It is proposed that the road infrastructure side should respond to autonomous cars, considering the development and dissemination of autonomous cars. In this study, by analyzing driving test data, the effects of lane marking conditions and global solar radiation on the detection performance of the lane departure warning system were investigated. We conducted a running experiment on a test track with lane markings that simulated the deterioration state, and measured detection result, the brightness of the lane markings and the road surface, and the amount of global solar radiation. From the analysis results, it was suggested that the luminance difference between the lane markings and the road surface and the amount of global solar radiation are factors affecting the detection performance, and that the progress of peeling of the lane markings also deteriorates the detection performance. Although the conditions were limited, we were able to suggest the quantification of the detection performance degradation conditions for the formulation of lane marking management guidelines.

Identifying an inundation area after a flood event is essential to developing an emergency response plan. This study proposes a method to automatically determine disrupted road segments by floods using image recognition technology, a deep learning method. The algorithm used in this study is YOLO (You only look once) v3, developed in 2018. A model developed using aerial photographs taken during a past flood event is applied to aerial photographs taken during another flood event to verify the conformity of inundation conditions of road segments. The model visualizes the inundation status of roads by determining the presence or absence of inundation on a 100-meter mesh basis using aerial photographs and integrating the information on whether the mesh includes road links. The results showed that the model developed in this study could discriminate road flooding with more than 80% accuracy. However, there are frequent misclassified images where it was difficult to determine the inundation status and images that included fields.

In recent years, the crime cases known to the police in Japan are on the decrease, but public anxiety about crime remains high. In particular, the number of bicycle thefts cases is high, but the arrest rate is extremely low. Hence, it is a critical issue to take measures to prevent bicycle thefts. In this study, we constructed a model to estimate the incidence of bicycle thefts using data on the case of bicycle thefts for the past 7 years and clarified the elements of urban form associated with the occurrence of bicycle thefts. First, the optimal bandwidth for predicting density of bicycle thefts in the kernel density estimation method was verified, and it is shown that 40m is optimal to predict the density. The result of multiple regression analysis shows that numbers of free municipal bicycle parking spaces, numbers of passengers at stations, and areas of commercial district had a particularly strong relation to the density of bicycle thefts.

To examine appropriate safety guidance and alerts for children when riding a bicycle, the authors conducted a web questionnaire survey to verify the characteristics of risky behavior and near-miss events when riding a bicycle. In this study, the authors investigated the awareness of parents because it is thought that children themselves are often unaware of their risky behavior. The responses were categorized into 3 types of quantification, and the characteristics of each age group and the relationship between risky behavior and risky events were analyzed. As a result of the analysis, it was found that risky behavior could be classified with two axes: "1. enjoyment of driving/ enjoyment other than driving" and "2. unconscious carelessness/ conscious conceit". In addition, it was found that elementary school students were more likely to take pleasure in driving and risky behavior associated with unconscious carelessness, while junior high school and high school students were more likely to take pleasure in activities other than driving and risky behavior associated with conscious conceit.

It is important to understand the influence of alteration of initial activity plans on pedestrian activity-travel behavior so as to suggest the relevant policies regarding the creation of more attractive city center to visitors. This study aims to develop pedestrian activity-travel behavior models considering alteration of visitors' initial activity plans, based on questionnaire survey in the central area of Kobe City. We first classify visitors' activity-travel behaviors with and without alteration of activity plans. The proposed models consist of choice of the number of activity locations initially planned, choice regarding whether continuing activity, activity type choice, activity location choice and activity duration choice. Second, we evaluate the influence of relocation of commercial facilities on alteration of visitors' activity plans. The results show that it is possible to reproduce visitors' activity-travel behavior with good accuracy using the models, and the increase in the attractiveness of city center can encourage alteration of visitors' activity plans, causing the number of activity locations and activity duration in the city center.

This study focuses on separated traffic signal control that separate pedestrians and vehicles, which have been confirmed to be effective in reducing traffic accidents and are being installed throughout Japan. In order to examine the conditions that would further increase the effectiveness of traffic accident reduction through the installation of such signals, we analyzed the factors that affect the number of traffic accidents caused by various conditions of separated traffic signal control. The analysis was conducted using actual traffic accident data and a negative binomial regression model for traffic accident risk assessment. As a result of this study, the main findings were: 1) different parties have different factors that affect traffic accident risk; 2) factors that reducing the traffic accident risk include shorter distance between stop lines and installing bicycle crossing lanes; and 3) factors that lessen the effect of reducing the traffic accident risk in interaction with installing separated traffic signal control include longer distance between stop lines for all accidents and vehicle accidents, and a larger population around the intersections for all accidents.

