Tomio Miwa

Driver’s Route Choice Behavior and its Implications on Network Simulation and Traffic Assignment

The principle of driver’s route choice has long been the shortest path with a fixed time penalty of a toll. It is also understood that traffic is assigned on the road network based on user equilibrium with perfect information assumption. This paper demonstrates two empirical studies that pose questions to these traditional assumptions, to better understand route choice behavior. Multi-class user equilibrium assignment with imperfectly informed drivers’ classes is applied to a metropolitan area network first. Next, route choice behavior is directly observed and analyzed using probe’ car data.

Updating Dynamic Origin-destination Matrices using Observed Link Travel Speed by Probe Vehicles

A method for updating dynamic O-D matrix using observed link travel speed of probe vehicles is developed. The method consists of two parts: link flow of total traffic is estimated from the link travel speed of probe vehicles, and dynamic O-D matrix is estimated using the estimated link flow and historical O-D matrix. The k-v function derived from the acceleration model by Gazis et al. (1961) is applied to the link flow estimate, and the entropy maximization model proposed by Willumsen (1984) is applied to the dynamic O-D demand estimate. By using Bayesian inference approach, the variance of the estimate as well as the point estimate of link flow is obtained. Also, the entropy maximization model is extended to incorporate the difference in the reliability of link flow estimates among links. The results of a case study show that the accuracy of the estimated dynamic O-D matrix is improved by the proposed method, and that the accuracy of the link flow estimates obtained in dynamic O-D demand estimation is improved by the proposed method, too.

Dynamic Route Choice Behavior Analysis Considering En Route Learning and Choices

This paper presents a method to model a drivers en route learning process and changes in route choice at each decision node. A model based on Bayesian networks is proposed to describe the en route updating of the drivers knowledge of the traffic state. A random utility–based model is developed to predict en route choices. A case study based on probe data is carried out to illustrate the development of the model and analyze the dynamic route choice problem. The results show that the model in which a drivers choice of making decisions en route is taken into account has a better goodness of fit. The probability of making a choice en route is related to the distance from the origin and the spatial scale of the intersection at the decision node.

Use of Private Probe Data in Route Choice Analysis to Explore Heterogeneity in Drivers' Familiarity with Origin-Destination Pairs

An exploratory analysis about the effect on route choice of heterogeneity in driver familiarity with origin–destination (O-D) pairs was carried out. This analysis was based on probe data collected by private vehicles in Toyota, Japan. The hypothesis test results showed that route choice behavior changed in relation to the level of familiarity with O-D pairs. Two specifications of choice models were proposed to consider the effect of familiarity. The estimation results showed that the models that considered familiarity fit the data better and suggested that trips between more familiar O-D pairs had larger error variances and less sensitivity to explanatory variables. The estimated models were applied to a specific choice situation, and the prediction results showed the potential biases introduced by not considering the heterogeneity in the familiarity with O-D pairs. In addition, when traveling between more familiar O-D pairs, drivers were less sensitive to the intersection count than to the free travel time.

Recursive bivariate response models of the ex-ante intentions to link perceived acceptability among charge and refund options for alternative road pricing schemes

Abstract Road pricing (RP) is currently under development as a way to mitigate the acute traffic congestion in Jakarta. The RP has been recognized as a powerful tool in delivering efficient road use. Despite of a well-established rationale that tackled congestion, RP policy faces lack of public support and is seen rather controversial due to charge required to enter charging zone. To encounter shortcomings of RP, we introduce a parking deposit system (PDS) as an alternative. The PDS is based on refunds to car users. Refunds are offered only on parking fees or as discounts on purchases within the charging zone. Thus, this paper explores the key influences and the extent to which RP and PDS schemes are acceptable to the public. Taking into account citizen’s consciousness based on a fictitious voting conducted in Jakarta, we estimated recursive simultaneous bivariate voting models to link perceived acceptability among RP and PDS bundles. We found that there is a complementary relationship between acceptance and consciousness, with PDS offers better approval and there is a substantially strong link between voting intentions and perceived acceptability.

Use of latent variables representing psychological motivation to explore citizens’ intentions with respect to congestion charging reform in Jakarta

The aim of this paper is to investigate the intentions of Jakarta citizens with respect to the electronic road pricing (ERP) reform proposed by the city government. Utilizing data from a stated preference survey conducted in 2013, we construct six variables representing latent psychological motivations (appropriateness of ERP adoption; recognition that ERP can mitigate congestion and improve the environment; car dependency (CDC); awareness of the problems of cars in society; inhibition of freedom movement caused by ERP; and doubts about the ability of ERP to mitigate congestion and environment problems). A multiple-indicators multiple-causes (MIMIC) model is developed to investigate the effects of respondents’ socio-demographics (causes) on the latent constructs in order to gain better understanding of the relationship between respondents’ intentions and the observed individual’s responses (indicators) obtained from the stated preference survey. The MIMIC model offers a good account of whether and how socio-demographic attributes and individual indicators predict the latent variables of psychological motivation constructs. Then, we further verify the influences of the latent variables, combining them with levy rate patterns and daily mobility attributes to investigate significant determining factors for social acceptance of the ERP proposal. A latent variable representations based on the generalized ordered response model are employed in our investigations to allow more flexibility in parameter estimation across outcomes. The results confirm that there is a strong correlation between latent psychological motivations and daily mobility attributes and the level of social acceptance for the ERP proposal. This empirical investigation demonstrates that the latent variables play more substantial role in determining scheme’s acceptance. Moreover, elasticity measures show that latent attributes are more sensitive compared to levies and daily mobility attributes. Therefore, in order to gain the acceptance of the majority of road users, a gradual introduction of ERP may be needed to allow users to gain understanding of the scheme, thereby gaining a more acceptable response. The findings from this work should provide insight for the Jakarta government in designing a more effective and acceptable policy aiming at promoting the adoption of ERP in Jakarta.

