Event-Based Modeling of Driver Yielding Behavior to Pedestrians at Two-Lane Roundabout Approaches.

Unlike other types of controlled intersections, drivers do not always comply with the "yield to pedestrian" sign at the roundabouts. This paper aims to identify the contributing factors affecting the likelihood of driver yielding to pedestrians at two-lane roundabouts. It further models the likelihood of driver yielding based on these factors using logistic regression. The models have been applied to 1150 controlled pedestrian crossings at entry and exit legs of two-lane approaches of six roundabouts across the country. The logistic regression models developed support prior research that the likelihood of driver yielding at the entry leg of roundabouts is higher than at the exit. Drivers tend to yield to pedestrians carrying a white cane more often than to sighted pedestrians. Drivers traveling in the far lane, relative to pedestrian location, have a lower probability of yielding to a pedestrian. As the speed increases the probability of driver yielding decreases. At the exit leg of the roundabout, drivers turning right from the adjacent lane have a lower propensity of yielding than drivers coming from other directions. The findings of this paper further suggest that although there has been much debate on pedestrian right-of-way laws and distinction between pedestrian waiting positions (in the street versus at the curb), this factor does not have a significant impact on driver yielding rate. The logistic regression models also quantify the effect of each of these factors on propensity of driver yielding. The models include variables which are specific to each study location and explain the impact size of each study location on probability of yielding. The models generated in this research will be useful to transportation professionals and researchers interested in understanding the factors that impact driver yielding at modern roundabouts. The results of the research can be used to isolate factors that may increase yielding (such as lower roundabout approach speeds), and can feasibly be incorporated into microsimulation algorithms to model driver yielding at roundabouts.

Simulator Study of Driver Responses to Pedestrian Treatments at Multilane Roundabouts.

Previous studies have shown that roundabouts - especially multilane roundabouts - pose accessibility challenges to pedestrians with vision impairments, in part due to a lack of yielding by drivers, especially on multilane roundabout exit legs. In this study, three different treatments are assessed in terms of their propensity for increasing driver yielding rate using a driving simulator. These are stop bar and crosswalk relocation away from the beginning of exit leg, and two types of beacons, namely a Pedestrian Hybrid Beacon (PHB) and a Rectangular Rapid Flashing Beacon (RRFB). The study shows that installation of any kind of beacon (PHB or RRFB) with or without crosswalk relocation increases driver yielding rates significantly. Relocating the crosswalk does not provide a significant increase in driver yielding rate for the base case, but appeared to further enhance the effectiveness of the PHB and RRFB treatments. The results of using an eye tracker on drivers to track their gaze pattern while exiting the roundabout shows that having a beacon installed with crosswalk relocation increases drivers' attention both on the beacon and the pedestrian along the road. However, a portion of participants failed to see and react to the pedestrian treatments, causing concern about the visibility of these treatments at the roundabout exit leg.

Development and Implementation of a Conflict-based Assessment of Pedestrian Safety (CAPS) to Evaluate Accessibility of Complex Intersections.

This paper develops and implements the Conflict-based Assessment of Pedestrian Safety (CAPS) methodology for evaluating pedestrian accessibility at complex intersections. In past years, a significant research has been done on pedestrian access to modern roundabouts and other complex intersection forms, including a significant focus on the accessibility for pedestrians who are blind. A majority of these studies have relied on actual street crossings by study participants under supervision of trained Orientation and Mobility (O&M) Specialist. These crossing studies were used to evaluate risk from a measurement of intervention events, where the O&M specialist had to physically stop the participant from crossing. While providing arguably the most accurate data for the crossing risk at a particular intersection, actual street crossings can be dangerous to the study participants, and are further very time consuming and expensive to conduct. The CAPS method presented in this paper emphasizes the use of conflict-based safety factors to quantify risk. The CAPS method relates pedestrian crossing decisions to advanced measurements of vehicle dynamics to estimate lane-by-lane conflicts. CAPS identifies the grade of conflict based on a score generated on a five-criterion rating scale. Each of these criteria or factors has different severity levels, and when combined, provides an overall risk rating of the crossing decision. The CAPS framework was applied to a study of blind pedestrian crossings at a multi-lane roundabout. The resulting risk scores were calibrated from actual O&M interventions observed during the study to give confidence in the CAPS performance. The calibrated CAPS framework correctly matched all (high risk) O&M intervention events, and further identifies other (lower risk) pedestrian-vehicle conflicts. The resulting method has the potential to allow for a faster and most importantly safer evaluation of complex intersections for pedestrian access. Since all factors are measured prior to the pedestrian stepping into the roadway, this approach is compatible with crossing indicator studies, where the participants merely indicate when they would cross, rather than actually stepping into the roadway. The CAPS framework therefore allows for a more objective and consistent safety assessment of pedestrian crossings in a research context without having pedestrians physically step into the roadway.

