National Transportation Center

Major transit infrastructure disruptions have become more frequent because of increasing maintenance needs for aging infrastructure, system failures, and disasters. Understanding travel behavior reactions to service disruptions on the basis of empirical observations is a fundamental step toward planning and operating an efficient and reliable transportation system. Few studies in the literature have investigated the behavioral and system impact of transit service disruptions. To bridge this gap in the literature, this research investigated travel behavioral reactions to transit service disruptions during the Metro SafeTrack projects in Washington, D.C., with the use of a unique panel survey. This study offers new insights on multimodal, multidimensional travel behavioral responses to major transit network disruptions, a critically theoretical prerequisite for developing and implementing effective strategies (e.g., how to deploy the reserve bus fleet optimally) that minimize system impact and improve transit system reliability and resilience.

Over the past several years, there has been considerable debate surrounding ethical decision making in situations resulting in inevitable casualties. Given enough time and all other things being equal, studies show that drivers will typically decide to strike the fewest number of pedestrians in scenarios where there is a choice between striking several versus one or no pedestrians. However, it is unclear whether drivers behave similarly under situations of time pressure. In our experiment in a driving simulator, 32 drivers were given up to 2 s to decide which group of pedestrians to avoid among groups of larger (5) or smaller (≤1) number of pedestrians. Our findings suggest that while people frequently choose utilitarian decisions in the typical, abstract manifestations of the Trolley Problems, drivers can fail to make utilitarian decisions in simulated driving environments under a restricted period of time representative of the time they would have to make the same decision in the real world (2 s). Analysis of eye movement data shows that drivers are less likely to glance at left and right sides of crosswalks under situations of time duress. Our results raise critical engineering and ethical questions. From a cognitive engineering standpoint, we need to know how long at minimum a driver needs to make simple, moral decisions in different scenarios. From an ethical standpoint, we may need to evaluate whether automated vehicle algorithms can aid decision making on our behalf when there is not enough time for a driver to make a moral decision.

Older drivers are at increased risk of intersection crashes. Previous work found that older drivers execute less frequent glances for detecting potential threats at intersections than middle-aged drivers. Yet, earlier work has also shown that an active training program doubled the frequency of these glances among older drivers, suggesting that these effects are not necessarily due to age-related functional declines. In light of findings, the current study sought to explore the ability of older drivers to coordinate their head and eye movements while simultaneously steering the vehicle as well as their glance behavior at intersections. In a driving simulator, older (M = 76 yrs) and middle-aged (M = 58 yrs) drivers completed different driving tasks: (1) travelling straight on a highway while scanning for peripheral information (a visual search task) and (2) navigating intersections with areas potential hazard. The results replicate that the older drivers did not execute glances for potential threats to the sides when turning at intersections as frequently as the middle-aged drivers. Furthermore, the results demonstrate costs of performing two concurrent tasks, highway driving and visual search task on the side displays: the older drivers performed more poorly on the visual search task and needed to correct their steering positions more compared to the middle-aged counterparts. The findings are consistent with the predictions and discussed in terms of a decoupling hypothesis, providing an account for the effects of the active training program.

The crash rate for young novice drivers is at least eight times higher than that of their experienced counterparts. Literature shows that the young novice drivers are not careless drivers but they are clueless drivers’ - clueless because of their inability to predict the risk ahead of time that might materialize on the forward roadway. Other error-feedback training programs exist that emphasize the teaching of risk awareness and perception skills to young drivers. In the current study, a Virtual reality based risk awareness and perception training program (V-RAPT) was developed on the Oculus Rift and evaluated on a driving simulator. The training program provides 360 degrees’ views of 6 high risk driving scenarios towards training the young driver to anticipate and mitigate latent hazards. Twenty-four participants in three experiment groups were trained on one of 3 training programs- VRAPT, RAPT and Control, and were evaluated on a driving simulator. Eye movements were collected throughout the experiment. The simulator evaluation drives included six near-transfer scenarios used in the training and four far-transfer scenarios not used in the training but validated previously in other similar studies. The young drivers trained on the V-RAPT were found to anticipate a significantly greater proportion (86.25%) of the potential latent hazards as compared to the RAPT trained young drivers (62.36%) and control trained drivers (30.97%). The VR-based training program is shown to be effective in improving young drivers’ ability to anticipate latent threats.

