In 2021, approximately 70% of traffic accident fatalities at pedestrian crossings were caused by drivers violating pedestrian crossing laws, indicating that pedestrians are not always given priority at pedestrian crossings. In response to this situation, the police have been stepping up enforcement, but the effectiveness of such measures is thought to be limited. In this study, we conducted a survey focusing on the standing position and behavior of pedestrians while waiting to cross the street, based on the idea that if pedestrians could clearly communicate their intention to cross the street to drivers, it would be possible to improve the rate of driver’s yielding. As a result, it revealed that standing position has more influence on the yielding rate of cars than behavior while waiting to cross the street. The study also suggested that the effects of standing position and behavior on the yielding rate could be enhanced if drivers were unable to take evasive action.

The majority of traffic fatalities in Japan occurs during road crossings, partly due to the low rate of vehicles stopping at unsignalized crosswalks. One of these measures to prevent such accidents is the installation of road markings. Road markings have an advantage that they can be installed at a low cost, but there is also a problem that the markings and color schemes are not standardized across regions. In this study, driving experiments using a driving simulator was conducted to clarify the effects of the proposed road markings on drivers’ driving behavior and driving awareness when approaching an unsignalized crosswalk, using “Slow Down”, “Be Careful of Pedestrians”, colored pavement, and no countermeasure. In addition, factor analysis was conducted to understand the influence of the countermeasures on drivers’ driving behavior, and it was shown that drivers mainly shift to deceleration behavior when they see the road markings and that the visibility of the road markings affects the promotion of deceleration behavior.

This research developed a fast and robust parameter estimation algorithm for activity simulators. The algorithm enables the real-time parameter estimation which is necessary especially in unstable situations like a recovery-period from a disaster when activity parameters change considerably. The developed algorithm is based on surrogate-based optimization, which can save much time for optimization with the help of machine learning. We proposed and validated three expansions of the existing algorithm. The study applied our proposed algorithm to estimate the parameters of a multi-agent activity simulator using the data collected during the 2016 Kumamoto Earthquake. The numerical example showed that the proposed algorithm could estimate well-calibrated parameters within ten minutes after the observed data arrived, while the conventional GA method takes nearly one day to estimate parameters. Additionally, the parameters updated every hour improved the accuracy of the prediction of the populations in 1 km square zones compared with the prediction by less updated parameters.

In our country automobile traffic census is generally conducted manually. However, the shortage of surveyors is becoming the norm due to the declining birthrate and aging population, and there is a need to make surveys more efficient and labor-saving. As a solution to this problem, AI-based traffic census has been attracting attention. But it is impossible to identify the same vehicle because most of the technology target cross-sectional traffic volume measurement. Therefore, it is thought that the same vehicle can be identified by recognizing the letters on the license plate. In this study, first, we devised a method for extracting and recognizing a series of designated license plate numbers and verified its accuracy. The devised method was then applied to video images taken between the two locations to verify whether the same vehicle could be identified. As a result, revealed the possibility of identifying identical vehicles, which could not be measured with existing technology.

The road network in Japan has been steadily developed until today, however, some issues are remaining such as low travel speeds on main roads and through-passing traffics into minor streets in residential areas. This is considered because of insufficient consideration of the hierarchical function of each road at the planning stage. For this reason, it is necessary to reconsider the road network from the functional hierarchy perspective of whether each road is hierarchically being used according to its target function. Therefore, the objective of this study is to clarify influencing factors of the functional hierarchy based on questionnaire surveys for experts in road design and planning. As a result, it was found that road conditions such as the trip length and the travel speed affect the functional hierarchy, and the combination of them is necessary for proper evaluation of the functional hierarchy.

This paper proposes an assignment model of three subjects (Solo Driver (SD), Ridesharing Driver (RD) and Rider (R)) under the circumstances where the ride-sharing service is available. The proposed model seeks the spatio-temporal flows of three subjects so as to achieve system optimum condition. The ridesharing services requires riders to move with ridesharing drivers’ vehicle, and hence there is a possibility that riders may not be able to return home. In order to take into account such effects, we developed two models; i) ‘trip based model’ that assign users by trip, and ii) ‘activity based model’ that can take into account accommodates changes, and even cancellations, in the timing of the members of the three subjects. The proposed models are applied to a toy network. We firstly perform sensitivity analysis of parameters in the trip based model. We then conduct a comparative analysis of the trip based model and the activity based model.

