2021

Connor Klassen, VHB

Throughout recent history, vehicular traffic has been prioritized over active forms of transportation such as walking and bicycling. When a highway divides a once formally connected patchwork of neighborhoods, the demand to cross the corridor increases significantly to access employment opportunities, social connections, and necessities. This study explores the Independence Boulevard corridor in Charlotte, North Carolina, an 11.5-mile partially limited access expressway and median divided highway between I-277 near uptown Charlotte and I-485 in Matthews. The research design uses a four-part process involving crash data analysis, pedestrian activity modeling, transit ridership investigation, and walkshed analysis. The methodological design is then used in the context of Independence Boulevard as a case study. Crash data analysis allowed for the identification of pedestrian crossing hotspots and areas of high safety concern. The pedestrian activity model validated these hotspots through roadway connectivity and employment density metrics to identify neighborhoods acting as ‘senders’ or ‘receivers’ across the corridor. Transit ridership analysis located high frequency destinations and stops of regional importance. Lastly, walkshed analyses allowed for potential recommendations to be simulated to calculate the impacts in terms of the number of destinations and households accessible within a 10-minute walkshed. This four-step approach aided in creating various short-term and long-term recommendations to improve pedestrian safety across the Independence Boulevard corridor.

Brian Peck, WSP
Rebecca Sanders (rebecca@safestreetsresearch.com), Arizona State University
Deborah Salon, Arizona State University

Phoenix, Arizona, is consistently identified as one of the most dangerous U.S. cities for pedestrians, prompting urgent research into pedestrian safety issues and potential improvements. This paper presents the results of a recent safety study aiming to identify key contributors to pedestrian crash severity in order to inform a strategy for pedestrian safety improvements in Phoenix both at hotspots and system-wide. Our study has two main contributions. Methodologically, we used crowd-sourced speed data to identify prevailing vehicle speeds at the time of the crashes in order to better understand the relationship between speed and pedestrian crash severity. Substantively, our logistic regression model indicates that the likelihood of a severe pedestrian injury in Phoenix is significantly impacted by vehicle speeds over 35 mph (56 kph), pedestrian and driver intoxication, darkness, age, and pedestrians crossing outside of the crosswalk. We also found a significant increased likelihood of crash severity during the morning rush hour, unexplained by light conditions, as well as a protective relationship between crash severity and higher amounts of commuting by transit. We also find higher rates of pedestrian crashes and severe pedestrian crashes in census tracts that are minority non-Hispanic white and that have a higher percentage of lower-income households, zero-vehicle households, and transit and walk commuters. These findings underscore the need to use an equity lens to substantially rethink roadway design in Phoenix and prioritize efforts to slow speeds and enable pedestrians to safely walk and cross the street.

Mallikarjun Patil, Birla Institute of Technology and Science, Hyderabad
Bandhan Majumdar, Birla Institute of Technology and Science, Pilani
Prasanta Sahu, Birla Institute of Technology and Science, Pilani

This study presents a methodology for evaluating a set of crash-prone sidewalk and crosswalk locations of an urban area with respect to their existing walkability condition and recommending improvement needs. Initially, a set of fifteen sidewalk- and ten crosswalk specific attributes relevant to India were identified. Analytical Hierarchy Process (AHP) was used to estimate relative weights associated with the attributes from the domain-specific expert’s perspective. Then, a weighted sum method was used to formulate sidewalk condition index (SCI) and crosswalk condition index (CCI) respectively. Ten locations across Hyderabad with highest pedestrian fatality were selected as study locations. Sidewalk and crosswalks at these ten locations were evaluated using SCI and CCI, respectively, and a set of measures are recommended for improving the walkability. Location-specific SCI and CCI estimates were used to prioritize the locations in terms of the existing condition and infrastructural requirements. Results indicated that sidewalk attributes such as sidewalk lighting, cleanliness, physical separation of traffic and traffic speed and crosswalk attributes such as conflicts with crossing traffic, crosswalk illumination, and intersection control to influence safety and walkability significantly. Measures such as the provision of exclusive right-of-way to pedestrians, maintaining the sidewalk quality, enforcing no jaywalking, re-design of signal timing with pedestrian phase, provision of the zebra crossing and refuge Island would improve the walkability at pedestrian crash-prone locations across Hyderabad. This proposed methodology and research findings could act as a critical tool to improve the overall safety and walkability of sidewalks and crosswalks across the city.

