Exploring the spatial impacts of human activities on urban traffic crashes using multi-source big data

Traffic crashes are geographical events, and their spatial patterns are strongly linked to the regional characteristics of road network, sociodemography, and human activities. Different human activities may have different impacts on traffic exposures, traffic conflicts and speeds in different transportation geographic areas, and accordingly generate different traffic safety outcomes. Most previous researches have concentrated on exploring the impacts of various road network attributes and sociodemographic characteristics on crash occurrence. However, the spatial impacts of human activities on traffic crashes are unclear. To fill this gap, this study attempts to investigate how human activities contribute to the spatial pattern of the traffic crashes in urban areas by leveraging multi-source big data. Three kinds of big data sources are used to collect human activities from the New York City. Then, all the collected data are aggregated into regional level (ZIP Code Tabulation Areas). Geographically Weighted Poisson Regression (GWPR) method is applied to identify the relationship between various influencing factors and regional crash frequency. The results reveal that human activity variables from multi-source big data significantly affect the spatial pattern of traffic crashes, which may bring new insights for roadway safety analyses. Comparative analyses are further performed for comparing the GWPR models which consider human activity variables from different big data sources. The results of comparative analyses suggest that multiple big data sources could complement with each other in the coverage of spatial areas and user groups, thereby improving the performance of zone-level crash models and fully unveiling the spatial impacts of human activities on traffic crashes in urban areas. The results of this study could help transportation authorities better identify high-risky regions and develop proactive countermeasures to effectively reduce crashes in these regions.     

A zonal level safety investigation of pedestrian crashes in Riyadh,Saudi Arabia

In the recent decade, walking has been encouraged as an active mode of transportation, which could reduce congestion and air pollution and also improve community health. However, pedestrians are more vulnerable to traffic crashes compared with other road users, especially in developing countries such as Saudi Arabia. This paper examines the association among traffic volume, land-use, socio-demographic and roadway characteristics factors, and the frequency of pedestrian crashes based on macro-level safety analysis using data from Riyadh, the Capital of Saudi Arabia. Two Bayesian spatial Poisson-lognormal models for total and severe pedestrian crashes are developed in this study. The results show that the factors that affect total pedestrian crash occurrence are different from those affecting severe pedestrian crash. Several implications for pedestrian safety policies in Riyadh are suggested based on the results.     

Macro and micro models for zonal crash prediction with application in hot zones identification

Zonal crash prediction has been one of the most prevalent topics in recent traffic safety research. Typically, zonal safety level is evaluated by relating aggregated crash statistics at a certain spatial scale to various macroscopic factors. Another potential solution is from the micro level perspective, in which zonal crash frequency is estimated by summing up the expected crashes of all the road entities located within the zones of interest. This study intended to compare these two types of zonal crash prediction models. The macro-level Bayesian spatial model with conditional autoregressive prior and the micro-level Bayesian spatial joint model were developed and empirically evaluated, respectively. An integrated hot zone identification approach was then proposed to exploit the merits of separate macro and micro screening results. The research was based on a three-year dataset of an urban road network in Hillsborough County, Florida, U.S.Results revealed that the micro-level model has better overall fit and predictive performance, provides better insights about the micro factors that closely contribute to crash occurrence, and leads to more direct countermeasures. Whereas the macro-level crash analysis has the advantage of requirement of less detailed data, providing additional instructions for non-traffic engineering issues, as well as serving as an indispensable tool in incorporating safety considerations into long term transportation planning. Based on the proposed integrated screening approach, specific treatment strategies could be proposed to different screening categories. The present study is expected to provide an explicit template towards the application of either technique appropriately.     

Investigating the associations between road network structure and non-motorist accidents

Road networks channel traffic flow and can impact the volume and proximity of walking and bicycling. Therefore, the structure of road networks—the pattern by which roads are connected—can affect the safety of non-motorized road users. To understand the impact of roads’ structural features on pedestrian and bicyclist safety, this study analyzes the associations between road network structure and non-motorist-involved crashes using data from 321 census tracts in Alameda County, California. Average geodesic distance, network betweenness centrality, and an overall clustering coefficient were calculated to quantify the structure of road networks. Three statistical models were developed using the geographically weighted regression (GWR) technique for the three structural factors, in addition to other zonal factors including traffic behavior, land use, transportation facility, and demographic features. The results indicate that longer average geodesic distance, higher network betweenness centrality, and a larger overall clustering coefficient were related to fewer non-motorist-involved accidents. Thus, results suggest that: (1) if a network is more highly centered on major roads, there will be fewer non-motorist-involved crashes; (2) a network with a greater average number of intersections on the shortest path connecting each pair of roads tends to experience fewer crashes involving pedestrians and bicyclists; and (3) the more clustered road networks are into several sub-core networks, the lower the non-motorist crash count. The three structural measurements can reflect the configuration of a network so that it can be used in other network analyses. More information about the types of road network structures that are conducive to non-motorist traffic safety can help to guide the design of new networks and the retrofitting of existing networks. The estimation results of GWR models explain the spatial heterogeneity of correlations between explanatory factors and non-motorist crashes, which can support regional agencies in establishing local safety policies.     