In recent years, an increase in the number of fatal accidents involving elderly drivers has become a problem. As a countermeasure, there is a license return policy, but there is a risk that it will bring about a drastic change in the lives of the elderly. Therefore, this research focused on "compensatory driving", which considers the deterioration of one's own driving skills and drives safely instead of returning the driver's license. A questionnaire survey was conducted on the actual situation of compensatory driving and measures to promote it, especially at night and on rainy days when fatal accidents are more likely to occur. Promotional measures include 1) providing information about the dangers of driving at night and in the rain, and 2) looking back on one's own driving and recognizing the decline in driving skills. As a result, we grasped the relationship between driving frequency, driving confidence, and the rate of those who performed compensatory driving. Both the information provision and the driving review influenced changes in intention to perform compensatory driving, and they increased the percentage of those who will perform compensatory driving statistically significantly.

The spread of advanced driver assistance systems, such as collision damage reduction brakes, is expected to improve the safety and smoothness of road traffic. On the other hand, there is concern in various quarters about the increase in collisions at the beginning of the works zone among the traffic accidents that have occurred in recent years. To address this problem, this study believes that driver-focused countermeasures should be prioritized, and attempted to devise and evaluate a regulatory information design for the beginning of the work zone that is easily visible and conveys instructions. Driving behavior data and interview data were collected from indoor driving experiments using a driving simulator. The results of the analysis suggest that a regulatory information design at the beginning of the work zone that includes a sign car with a red and white checkerboard background and a large arrow highlighting a lane change warning is expected to reduce the risk of accidents.

The concept of a functional hierarchical road network is important as one of the solutions to the road traffic problems that still occur in Japan, where the road network has been steadily developed. This study aims to propose a method to evaluate in which degree of functional hierarchy the road network is being used. In a virtual road network with three levels of hierarchy, the shortest routes are searched under the free flow conditions, and the road use characteristic indices are calculated such as trip lengths and use rates of each hierarchical level. Based on the conditions of the road network represented by these index values, a method to evaluate its functional hierarchy in four degrees is proposed. As a case study, the degree of hierarchy was calculated for the road networks in southern Gifu and western Shizuoka. The results revealed that the current evaluations of the functional hierarchy in these areas were low, and that the mid-level roads are required for improving the current situation.

In this study, the merging behaviors of vehicles due to a roadwork lane restriction and the behaviors of vehicles that were influenced by the merging vehicles at a three-lane outbound section of Chuo Expressway were analyzed based on video images framed by plural cameras. Conflicts extracted by car brake lights and by rapid changes in vehicle running trajectories as rough criteria were classified into six categories. The occurrence processes of conflicts in the six categories were analyzed in detail by estimated the running trajectories of merging vehicles and of the vehicles they influenced at every 1/30 second. As a result, high risks were estimated to be primarily caused by delays in finding lane reductions and by insufficient communication among related drivers. A Polygon Index for Conflicts (PIC) that evaluated risks by areas composed of polygons that consisted of the point of lane reduction and the positions of related vehicles at every 1/30 second as vertexes was proposed and validated.

This study aimed to show similarities between traffic volume fluctuations of a small street and the main road nearby to develop a method to estimate the traffic volume of a narrow street from that of the main road, where traffic volume was observed. As a result of the analysis, it was found out as follows: First, there was a high correlation between the traffic volume fluctuations of a small street and the main road nearby. Secondly, the correlation between a small street and the main road nearby was higher than that between small streets or main roads. Thirdly, the correlation between a narrow street and the main road nearby was higher than that between a narrow street and the main road more than 5km apart. In addition, the correlation between the two parallelly positioned was higher than that between the two intersectionally positioned in most cases. These findings show the possibility of estimating the traffic volume of a narrow street from that of the main road nearby.

Bicycles are expected to play a significant role on urban transportation for their various advantages, such as reducing environmental load and improving users’ health. However, road space for cycling in Japan is still insufficient and needs to be improved urgently. In addition, it will be essential for cycling to coexist with automated vehicles, which are expected to be realized in the future. Particularly, efficient curbside management is required to meet the demand for on-street parking. In this study, we focused on the space for picking up and dropping off from automated vehicles and the space for bicycles in an automated society and examined the road space where they can coexist. Traffic simulations were conducted for the organized road design patterns by changing the amount of vehicle traffic and stopping frequency. the smoothness and safety of traffic flow were evaluated to propose road spaces which are sufficient to various road characteristics based on quantitative evidence.