Ahadul Islam, Utah State University
Michelle Mekker, Utah State University
Patrick Singleton (patrick.singleton@usu.edu), Utah State University

The focus of this study is twofold: (1) to estimate models of pedestrian crash frequency and severity at signalized intersections, using pedestrian and traffic volumes and other predictor variables; and (2) to examine whether the “safety in numbers” effect applies to pedestrian safety the US using robust measures of pedestrian exposure. Specifically, the analysis used pedestrian crossing volumes estimated from one year of pedestrian push-button data, and ten years of crash data at signalized intersections in Utah. Data from 919 signalized intersections were used to calibrate a zero-inflated negative binomial model for crash frequency analysis. The model results indicated that signals with longer crossing distances, far-side bus stops, larger shares of residential and commercial land uses, and in neighborhoods with lower-income and larger households saw more pedestrian crashes. To analyze injury severity in pedestrian crashes, an ordered logit model was fitted with 1,483 pedestrian crash observations. The model results indicated that speed limit, vehicle size, number of vehicles, vehicle maneuvering direction, and involvement of DUI/drowsy/distracted driving in crashes had significant effects on severity. The study also found a non-linear relationship where pedestrian-vehicle crash rates decreased with an increase in pedestrian volumes, supporting the “safety in numbers” effect. The authors suggest potential countermeasures, policy alterations, and scope of future research for improving pedestrian safety at signalized intersections.

Shahnewaz Hasanat-E-Rabbi (shahnewaz.rabbi@gmail.com), BUET
Md Asif Raihan, Bangladesh University of Engineering and Technology
S M Mahmud, Accident Research Institute (ARI)
Md. Shamsul Hoque, Bangladesh University of Engineering and Technology

Application of injury severity modeling approach for identifying exogenous factors and their impact on injury severities in a crash is a state of art practice among the safety researchers. However, in the context of urban metropolis of developing countries, this practice is very scarce particularly for pedestrian injury severity- the most vulnerable road users in terms of crashes and injury severities. In this study, a comprehensive comparison exercise has been made of the performance of unordered and ordered response models including Multinomial Logit (MNL), Ordered Logit (OL)  and Partial Proportional Odds (PPO) model to identify and examine the impact of exogenous factors on pedestrian injury severity in the context of a developing urban metropolis. Five years reported pedestrian crash data from Dhaka metropolitan city are used in this research. The comparative analysis revels that among these models PPO model identifies more risk factors and performed relatively better for the current data set. The study identifies a range of risk factors that significantly affect the probability of pedestrian injury outcomes. Elasticity impact of those factors are also evaluated which have significant policy implication for improving pedestrian safety in developing cities.

Mankirat Singh, California State Polytechnic University, Pomona
Yongping Zhang, California State Polytechnic University, Pomona
Wen Cheng, California State Polytechnic University, Pomona
Hairui Tang, California State Polytechnic University, Pomona
Edward Clay, California State Polytechnic University, Pomona

Walking and cycling for transportation provide immense benefits associated with health, environment, mobility, and many more. However, non-motorists are the most vulnerable segment of traveling public from motorized traffic due to the lack of protective structure and difference in body mass when compared with vehicles. Numerous studies in the past are dedicated to enhancing the safety of active transportation modes, but very few studies are devoted to safety analysis of the transit stations, which serve as the important modal interface for pedestrians and bicyclists. The current study bridges the gap by developing joint models based on the multivariate conditionally autoregressive (MCAR) priors with distance-oriented neighboring weight matrix. For this purpose, transit station-centered data in Los Angeles County were used for model development. Feature selection relying on both random forest and correlation analyses were employed which leads to different covariate inputs to each of the two jointed models, resulting in increased model flexibility. Integrated nested Laplace approximation (INLA) algorithm was adopted due to its fast yet robust analysis. For a comprehensive comparison of the predictive accuracy of models, different evaluation criteria were utilized which include deviance information criterion (DIC), widely applicable information criterion (WAIC), posterior mean deviance (Dbar) and effective number of parameters (Pd). The results demonstrate that models with correlation effect perform much better than the models without correlation of pedestrian and bicyclists. The joint models also aids in the identification of the significant covariates contributing to the safety of each of the two active transportation modes.