The geography of freight-related accidents in the era of E-commerce: Evidence from the Los Angeles metropolitan area

Due to the burgeoning demand for freight movement in the era of e-commerce, freight related road safety threats have been growing in both urban and suburban areas, despite the improved general traffic safety over the past decades. The empirical evidence on how freight trucks related crashes are distributed across neighborhoods and correlated to spatially varying factors is, however, highly limited. This article uses data from the Los Angeles region in 2018 to analyze the spatial patterns of freight trucks related traffic crashes and examines the major factors that contribute to those patterns using spatial econometric models. Maps show that freight trucks related crashes are highly associated with major freight generators but less clustered than the overall traffic crashes. Results from the spatial Durbin model indicate that access to freight generators, economic attributes, land uses, road infrastructure, and road network variables all contribute to the spatial distribution of freight trucks related crashes. The findings could help transport planners understand the dynamics of freight trucks related traffic safety and develop operational measures for mitigating the impacts of growing goods movement on local communities.     

A new zone system to analyze the spatial relationships between the built environment and traffic safety

This study aims to investigate the impacts of the built environment on traffic safety at a zonal level using a newly developed crash-related zone system. Traffic analysis zones (TAZs) have been widely employed to analyze traffic safety at a macroscopic level. However, this zone system use may present problems. Unlike previous studies, in which new zoning systems were created from aggregating TAZs, in this study the new zone system, formed by traffic safety analysis zones (TSAZs), is created from the smallest available census units. Geographically Weighted Negative Binomial Regression (GWNBR) models are used and a comparative analysis between non-spatial global crash prediction models and spatial local GWPR (Geographically Weighted Poisson Regression) and GWNBR models using the two zonal systems is presented. We find that TSAZs based models performed better than TAZs based models, especially when combined to the GWNBR technique. Our results show that several features of the built environment are significant crash predictors, and that the relationships among these features and traffic safety vary across space. By combining a crash-related zonal system with spatial GWNBR models to understand the built environment effects on traffic safety, the results of the analysis can help urban planners to consider traffic safety proactively when planning or retrofitting urban areas.     

The safety impact of land use changes resulting from bypass road constructions

This paper analyzes the effect of bypass construction on road safety, specifically crash rates on bypass segments and in bypassed communities. It further investigates the effect of bypass construction on those communities’ economic development through changes in the number and spatial distribution of businesses and residential development, and examines whether these changes affect the road-safety level. Finally, it evaluates the effect on road safety of various traffic arrangements implemented in such communities. Drawing on an extensive data collection, the study focuses on Arab cities in the Galilee region of Israel, using cross-sectional and longitudinal analysis.The results show that bypass construction does not necessarily reduce overall crash frequencies or crash rates. It merely shifts road crashes from the highways (the bypass roads) to inner roads (bypassed) and from there to local traffic, with no significant reduction. The findings indicate a large variance in the effects of bypass constructions on safety.     

Effects of winter precipitation on automobile collisions,injuries, and fatalities in the United States

To better understand the links between winter precipitation (snow, sleet, and freezing rain) and travel risk, data on weather conditions and vehicle crashes, injuries and fatalities are gathered for 13 U.S. cities. A matched pair analysis is used to construct event-control pairs to determine the relative risk of crash, injury, and fatality. Winter precipitation is associated with a 19% increase in traffic crashes and a 13% increase in injuries compared to dry conditions. The type of winter precipitation (snowfall vs. freezing rain, ice pellets, or sleet) had no significant impact on the relative risk of crash. The relative risk of crash was significantly higher during the evening (1800–2359 local time) than during other times of the day. More intense precipitation led to increased relative risk of crash and injury compared to less intense precipitation. Relative risk of crash, injury, or fatality was not significantly higher during the first three winter precipitation events of the year as compared to subsequent events. The relative risk of both winter precipitation crash and injury showed no significant trend during the 1998–2008 period. Sensitivity of U.S. cities to winter precipitation varies from city to city in a manner that is not easily explained. Future research will be required to determine which safety interventions are most effective in each city and revise or expand safety programs appropriately.     