This study employs structural equation modeling techniques to investigate the relationship between demand for taxi, Uber, and Lyft services and socioeconomic, built environment characteristics, and access to transit and bike-sharing modes, for taxi zones within New York City. The results show that income and car ownership levels influence demand for these for-hire modes. Additionally, higher activity density and higher extent of mixed land-use are associated with increased demand for for-hire modes, while pedestrian-friendly street networks are associated with lower demand levels. Temporal destination accessibility also has an impact on demand for taxi, Uber, and Lyft. Further, accessibility to transit and bike-sharing significantly influences demand for for-hire modes. These findings provide a better understanding of the link between for-hire modes and built environment as well as accessibility to other modes, which can be used to improve demand forecasting of taxi, Uber, and Lyft services in large cities.

A Ground Based Detect and Avoid (GBDAA) system, derived from the Terminal Automation Modernization Replacement (TAMR) system, the terminal air traffic control (ATC) automation system currently in use in US National Airspace (NAS) continues to evolve to provide enhanced detect and avoid decision making for unmanned aircraft (UA) operating in the NAS. GBDAA capabilities have been developed as a partnership between the USAF, State of Ohio, DOT Volpe Center, MITRE and Raytheon. The system displays airborne tracks provided by a collection of sensors for situational awareness and proximity alerts to crew and mission planners. Large and small unmanned aircraft systems (sUAS) operations are supported using both fixed and mobile facilities. The GBDAA system allows the pilot to make decisions about flight maneuvers without using ground observers or chase planes during day and night operations. Recent enhancements include the use of a separate GBDAA Operator (GO) display, synchronized with the UA's Pilot in Command's (PIC) display to allow for the safe passage of the UA throughout its operational airspace. The GO supports the PIC with alert prioritization and maneuver recommendations, allowing the PIC to focus on piloting tasks and reduce the time needed for avoidance maneuvers. Several types of UAS operations are described that include: · Table top sUAS operations at a small airfield · Medium sized unmanned aircraft (UA) Ground Control Systems (GCS) such as those used for Predators/Reapers and mobile cart technology. · Larger UAS operations such as those used for the Global Hawk with a separate mission operations center · Airspace utilization, such as transit corridor operations, transition from terminal operations to Class A airspace, and free flight considerations The availability of surveillance coverage and the size of the UAS operational airspace that can be supported can be determined based on available surveillance assets and airspace environment at each location. In many cases, existing surveillance assets are used to provide the information needed on the airspace being monitored. New radar equipment and modifications to existing surveillance assets can be used to augment existing ATC sensors and close surveillance coverage gaps. GBDAA decision making is provided by new software and the use of multisensor fusion tracking software that supports the display of track information for both cooperative and noncooperative targets. These are displayed to both the GO and PIC using a common situation display showing the operational airspace. Several ever increasingly urgent levels of proximity alerts are displayed to both flight crew members as they perform their respective tasks. This paper describes important aspects of the development of a mobile facility as the team integrated access to FAA surveillance radar assets, radio and telecommunications gear to create a complete set of GBDAA services that can be installed at different sites. The necessary airport field work, integration of surveillance, radio and phone connections and the installation of a fully functional GBDAA system into a 36-foot modified recreational vehicle is described. This “GBDAA Bus” provides a working environment to support operations for sUAS stakeholders carrying out different missions for both AFRL and the State of Ohio's UAS Center. Where no access is available to certified FAA/DoD airport surveillance and long-range radar sensors, several different types of towable sensors can be used, some even offering elevation approximation for non-cooperative targets.