Understanding pedestrian route choice is essential for current urban planning needs, such as the liveliness of urban central areas. However, pedestrian routes are not explored by existing PT surveys. In this study, a route choice model was applied using commercially available mobile GPS data. The pedestrian route choice analysis was applied using data at a certain city with the aim of identifying the factors that influence route choice. When the Cross Nested Logit model was applied, it was shown that routes with characteristics such as "short route length, many stores nearby, and few right/left turns" were more likely to be selected. Pedestrian route selection characteristics similar to existing studies using GPS data with an acquisition time interval of 1 second were obtained. It was suggested that it is possible to grasp the route with the accuracy that can obtain the walking characteristics even using commercially available GPS, and that it is possible to analyze pedestrian flow lines even if using commercially available GPS data.

This study proposes the decomposition of the action-value in Markov Decision Process (MDP) for the decentralized network-wide signal control by the multi-agent reinforcement learning. The proposed decentralized control is theoretically proved to yield the optimum signal control for an entire network. Recent trends in the reinforcement learning for the signal control are the cooperative multi-agent system to capture interactions among multiple intersections so as to establish the system optimum control. However, cooperative arrangements still do not guarantee to yield the system optimum control. This study therefore reconsiders the structure of MDP and shows that the action-value in MDP can be decomposed so that the system optimum control is obtained in the decentralized fashion in which an individual intersection can decide its own action and update its action-value only using the state and action associated with itself. The proposed decentralized control is tested in an arterial signal control, and it is confirmed that the decentralized control successfully produces the system optimum signal control. Also, issues on control parameters obtained from the reinforcement learning are summarized.

This paper clarifies how forms the Personal Space of a resting person in a street space to a pedestrian. The experiments show that the Personal Space formed by a resting person tends to be different from the Personal Space investigated in previous research. Therefore, we defined it as "Dynamic Personal Space" as the Personal Space formed by a person resting in a sitting position concerning a pedestrian and analyzed the characteristics of this Personal Space in this paper. It was found that the formation of the resting person's Personal Space to pedestrians was related to the environment of the street space, such as "Location", "Posture", and " Body orientation". In addition, we were able to confirm the characteristic that the range of the resting person's Personal Space changes as these environmental conditions. In the future, we would like to investigation in the actual street space and we can identify the position where the boundary of a Dynamic Personal Space becomes zero, and we would like to conduct the experiment without bias in the gender of the subjects to clarify the characteristics of the individual distance for each attribute.

This study evaluated the precision of vehicle probe data used to estimate average delay times due to traffic signal control. The data that were collected and processed by a private company are to be used to evaluate a developing delay time feedback control system for traffic signals by traffic simulation. Average delay times as evaluation criteria were calculated based on the times when stop lines were crossed by vehicles that were matched by license plate numbers at a rate of 70 to 80% from video images acquired at 14 intersections on Komazawa-Dori Street and its vicinity during a morning peak. As a result, the sample rates of the probe data were a few percent to a dozen, and the root mean square errors and mean absolute errors of the average delay times were less than 60 and 40 seconds per vehicle respectively. On the other hand, the precision of the delay time estimates was potentially decreased due to probe vehicles that detoured atypically heavy traffic congestion. In addition, delays in transmission and processing of the probe data were estimated to be at most 5 minutes.

It is important to reveal a relationship of longitudinal alignment to bottleneck section to consider a mechanism of breakdown phenomena. First, this paper confirms that a ratio of the change of the longitudinal gradient is an appropriately alternative index of the radius of a curvature using traffic counter data and longitudinal alignment data on Tohoku Expressway. Second, a logistic regression model to express break-down in flow is calibrated using traffic flow rate and a ratio of the change of the longitudinal gradient of adjacent sags on Tomei, Kanetsu, Joban as well as Tohoku Expressways. Third, the authors explore alternative models where a ratio of the change of the longitudinal gradient is non-linearly transformed as the transformed ratio has a peak at a specific ratio. When the specific ratio is 0.5%/km, the fitness of the model is the best. This means that a sag whose ratio of the change of the longitudinal gradient at is moderate and whose adjacent sags are sharp, has a tendency to break down to congestion.