MD MEHEDI HASAN, AECOM
Jun-Seok Oh, Western Michigan University
Valerian Kwigizile, Western Michigan University

Walking is one of the main modes of active transportation, which facilitates a healthy and livable community by reducing the risk of health-related issues. Since the walking experience depends on sidewalk characteristics and surrounding built environment, there is a need to assess the impact of built environment on pedestrians’ perception on walkability. In this study, we evaluated the pedestrians’ perception on segments with various environment. A web survey was conducted to collect the pedestrians’ perception on the walking environment in terms of sidewalk comfort, safety, street view, vitality, and overall performance. A total of 130 participants in Michigan and Texas provided their evaluations on a total of 22 walking environments. The survey output showed that the walkability was positively correlated with presence of buffer and high tree/greenery density. On the other hand, negative walking experience was observed in walking environment with high pedestrian volume, high bicycle/e-scooter density, high building density, and heavy traffic.

Yunchang Zhang, Purdue University
Jon Fricker (fricker@purdue.edu), Purdue University

At “semi-controlled” crosswalks with yield signs and markings, negotiations as to the right-of-way occur frequently between pedestrians and motorists, to determine who should proceed first. This kind of “negotiation” often leads to traffic delay and potential conflicts. To minimize misunderstandings between pedestrian and motorist that can have serious safety consequences, it is essential that we understand the decision-making process as they interact in real street-crossing situations. This paper employs a game-theoretic approach to investigate the joint behaviors of pedestrians and motorists from the perspective of safety. Assuming bounded rationality for each player, the quantal response equilibrium is a special kind of game with incomplete information. Explanatory variables such as conflicting risks and time savings can be incorporated into the payoff functions of the “players” via expected utility functions. Finally, model parameters can be estimated using an expectation maximization algorithm. The game-theoretic framework is applied to model pedestrian-motorist interactions at a semi-controlled crosswalk on a university campus. The estimation results indicate that the likelihood of pedestrian-vehicle conflict can be quantified. The results can lead to control measures that facilitate the negotiation between pedestrian and motorist and reduce the conflict risk at semi-controlled crosswalks.

Mazimba Mazimba, Shanghai Jiao Tong University
Zhong-Ren Peng, University of Florida
Hong-Di He, Shanghai Jiao Tong University
Hongmei Zhao, Shanghai Maritime University
Kaifa Lu, Shanghai Jiao Tong University

As pedestrians move through cities, they are faced with different path choices. Most of the time, path choice is influenced by the length of that path or the time they spend on that path to a specific destination. When at an intersection, the same notion applies. Possible pollution exposure levels that pedestrians might face on the path choices provided at an intersection have been overlooked. In this paper, pedestrians’ exposure to Traffic Related Air Pollution (TRAP) is yet again put under the microscope. How does traffic related air pollution exposure vary, depending on the path taken at an intersection? This paper therefore, assesses pedestrians’ exposure to Particulate matter (PM) and Black Carbon (BC) on two pedestrian walking facilities at two intersections in Shanghai, China. PM and BC data was collected using mobile monitors. A prototype empirical equation that models the distribution of pollutants, without any requirement for advanced analysis was developed. The results were visualized using the Paraview software. PM concentration must be the highest on the path closest to the source; the traffic. However, there is a deviation from this hypothesis as the data shows the concentrations and exposure being higher on the footbridge in some cases. There is a high correlation between the traffic volume and the PM concentrations. Section views of the distribution of pollutants on the footbridges shows that the height of the footbridge is inversely proportional to the change in concentration of pollutants.  Keywords: Pedestrians, exposure, intersection, particulate matter, black carbon

