Land use and traffic collisions: A link-attribute analysis using Empirical Bayes method

Identification of Vehicle-Pedestrian Collision Hotspots at the Micro-Level Using Network Kernel Density Estimation and Random Forests: A Case Study in Shanghai, China

“…The mean feature importance of the two variables among the five samples was 0.3 and 0.15, respectively, indicating their ability to predict the occurrence of vehicle-pedestrian collisions. As mentioned before, previous studies have already investigated the relationship between land use characteristics and the occurrence of traffic crashes involving pedestrians [30,51], and it was found that vehicle-pedestrian collisions were more likely to happen in commercial areas. In this study, the commercial land was further segmented into different types of places such as retail shops and restaurants.…”

“…In the absence of detailed pedestrian flow data, we employed a set of variables that comprehensively reflect characteristics of pedestrian flow. As different uses of land may suggest diverse activities of human beings, which influence different features of pedestrian flow [51][52][53], we employed land use data to reflect the spatial variation in pedestrian exposure. Point of Interest (POI) As mentioned earlier, to determine the potential for vehicle-pedestrian collision reduction, the vehicle-pedestrian collision density should be modeled by RF with variables that describe both vehicle and pedestrian volume.…”

A Combined Approach to Address Road Traffic Crashes beyond Cities: Hot Zone Identification and Countermeasures in Indonesia

“…Various spatial statistics of local (i.e., Kernel Density Estimation, K-function, Local Moran's I, Local Getis and Ord G and spatial autocorrelation) and global measures (i.e., Quadrant methods and K-functions) are used to illustrate the clustering pattern of road traffic crashes [12]. In relation, some studies suggest using Empirical Bayes and Full Bayes methods to reduce the regression-to-mean (RTM) bias [13]. Although there is no consensus on the best approach to detect hot zones, the use of local statistical measures can generally generate better results, as global measures only capture the average spatial pattern in an entire region [3].…”

“…In traffic crash analysis, many empirical findings indicate that the spatial pattern of traffic collisions can vary according to a wide range of factors. The distribution of hot spots or hot zones can, for example, vary based on temporal change [20], road types (e.g., highways and urban roads) [21], land uses [13] and pedestrians or other vulnerable groups [22]. Putting the spatial context into the perspectives of hotspot and hot zone identification is therefore conducive to formulating geographical-specific and effective road safety measures.…”

“…A hot spot generally refers to a dangerous road location where the actual crash rate is higher than the critical crash rate [2][3][4], whereas a hot zone refers to a dangerous road location made of two or more continuous segments that contain an actual crash rate higher than a threshold value [2,[5][6][7]. There is a rich literature that examines the spatial statistics of identifying hot spots and hot zones of traffic crashes [6][7][8][9][10][11][12][13]. However, the application of suitable identification methods is largely dependent on the spatial context.…”