This research focuses on the low compliance rate of stoppages at an entrance of parking facilities stipulated by the Road Traffic Act. It is examined the impact of increased stoppage rate at an entrance of parking facilities to the traffic flows if autonomous vehicles were socialized. The survey was conducted to reveal the most of the vehicles were passing slowly at present, while the number of temporary stops when there are no pedestrians was low at entrances and exits of many parking facilities. In addition to the result of a micro-traffic flow simulation, it is clear that increasing autonomous vehicles caused improving the stoppage ratio and increasing the passing time of straight on the surfaced roads. In addition, the pedestrian traffic volumes were also affected to the delay time of traffic flows. Therefore, it is necessary to implement a traffic assessment at the entrance and exit of the parking facilities.

In response to the traffic fatal accident in Kameoka City, Kyoto Prefecture in 2012, the three ministries informed prefectural governors and other relevant agencies to cooperate with each other to conduct safety inspections and ensure the safety of school routes in their jurisdictions, and joint inspections were conducted in cities, wards, towns, and villages nationwide. Based on the results of the joint inspections, the government recommended that the PDCA cycle based on the School Road Traffic Safety Program be continued, as this would lead to improved safety on school routes. Although the program is thought to help prevent accidents from occurring in the first place, the government again requested the implementation of joint inspections in response to a fatal traffic accident in Yachimata city, Chiba Prefecture, in 2021. In this study, we investigated the involvement of prefectures and the joint inspections and programs implemented by cities, wards, towns, and villages, based on a website survey and questionnaire survey, in response to the government's request for joint inspections to be conducted by prefectures and other entities.

The traffic information design for mitigating congestion has been studied under the decentralized traffic management. Iwase et al. focused on oscillatory phenomena called hunting in traffic flow and proposed the concept of “self-fulfilling signal” as traffic information design to suppress hunting. Ukai and Fukuda conducted laboratory experiments to examine the suppression of hunting by providing self-fulfilling signal on drivers. However, the experiments made was small and insufficient for a proper understanding of the effect of the signal. This study conducted a large-scale interactive online experiment to analyze the inhibitory effect of self-fulfilling signal on hunting. The experimental results showed that hunting was well suppressed when the diversity of traffic information based on self-fulfilling signals was large enough. The individual route choice data was further investigated to understand information compliance behavior of drivers using the mixed logit model.

In efforts to increase drivers yielding right-of-way laws at uncontrolled pedestrian crosswalks, few studies have focused on road traffic factors associated with drivers yielding behavior. In this study, a field observation was conducted to investigate drivers yielding behavior at an untreated road improvement location in Tokyo. In total, 666 pedestrian-vehicle encounters at the observation site were observed over the three days. The results showed that pedestrians were more likely to be yielded when they were waiting on the near-side sidewalk than the far-side sidewalk. It was also found that yielding rate of pedestrians behavior in the pre-crossing phase was lower when waiting on the sidewalk than on the roadway. In regard to road traffic environment, drivers yielding was lower when the signal indication was green than red at the crosswalk near signalized intersections. Finally, a lower rate of yielding to pedestrians occurred when there were following vehicles in the same direction or oncoming vehicles on the opposite side of the road.

The accident in Yachimata City, Chiba Prefecture, in June 2021 has once again made us aware of the need for countermeasures against accidents involving elementary school students on their way to and from school. While the number of school crossing bridges installed along school routes is increasing, it is expected that the number of cases where pedestrian crossings are designated as school routes will increase in the future. In this study, we investigated the crossing behavior of elementary school students at pedestrian crossings with traffic signals. As a result, it was found that there were crosswalks where children crossed in a manner that could cross with motor vehicles even when adults did not cross in a manner that could cross with motor vehicles. The results also suggest that factors other than the road environment, such as the priority of group behavior, lack of attention, and adult violations, may be influencing this behavior.

This study develops an urban economic analysis method that can measure changes in the spatial distribution of economic activity due to increased risk of flooding, which contributes to the assessment of the long-term impact of transportation infrastructure development. For this purpose, we construct a model that can take into account the risk of flooding, based on the models of previous studies that consider agglomeration economies. It is shown that the model has a potential function that enables large-scale numerical calculations. In addition, we present a systematic parameter setting method that enables analysis in real space. Furthermore, we conduct a counterfactual experiment on the increased risk of flooding in the Kanazawa urban employment zone. We confirm that the results reflect the characteristics of the model and that it is possible to measure the long-term impact of transportation infrastructure development. We also discuss measures to reduce flood damage based on the changes in the flooded area population.