Factors contributing to injury severity in work zone related crashes in New Zealand

Roadworks take place frequently on existing roads in New Zealand. The adverse effects of poor road conditions and reduced road width due to the presence of a work zone on the safety of road users and workers at the work zone have been a matter of concern. Several studies have been conducted to examine the risk factors contributing to the occurrence of road crashes in work zones in different countries. Slow-moving and stopped vehicles near work zones have been found to be the primary cause of crashes and casualties in the work zones. Excessive speed of passing traffic has also been recognized as a crucial factor contributing to work zone related crashes in New Zealand. This study examined the effect of possible risk factors contributing to severe injury and fatality in work zone related crashes in New Zealand. A multinomial logistic regression model was established to determine the association between crash severity and factors such as road environment, vehicle attributes, driver behavior, and crash circumstances, based on the information available on 453 road crashes during the period from 2008 to 2013. The results indicated that the time period, vehicle involvement, and presence of vulnerable road users were the factors that determined the crash severity in work zones. This implies that improvements are required in traffic control and management measures in work zones to enhance road safety in the long run.     

Development of zone system for macro-level traffic safety analysis

Various geographic units have been used in macro-level modeling. Amongst these units, traffic analysis zones (TAZs) have been broadly employed in many macroscopic safety studies. Nevertheless, no studies questioned the validity of TAZs for crash analysis at the macro-level crash modeling. In this study, we point out several possible problems of TAZs as spatial units for macroscopic safety studies. Current TAZs with homogenous crash rates were combined into new single zones. Then we created ten new zonal systems by different zone aggregation levels. The optimal zonal scale for traffic safety analysis zones (TSAZ) was determined using the Brown-Forsythe test. It was found that the zone system with about 1:2 aggregation was the optimal zone system for macroscopic safety modeling. Thus we develop what we call traffic safety analysis zones (TSAZs) that has the potential of reducing several possible problems of TAZs. Also it was shown that TSAZ based models had better fit compared to TAZ based models.     

A driver focused truck crash prediction model

This paper advances a driver focused truck crash prediction model. The model investigates the contribution of driver factors on the number of state reportable crashes in which the driver was involved. The findings suggest that the following factors are significantly related to the likelihood of a crash occurrence: driver age, weight, height, gender, and employment stability as well as previous driver and vehicle violations and past crashes. The results have significance regarding the Federal Motor Carrier Safety Administration imperative to improve safety.     

Using odometer readings to assess VKT changes associated with a voluntary travel behaviour change program

In order to detect changes in daily average vehicle kilometres travelled (VKT) induced by a large-scale TravelSmart intervention in Melbourne, a panel of households was asked to complete before and after surveys, which included week-long odometer readings. In contrast to results reported from previous TravelSmart applications, the Melbourne program did not induce a statistically significant change in the average daily VKT when measured 1 year after the intervention. Multiple regressions revealed that the variability in change in VKT was better explained by socio-demographic variables than by the TravelSmart treatment. The change in VKT was also found to be strongly negatively correlated with the average daily vehicle kilometres recorded in the before survey—indicating the possibility of the ‘regression-to-the-mean’ effect well known in the road safety literature. The conditions under which the regression-to-the-mean effect may create the illusion of a positive TravelSmart program impact on the reduction in daily average VKT are examined. It is concluded that, in the context of voluntary travel behaviour change evaluations, greater attention should be paid to instrument reactivity arising from the impact of the before travel survey on TravelSmart uptake and/or on change in VKT, and to regression-to-the-mean effects.     

Goods movement,road safety,and spatial inequity: Evaluating freight-related crashes in low-income or minority neighborhoods

The significant growth in freight traffic and relevant crashes has aroused increasing concerns about road safety threats in local communities. We use data from the Minneapolis-St. Paul Metropolitan Area and examine the spatial relationship between freight-related crashes and neighborhoods with low-income and minority populations. We find that both household income and percentage of minority population are significantly correlated with the density of both freight-related crashes and freight-related crashes causing severe injuries and fatalities. The results indicate that freight-related crashes are subject to a spatial inequity problem. The findings underscore the importance of incorporating freight-related safety improvement within these low-income and minority neighborhoods.     