The purpose of this study was to analyze and discuss residents' attitudes at multiple time points regarding the relationship between their sense of security in using public transportation and their use of such facilities during an infectious disease epidemic in order to obtain a basic knowledge of appropriate public transportation use. An analysis of covariance was conducted among residents of Aichi Prefecture (n=1,242, n=1,685) to examine the influence of the sense of security about the facility on the frequency of public transportation use at two-time points (July 2020 and October 2021) during the Covid-19 epidemic, and the influence of vaccination on this sense of security. The following findings were obtained. (1) The provision of a sense of security impacted the frequency of public transportation use, and the impact of the provision of a sense of security on bus use was significant at the time of the Covid-19 epidemic. (2) Vaccination did not affect the frequency of public transportation use or the provision of a sense of security.

For preventing fatigue driving it has to induce the driver to enter rest area to take a rest. It is necessary to analyze drivers' resting behavior on expressway as the first step for suggesting better measures. This study conducted a survey for drivers' resting behavior and stopover at SA/PA, and constructed a rest behavior model on expressway in consideration of rest area facilities and fatigue caused by driving. The valid of the model was confirmed based on statistical test results and hitting ratio of SA/PA choice. Modeling driving fatigue due to driving time in this study, will contribute to prediction for changes in traffic behavior with the popularity of automatic vehicle, especially with consideration of driving fatigue reduction by automation.

This study examined the results of an attitude survey in Akita City, Akita Prefecture, Japan, regarding the expansion of the range of activities of elderly people through discounted cab and bus transit. The results indicated that being able to reserve a cab on the same day and transferring between cabs and buses at facilities such as supermarkets influenced the intention to use a transfer plan between cabs and buses. It was also indicated that it is important to set lower fares and shorter transfer times between cabs and buses. Regarding the plan for a discount for transferring between cabs and buses, the results showed that the intention to use the plan was high among the elderly over 75 years old and among those who currently use only buses and trains. For example, 25.2% of the respondents would like to use a plan that offers a 30-minute transfer time and a cab fare of 300 yen to seniors aged 75 or older who currently use only buses and trains.

The occurrence of traffic accidents in Japan has continued to decrease, but in recent years it has bottomed out. This trend is more pronounced on community roads. Compared to highway, it has been difficult to obtain data on traffic volume and other data on community roads, limiting the analysis of safety measures. In recent years, however, the availability of data on community roads has been increasing due to the widespread use of ETC2.0. On the other hand, in terms of analysis methods for traffic accident prediction, multivariate analysis methods are expected to be applied to AI (Artificial Intelligence) for effective traffic accident analysis. Therefore, this study utilized ETC data and inverse reinforcement learning, a type of AI, to estimate the factors of potential hazards on community roads. We can obtaine knowledge on the possibility of using inverse reinforcement learning to estimate the effect of safety measures.

The purpose of this study is to investigate the influence of two factors, “road surface condition” and “traffic environment”, on the visual attention of pedestrians, based on the hypothesis that these factors influence the gazing behavior of students walking along the school route. The following 3 points were clarified by analyzing gaze data and questionnaire survey results obtained by the moving image viewing experiment in the real image VR environment where the experimental condition can be controlled. (1) The effect of the gazing action on immovable objects was limited when snow cover and movable objects separately increased the load. (2) When snow cover and the presence or absence of movable objects simultaneously increased the load, the average gaze time to immovable objects might have increased due to the lowering of visual attention by the loss of margin in cognitive resources. (3) The relationship between traffic safety consciousness for immovable objects and visual attention for immovable objects is clarified.

