Analyzing vehicle-pedestrian interactions: Combining data cube structure and predictive collision risk estimation model.

Road traffic accidents are a severe threat to human lives, particularly to vulnerable road users (VRUs) such as pedestrians causing premature deaths. Therefore, it is necessary to devise systems to prevent accidents in advance and respond proactively, using potential risky situations as one of the surrogate safety measurements. This study introduces a new concept of a pedestrian safety system that combines the field and the centralized processes. The system can warn of upcoming risks immediately in the field and improve the safety of risk-frequent areas by assessing the safety levels of roads without actual collisions. In particular, this study focuses on the latter by introducing a new analytical framework for a crosswalk safety assessment with various behaviors of vehicles/pedestrians and environmental features. We obtain these behavioral features from actual traffic video footages in the city with complete automatic processing. The proposed framework mainly analyzes these behaviors in multi-dimensional perspectives by constructing a data cube structure, which combines the Long Short-Term Memory (LSTM)-based predictive collision risk (PCR) estimation model and the on-line analytical processing (OLAP) operations. From the PCR estimation model, we categorize the severity of risks as four levels; "relatively safe," "caution," "warning," and "danger," and apply the proposed framework to assess the crosswalk safety with behavioral features. With the proposed framework, the various descriptive results are harvested, but we aim at conducting analysis based on two scenarios in our analytic experiments; the movement patterns of vehicles and pedestrians by road environment and the relationships between risk levels and car speeds. Consequently, the proposed framework can support decision-makers (e.g., urban planners, safety administrators) by providing the valuable information to improve pedestrian safety for future accidents, and it can help us better understand cars' and pedestrians' proactive behavior near the crosswalks. In order to confirm the feasibility and applicability of the proposed framework, we implement and apply it to actual operating CCTVs in Osan City, Republic of Korea.

SafetyCube: Framework for potential pedestrian risk analysis using multi-dimensional OLAP.

In the past decade, the number of road traffic accidents and fatalities has remained about the same level. One of strategies to protect vulnerable road users (VRUs) is to analyze the factors that cause traffic accident and then to deploy safety facilities. However, most traffic safety systems currently in operation rely on historical data, which is post-facto approach. Thus, it is necessary to prevent accident in advance and to respond in proactive manner before the accident. In this study, we propose a framework for potential pedestrian risk analysis using a multi-dimensional on-line analytical processing (OLAP), called SafetyCube, which enables decision-makers to understand the situations by scrutinizing interactive behaviors between vehicle and pedestrian. First, we collect the behavioral features of traffic-related objects (e.g., vehicles and pedestrians) extracted from closed circuit televisions (CCTVs) deployed on crosswalks throughout the overall urban, and accumulate them in a data warehouse over an extended period in order to construct a data cube model. Then, we conduct comprehensive analyses in multi-dimensional perspective using OLAP operations by varying the abstraction levels. Our analytical experiments are based on three scenarios, and the results show that the vehicle's movement patterns before entering the crosswalk, patterns of changes in speed of vehicles approaching to pedestrians, and so on. Through these results from the proposed analytical system, decision-makers can gain a better understanding of how the vehicles and pedestrians behave near the crosswalk by visualizing their interactions. Further, these insights would be reflected to improve the road environment safer. In order to validate the feasibility and applicability of the proposed system, we apply it to various crosswalks in Osan city, South Korea.

Design of reinforcement learning for perimeter control using network transmission model based macroscopic traffic simulation.

Perimeter control is an emerging alternative for traffic signal control, which regulates the traffic flows on the periphery of a road network. Some model-based approaches have been suggested earlier for the optimization of perimeter control based on macroscopic fundamental diagrams (MFDs). However, there are several limitations when considering their application to a large-scale urban area because the model-based approaches may not be scalable to multiple regions and inappropriate for handling various effects caused by the shape change of MFDs. Therefore, we propose a model-free and data-driven approach that combines reinforcement learning (RL) with the macroscopic traffic simulation based on the recently developed network transmission model. First, we design four perimeter control models with different macroscopic traffic variables and parametrizations. Then, we validate the proposed models by evaluating their performances with the test demand scenarios at different levels. The validation results show that the model containing travel demand information adapts to a new demand scenario better than the model containing only density-related factors.

Measuring health of highway network configuration against dynamic Origin-Destination demand network using weighted complex network analysis.

Ideal configuration or layout of highways should resemble the actual demands for the roads represented by Origin-Destination (OD) information. It would be beneficial if existing highways can be evaluated for their configurational fitness against the current demands, and newly planned highways can carefully be designed in terms of their layouts and topologies that would reflect the demands. Analysis techniques used for complex networks in the matured field of network theory can be applied for the highway layout health monitoring against the current OD information. This paper proposes a methodology of measuring the fitness of existing highways by comparing their structural configuration against conceptual OD networks using well-established techniques in network theory for complex networks. In the first phase, this paper conducts an empirical analysis and finds that both structural highway network and OD network follow the "power law" distribution as they are weighted by capacity and traffic volume respectively. It is also found that the power law coefficient of the OD network dynamically changes throughout the day and week. In the second phase, a noble methodology of weighting and measuring the health, of structural highway networks against OD networks by means of comparing their power law coefficients is proposed. It is found that the proposed method is effective at detecting deviations from ideal structural configurations associated with actual demands.

Correction: Enhancing healthcare accessibility measurements using GIS: A case study in Seoul, Korea.

[This corrects the article DOI: 10.1371/journal.pone.0193013.].

Enhancing healthcare accessibility measurements using GIS: A case study in Seoul, Korea.