Uncertainty of emergency situations such as disasters result in very different behaviors of evacuees from that of ordinary situations. The decision process follows a different structure, and even the choice set of each decision maker may change drastically. The uncertainty in future choice set also plays a major role in making the series of decisions, but past literatures overlook the effect of assessing choice set on preference. In this research, we constructed a dynamic discrete choice model that expresses dynamic and probabilistic change of choice set. At the choice stage, each decision maker chooses his/her next choice from a non-compensatory choice set. Since choice set of the uncertain future cannot be determined at the choice stage, a probabilistic and compensatory structure is implemented using the survival probability of alternatives. Our proposed model is applied to empirical data of location choice during a heavy disaster, which proved the validity of the model.

This study focuses on the long-term closure of Chugoku Expressway due to its major renewal project. It attempts to analyze the willingness to detour to the Shin-Meishin Expressway, the recommended detour route, and to evaluate factors affecting route choice on the expressway. A discriminant analysis using discrete variables was conducted to analyze the characteristics of drivers who were willing in detouring to the Shin-Meishin Expressway. In the evaluation of route selection factors, weights of the route evaluation criteria for each subject were calculated using AHP, and a cluster analysis was conducted to categorize the weights of the route evaluation criteria. In addition, by showing the relationship between the variables which influence the willingness in detouring to Shin-Meishin Expressway and the cluster analysis of the weights of the evaluation criteria for route choice, a method to expect the effectiveness of measures for drivers who need detouring is proposed in the paper.

Recently, human-centered and local-context-oriented streets have been more focused, in addition to the conventional vehicle-oriented street smoothing traffic flow of vehicles. Especially in the new normal era, new local design is required to enhance people's sense of self-efficacy and happiness. The quality of pedestrian space from the viewpoint of walkability and lingerability needs to be improved to realize human-centered transportation machizukuri. Therefore, this study aims to evaluate the spatial performance of the streets and analyze the factors influencing walkability and lingerability by utilizing image recognition technology. An image recognition using fine-tuned AI model was conducted to evaluate five pedestrian spaces. Then, spatial elements influencing spatial performance were visualized and identified by using Grad-CAM. After that, respective pedestrian space is positioned in 2D coordinates with walkability and lingerability axes, and the spatial performances of close streets are compared. Finally, the potential contribution of the proposed model to improve the walkability and lingerability of streets was summarized.

Although the number of traffic accidents and fatalities in Japan has been decreasing year by year, about 60% of pedestrians are involved in fatal accidents due to pedestrian crossing violations, and it is necessary to clarify the mechanism of crossing outside the pedestrian crossing. Since no existing study has clarified the mechanism of danger occurrence at locations where crossing outside the pedestrian crossing occurs despite the proximity of a pedestrian crossing, this study conducted a video observation survey at an unsignalized pedestrian crossing and analyzed the behavior of both pedestrians and drivers. The results showed that pedestrians crossing outside the pedestrian crossing tended not to check left and not to bail before crossing. It was also found that pedestrians tended to use crosswalks when automobile traffic was flowing in the far lane. In terms of the roadway environment, it was found that the walkable area is a factor in crossing outside the pedestrian crossing, suggesting the need to improve the location of pedestrian crosswalks and the walkable area together.

In this study, we proposed a cooperative control method using right-turn lane and conducted simulation experiments, assuming that the vehicles are automatic connected vehicles which share the information about speed, position, and route. In the method, vehicles that are not in a conflict relationship can enter an intersection simultaneously, and the group of vehicles is called a "generation”. The generation is updated each time a vehicle enters the intersection. It is also possible for a vehicle to go straight out of the right-turn lane to accommodate different intersection inflow timing. Simulation experiments were conducted by varying the right/left turn rate of the traffic flow. The results showed the cooperative control was as efficient as or more efficient than signal control when the generation was updated and vehicles could proceed straight ahead from the right-turn lane. The results also showed that the efficiency tended to increase as the right/left turn rate increased.