Event-Based Modeling of Driver Yielding Behavior at Unsignalized Crosswalks.

This research explores factors associated with driver yielding behavior at unsignalized pedestrian crossings and develops predictive models for yielding using logistic regression. It considers the effect of variables describing driver attributes, pedestrian characteristics and concurrent conditions at the crosswalk on the yield response. Special consideration is given to 'vehicle dynamics constraints' that form a threshold for the potential to yield. Similarities are identified to driver reaction in response to the 'amber' indication at a signalized intersection. The logit models were developed from data collected at two unsignalized mid-block crosswalks in North Carolina. The data include 'before' and 'after' observations of two pedestrian safety treatments, an in-street pedestrian crossing sign and pedestrian-actuated in-roadway warning lights.The analysis suggests that drivers are more likely to yield to assertive pedestrians who walk briskly in their approach to the crosswalk. In turn, the yield probability is reduced with higher speeds, deceleration rates and if vehicles are traveling in platoons. The treatment effects proved to be significant and increased the propensity of drivers to yield, but their effectiveness may be dependent on whether the pedestrian activates the treatment.The results of this research provide new insights on the complex interaction of pedestrians and vehicles at unsignalized intersections and have implications for future work towards predictive models for driver yielding behavior. The developed logit models can provide the basis for representing driver yielding behavior in a microsimulation modeling environment.

Empirical Behavioral Models to Support Alternative Tools for the Analysis of Mixed-Priority Pedestrian-Vehicle Interaction in a Highway Capacity Context.

This paper presents behavioral-based models for describing pedestrian gap acceptance at unsignalized crosswalks in a mixed-priority environment, where some drivers yield and some pedestrians cross in gaps. Logistic regression models are developed to predict the probability of pedestrian crossings as a function of vehicle dynamics, pedestrian assertiveness, and other factors. In combination with prior work on probabilistic yielding models, the results can be incorporated in a simulation environment, where they can more fully describe the interaction of these two modes. The approach is intended to supplement HCM analytical procedure for locations where significant interaction occurs between drivers and pedestrians, including modern roundabouts.

Working Concept of Accessibility: Performance Measures for Usability of Crosswalks by Pedestrians with Vision Impairments.

This research presents an analysis framework and associated performance measures for quantifying the accessibility of pedestrian crossings at modern roundabouts for pedestrians who are blind. The measures, developed under two ongoing national research projects, NCHRP Project 3-78A and a bioengineering research grant from the National Institutes of Health-National Eye Institute, attempt to isolate the components of the crossing task for a blind pedestrian into computable and replicable quantities that allow the comparison of accessibility across individuals or sites. The framework differentiates between crossing opportunities in the form of yields and crossable gaps and the utilization of these opportunities by the pedestrian. It further accounts for the amount of delay and risk involved in the crossing. The analysis framework and measures are demonstrated for two single-lane roundabouts in North Carolina evaluated under the aforementioned research projects. The application shows that the accessibility of a pedestrian crossing to a blind pedestrian is characterized by a combination of different measures and further depends on crossing geometry, traffic volume, driver behavior, and the travel skills and risk-taking behavior of the individual. With successful demonstration at roundabout crosswalks, the analysis framework is hypothesized to have broader application to unsignalized pedestrian crossings, including midblock locations.

Vehicle-specific emissions modeling based upon on-road measurements.

Vehicle-specific microscale fuel use and emissions rate models are developed based upon real-world hot-stabilized tailpipe measurements made using a portable emissions measurement system. Consecutive averaging periods of one to three multiples of the response time are used to compare two semiempirical physically based modeling schemes. One scheme is based on internally observable variables (IOVs), such as engine speed and manifold absolute pressure, while the other is based on externally observable variables (EOVs), such as speed, acceleration, and road grade. For NO, HC, and CO emission rates, the average R(2) ranged from 0.41 to 0.66 for the former and from 0.17 to 0.30 for the latter. The EOV models have R(2) for CO(2) of 0.43 to 0.79 versus 0.99 for the IOV models. The models are sensitive to episodic events in driving cycles such as high acceleration. Intervehicle and fleet average modeling approaches are compared; the former account for microscale variations that might be useful for some types of assessments. EOV-based models have practical value for traffic management or simulation applications since IOVs usually are not available or not used for emission estimation.