The present study is to evaluate drivers' evasive behavior focusing on the timing of visual perception for pedestrians when turning right at signalized intersections. The Virtual Reality headset-based Driving Simulator (VRDS) was used to measure drivers' timing of visual perception for pedestrians. As a result of experiments with 10 participants using VRDS, the drivers in the early timing to detect pedestrians can reduce the probability of severe conflicts with pedestrians. And when there were many pedestrians crossing at the time of conflict, it was difficult to detect early when drivers turned right at a gap. Based on the results of the eye measurement, we also proposed a safety level that evaluates the probability of conflict between a driver and a pedestrian. The proposed safety level was evaluated, and its validity was shown as an indicator to evaluate conflict probability. It was also suggested that for a more reliable safety level, time intervals should be included in addition to the results of eye measurement. Furthermore, we considered the timing and content of information provisions to avoid conflict with pedestrians when turning right.

In recent years, several studies have pointed out changes in traffic flow on expressways over time, and the investigation of the causes of these changes has become an urgent issue. In light of this, the purpose of this study is to clarify the changes over time in traffic flow at several points on the Metropolitan Expressway by comparing three different period and to examine the causes of these changes. From a macro perspective, a comparative analysis of QV diagrams and 99% tile value traffic volumes and speeds was conducted, and from a micro perspective, a comparative analysis of speeds, traffic volumes, and time headway within the platoon before and after the breakdown was conducted. As a result, in general, the QV diagram showed a shrinking trend, the 99% tile value traffic volume and the traffic volume before and after the breakdown showed a decreasing trend, while the rate of platoon generation before and after the breakdown showed an increasing trend. The statistical analysis suggested that changes in traffic volume are affected by changes in average time headway, heavy vehicle ratio, number of vehicles constituting a platoon, and platoon generation rate.

Traffic congestion in the south-central urban area of Okinawa Island has become a serious social problem, and there is a need to promote the use of public transportation such as buses to solve this problem. According to a questionnaire survey, there is a strong need to improve the current bus routes and schedules, and a bus network that takes into account the actual conditions of travel is required. As a first step toward this goal, this study compared the number of buses in operation with the number of people living in the target area, which was obtained from cell phone GPS data, across a 500-m mesh to determine the number of buses in operation. As a result, the number of buses running in relation to the flow of people was higher in areas where long-distance routes such as National Route 58, National Route 330, and Prefectural Route 75 Okinawa-Ishikawa Line were running. On the other hand, in Urasoe City and Tomigusuku City, which are adjacent to Naha City, there were many areas where the number of services was small in relation to the flow of people.

In this study, we surveyed Sakai City and Awara City, tourist destinations with a focus on history and culture in Reihoku region of Fukui Prefecture. Specifically, we conducted research on the potential for reducing car-dependent tourism. The results suggest that visitors who use public transportation tend to spend approximately 1.3 times longer per day and spend more than those who use cars. Furthermore, we found respondents would be willing to use a new mobility service that 62.0% if it introduced shared bicycles and 47.5% if it introduced E-scooter sharing. In addition, it is revealed that car-using visitors were significantly more willing to use the E-scooter sharing than public transportation. The findings of this study will assist in the consideration of measures to support transportation in regional cities with high car dependence and measures to restore tourism, which supports the local economy. We intend to conduct a more in-depth study in the future.

In recent years, over-tourism has been a global problem due to the increasing number of tourists. Congestion pricing is attracting attention as a solution to over-tourism. However, it is difficult to predict the effects of pricing in a sightseeing area by referring to past cases because most cases were conducted in city centers where commuter traffic is dominant. In this study, the activity-based model for tourist travel behavior is developed by extending the framework of a mixed integer linear programming problem. Then, we predict the change of transportation demand in a sightseeing area by considering the dynamic user equilibrium of tourists ’schedules to take into account the change in travel time due to dynamic congestion. The results of the analysis in Kamakura city show that congestion pricing would prompt some private car users to shifting toward public transportation, thereby network congestion would be reduced. By contrast, we find that the increase in the amount charged caused decrease in tourist utility.

The North Kanto region is the most advanced car society in Japan. In low density urban areas, transportation demand is widely and thinly distributed, and it is not efficient to provide transportation services by fixed and scheduled route public transportation. According to Gunma Prefecture person trip survey data, for the elderly and others who are unable to drive a car by themselves, pick up and drop off by car is an important means of transportation. This paper focuses on "an inter household ridesharing" in which residents of other households serve as drivers for the elderly as a means of daily transportation, and appropriate analytical methods using Person Trip survey data, as well as analysis focusing on driver's license and car ownership status and household composition. The purpose of this study is to understand how an inter household ridesharing supports the elderly who do not have free access to a car.