To ensure safe unmanned aircraft (UA) operations in the US National Airspace System, Ground Based Sense and Avoid (GBSAA) capabilities have been added to the Standard Terminal Automation Replacement System (STARS) for the United States Air Force (USAF) at Cannon Air Force Base (AFB) to support Predator and Reaper UA operations.

To provide quantitative support for the Simultaneous Offset Instrument Approach (SOIA) procedure, an extensive data collection effort was undertaken at San Francisco International Airport by the Federal Aviation Administration (FAA, U.S. Dept. of Transportation). During the time period from March 2000 to October 2002, wake vortex data was measured for over 260,000 landing aircraft. The data set includes wake vortex measurements from Small, Large, and Heavy category aircraft. The measurements consist of cross-runway wind speed recorded every two seconds from three windlines, comprised of a series of propeller anemometers on three-foot poles near the threshold of runways 28L and 28R. The resulting data set is being used to demonstrate the feasibility of SOIA and to guide the improvement of the wake vortex model in the FAA airspace simulation tool ASAT (Airspace Simulation and Analysis for TERPS). We show that a slightly modified version of the AVOSS (Aircraft VOrtex Spacing System) Prediction Algorithm (APA) produces lateral position predictions that agree well with the windline\ndata. We also show that the data and the APA results agree with results produced by TASS (Terminal Area Simulation System), a numerical code developed by NASA. These comparisons between code and data provide an independent validation of the lateral transport estimates using windline sensors and give us increased confidence in both the data obtained from the windline sensors and our ability to predict vortex evolution using numerical simulations.\n

Newly licensed teenage drivers experience a higher risk of crashing compared to other age cohorts. Literature reveals that novice drivers exhibit poor hazard mitigation skills. The current study assesses the effectiveness of a training program at improving novice divers’ hazard mitigation and speed selection behaviors on curves. In this study, drivers are randomly assigned to two training cohorts (ACT and placebo), and were exposed to 2 different scenarios of interest, one scenario contained a moderate curve left and the other included a tightening curve right. ACT trained drivers made more glances to the far extent of the curve, than the placebo-trained drivers. ACT (Anticipate, Control, and Terminate) trained drivers were also significantly more likely to slow to the target speed before the curve, when compared to the placebo trained drivers. The results indicate the effectiveness of ACT as a countermeasure, at training novice drivers to select better glancing and speed management strategies.

This paper investigates the influence of spatial distribution of employment centers on travel behavior using data from two large metropolitan areas: Atlanta, GA, and Phoenix, AZ. Up-to-date fine-grained land-use data and a consistent GIS-based framework were employed to first identify regional employment subcenters and statistical models were developed to analyze the travel outcomes of employment subcenters. Results suggest that a polycentric urban form encourages transit mode choice more than monocentric pattern especially when transit accessibility is improved in polycentric pattern. We found evidence that trips ending in subcenters are shorter and have a higher probability of transit mode choice compared to trips originating from or destined to the city center. These findings suggest that any form of employment concentration would be more efficient in multiple smaller clusters distributed throughout the entire region rather than in a single large cluster.