Mixed-Priority Pedestrian Delay Models at Single-Lane Roundabouts.

This paper presents an approach for developing mixed-priority pedestrian delay models at single-lane roundabouts using behavioral crossing data. Mixed-priority refers to crosswalk operations where drivers sometimes yield to create crossing opportunities, but where pedestrians sometimes have to rely on their judgment of gaps in traffic to cross the street. The models use probabilistic behavioral parameters measured in controlled pedestrian crossings by blind pedestrians as part of NCHRP project 3-78a. While blind pedestrians clearly represent a special population of pedestrians, the developed delay model is structured to be applicable to any pedestrian population. Delay is predicted as a function of the probability of encountering a crossing opportunity in the form of a yield or crossable gap, and the probability of utilizing that opportunity, which are combined to produce an overall probability of crossing. The paper presents the theoretical approach to estimating the probability parameters and uses a multi-linear log-transformed regression approach to predict the average pedestrian delay. The final delay model explains 64% of the variability in the observed data and therefore represents a reasonable model for predicting pedestrian delay at single-lane roundabouts. The paper concludes with a discussion of how agencies can estimate the underlying probability parameters for existing or proposed roundabouts using empirical and theoretical approaches, and how pedestrian crossing treatments can be used in the context of the model to reduce average pedestrian delay. The research is important in light of the ongoing debate of the accessibility of modern roundabouts to pedestrians who are blind.

Regional on-road vehicle running emissions modeling and evaluation for conventional and alternative vehicle technologies.

This study presents a methodology for estimating high-resolution, regional on-road vehicle emissions and the associated reductions in air pollutant emissions from vehicles that utilize alternative fuels or propulsion technologies. The fuels considered are gasoline, diesel, ethanol, biodiesel, compressed natural gas, hydrogen, and electricity. The technologies considered are internal combustion or compression engines, hybrids, fuel cell, and electric. Road link-based emission models are developed using modal fuel use and emission rates applied to facility- and speed-specific driving cycles. For an urban case study, passenger cars were found to be the largest sources of HC, CO, and CO(2) emissions, whereas trucks contributed the largest share of NO(x) emissions. When alternative fuel and propulsion technologies were introduced in the fleet at a modest market penetration level of 27%, their emission reductions were found to be 3-14%. Emissions for all pollutants generally decreased with an increase in the market share of alternative vehicle technologies. Turnover of the light duty fleet to newer Tier 2 vehicles reduced emissions of HC, CO, and NO(x) substantially. However, modest improvements in fuel economy may be offset by VMT growth and reductions in overall average speed.

Comparison of flexible fuel vehicle and life-cycle fuel consumption and emissions of selected pollutants and greenhouse gases for ethanol 85 versus gasoline.

The objective of this research is to evaluate differences in fuel consumption and tailpipe emissions of flexible fuel vehicles (FFVs) operated on ethanol 85 (E85) versus gasoline. Theoretical ratios of fuel consumption and carbon dioxide (CO2) emissions for both fuels are estimated based on the same amount of energy released. Second-by-second fuel consumption and emissions from one FFV Ford Focus fueled with E85 and gasoline were measured under real-world traffic conditions in Lisbon, Portugal, using a portable emissions measurement system (PEMS). Cycle average dynamometer fuel consumption and emission test results for FFVs are available from the U.S. Department of Energy, and emissions certification test results for ethanol-fueled vehicles are available from the U.S. Environmental Protection Agency. On the basis of the PEMS data, vehicle-specific power (VSP)-based modal average fuel and emission rates for both fuels are estimated. For E85 versus gasoline, empirical ratios of fuel consumption and CO2 emissions agree within a margin of error to the theoretical expectations. Carbon monoxide (CO) emissions were found to be typically lower. From the PEMS data, nitric oxide (NO) emissions associated with some higher VSP modes are higher for E85. From the dynamometer and certification data, average hydrocarbon (HC) and nitrogen oxides (NOx) emission differences vary depending on the vehicle. The differences of average E85 versus gasoline emission rates for all vehicle models are -22% for CO, 12% for HC, and -8% for NOx emissions, which imply that replacing gasoline with E85 reduces CO emissions, may moderately decrease NOx tailpipe emissions, and may increase HC tailpipe emissions. On a fuel life cycle basis for corn-based ethanol versus gasoline, CO emissions are estimated to decrease by 18%. Life-cycle total and fossil CO2 emissions are estimated to decrease by 25 and 50%, respectively; however, life-cycle HC and NOx emissions are estimated to increase by 18 and 82%, respectively.

