State departments of transportation (DOTs) are investing significantly more resources to enhance pedestrian safety. However, there is still a need to effectively and systematically address the experiences of pedestrians in low-income areas. A Governing analysis of pedestrian crashes occurring in 2008—2012 found that pedestrians were killed at disproportionately higher rates in the nation’s poorer neighbourhoods and that within metro areas, low-income tracts recorded pedestrian fatality rates approximately twice those of more affluent neighbourhoods. Examining Census tract poverty rates yielded a similar pattern–the country’s poorest neighbourhoods have the highest per-capita pedestrian fatalities. Low-income areas generally contain a problematic mix of politically-underrepresented populations and pedestrians with little driving experience to inform their decision-making in traffic environments. These low-income areas are often sectioned by high volume/high speed arterials, which compounds the problem. To develop proactive and effective countermeasures for pedestrian safety in low-income areas, it is important to investigate and understand major contributing causes and a combination of the “pre-conditions” for pedestrian crash frequency and its resulting injury severities. Section 1 of this final report details the project background and objectives. Realizing the challenges of pedestrian safety in low-income areas, the objectives of this research project are to (1) develop a demographics-based methodology that identifies low-income areas that possess a combination of “pre-conditions” for greater pedestrian hazard, (2) identify major factors associated with pedestrian crash frequency and injury severity and quantify their relationships, and (3) produce recommendations for both engineering countermeasures and pedestrian safety education or outreach plans that will resonate with a given area’s demographics. Section 2 addresses the first task of this project and includes a comprehensive literature review that identifies variables associated with pedestrian crash frequency and injury severity, methodologies for pedestrian crashes analysis, and existing GIS databases and tools. This section also includes input and feedback regarding the variables, outputs, and outcomes for pedestrian safety analysis from key Florida Department of Transportation (FDOT) representatives to support the objectives of this project. Section 3 addresses a methodological flowchart and a brief description regarding the major steps for the demographic-based approaches to dealing with pedestrian safety in low-income areas. The technical approach flowchart consists of three key components: inputs, outputs, and outcomes. The variables (inputs) identified in Section 2 should be used in geographic analysis and statistical modelling (both crash frequency modelling and injury severity modelling) to generate outputs, and those outputs can be used to produce recommendations (outcomes) for engineering and education countermeasures. Section 4 provides details of the methodology test, including testing the methodological flowchart developed in Section 3 for demographic analysis, providing the analysis results and findings for FDOT District 4 to demonstrate the kind of outputs and outcomes to be generated by following the developed methodology, and verifying that the methodology is implementable by using available datasets such as FDOT GIS databases, Census data, and other easily-available data sources. The following summarizes the major findings in methodology test: • Pedestrian crashes are more frequent in low-income block groups (BGs) that have more population, a smaller proportion of older adults, are minority-dominated, have zero-car ownership neighbourhoods, and are among populations with a low education level. - Average marginal effects indicate that the top four influential demographic variables are the proportion of older adults (negative effect), proportion of commuters using public transit or biking, proportion of people with a low education level (less than high school), and proportion of zero-car ownership. • Pedestrian crashes are more frequent in low-income BGs with more intersections, traffic signals, and bus stops and a larger proportion of roads with higher speed limits. - Average marginal effects indicate that the most influential roadway factor is the number of traffic signals per BG, followed by the number of bus stop per mile. The third most influential variable is the proportion of lower-speed roads (negative effect); an increase in the proportion of lower-speed roads in a low-income BG can help decrease pedestrian crashes. • Pedestrian crashes occur more frequently in low-income BGs with the presence of a Walmart store and with greater densities of discount department stores, fast-food restaurants, convenience stores, grocery stores, and barber shops. - Average marginal effects indicate that the most influential variable related to land use types is density of discount stores, followed by density of convenience stores and density of fast-food restaurants. • Individual characteristics, including the involvement of older pedestrians, non-crosswalk locations of pedestrians, improper pedestrian action (dart/dash), impaired pedestrians, and aggressive drivers, have positive effects on severe injuries in pedestrian crashes. - Average marginal effects indicate that alcohol or drug involvement of a pedestrian is the most influential variable for severe injury in a pedestrian crash, followed by the involvement of aggressive drivers and older pedestrians. • Environmental factors including lighting conditions, roadway speed limits, and the presence of traffic control devices have significant effects on the injury severity of a pedestrian crash. - Average marginal effects indicate that a dark—not lighted condition is the most influential variable for severe injury in pedestrian crashes, followed closely by dark— lighted condition. The third most influential variable is higher speed limit. A dark— lighted condition seems to indicate that various lighting levels could have different impacts on injury severity in a pedestrian crash. • Pedestrian crashes are more frequent in segments in which the average number of fast-food restaurants, department stores, and banks is higher than average for the corridor. • Pedestrian crashes are more frequent in segments in which the average number of bus stops and intersections is higher than average for the corridor. • A proximity analysis illustrated that impaired pedestrian crashes tend to be more frequent in alcohol availability buffers (near the location of bars and alcohol retail) in low-income areas. Section 5 illustrates how the outcomes provided in the methodology developed in Section 3 and tested in Section 4 connect with the target area demographics. Engineering countermeasures are recommended based on crash analysis and types of existing facilities, and education/outreach countermeasures are recommended based on demographics, land use, and other data. This section also provides strategies for implementation of the countermeasures in a systematic approach. Section 6 summarizes the conclusions and recommendations of this research project and includes a summary of identified related databases, the proposed and tested methodological flowchart, and the major findings to recommend implementation strategies for pedestrian safety. The recommended engineering countermeasures include the following: • Roadway lighting and lighting levels — presence of lighting, adequate lighting level and uniformity, proper pedestrian lighting placement • Treatments at non-intersection locations — midblock pedestrian crossing signals (HighIntensity Activated Crosswalk [HAWK] and Rectangular Rapid Flashing Beacon [RRFB]), high-visibility crosswalks, medians and crossing islands, appropriate landscaping • Bus stop improvements — bus stop reallocation, transit stop request lights • Speed reduction treatments — slow speed zones, road diets, roundabouts, traffic calming on residential streets • Road Safety Audits (RSA) The recommended education and outreach plan includes the following: • WalkWise safety education; • Distribution of education tip cards; • Social media outreach; • Community networking; • Business sweeps; • Law enforcement role call training; • Public-private partnerships. The implementation of an education and outreach plan along with targeted High-Visibility Enforcement (HVE) has great potential in reducing both crash and injury frequency and severity. The combined engineering, education, and enforcement approach could produce the most benefits in reducing pedestrian fatalities, injuries, and crashes with a given area’s demographics. (Author/publisher)
Samenvatting