Factors associated with disability paratransit's travel time reliability

This paper identifies some of the characteristics of trips and pick-up and drop-off locations that are associated with paratransit's travel time reliability. Following convention, reliability has been defined as the inverse of variability. Four measures of travel time variability have been used to examine reliability: Standard Deviation, Percent Variation, Misery Index, and Buffer Index. Regression models have been used to estimate these four variables with trip data from Access Link, the paratransit service provided by NJ TRANSIT pursuant to the Americans with Disabilities Act (ADA). A number of characteristics of the pick-up and drop-off locations as well as selected characteristics of the trips were used as independent variables of the models. The statistical significance of the independent variables varied depending on which measure of reliability was estimated, but a few variables were consistently associated with reliability in all four models. These variables were trip distance, booking type, winter season, density of motor vehicle crashes in pick-up and drop-off locations, and whether pick-ups occurred in suburban bus corridors or urban core areas. Because of the significance of the variables on motor vehicle crash density in pick-up and drop-off locations, an additional regression model was used to examine the effect of crash incidents on trip duration by considering drop-offs that occurred in locations immediately after a crash. The model showed that trips take 4 to 5% longer when crashes occur in locations prior to a drop off. Planning implications of the findings are discussed.     

A GIS-based Bayesian approach for analyzing spatial–temporal patterns of intra-city motor vehicle crashes

This paper develops a GIS-based Bayesian approach for intra-city motor vehicle crash analysis. Five-year crash data for Harris County (primarily the City of Houston), Texas are analyzed using a geographic information system (GIS), and spatial–temporal patterns of relative crash risks are identified based on a Bayesian approach. This approach is used to identify and rank roadway segments with potentially high risks for crashes so that preventive actions can be taken to reduce the risks in these segments. Results demonstrate the approach is useful in estimating the relative crash risks, eliminating the instability of estimates while maintaining overall safety trends. The 3-D posterior risk maps show risky roadway segments where safety improvements need to be implemented. Results of GIS-based Bayesian mapping are also useful for travelers to choose relatively safer routes.     

Driving demand: What can gasoline refueling patterns tell us about planning an alternative fuel network?

Knowing which variables predict gasoline demand can help inform which are useful in determining future demand at an alternative fuel station such as those for bio-fuels, natural gas, hydrogen, or fast-charge electricity. This study explores the spatial distribution of demand by comparing two main classes of variables: those without a displacement component such as population in a census block group, and those that imply a vector or directionality such as vehicle kilometers traveled. The spatial distribution of these variables is compared to the spatial distribution of demand for gasoline using regression. Many models examining the transition from gasoline to an alternative fuel assume a demand pattern for fuel a priori in order to estimate potential demand at a future alternative fuel station. This paper studies not the models themselves but the variables used to predict demand. The results indicate that vehicle kilometers traveled (VKT) is the best variable to pinpoint where demand for fuel will occur. However, travel to the central business district of the metropolitan area does not appear to translate into demand for fuel in proportion to the VKT. While gasoline demand does appear to vary with population as well, the location of demand is much less specific than that predicted by VKT. The results also suggest that the route between home and the nearest freeway entrance may help predict a large portion of refueling and merits further investigation. This possible tendency can be used to create a new variable called “population-traffic” which appears to describe the spatial distribution of demand well. The good performance of this independent variable in regressions suggests that stations sited along the freeway may serve customers needs and provide the necessary concentration of demand for initial alternative fuel stations. A practical application of this work would be to help define refueling demand patterns in a rollout of alternative fueled vehicles in a neighborhood or town.     

Macro-level analysis of bicycle safety: Focusing on the characteristics of both crash location and residence

Over the last decade, bicycle ridership has been encouraged as a sustainable mode of transportation as it is economic and has less impact on the environment. Still, higher crash risk for bicyclists remains a deterrent for people to choose bicycling as their major mode of travel. As a first step in investigating bicycle safety, it is essential to identify not only the characteristics of the areas with the excessive number of bicycle crashes; but also those of the areas where crash-prone bicyclists reside. Therefore, this study aims to identify contributing factors for two subjects: (1) the number of bicycle crashes in the crash location's ZIP code and (2) the number of crash-involved bicyclists in their residence's ZIP. In order to achieve these objectives, a multivariate Bayesian Poisson lognormal CAR (conditional autoregressive) model was developed to identify the contributing factors for each subject. Regarding the model performance, the multivariate model outperformed its univariate counterpart in terms of DIC (deviance information criterion). Subsequently, hot zones for the two target zones were identified based on the modeling results. It is expected that practitioners are able to understand the contributing factors for bicycle crashes and identify hotspots from the results suggested in this study. In addition, they could implement safety countermeasures for the identified problematic locations to effectively reduce bicycle crashes.     