In developing countries, the rapid expansion of urban areas along with economic development results in various urban problems such as traffic jams. In order to solve these problems, appropriate urban planning based on the updated travel demand forecasts considering rapid growth is necessary. However, it is difficult to conduct person trip survey frequently. In this context, the planners expect the use of GPS probe data as an alternative travel survey. This study aims to propose the availability of GPS probe data for that purpose. We detected trips, then derived travel purposes and travel modes from the commercially available data in Phnom Penh, Cambodia. Through comparison with existing person trip surveys, we summarized the characteristics and challenging issue of GPS data, and then validated the possibility of replacing and complementation of existing person trip surveys.

This study aims to develop a next-generation traffic signal control system, which adapts to the observed real-time traffic states obtained by probe observations. In this paper, 4 types of traffic states, for which different control policies should be applied for efficient signal control, are defined. Then, a method is proposed to classify the intersection traffic states into the 4 types based on the real-time probe travel time. Specifically, the probe travel time is utilized to determine whether or not the discharge flow rate of the intersection is decreased and whether or not congestion appears at each signal phase; then the current traffic states are classified into one of the 4 states based on these results. The accuracy of the detection of discharge flow rate reduction is verified using actual observation data, and differences in the accuracy are observed at different intersections. In the future study, factors that affect the classification accuracy will be examined to improve the accuracy.

Recently, unsignalized crossing facilities including two-stage crossings are focused to reduce crossing outside of crosswalks (jaywalking) and to increase safe crossing opportunities. However, the quantitative effect of such facilities by changing the location and type for reducing jaywalking is not clear. This research conducted a virtual reality experiment to analyze the impact of locations and types of unsignalized crossing facilities on individual pedestrians’ crossing location choice. The results revealed that pedestrians chose unsignalized crossing facilities as their crossing locations, especially when zebra markings were available rather than when the facilities were only refuge islands without pavement markings. Furthermore, as compared to the ratio of jaywalking declared in advance of the experiment, the ratio of actual jaywalking was higher. The sequential choice model showed that the existence of the vehicles when pedestrians arrive at candidate locations of crossing influences jaywalking behavior. This would make the difference between pre-declared and actual crossing location choices.

Although autonomous vehicles (AVs) are expected to improve the safety and speed of traffic flow, traffic speed may decrease because of the mixture of AVs, manually driven vehicles, and public transportation. In this study, we analyze the impact of the mixture on traffic flow through a multimodal micro traffic simulator. We conducted a traffic survey on vehicles, buses, and trams in downtown Kumamoto, Japan. Subsequently, the survey results were used to develop a multimodal traffic simulator used to analyze the effect of the mixture of vehicles and the settings of parameters in the driving-behavior model on multimodal traffic flow. Simulation results revealed that the increased share of AVs improved the average speed of traffic flow. However, the average speed of trams decreased owing to increased AVs. Therefore, we also conducted a simulation to verify that a change in the intersection structure and signal control can prevent the decrease in the speed of trams.

This paper aims to show an overview of the projects’ features of consolidated transportation of passengers and cargos during 2011 - 2022 in Japan, based on the data published by concerned companies and local governments. The major results are as follows: 1) The projects were significantly increased after the revision of the regulation of Road Transportation Act in 2017, accompanied with diversification of project objectives and transportation modes. 2) The projects to promote efficient cargo transportation and/or to sustain public transportation account large share of whole projects, which is more significant in depopulated municipalities. 3) The projects cooperated with local administrations are minority. 4) Although the features of the projects operated by the major three companies are basically same to those of whole projects, major three companies tend to operate more projects in depopulated municipalities. 5) Typical features of the projects for cargo and/or public transportation in depopulated municipalities by three major companies are one-way, once a day consolidated service among 5-12 ordinary transportation services at 20-40km distance routes.

This study conducts an empirical analysis of "capacity drop (CD)" phenomenon at multiple sag and tunnel bottlenecks on expressway in Japan based on the recent advanced continuum traffic flow theory. Specifically, jam sections (and jam events) were selected with reference to jam event data compiled by expressway companies in Japan during the year 2019, and the model was calibrated using ETC2.0 probe data and vehicle detector data. Then we estimate the safe time gap, acceleration parameter at the downstream end of the bottleneck, and other parameters. We then show the following two results with confirmation the generality of the theory and the trends of the estimated parameters. Firstly, where in the sag the bottleneck section appears is various. Secondly, the safety time gap between vehicles determines the absolute level of the Queue Discharge Flow rate (QDF).