Emmanuel James (elj52@nau.edu), Northern Arizona University
Brendan Russo, Northern Arizona University

Pedestrian safety has been a prevalent issue across the United States, and especially in  regions that experience higher pedestrian activity year-round due to warm climates, such as the state of Arizona. In the past several years, recent trends indicate a steady increase in pedestrian crashes in the city of Phoenix, Arizona. To address this issue, the objective of this study was to investigate factors associated with the frequency of vehicle-pedestrian crashes at the census block level by amalgamating four datasets consisting of crash data, demographic data, land use data, and roadway characteristics. A negative binomial model was estimated to identify factors significantly associated with pedestrian crash frequency. Numerous parameters in the model were found to influence pedestrian crashes, and the general effect (though with differing magnitudes) of most variables were similar to previous studies conducted in other regions, however two specific land use and demographic characteristics were found to be unique to the city of Phoenix. Percent of industrial land use type and percent of persons aged 65 or older were found to have differing effects compared with those found in previous studies. Ultimately, the findings of this study provide new insights that can help frame or amend policies and countermeasures aimed at improving pedestrian safety.

Benson Long (blong2@unm.edu), University of New Mexico
Nick Ferenchak, University of New Mexico

Over the past decade, the United State has experienced a 53% increase in pedestrian fatalities, with 2018 having a 3.4% increase from 2017. Of these fatalities, 76% occurred in dark/nighttime lighting conditions, with 50% occurring between 6:00 PM and 11:59 PM. Despite past research exploring several contributing attributes for nighttime crashes, there is limited research that investigates spatial factors of land use attributes and socio-economic factors. Have these trends been concentrated in certain land uses? Could an establishment with the capacity to serve alcohol invoke a greater risk for pedestrian crashes? Does socioeconomic status correlate with clustering for fatal and/or severe crashes? To better understand the spatial characteristics of this trend, we analyzed crash data from Albuquerque, New Mexico for pedestrian fatalities and severe injuries from 2013-2018. We used confidence intervals to verify the statistical integrity of the trends. Findings suggest that the trend is most prevalent near bars at night in lower socio-economic areas with elevated concentrations in minority populations.

Sarah Plonka, Government of Ontario Ministry of Transportation
Patrick Byrne, Government of Ontario Ministry of Transportation
Sara Volo, Government of Ontario Ministry of Transportation
Ian Sinclair, Government of Ontario Ministry of Transportation
Thadsha Prabha, Government of Ontario Ministry of Transportation

Pedestrian-involved collisions are a key contributor to roadway fatalities in Ontario, and pedestrian deaths have been growing as a proportion of total road fatalities. This study aimed first to determine trends in the pedestrian fatality rate in Ontario over time, and second to assess the impact of select large-scale trends on pedestrian fatalities. Large-scale trends were identified through a review of literature and hypotheses were tested using Ontario collision data from 2002-2016. The following four key areas were assessed for their impact: 1) The ageing demographic; 2) the impact of increasing consumer preference for light trucks; 3) the potential for an increase in alcohol-consuming pedestrians associated with a decrease in alcohol-consuming drivers, and; 4) increasing inattention, in part by electronic device use, by pedestrians and drivers. A quadratic model with a minimum at 2010 best described changes in Ontario’s pedestrian fatality rate, suggesting a transition from a decreasing to increasing trend at that time. Results of the four key areas were: 1) The proportion of killed pedestrians aged 75 and older has been increasing over time, a trend that can be fully explained by their increased representation in Ontario’s population. This trend is expected to continue. 2) Similarly, the increase in the proportion of pedestrians killed by a light truck can be explained by their increased representation in Ontario’s registered vehicle population. 3) The odds of a pedestrian being alcohol positive have been decreasing over time. 4) The odds are higher that a driver who kills a pedestrian is inattentive.