Bilevel Generalized Least Squares Estimation of Dynamic Origin-Destination Matrix for Urban Network with Probe Vehicle Data

Methods of estimating dynamic origin–destination (O-D) matrices for urban networks from probe vehicle data are explored. A speed–density function is derived and fitted for different types of roads with the use of the maximum likelihood method. Both a Bayesian method that carefully incorporates prior information and an ordinary method are used to estimate link flows from probe vehicle speed. A dynamic traffic assignment–based bilevel generalized least squares (GLS) estimator considering the distance between the estimated and target O-D matrices as well as the distance between the calculated and observed link flows is formulated to estimate dynamic O-D matrices from estimated link flows. In the iterative solution procedure, the upper level is solved with the extended Bell algorithm, and the microscopic dynamic traffic assignment system VISSIM is applied to produce the assignment matrix in the lower level. A medium-sized signalized network in Tokyo is modeled in a case study in which Bayesian and ordinary methods are compared both in link flow estimation and O-D matrix estimation. Further, the bilevel GLS estimator and bilevel ordinary least squares estimator are implemented and then compared in O-D estimation. The results validate the proposed bilevel GLS estimator.

Explaining Differences in Acceptance Determinants Toward Congestion Charging Policies in Indonesia and Japan

AbstractComprehensive stated choice (SC) experiments were conducted in Jakarta, Indonesia, and Nagoya, Japan, where proposals for congestion charges have been introduced and remain under consideration as a way to reduce acute car dependence, particularly in Jakarta. Causal paths among psychological determinants and their strengths are measured and analyzed along with proposal acceptability from a cross-country perspective based on the similar context of the SC experiments. The findings from analysis with a multiple-sample, multiple-indicator, multiple-cause (MS-MIMIC) model show that a number of psychological determinants provide an explanation for the acceptability of the proposed scheme in both cities. Psychological motivations, including awareness of the city’s environment and awareness of the problem of cars in society, appear to be the most important direct determinants leading to recognition of the effects of a congestion charging scheme and they are indirect determinants of policy acceptance in bot...

An optimal mandatory lane change decision model for autonomous vehicles in urban arterials

ABSTRACT Autonomous driving has become a popular topic in both industry and academia. Lane-changing is a vital component of autonomous driving behavior in arterial road traffic. Much research has been carried out to investigate discretionary lane changes for autonomous vehicles. However, very little research has been conducted on assisting autonomous vehicles in making mandatory lane changes (MLCs), which is the core of optimal lane-specific route planning for autonomous vehicles. This research aims to determine the best position for providing MLC instruction to autonomous vehicles. In this article, an optimization model is formulated to determine the optimal position at which an instruction to change lanes should be given through automotive navigation systems. First, the distribution of time spent waiting for safe headway to make a lane change is modeled as an exponential distribution. Lane-specific travel times are then calculated for vehicles in various situations by applying traffic shockwave theory and horizontal queuing theory. Finally, the expected travel time is derived for a vehicle receiving a lane change instruction to change lanes at an arbitrary position along the road. The proposed model is validated by a comparison with a simulation model in VISSIM. Additional experiments show that the instruction should be given earlier in the case of denser traffic or higher travel speed in the target lane and that vehicles can save considerable time, if they follow the guidance provided by the proposed model. The proposed model can be applied to guide autonomous vehicles to travel an optimal route.

Allocation Planning for Probe Taxi Devices Aimed at Minimizing Losses to Travel Time Information Users

In many cities, taxis equipped with global positioning system (GPS) devices are used as probe vehicles. However, there are cases when the data from such taxis are not suitable for use as traffic information because of long data polling intervals. There are also developing countries in which taxis are not equipped with GPS devices. In such cases, a probe vehicle system has to be established by allocating probe vehicle devices to taxi dispatch centers (TDCs). In this study, the authors analyze the efficient allocation of such probe vehicle devices to TDCs under the cost constraint and the assumption of independent link travel times, adopting a definition of efficiency based on the loss to travel time information users resulting from information errors. The travel time information obtained using a probe vehicle system is the sample mean of a statistical distribution and should be treated as a statistical variable. Thus, the information error, which leads to losses to information users, is also a statistical variable. The allocation problem minimizes such losses. The results from a case study show that the loss to information users changes with the time of day and among different city areas. It is also shown that probe devices should be evenly allocated among TDCs. If either the number of probe devices or the data collection period is limited, more probe devices should be allocated to the central urban area where the traffic is heavier.