A vehicle-specific power approach to speed- and facility-specific emissions estimates for diesel transit buses.

Emissions during a trip often depend on transient vehicle dynamics that influence the instantaneous engine load. Vehicle specific power (VSP) is a proxy variable for engine load that has been shown to be highly correlated with emissions. This study estimates roadway link average emission rates for diesel-fueled transit buses based on link mean speeds, using newly defined VSP modes from data gathered by a portable emissions monitoring system. Speed profiles were categorized by facility type and mean travel speed, and stratified into discrete VSP modes. VSP modal average emission rates and the time spent in the corresponding VSP modes were then used to make aggregate estimates of total and average emission rates for a road link. The average emission rates were sensitive to link mean speed, but not to facility type. A recommendation is made regarding the implementation of link average emission rates in conjunction with transportation models for the purpose of estimating regional emissions for diesel transit buses.

Fuel use and emissions comparisons for alternative routes, time of day, road grade, and vehicles based on in-use measurements.

The objective here is to quantify the variability in emissions of selected light duty gasoline vehicles by routes, time of day, road grade, and vehicle with a focus on the impact of routes and road grade. Field experiments using a portable emission measurement system were conducted under real-world driving cycles. The study area included two origin/destination pairs, each with three alternative routes. Total emissions varied from trip to trip and from route to route due to variations in average speed and travel time. On an average trip basis, the total NO emissions differed by 24% when comparing alternative routes and by 19% when comparing congested travel time with less congested traffic time. Positive road grades were associated with an approximately 20% increase in localized emissions rates, while negative road grades were associated with a similar relative decrease. The average vehicle-specific power based NO modal emission rates differed by more than 2 orders of magnitude when comparing different vehicles. The results demonstrate that alternative routing can significantly impact trip emissions. Furthermore, road grade should be taken into account for localized emissions estimation. Vehicle-specific models are needed to capture episodic effects of emissions for near-road short-term human exposure assessment.

Quantification of highway vehicle emissions hot spots based upon on-board measurements.

The purpose of this study is to demonstrate a methodology for quantification of high emissions hot spots along roadways based upon real-world, on-road vehicle emissions measurements. An emissions hot spot is defined as a fixed location along a corridor in which the peak emissions are statistically significantly greater by more than a factor of 2 than the average emissions for free-flow or near free-flow conditions on the corridor. A portable instrument was used to measure on-road tailpipe emissions of carbon monoxide, nitric oxide, hydrocarbons, and carbon dioxide on a second-by-second basis during actual driving. Measurements were made for seven vehicles deployed on two primary arterial corridors. The ratio of average emissions at hot spots to the average emissions observed during a trip was as high as 25 for carbon monoxide, 5 for nitric oxide, and 3 for hydrocarbons. The relationships between hot spots and explanatory variables were investigated using graphical and statistical methods. Average speed, average acceleration, standard deviation of speed, percent of time spent in cruise mode, minimum speed, maximum acceleration, and maximum power have statistically significant associations with vehicle emissions and influence emissions hot spots. For example, stop-and-go traffic conditions that result in sudden changes in speed, and traffic patterns with high accelerations, are shown to generate hot spots. The implications of this work for future model development and applications to environmental management are discussed.

On-road measurement of vehicle tailpipe emissions using a portable instrument.

A study design procedure was developed and demonstrated for the deployment of portable onboard tailpipe emissions measurement systems for selected highway vehicles fueled by gasoline and E85 (a blend of 85% ethanol and 15% gasoline). Data collection, screening, processing, and analysis protocols were developed to assure data quality and to provide insights regarding quantification of real-world intravehicle variability in hot-stabilized emissions. Onboard systems provide representative real-world emissions measurements; however, onboard field studies are challenged by the observable but uncontrollable nature of traffic flow and ambient conditions. By characterizing intravehicle variability based on repeated data collection runs with the same driver/vehicle/route combinations, this study establishes the ability to develop stable modal emissions rates for idle, acceleration, cruise, and deceleration even in the face of uncontrollable external factors. For example, a consistent finding is that average emissions during acceleration are typically 5 times greater than during idle for hydrocarbons and carbon dioxide and 10 times greater for nitric oxide and carbon monoxide. A statistical method for comparing on-road emissions of different drivers is presented. Onboard data demonstrate the importance of accounting for the episodic nature of real-world emissions to help develop appropriate traffic and air quality management strategies.