Modeling bicycle crash costs using big data: A grid-cell-based Tobit model with random parameters

Bicyclists are among the most vulnerable road users in the urban transportation system. It is critical to investigate the contributing factors to bicycle-related crashes and to identify the hotspots for efficient allocation of treatment resources. A grid-cell-based modeling framework was used to incorporate heterogeneous data sources and to explore the overall safety patterns of bicyclists in Manhattan, New York City. A random parameters (RP) Tobit model was developed in the Bayesian framework to correlate transportation, land use, and sociodemographic data with bicycle crash costs. It is worth mentioning that a new algorithm was proposed to estimate bicyclist-involved crash exposure using large-scale bicycle ridership data from 2014 to 2016 obtained from Citi Bike, which is the largest bicycle sharing program in the United States. The proposed RP Tobit model could deal with left-censored crash cost data and was found to outperform the Tobit model by accounting for the unobserved heterogeneity across neighborhoods. Results indicated that bicycle ridership, bicycle rack density, subway ridership, taxi trips, bus stop density, population, and ratio of population under 14 were positively associated with bicycle crash cost, whereas residential ratio and median age had a negative impact on bicycle crash cost. The RP Tobit model was used to estimate the cell-specific potential for safety improvement (PSI) for hotspot ranking. The advantages of using the RP Tobit crash cost model to capture PSI are that injury severity is considered by being converted into unit costs, and varying effects of certain explanatory variables are accounted for by incorporating random parameters. The cell-based hotspot identification method can provide a complete risk map for bicyclists with high resolution. Most locations with high PSIs either had unprotected bicycle lanes or were close to the access points to protected bicycle routes.     

Integrating police reports with geographic information system resources for uncovering patterns of pedestrian crashes in Denmark

Promoting walking goes a long way in contributing to the sustainability and health of future cities and regions, and improving pedestrian safety is essential for building more sustainable and healthier communities. As the problem is multifaceted in nature, this study looks at patterns of pedestrian crashes from a perspective that goes beyond the traditional investigation of pedestrian characteristics and behaviour by analysing the contribution of built environment, land use, and traffic conditions. Moreover, this study goes beyond the traditional analysis of traditional police reports by integrating them with rich geographic information system resources. This study analysed a sample of 7469 crashes between a pedestrian and another road user that occurred in Denmark between 2006 and 2015. The crash locations were geocoded and matched to a detailed traffic network, a transport planning model, and several resources detailing building and land use composition. Latent class analysis uncovered patterns of pedestrian crashes for both the fully identified records and the substantial amount of hit-and-run records. Findings from this study reveal a major red thread in the lack of hazard awareness for both pedestrians and road users and suggest solutions from both the behavioural and the infrastructure perspectives. Major needs are (i) educating pedestrians about the risks related to drinking and then walking along major roads in the darkness, (ii) making crossings for pedestrians and approaches for road users easier to understand and to access in order to reduce unnecessary conflicts, and (iii) designing traffic calming solutions around major shopping and leisure locations in dense city centres.     

Spatial patterns and temporal dynamics of urban bicycle crashes—A case study from Salzburg (Austria)

Most bicycle crash analyses are designed as explanatory studies. They aim to identify contributing risk factors and calculate risk rates based on – most of the time – highly aggregated statistical data. In contrast to such explanatory study designs, the presented study follows an exploratory approach, focusing on the absolute number of crashes. The aim is to reveal and describe patterns and dynamics of urban bicycle crashes on various spatial scale levels and temporal resolutions through a multi-stage workflow. Spatial units are delineated in the network space and serve as initial units of aggregation. In order to facilitate comparisons among regions and quantify temporal dynamics, a reference value of crash frequency is simulated for each unit of the respective spatial scale level and temporal resolution.For the presented case study, over 3000 geo-coded bicycle crashes in the city of Salzburg (Austria) were analyzed. The data set covers 10 years and comprises all bicycle crashes reported by the police. Distinct spatial and temporal patterns with clusters, seasonal variations, and regional particularities could be revealed. These insights are indicators for urban dynamics in the transport system and allow for further, targeted in-depth analyses and subsequent counter measures. Moreover, the results prove the applicability of the proposed multi-stage workflow and demonstrate the added value of analyses of small aggregates on various scale levels, down to single crashes, and temporal resolutions.