This study compared the observed path flows and the estimated path flows for the period before and after the bus exclusive lane regulation was implemented in Miyazaki City. The aggregated probe vehicle data was utilized as the observed path flows. As a result, we confirmed the common tendency in February, after the implementation of the bus exclusive lane regulation, that the path flows not only in the bus exclusive lane regulation section but also in the vicinity of the regulation section decreased during the regulation period while the path flows in the regulation section increased during the period before and after the regulation period. We also confirmed the different tendency that the observed path flows on toll roads located further away from the regulation section increased while the estimated path flows on that section did not increased during the regulated period after the implementation of the bus exclusive lane regulation.

Elderly license holders now account for more than 23% of the total number of license holders in Japan, and the percentage of traffic accidents by the elderly is increasing. It is important to establish traffic safety measures for the elderly. In this study, a cooperative simulator was used to compare the safety of elderly and young drivers at intersection when a car turns left and a bicycle goes straight in three paths (running on the road, shortcut on the sidewalk, and running on the sidewalk). The following findings were obtained: 1. The elderly tended to have a greater distance to the nearest point of approach in all travel paths of a bicycle; 2. minimum TTC is smaller for younger drivers.; 3. Younger drivers are more likely than elderly drivers to feel safe; 4. Comparing anxiety for cyclists and drivers, the percentage of anxiety in cyclists was lower than that in drivers.

Traffic congestion on ordinary roads including national highways remains an important problem in traffic engineering. Although various kinds of big data have been accumulated through the development of ICT in recent years, their utilization is not necessarily sufficient. In this study, traffic conditions are investigated mainly at signalized intersections, which are the main bottlenecks on ordinary roads, using ETC2.0 probe data, and road structure and land use data are organized using other big data such as digital national land information. Integrating and using these multiple data, the factors of traffic congestion on ordinary roads are quantitatively analyzed. Machine learning model was constructed using XGBoost and interpreted by SHAP. As a result, it was confirmed that one-hour traffic volume and other factors affect the occurrence of congestion and left-turn lane length and other factors may lead to the reduction of congestion. In addition, the effect of extending the right-turn lane on reducing congestion was confirmed by predicting the effect of congestion countermeasures focusing on individual signalized intersections.

As a basic study on the contribution of traffic parks to traffic safety education, this study aims to understand the influence of children's experience of using traffic parks on their knowledge of traffic rules. Specifically, we analyze the relationship between the frequency of using a traffic park and the knowledge of traffic rules through a questionnaire survey targeting children and their parents in Toyohashi city. The results show that the children's experience of using traffic parks before entering an elementary school has a positive influence on their knowledge of traffic rules: compared to children who had never used the traffic park before entering school, children who had used it are more likely to answer to stop at a "stop" sign when riding a bicycle and more likely to answer to stop at uncontrolled crosswalk to yield the right of way to pedestrians. Regarding “riding a bicycle on left-hand side”, however, the frequency of using the traffic park before entering an elementary school increases the awareness of the rule but does not lead them to take the behavior.

This paper explores the improvement mechanisms of traffic performance such as the traffic capacity before congestion and the queue discharge flow rate by the audible speed recovery information provision introduced as a congestion measure at the Kobotoke tunnel of Chuo Expressway, Japan. Specifically, we apply the recent advanced traffic flow theory for endogenously describing the capacity drop phenomenon at sag and tunnel bottlenecks (Jin, 2018; Wada et al., 2020) to data measured before and after the introduction of the system and compare results. We then show that (i) the improvements in traffic capacity before congestion and the queue discharge flow rate can be explained by a common mechanism “reduction of safety time gap” within the bottleneck section; (ii) the improvement in acceleration during the congestion at the end of the bottleneck section does not result in that in queue discharge flow rate, and thereby there is room for further improvement.

Many local governments in Japan have introduced community bus services to improve the accessibility of public transport and the mobility of vulnerable road users. Recently, however, many local governments are making a great effort to increase community bus users, improve profitability and reduce subsidies because of financial difficulties. Therefore, identifying the factors that influence community bus origin-destination (OD) demand is one of the most important research issues. This study mainly aims to identify the factors that influence community bus OD demand by focusing on (1) the characteristics of catchment areas of community bus stops, (2) the distances between community bus stops and other public transport stops, and (3) the distance between origin and destination community bus stops. As a result, we found that the following are important to increase community bus OD demand: (1) integrate community bus services and other public transports, (2) install community bus stops in the areas where many urban amenities are provided, (3) provide many urban amenities in community bus stop catchment areas, and (4) shorten the distance between origin and destination community bus stops in the areas where many amenities are provided.

