Chen, Peng. (2015). Built Environment Factors in Explaining the Automobile-Involved Bicycle Crash Frequencies: A Spatial Statistic Approach. Safety Science, 79, 336 – 343.
View Publication
Abstract
The objective of this study is to understand the relationship between built environment factors and bicycle crashes with motor vehicles involved in Seattle. The research method employed is a Poisson lognormal random effects model using hierarchal Bayesian estimation. The Traffic Analysis Zone (TAZ) is selected as the unit of analysis to quantify the built environment factors. The assembled dataset provides a rich source of variables, including road network, street elements, traffic controls, travel demand, land use, and socio-demographics. The research questions are twofold: how are the built environment factors associated with the bicycle crashes, and are the TAZ-based bicycle crashes spatially correlated? The findings of this study are: (1) safety improvements should focus on places with more mixed land use; (2) off-arterial bicycle routes are safer than on-arterial bicycle routes; (3) TAZ-based bicycle crashes are spatially correlated; (4) TAZs with more road signals and street parking signs are likely to have more bicycle crashes; and (5) TAZs with more automobile trips have more bicycle crashes. For policy implications, the results suggest that the local authorities should lower the driving speed limits, regulate cycling and driving behaviors in areas with mixed land use, and separate bike lanes from road traffic. (C) 2015 Elsevier Ltd. All rights reserved.
Keywords
Injury Crashes; Risk Analysis; Models; Infrastructure; Dependence; Counts; Level; Bicycle Crash Frequency; Hierarchal Bayesian Estimation; Poisson Lognormal Random Effects Model; Built Environment; Traffic Analysis Zone
Chen, Peng; Shen, Qing. (2016). Built Environment Effects on Cyclist Injury Severity in Automobile-Involved Bicycle Crashes. Accident Analysis & Prevention, 86, 239 – 246.
View Publication
Abstract
This analysis uses a generalized ordered logit model and a generalized additive model to estimate the effects of built environment factors on cyclist injury severity in automobile-involved bicycle crashes, as well as to accommodate possible spatial dependence among crash locations. The sample is drawn from the Seattle Department of Transportation bicycle collision profiles. This study classifies the cyclist injury types as property damage only, possible injury, evident injury, and severe injury or fatality. Our modeling outcomes show that: (1) injury severity is negatively associated with employment density; (2) severe injury or fatality is negatively associated with land use mixture; (3) lower likelihood of injuries is observed for bicyclists wearing reflective clothing; (4) improving street lighting can decrease the likelihood of cyclist injuries; (5) posted speed limit is positively associated with the probability of evident injury and severe injury or fatality; (6) older cyclists appear to be more vulnerable to severe injury or fatality; and (7) cyclists are more likely to be severely injured when large vehicles are involved in crashes. One implication drawn from this study is that cities should increase land use mixture and development density, optimally lower posted speed limits on streets with both bikes and motor vehicles, and improve street lighting to promote bicycle safety. In addition, cyclists should be encouraged to wear reflective clothing. (C) 2015 Elsevier Ltd. All rights reserved.
Keywords
Cycling Injuries; Traffic Accidents; Transportation Planning; Data Analysis; Employment; Built Environment; Cyclist Injury Severity; Generalized Additive Model; Generalized Ordered Logit Model; Ordered Response Model; United-states; Helmet; Frameworks; Driver; Risk
Chen, Peng; Zhou, Jiangping. (2016). Effects of the Built Environment on Automobile-involved Pedestrian Crash Frequency and Risk. Journal Of Transport & Health, 3(4), 448 – 456.
View Publication
Abstract
This area-based study explores the relationship between automobile-involved pedestrian crash frequency versus risk and various built environment factors such as road network and land use. The methodology involves the use of Bayesian hierarchical intrinsic conditional autoregressive model, which accounts for unobserved heterogeneities and spatial autocorrelations. The city of Seattle is selected for this empirical study, and the geospatial unit of analysis is traffic analysis zone. The primary data were obtained from collision profiles available at the Seattle Department of Transportation. The major findings of this study include: (1) the densities of 4-way intersections and more than 5-way intersections and land use mixture are positively correlated with the pedestrian crash frequency and risk; (2) sidewalk density and the proportion of steep areas are negatively associated with the pedestrian crash frequency and risk; (3) areas with a higher bus stop density are likely to have more pedestrian crashes; (4) areas with a greater proportion of industrial land use have lower pedestrian crash frequency; (5) areas with an averagely higher posted speed limit has higher pedestrian crash risk; (6) areas with a higher employment density has lower pedestrian crash risk; (7) the mode share of walking and the total number of trips are positively correlated with the pedestrian crash frequency, and the total number of trips is negatively correlated with the pedestrian crash risk. These findings provide support for planning policy making and road safety programs. Local authorities should improve walkability by providing more sidewalks and separate travel lanes for motorized traffic and pedestrians in areas with different land use purposes. Compact development should be encouraged to support building a safe walking environment. (C) 2016 Elsevier Ltd. All rights reserved.