With recent aging demographic trends, the needs for enhancing geo-spatial analysis capabilities and monitoring the status of accessibilities of its citizens with healthcare services have increased. The accessibility to healthcare is determined not only by geographic distances to service locations, but also includes travel time, available modes of transportation, and departure time. Having access to the latest and accurate information regarding the healthcare accessibility allows the municipal government to plan for improvements, including expansion of healthcare infrastructure, effective labor distribution, alternative healthcare options for the regions with low accessibilities, and redesigning the public transportation routes and schedules. This paper proposes a new method named, Seoul Enhanced 2-Step Floating Catchment Area (SE2SFCA), which is customized for the city of Seoul, where population density is higher and the average distance between healthcare-service locations tends to be shorter than the typical North American or European cities. The proposed method of SE2SFCA is found to be realistic and effective in determining the weak accessibility regions. It resolves the over-estimation issues of the past, arising from the assignment of high healthcare accessibility for the regions with large hospitals and high density of population and hospitals.

An Analytical Planning Model to Estimate the Optimal Density of Charging Stations for Electric Vehicles.

The charging infrastructure location problem is becoming more significant due to the extensive adoption of electric vehicles. Efficient charging station planning can solve deeply rooted problems, such as driving-range anxiety and the stagnation of new electric vehicle consumers. In the initial stage of introducing electric vehicles, the allocation of charging stations is difficult to determine due to the uncertainty of candidate sites and unidentified charging demands, which are determined by diverse variables. This paper introduces the Estimating the Required Density of EV Charging (ERDEC) stations model, which is an analytical approach to estimating the optimal density of charging stations for certain urban areas, which are subsequently aggregated to city level planning. The optimal charging station's density is derived to minimize the total cost. A numerical study is conducted to obtain the correlations among the various parameters in the proposed model, such as regional parameters, technological parameters and coefficient factors. To investigate the effect of technological advances, the corresponding changes in the optimal density and total cost are also examined by various combinations of technological parameters. Daejeon city in South Korea is selected for the case study to examine the applicability of the model to real-world problems. With real taxi trajectory data, the optimal density map of charging stations is generated. These results can provide the optimal number of chargers for driving without driving-range anxiety. In the initial planning phase of installing charging infrastructure, the proposed model can be applied to a relatively extensive area to encourage the usage of electric vehicles, especially areas that lack information, such as exact candidate sites for charging stations and other data related with electric vehicles. The methods and results of this paper can serve as a planning guideline to facilitate the extensive adoption of electric vehicles.

Intelligent advisory speed limit dedication in highway using VANET.

Variable speed limits (VSLs) as a mean for enhancing road traffic safety are studied for decades to modify the speed limit based on the prevailing road circumstances. In this study the pros and cons of VSL systems and their effects on traffic controlling efficiency are summarized. Despite the potential effectiveness of utilizing VSLs, we have witnessed that the effectiveness of this system is impacted by factors such as VSL control strategy used and the level of driver compliance. Hence, the proposed approach called Intelligent Advisory Speed Limit Dedication (IASLD) as the novel VSL control strategy which considers the driver compliance aims to improve the traffic flow and occupancy of vehicles in addition to amelioration of vehicle's travel times. The IASLD provides the advisory speed limit for each vehicle exclusively based on the vehicle's characteristics including the vehicle type, size, and safety capabilities as well as traffic and weather conditions. The proposed approach takes advantage of vehicular ad hoc network (VANET) to accelerate its performance, in the way that simulation results demonstrate the reduction of incident detection time up to 31.2% in comparison with traditional VSL strategy. The simulation results similarly indicate the improvement of traffic flow efficiency, occupancy, and travel time in different conditions.

Evaluating the performance of network screening methods for detecting high collision concentration locations on highways.

This paper documents findings from evaluating performances of three different methods for segmenting freeway sites for the purpose of identifying high collision concentration locations: Sliding Moving Window (SMW), Peak Searching (PS) and Continuous Risk Profile (CRP). The traffic collision data from sites segmented in each method under two different roadway definitions were used to estimate excess expected average crash frequency with Empirical Bayes adjustment with respect to two different sets of Safety Performance Functions (SPFs). The estimates from each of the methods were then used to prioritize the detected sites for safety investigation and these lists were compared with previously confirmed high collision concentration locations (or hot spots). The input requirements for each of three methods were identical, yet their performance markedly varied. The findings revealed that CRP method has the lowest false positive (i.e., requiring a site for safety investigation while it is not needed) rate. The performances of SMW and PS significantly varied when different sets of SPFs were used while that of CRP was less affected.

Impact of traffic states on freeway crash involvement rates.

Freeway traffic accidents are complicated events that are influenced by multiple factors including roadway geometry, drivers' behavior, traffic conditions and environmental factors. Among the various factors, crash occurrence on freeways is supposed to be strongly influenced by the traffic states representing driving situations that are changed by road geometry and cause the change of drivers' behavior. This paper proposes a methodology to investigate the relationship between traffic states and crash involvements on the freeway. First, we defined section-based traffic states: free flow (FF), back of queue (BQ), bottleneck front (BN) and congestion (CT) according to their distinctive patterns; and traffic states of each freeway section are determined based on actual measurements of traffic data from upstream and downstream ends of the section. Next, freeway crash data are integrated with the traffic states of a freeway section using upstream and downstream traffic measurements. As an illustrative study to show the applicability, we applied the proposed method on a 32-mile section of I-880 freeway. By integrating freeway crash occurrence and traffic data over a three-year period, we obtained the crash involvement rate for each traffic state. The results show that crash involvement rate in BN, BQ, and CT states are approximately 5 times higher than the one in FF. The proposed method shows promise to be used for various safety performance measurement including hot spot identification and prediction of the number of crash involvements on freeway sections.