caused by link exit capacity. The link performance function and link exit function are given as functions of the arrival rate and the queue length, where the first-in-first-out (FIFO) rule is shown to be satisfied. The DUE problem is then formulated as a variational inequality (VI) problem over a polyhedral set by constructing a new network so that the path enumeration is avoided. Therefore the proposed model is applicable for large-scale networks. The existence of solution for VI problem is proved, and an iteration method is discussed in detail. A static capacity constrained model is employed to initialize the network and avoid the zero-flow network at initial time. Finally, the advantages of the new model and method proposed are tested by numerical examples.

Modeling disaggregate behavioral modal switching models based on intention data

The purpose of this paper is to develop a disaggregate behavioral model which directly predicts travelers modal switching behavior due to a transport service change. Based on the random utility theory, several disaggregate modal switching models are formulated, which incorporate resistance to change in modal switching behavior. These models are estimated using stated preference data obtained from an originally designed questionnaire survey. An empirical examination shows that it is very important to incorporate state dependence effects into disaggregate modal choice models.

Estimation of origin-destination matrices from link traffic counts considering the interaction of the traffic modes

This paper proposes a new optimization model in the form of a combined distribution, modal split, and traffic assignment method. This model extends Fisks model by introducing an entropy constraint condition with respect to the traffic mode choice. The origin-destination (O-D) trip matrices by different traffic modes can be estimated by solving the proposed model for the given link traffic counts and traffic volumes generated from each zone. From the optimal conditions of the proposed model, a demand model is derived. The proposed model is applied to a Nagoya road network to determine feasibility and efficiency. The estimated results show that the proposed model is feasible and that accurate results can be obtained from it.

Reactive user optimum and predictive user optimum in dynamic traffic assignment

Outlines a simulation method developed strictly according to the behavioral assumption of the reactive user optimum, and the results obtained from the simulation method show the correct dynamic assignment for the corresponding user optimum. The test results also show the same conclusion. The authors calculation also shows that there is no problem in computation time and stability. Therefore, this method affords an efficient dynamic assignment method. It can be applied to practical dynamic traffic networks (large scale, multi-origin, multi-destination). Furthermore, this method has the advantage of simplicity, a computer program can be developed according to the procedures of the simulation method.<<ETX>>

Incorporating a Disaggregate Model for Trip Generation, Trip Distribution and Mode Choice

A major innovation in the analysis of transportation demand since the mid-1970s is the development of disaggregate travel demand models based on discrete choice analysis. In the modelling of travel demand, disaggregate models allow us to introduce an individual’s attributes that influence travel behaviour, and to make more efficient use of the available individual data.

Travel Mode Choice Models Considering Heterogeneity of Decision Maker using SP, RP and Attitudinal Data

The purpose of this study is investigating efficient methods incorprating the heterogeneity of individual into the travel mode choice models. In this study, we propose two different market-segmentation approaches. One approach is a market segmentation by individual specific estimates obtained from a number of SP data of an individual. Another is a market segmentation by latent attitudinal variables. Then we applied these proposed methods to the mode choice SP data setting up virtual high-speed ferry and mode choice RP data. Finally, we got efficiently and interesting results, however some investigations are acquired.

A Study on Parking Place Choice and Parking Demand Forecast by Traffic Assignment Models

Abstract This paper presents an idea to forecast the parking demand to each parking lot by traffic assignment model. It is assumed that each traveler chooses parking lot to minimize his or her generalized cost, including running time by car, waiting time in parking lot and walk time. The model can be used to estimate the parking demand to each parking lot, and hence to evaluate the effect of parking policy and new parking lot.

A Study on Formulation of a Combined Modal Split and Assignment Equilibrium Model: — with Analysis on Practical Application

Abstract Based on the analysis of asymmetry effect of bus and car, a combined modal split and assignment users equilibrium model with varied demand is proposed. The concept of generalized cost is introduced into the model. The time values are divided into four kinds in order to get the tradeoff between time and cost exactly. In the process of analyzing mass transit network, a minimum generalized cost route searching technique with considering of passenger fare is also proposed. The model is verified by applying to Nagoya City and is proved to be practicable.

A Combined Disaggregate Model System for Travel Demand Forecasting

Abstract A combined disaggregate model system is developed for urban travel demand forecasting. By using the nested logit model, the first three steps of the four-step procedure for travel demand forecasting, that is, trip generation, trip distribution and travel mode choice, were modeled. For trip assignment of the forth step, the logit based stochastic user’s equilibrium model is applied.