Feifei Xin, Tongji University
Chongjing Sun, Tongji University
Xiaobo Wang, Tongji University

In recent years, the conflicts between crossing pedestrians and right-turning vehicles have become more severe at intersections in China, where right-turning vehicles are usually not controlled by traffic signals. Pedestrian safety is increasingly important, but few studies have addressed this issue. In this study, we propose a quantitative method for evaluating the conflict risk between pedestrians and right-turning vehicles at intersections based on micro-level behavioral data obtained from field surveys and video recordings. A typical intersection in Shanghai was selected as the study site. In total, 670 min of video were recorded during the peak hours from 7:30 AM to 9:30 PM. After processing the video information, we obtained vehicle and pedestrian tracking data, including the velocity, acceleration, deceleration, time, and location coordinates. Based on these data, we proposed several conflict indicators and extracted these indicators automatically using MATLAB to identify pedestrian–right-turning vehicle conflicts and to determine the severity of the conflicts identified. We identified 93 conflict examples. The conflict risks were quantitatively classified using the K-means fuzzy clustering method and all of the conflicts were assigned to five grades. We also analyzed the characteristics of the conflict distribution and the severity of different types of conflict, which showed that conflicts on different areas on the crosswalk differed in their severity.

Mohammad Jalayer, Rowan University
Deep Patel, Rowan University
Patrick Szary, Rutgers University
Keith Hamas, North Jersey Transportation Planning Authority

Pedestrian safety at intersections, where motor vehicles cross paths with people walking, is a serious matter of concern for traffic and road safety engineers and professionals. Based on the National Highway Traffic Safety Administration, there were 5,977 pedestrian fatalities and more than 85,000 pedestrian injuries in the United States as a result of traffic crashes in 2017. More specifically, the state of New Jersey ranked second in the nation concerning the ratio of pedestrian fatalities to the total number of motor vehicle deaths, necessitating further investigations. This paper depicts the outcomes of the observational study to gauge the effectiveness of the Street Smart NJ pedestrian safety campaign— an awareness, public education, and behavioral change campaign program to improve compliance with pedestrian and motorist laws. The effectiveness of the campaign was evaluated by comparing the rates of non-compliant pedestrian and driver behaviors before and after the campaign. The studied non-compliant behaviors include unsafe crossing and crossing against a signal, failing to stop for pedestrians when turning, failing to stop before turning at a red light or stop sign, and running the red light signal or stop sign. Where, video data were collected in eight communities across the New Jersey (i.e., Asbury Park, Garfield, Morris Plains, Newark, Princeton, Rutherford, Teaneck, and Woodbridge) in 2018-2019. Overall, the results of the study show significant improvements in pedestrian and driver behaviors following the safety campaign.

Seth LaJeunesse, University of North Carolina
Paul Ryus, Kittelson & Associates, Inc. (KAI)
Wesley Kumfer, University of North Carolina, Chapel Hill
Sirisha Kothuri, Portland State University
Krista Nordback, University of North Carolina

Walking is the most basic and sustainable mode of transportation, and many jurisdictions desire increased walking rates to reduce congestion and emission levels and to improve public health. In the U.S., walking trips account for 10.5% of all trips undertaken. To increase this rate, additional research on what makes people feel more comfortable while walking is needed. Research on pedestrian quality of service (QOS) has sought to quantify the performance of the pedestrian facilities from a pedestrian’s perspective. However, the impact of pedestrian safety countermeasures on pedestrian QOS for roadway crossings is largely unknown.  The objective of this study is to discern pedestrian QOS based on physiological measurements of pedestrians performing normal walking activities in different traffic contexts. The naturalistic walking study described in this paper recruited 15 pedestrians and asked each to wear an instrumented wristband and GPS recorder on all walking trips for one week. Surprisingly, the findings from the study showed no correlation between participants’ stress levels and individual crossing locations. Instead, stress was associated with roadway conditions. Higher levels of stress were generally associated with walking in proximity to collector and arterial streets and in areas with industrial and mixed (e.g., offices, retail, residential) land uses. Stress levels were tempered in lower-density residential land uses, as well as in forest, park, and university campus environments. The outcomes from this study can inform how planners design urban environments that reduce pedestrian stress levels to promote walkability.