A recent Volpe Center report [1] for the Federal Railroad Administration’s (FRA) Rail Energy, Environment, and Engine (E3) Technology research and development program reviewed rail industry best practices (BPs) and strategies for improving energy efficiency (E2) and environmental sustainability. The review included examples of and opportunities for adoption of international transferrable BPs, and US technologies for equipment, operations and logistics software tools that have measurably improved E2 performance for passenger and freight railroads. Drivers providing renewed impetus for rail industry E2 advances include environmental compliance requirements with US Environmental Protection Agency (EPA) locomotive emission standards, US Department of Transportation Congestion Mitigation and Air Quality improvement program grants, state, regional and urban clean diesel campaigns, as well as the FRA National Rail Plan, and High-Speed Intercity Passenger Rail (HSIPR) initiatives. The report presented comparative rail system energy efficiency data and trends relative to competing modes, illustrated the benefits of energy-efficient technologies, and of alternative fuels use. Based on a comprehensive literature review and on experts’ inputs, the report highlighted models of corporate rail sustainability plans and system-wide BPs and success stories. Available rail equipment and operational practices proven to improve E2 with environmental and economic benefits for all rail industry segments were illustrated. Findings and recommendations for further improving rail E2 and sustainability were tailored to the specific needs and goals of intercity and commuter passenger rail, and freight railroads (Class I-III). Key opportunities highlighted included: public-private partnerships (P3) with Federal agencies (FRA, EPA/SmartWay) for joint research, development test and evaluation (RDT&E)on advanced equipment (electric and hybrid, or dual fuel locomotives), or alternative fuels (biodiesel, CNG/LNG, Fuel cells/Hydrogen); participation in international rail organizations (UIC) and trade associations (AAR, AREMA, APTA, AASHTO), and partnering with regional and State environmental protection agencies for cross-enterprise E2 and sustainability improvements.

Automated driving systems can support driver performance at varying levels (L2 to L5) but potentially impair attentional performance of drivers. The experiment examined drivers' latent hazard anticipation abilities across three levels of automation, L0, L2 and L3. The study was conducted on a driving simulator and data were collected from 36 young licensed drivers. Data indicated that drivers in the L3 condition were less likely to anticipate latent hazards than those in the L0 condition. However, data showed no reliable differences in the latent hazard anticipation for drivers between the L2 and L0 conditions and between L2 and L3 conditions. The results imply that drivers of high-level vehicle automation may display diminished ability to anticipate imminent hazard on the forward roadway. Additional studies need to be conducted to further understand changes in the mechanisms of drivers' latent hazard anticipation ability across different levels of vehicle automation. Better understanding of changes in drivers' latent hazard anticipation can lead to the design of effective countermeasures to support drivers' latent hazard anticipation in automated driving systems.

The paper addresses the need to examine the trade-offs between passenger safety and independence in travel by people who use wheeled mobility devices on passengers trains. It has been the practice in Asia, North America and Europe to not require passengers in wheeled mobility devices (WhMDs) such as wheelchairs to secure their wheeled devices when traveling by rail. There are several motivations for examining the need for containment of WhMDs on passenger trains. In general the population is aging and getting larger, and this is reflected in the types of WhMDs that passengers are trying to bring on board trains. The US Federal Railroad Administration (FRA) and members of the Rail Vehicle Access Advisory Committee (RVAAC) requested a feasibility study on the economic impacts of accommodating two or more wheeled mobility devices in the accessible seating area [1]. The feasibility study indicated that there is space to accommodate two WhMDs without significant impact on revenue seat loss, however safety issues have emerged, and are the basis of this paper. The three research questions that are addressed include: I. What is the appropriate interior space that accounts for WhMD maneuvering? II. What are the appropriate levels of deceleration and jerk to be considered in the vehicle interior for passenger rail vehicles under severe braking? III. What is the appropriate level of containment for occupied wheeled mobility devices on passenger rail vehicles? The paper examines research literature and other findings from both North America and Europe that address in part the research questions.

Forward roadway collision warning systems can reduce rear-end collisions, among other unsafe behaviors. Previous studies have shown that young drivers fail to scan adequately for latent hazards. The current driving simulator study investigates the effect of visual collision warning messages on drivers’ hazard anticipation ability, when presented either 2s, 3s or 4s in advance of a potential threat. This experiment examined the latent hazard anticipation behavior of forty-eight young drivers aged 18-25 across eight unique scenarios both, in the presence, and absence of visual collision warning alerts. The analysis of glance data captured using an eye tracker, show that visual warning messages significantly increased the proportion of latent hazards anticipated regardless of hazard type (pedestrian or vehicle). The 2s warning duration was found to statistically have the same effect on hazard anticipation compared to the 3s and 4s warning thresholds. The study has potential implications for the effective design of forward collision warning systems.