Keywords
Spatial-analysis; Urban Form; Land-use; Model; Counts; Transportation; Severity; Bicycle; Safety; Travel; Pedestrian Crash Frequency; Pedestrian Crash Risk; Built Environment; Spatial Autocorrelation; Road Network; Land Use
Pan, Haixiao; Li, Jing; Chen, Peng. (2016). Study on the Ownership of Motorized and Non-Motorized Vehicles in Suburban Metro Station Areas: A Structural Equation Approach. Urban Rail Transit, 2(2), 47 – 58.
View Publication
Abstract
As Chinese megacities are experiencing a large-scale motorization and suburbanization, an ever greater number of households are relocated to suburban towns. The increasing average travel distance surely encourages car growth. China is now the world's largest car consumer, resulting in a series of unforeseen environmental and public health issues. On the other hand, scooters, electric bikes, and motorcycles become attractive options to substitute non-motorized bicycles. The ongoing demographic changes should also be taken in account. China has a rapidly aging population and a higher birth rate following reforms to the one-child policy allowing couples to have a second child. These changes will lead to a dramatic alteration of the household composition in the near future. Under above emerging contexts, this study aims to understand what implies the ownership of motorized and non-motorized vehicles in suburban metro station areas by means of a structural equation model. The data employed in this study are based on a household survey collected from three neighborhoods in Shanghai suburban metro station areas in 2010. The major findings include: (1) Income is a decisive element in car ownership. Specifically, high-income households have higher propensity to own a car, while middle and poor income families tend to own scooters, electric bikes, motorcycles, or bicycles. (2) Workplace built environment features or mode preferences are not essential to understanding vehicle ownership in Chinese context. (3) Stem families are more likely to own cars; the presence of a child or a senior family member increases the probability of owning a car by enlarging the household. (4) The results estimated for core family and DINK (couple with no child) family are highly consistent, and these families are less likely to own cars. Therefore, transport policies may focus more on households. Providing safe, pleasant, and efficient pedestrian and bicycle paths for children and seniors may decrease the attractiveness of owning cars.
Keywords
Suburban Metro Station Areas; Ownership Of Motorized And Non-motorized Vehicles; Built Environment; Mode Preferences; Family Composition; Structural Equation Model
Shen, Qing; Chen, Peng; Pan, Haixiao. (2016). Factors Affecting Car Ownership and Mode Choice in Rail Transit-Supported Suburbs of a Large Chinese City. Transportation Research Part A: Policy & Practice, 94, 31 – 44.
View Publication
Abstract
As Chinese cities continue to grow rapidly and their newly developed suburbs continue to accommodate most of the enormous population increase, rail transit is seen as the key to counter automobile dependence. This paper examines the effects of rail transit-supported urban expansion using travel survey data collected from residents in four Shanghai suburban neighborhoods, including three located near metro stations. Estimated binary logit model of car ownership and nested logit model of commuting mode choice reveal that: (1) proximity to metro stations has a significant positive association with the choice of rail transit as primary commuting mode, but its association with car ownership is insignificant; (2) income, job status, and transportation subsidy are all positively associated with the probabilities of owning car and driving it to work; (3) higher population density in work location relates positively to the likelihood of commuting by the metro, but does not show a significant relationship with car ownership; (4) longer commuting distance is strongly associated with higher probabilities of riding the metro, rather than driving, to work; (5) considerations of money, time, comfort, and safety appear to exert measurable influences on car ownership and mode choice in the expected directions, and the intention to ride the metro for commuting is reflected in its actual use as primary mode for journey to work. These results strongly suggest that rail transit-supported urban expansion can produce important positive outcomes, and that this strategic approach can be effectively facilitated by transportation policies and land use plans, as well as complemented by timely provision of high quality rail transit service to suburban residents. (C) 2016 Elsevier Ltd. All rights reserved.
Keywords
Railroads; Public Transit; Choice Of Transportation; Automobile Ownership; Transportation; Suburbanization; China; Automobile Dependence; Large Chinese Cities; Rail Transit; Shanghai; Urban Expansion; Built Environment; Travel Behavior; Self-selection; Impact; Areas
Chen, Peng; Shen, Qing; Childress, Suzanne. (2018). A GPS Data-based Analysis of Built Environment Influences on Bicyclist Route Preferences. International Journal Of Sustainable Transportation, 12(3), 218 – 231.