In the recent years, bicycles have played an important role as a familiar and useful transportation. In addition, for bicycles to ride safely, it is important to reduce the chances of accident between bicycle and automobile, between bicycles, and between bicycle and pedestrian. In this study, targeting road with bicycle traffic spaces in Aichi Prefecture, we examined the impact on bicycle accidents from changes in the number of accidents and factor analysis. As a result, it cannot confirm an improvement effect in the change in the number of bicycle accidents for one year before and after the improvement of the bicycle traffic space. However, by the factor analysis, it was found that the number of bicycle accidents decreased when it was improvement of the bicycle traffic space. In addition, we pointed to the number of accidents is increase in which bicycles are the primary party in the improvement of bicycle track within carriageway, and in which bicycles are the secondary party in the improvement of bicycle lane.

The introduction of self-driving cars, electric vehicles, and the spread of sharing services is expected to significantly change the automobile usage environment. In order to facilitate smooth transfers between various modes of transportation, it will be necessary to develop transportation hubs that have multiple functions such as transfers and EV charging, in addition to the development of transportation systems that include terminal transportation connecting stations to residential areas. In this study, examples of mobility hubs were organized in order to understand the current status of transportation hubs. And the purpose of this study is to quantitatively evaluate the impact of the placement of transportation hubs on society. As the result of the simulation, it was quantitatively showed that the more transportation hubs are located, the greater the impact of carpooling becomes, and that there is the trade-off relationship among the indicators such as the distance traveled and the cost of installing vehicles.

Many researches on capacity bottleneck phenomena in access-controlled sections point out that reaction-delay on car-following behavior is one major factor in break-down phenomena. This study estimates distributions of parameters in car-following behavior models proposed by Chandler et al. and Gazis et al. using Zen Traffic Data (ZTD) produced by Hanshin Expressway. The Chandler’s model and the Gazis’ model are chosen as representatives of a linear model and non-linear model, respectively. To discuss on spatial heterogeneity on distributions of parameters, the stretch of the data collection is divided into four sections every 500 m. This study applies the method of weighted least square, where an order of magnitude larger weight of a focused section than the other sections is given. The estimated distributions of the reaction-delay in the model by Gazis et al. differ from section to section. However, the distribution on the bottleneck section shifts to shorter than the other sections. In congested flow state, the reaction delay at downstream of the bottleneck section is larger than the other sections.

As a driverless vehicle become widely used, pedestrians cannot communicate with the driverless vehicle using eye contact or gesture. Therefore, pedestrians cannot understand behavior of the driverless vehicle. In the present study, we attempted to verify the effect of e-HMI to enable two-way communications between the pedestrians and the driverless vehicles at unsignalized crosswalk to convey the intention of approaching driverless vehicles. Three types of e-HMI were proposed as follows; "Icon" is that encourage pedestrians to cross the street, "Front brake light" is that signal deceleration, and "Gaze-based Icon" is that take two-way communication into account. We conducted with 14 participants to evaluate the deceleration recognition time and crossing decision time for approaching driverless vehicles under seven information conditions displayed by the driverless vehicles, which included no information and combined three proposed types of e-HMIS. Results of the experiment, the deceleration recognition times of the approaching vehicle by displaying “Front brake light” were shorter than those by other information conditions. The crossing decision times in case of “Gaze-based Icon” were shorter than those for other information.

Since road markings occupy a large portion of the driver's field of vision, they are effective in conveying messages, and a variety of character markings have been laid down as nonlegal markings. We are developing visual 3D character markings, the advantage of which is their high legibility. However, we have not made a quantitative comparison with 2D character markings. Therefore, for the purpose of accumulating basic data, this study, we conducted an experiment to evaluate the recognizability and legibility from the driver's viewpoint for each type of road character markings, including 3D markings. The results showed that recognition distances ranged from 63 to 75 m and were less influenced by the type of road surface markings. Legibility distance was affected by the type of markings in the range of 19 to 42 m, with 3D markings having a greater distance of 13 to 20 m compared to 2D markings.