View Publication
Abstract
This study examines the effects of built environment features, including factors of land use and road network, on bicyclists' route preferences using the data from the city of Seattle. The bicycle routes are identified using a GPS dataset collected from a smartphone application named CycleTracks. The route choice set is generated using the labeling route approach, and the cost functions of route alternatives are based on principal component analyses. Then, two mixed logit models, focusing on random parameters and alternative-specific coefficients, respectively, are estimated to examine bicyclists' route choice. The major findings of this study are as follows: (1) the bicycle route choice involves the joint consideration of convenience, safety, and leisure; (2) most bicyclists prefer to cycle on shorter, flat, and well-planned bicycle facilities with slow road traffic; (3) some bicyclists prefer routes surrounded by mixed land use; (4) some bicyclists favor routes which are planted with street trees or installed with street lights; and (5) some bicyclists prefer routes along with city features. This analysis provides valuable insights into how well-planned land use and road network can facilitate efficient, safe, and enjoyable bicycling.
Keywords
Cyclists; Mobile Apps; Multiple Correspondence Analysis (statistics); Traffic Engineering; Cycling; Bicycle Route Choice; Built Environment; Labeling Routes; Mixed Logit Model; Principal Component Analysis; Smartphone GPS Data; Choice Sets; Safe Routes; Walking; Models; Health; Infrastructure; Facilities; California; Networks
Chen, Peng; Hu, Songhua; Shen, Qing; Lin, Hangfei; Xie, Chi. (2019). Estimating Traffic Volume for Local Streets with Imbalanced Data. Transportation Research Record, 2673(3), 598 – 610.
View Publication
Abstract
Annual average daily traffic (AADT) is an important measurement used in traffic engineering. Local streets are major components of a road network. However, automatic traffic recorders (ATRs) used to collect AADT are often limited to arterial roads, and such information is, therefore, often unavailable for local streets. Estimating AADT on local streets becomes a necessity as local street traffic continues to grow and the capacity of arterial roads becomes insufficient. A challenge is that an under-represented sample of local street AADT may result in biased estimation. A synthetic minority oversampling technique (SMOTE) is applied to oversample local streets to correct the imbalanced sampling among different road types. A generalized linear mixed model (GLMM) is employed to estimate AADT incorporating various independent variables, including factors of roadway design, socio-demographics, and land use. The model is examined with an AADT dataset from Seattle, WA. Results show that: (1) SMOTE helps to correct imbalanced sampling proportions and improve model performance significantly; (2) the number of lanes and the number of crosswalks are both positively associated with AADT; (3) road segments located in areas with a higher population density or more mixed land use have a higher AADT; (4) distance to the nearest arterial road is negatively correlated with AADT; and (5) AADT creates spatial spillover effects on neighboring road segments. The combination of SMOTE and GLMM improves the estimation accuracy on AADT, which contributes to better data for transportation planning and traffic monitoring, and to cost saving on data collection.
Keywords
Average; Prediction; Network; County
Chen, Peng; Shen, Qing. (2019). Identifying High-risk Built Environments for Severe Bicycling Injuries. Journal of Safety Research, 68, 1 – 7.
View Publication
Abstract
Introduction: This study is aimed at filling part of the knowledge gap on bicycling safety in the built environment by addressing two questions. First, are built environment features and bicyclist injury severity correlated; and if so, what built environment factors most significantly relate to severe bicyclist injuries? Second, are the identified associations varied substantially among cities with different levels of bicycling and different built environments? Methods: The generalized ordered logit model is employed to examine the relationship between built environment features and bicyclist injury severity. Results: Bicyclist injury severity is coded into four types, including no injury (NI), possible injury (PI), evident injury (El), and severe injury and fatality (SIF). The findings include: (a) higher percentages of residential land and green space, and office or mixed use land are correlated with lower probabilities of El and SIF; (b) land use mixture is negatively correlated with El and SIF; (c) steep slopes are positively associated with bicyclist injury severity; (d) in areas with more transit routes, bicyclist injury is less likely to be severe; (e) a higher speed limit is more likely to correlate with SIF; and (f) wearing a helmet is negatively associated with SIF, but positively related to PI and El. Practical applications: To improve bicycle safety, urban planners and policymakers should encourage mixed land use, promote dense street networks, place new bike lanes in residential neighborhoods and green spaces, and office districts, while avoiding steep slopes. To promote bicycling, a process of evaluating the risk of bicyclists involving severe injuries in the local environment should be implemented before encouraging bicycle activities. (C) 2018 National Safety Council and Elsevier Ltd. All rights reserved.
Keywords
Motor Vehicle; Land-use; Crashes; Severities; Facilities; Frameworks; Frequency; Cyclists; Bike; Bicyclist Injury Severity; Built Environments; Generalized Ordered Logit Model; Us Cities; Bicycles; Urban Environments; Injuries; Neighborhoods; Land Use; Urban Areas; Paths; Protective Equipment; Bicycling; Fatalities; Correlation; Residential Areas; Traffic Accidents & Safety; Safety; Logit Models; Ecological Risk Assessment; Slopes; Health Risks; Urban Transportation; Studies; Environments